UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pressure acid leaching of nickel laterite Chen, Guizhen 1998

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

Item Metadata

Download

Media
831-ubc_1998-0390.pdf [ 10.34MB ]
Metadata
JSON: 831-1.0078581.json
JSON-LD: 831-1.0078581-ld.json
RDF/XML (Pretty): 831-1.0078581-rdf.xml
RDF/JSON: 831-1.0078581-rdf.json
Turtle: 831-1.0078581-turtle.txt
N-Triples: 831-1.0078581-rdf-ntriples.txt
Original Record: 831-1.0078581-source.json
Full Text
831-1.0078581-fulltext.txt
Citation
831-1.0078581.ris

Full Text

P R E S S U R E A C I D L E A C H I N G O F N I C K E L L A T E R I T E by GUIZHEN CHEN B.A.Sc, Huazhong University of Science and Technology, 1983 M.A.Sc, Huazhong University of Science and Technology, 1986 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Metals and Materials Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1998 ©Guizhen Chen, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) 11 A B S T R A C T Pressure acid leaching is an acceptable process for the treatment of nickel laterites. This process offers many advantages such as low energy consumption, low capital cost and high nickel and cobalt recoveries. The dissolution process is coupled to selective thermal hydrolysis of iron and aluminum. The dissolution of laterite and precipitation of iron and aluminum are strongly dependent on the temperature and the acid-to-ore ratio, which is determined by ore compositions and mineralogy. Because of the complex nature of the laterites, a knowledge of the mineralogy of laterites, the nickel association with the various constitutional minerals and the reactivity of these mineral phases is fundamental to an understanding of the kinetics of nickel extraction and to the process optimization. The mineralogy of the laterite ore samples from two ore bodies in New Caledonia and a variety of locations in an Ivory Coast ore body was examined by X-ray diffraction, scanning electron microscope (SEM) and energy dispersive X-ray spectrometer (EDX). X-ray diffraction and SEM-EDX studies showed that goethite, maghemite, quartz and lithiophorite are common minerals in these laterite samples. Spinel and gibbsite are present as minor phases in the New Caledonia laterite samples. Silicates such as talc, kaolinite and clinochlore were found in the Ivory Coast laterites. EDX results indicated that goethite is the major host mineral for nickel. Nickel also showed association with maghemite, lithiophorite and silicates (to various extents in Ivory Coast laterites). I l l Interruptive autoclave leaching experiments were carried out on a blended sample from the Ivory Coast ore body. With a 30% pulp density slurry treated at 260°C, with acid/ore = 0.19, nickel and cobalt extractions reached 96% within 15 minutes as goethite and maghemite had completely dissolved. Hydronium alunite was detected by XRD in the leach residue after 30 minutes of leaching, indicating that the amount of alunite increased with leaching time. Interruptive pressure acid leaching experiments were also performed at 250°C on the samples from five different locations of the Ivory Coast ore body. Process conditions such as acid/ore ratio and pulp density were calculated from a spreadsheet provided by Falconbridge Limited. Nickel and cobalt extractions obtained from 30 minute leaching tests as well as the free acid concentrations in the filtrates differed significantly among the five samples. The goethite dissolution rate also varied among samples and showed a dependence on free acid concentration. Goethite fully dissolved from samples LD-97-13, LD-97-18 and LD-97-32 within 30 minutes at 250°C when the free acid concentration was >32 g/1 in the filtrate. However some of the goethite was detected in the leaching residue of LD-97-24 after 1 hour of leaching with a filtrate free acid concentration of 27.15 g/1. In the experiments, maghemite exhibited the same reactivity as goethite. Clinochlore appeared to be dissolved at free acid concentration of approximately 40 g/1 from sample LD-97-13. Talc, kaolinite, and quartz showed resistance to pressure acid leaching at prescribed leaching conditions. The results of the free acid measurement showed that acid consumption during leaching was dependent not only on the chemical composition, but also the mineralogy of the ore. iv T A B L E O F C O N T E N T S ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES viii ACKNOWLEDGEMENT x CHAPTER 1 INTRODUCTION 1 CHAPTER 2 LITERATURE REVIEW 4 2.1 Mineralogy of Nickel Laterite 4 2.1.1 The Ore Sequence in Laterite Deposits 4 2.1.2 Natural of Minerals in Nickel Laterite 10 2.2 Nickel Laterite Ore Processing 15 2.2.1 General 15 2.2.2 Ferronickel Smelting and Matte Smelting 17 2.2.3 Ammonia Leaching (Caron Process) 20 2.2.4 Pressure Acid Leaching (Moa Bay Process) 23 2.3 Chemistry of Pressure Acid Leaching 27 2.3.1 General 27 2.3.2 Dissolution of Goethite and Precipitation of Hematite 30 2.4 Leaching Kinetics and Nickel Extraction during Pressure Acid Leaching 33 2.5 Leach Residue: 37 2.6 The Scope of the Present Study 38 CHAPTER 3 EXPERIMENTAL PROCEDURES 39 3.1 Sample Descriptions and Head Assay Experimental Procedure 39 3.1.1 Sample Descriptions 39 3.2.2 Head Assay Experimental Procedure 40 3.2 Pressure Acid Leaching Experimental Program 40 3.2.1 Pressure Acid Leaching Apparatus and Procedures 40 3.2.2 Analytical Methods and Calculation of Metal Extractions 42 V 3.3 Determination of Free Acid Concentration 43 3.4 Methods Used for Mineralogical Investigation 47 3.4.1 Removal of Iron Oxide 47 3.4.2 Calculations of Metal Concentrations in DCB-soluble Fraction 49 3.4.3 Electron Microanalysis (SEM-EDX) 50 3.4.4 X - Ray Diffraction Analysis 50 CHAPTER 4 RESULTS AND DISCUSSION 52 4.1 Head Assays and Selective Dissolution Results of Ore Samples 52 4.1.1 Head Assays 53 4.1.2 Selective Dissolution Results for Ore Samples 54 4.2 Mineralogical Examination Results 56 4.2.1 Mineralogy of Ouaco LD-95-75 56 4.2.2 Mineralogy of Poum 8 61 4.2.3 Mineralogy of Ivory Coast Blend 64 4.2.3 Mineralogy of LD-97-13 68 4.2.4 Mineralogy of LD-97-18 73 4.2.5 Mineralogy of LD-97-24, LD-97-28 and LD-97-32 76 4.2.6 Summary of Mineralogical Examinations 84 4.3 Pressure Acid Leaching Experimental Results 86 4.3.1 Interruptive Leaching of Ivory Coast Blend 86 4.3.2 Leaching of LD-97-13 95 4.3.3 Leaching of LD-97-24 and LD-97-28 99 4.3.4 Leaching of LD-97-18 and LD-97-32 101 4.3.5 Selective Dissolution Results of Leach Residues 104 4.4 Morphology of the Hematite 109 4.5 Discussion of Results 112 4.4.1 Metal Extractions during Pressure Acid Leaching 112 4.4.2 Dissolution of Various Minerals during Pressure Acid Leaching 115 4.4.3 Terminal Acid Concentration after Leaching 117 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 119 5.1 Conclusions 119 5.2 Recommendations for Further Work 121 BIBLIOGRAPHY 124 v i A P P E N D I X I F O R M U L A S O F M I N E R A L S I N L A T E R I T E S 129 A P P E N D I X II C A L C U L A T I O N O F A C I D A D D I T I O N 130 A P P E N D I X III M A S S B A L A N C E S H E E T S 132 vii LIST O F T A B L E S Table 2-1. Nickel laterite ore processing 16 Table 2- 2. New laterite projects using pressure acid leaching in the South West Pacific region 17 Table 2- 3. Kinetic nickel extraction data of Murrin Murrin laterite samples 34 Table 4- 1. Head assays for laterite samples used in the present study 53 Table 4- 2. DCB-treatment results 55 Table 4- 3. The metal concentrations in DCB-dissolvable fraction in the laterites 55 Table 4- 4. Minor phase compositions in Ouaco Ore 61 Table 4- 5. Compositions of particles in +50 mesh fraction of Poum ore 64 Table 4- 6. Particle compositions in LD-97-13 73 Table 4- 7. Compositions of kaolinite in LD-97-18 76 Table 4- 8. Compositions of Mn-oxide in LD-97-18 76 Table 4- 9. Compositions of Al-Mn oxide and magnetic particles in LD-97-28 84 Table 4-10. Summary for mineralogical examination results 85 Table 4-11. Metal extractions for pressure acid leaching experiments of the Ivory Coast blend. 87 Table 4- 12. Metal extractions after pressure acid leaching of LD-97-13 95 Table 4-13. Leaching results for LD-97-24 and LD-97-28 99 Table 4- 14. Leaching conditions and results for LD-97-18 and LD-97-32 ...101 Table 4-15. The compositions of DCB-dissolvable fraction in the leach residues 109 Table 4- 16. Sulfuric acid leaching conditions generating residues 109 Table 4- 17. Comparison of iron extraction obtained from the leaching of five drill samples.... 112 Table 4- 18. Comparison of the goethite dissolution kinetics at a variety of leaching conditions 116 viii LIST OF FIGURES Figure 2- 1. Effect of climate on typical profiles of nickeliferous laterite orebodies 9 Figure 2- 2. The ideal model for goethite 10 Figure 2- 3. The ideal model for hematite 11 Figure 2- 4. Diagrammatic sketch of kaolinite structure 13 Figure 2- 5. Diagrammatic sketch of talc structure 14 Figure 2- 6. Schematic flow sheet of the Moa Bay pressure acid leaching 24 Figure 3- 1. The autoclave setup for pressure acid leaching experiments 41 Figure 4- 1. X-ray diffraction pattern for bulk sample of Ouaco LD-95-75 57 Figure 4- 2. X-ray diffraction pattern for residue obtained from partial dissolution of Ouaco LD-95-75 by hydrochloric acid 58 Figure 4- 3. Secondary electron images of particles in Ouaco LD-95-75 60 Figure 4- 4. X-ray diffraction pattern for Poum ore bulk sample 62 Figure 4- 5. X-ray diffraction pattern for +50 mesh fraction of Poum ore 63 Figure 4- 6. EDX-spectrum of gibbsite in Poum ore sample 65 Figure 4- 7. Secondary electron image of goethite in Poum ore 65 Figure 4- 8. X-ray diffraction pattern for Ivory Coast blend bulk sample 66 Figure 4- 9. X-ray diffraction pattern for +400 mesh of Ivory Coast blend sample 67 Figure 4-10. Relationship between A1203 and NiO content in Al-Mn oxide In Ivory Coast blend 69 Figure 4-11. Secondary electron image of goethite particles in Ivory Coast blend 69 Figure 4- 12. X-ray diffraction pattern for LD-97-13 bulk sample 70 Figure 4-13. X-ray diffraction pattern for DCB treated LD-97-13 sample 71 Figure 4- 14. X-ray diffraction pattern for LD-97-18 bulk sample 74 Figure 4- 15. X-ray diffraction pattern for DCB treated LD-97-18 sample 75 Figure 4- 16. X-ray diffraction pattern for LD-97-24 bulk sample 77 Figure 4- 17. X-ray diffraction pattern for DCB treated LD-97-24 sample 78 Figure 4- 18. X-ray diffraction pattern for LD-97-28 bulk sample 80 Figure 4- 19. X-ray diffraction pattern for DCB treated LD-97-28 sample 81 Figure 4- 20. X-ray diffraction pattern for LD-97-32 bulk sample 82 ix Figure 4-21. X-ray diffraction pattern for DCB treated LD-97-32 sample 83 Figure 4- 22. The acid consumption during interruptive pressure acid leaching 88 Figure 4- 23. Mineralogical change of Ivory Coast blend after interruptive pressure acid leaching, a: heating up to 260°C; b: 5 minutes; c: 10 minutes; d: 15 minutes; e: 30 minutes 90 Figure 4- 24. X-ray diffraction patterns for DCB treated leaching residues of Ivory Coast blend, a: 5 minutes; b: 15 minutes; c: 30 minutes 91 Figure 4- 25. The secondary electron images of leaching residues produced after pressure acid leaching of Ivory Coast blend with different leaching time. a: 5 minute; b: 10 minutes; c:15 minutes; d: 30 minutes 94 Figure 4- 26. X-ray diffraction pattern for leaching residues of LD-97-13 sample 97 Figure 4- 27. X-ray diffraction pattern for DCB treated leaching residues of LD-97-13 sample 98 Figure 4- 28. X-ray diffraction pattern for leaching residues of LD-97-24 sample 100 Figure 4- 29. X-ray diffraction pattern for leaching residues of LD-97-28 sample 102 Figure 4-30. X-ray diffraction pattern for DCB treated leaching residues of LD-97-28 sample 103 Figure 4-31. X-ray diffraction pattern for leaching residue of LD-97-18 sample 105 Figure 4- 32. X-ray diffraction pattern for DCB treated leaching residue of LD-97-18 sample 106 Figure 4- 33. X-ray diffraction pattern for leaching residue of LD-97-32 sample 107 Figure 4- 34. X-ray diffraction pattern for DCB treated leaching residues of LD-97-32 sample 108 Figure 4-35. The secondary electron images of the hematite particles obtained from pressure acid leaching of various laterite samples, a: Ivory Coast blend; b: Ouaco ore; c,d: Poum ore I l l ACKNOWLEDGEMENT The present study would not have been possible without the help of many dedicated people. First of all, I would like to express my deepest gratitude to my supervisor, Dr. David Dreisinger, for providing me the opportunity to study at UBC and giving me an unforgettable experience in Canada. I am especially indebted to the people at Falconbridge Limited who generously sponsored my study. My sincere thanks go to Dr. Be Wassink for his patient help in developing the experimental procedure for the acid measurement in the laterite leaching liquor and valuable advice during my lab experiments. I am also grateful to Mary Mager for her assistance in the SEM investigations and Dr. Mati Raudsepp for his advice on XRD analysis. Many thanks are due to Hu Long for help in the autoclave leaching experiments and providing assistance whenever I needed it and Anita Lam for setting up the selective dissolution procedure and giving me help on XRD experiments. The help of Ross McLeod, Carl Ng with the maintenance of the autoclave is also greatly appreciated. Finally, I feel indebted to my husband, Daqing, whose encouragement supports me to accomplish this thesis and to my children, Shan and Jonas, whose vitality sustains my spirit. 1 C H A P T E R 1 I N T R O D U C T I O N Lateritic deposits constitute two thirds of the earth's known nickel resources and contain an even greater proportion of the world cobalt resources (Krause, 1997). These ores, formed by the weathering of ultra basic rocks, typically consist of a limonite zone and a saprolite zone. Nickel evenly distributes throughout goethite in the limonite zone and serpentine in the saprolite zone. No discrete Ni-mineral can be found in laterite deposits. Therefore it is impossible to concentrate the ore by any present mineral processing technology. The whole ore is therefore treated for nickel recovery. Because of the complex mineralogy of the ores, a variety of possible extractive techniques have been developed. Among these technologies, matte smelting, ferronickel smelting, high temperature sulphuric acid leaching and reduction-ammonia leaching are currently in commercial operation (Canterford, 1975, Kerfoot, 1988, Monhemius, 1987). Smelting presently accounts for 80% of nickel production from laterites. It has been used for treating the lower horizon saprolite ores. Saprolite ore contains nickel, iron, silica and magnesia in such proportions that when heated to over 1500°C it melts and separates into a heavy ferronickel phase and a lighter silica-magnesium slag phase. The relatively high nickel and low iron contents of saprolite ore ensure a ferronickel product containing more that 20% nickel. It is impractical to use limonite ore for ferronickel production since the product would contain less 2 than 10% nickel, reducing the flexibility of its use by stainless steel mills and increasing transportation cost (Reid, 1996). Since limonite ore is not suitable for pyrometallurgical processing, two commercial hydrometallurgical processes, ammonia leaching and sulphuric acid leaching, have been developed to treat the limonite ores. The ammonia leaching process is highly sensitive to energy cost due to the need to dry the ore prior to reduction roasting. Pressure acid leaching has two major advantages over ammonia leaching: (1) the energy-intensive drying and reduction steps are eliminated, since the ore can be directly treated, and (2) recoveries of over 90% for both nickel and cobalt can be achieved. Pressure acid leaching is a process of dissolution and selective thermal hydrolysis. As ores dissolve in the sulphuric acid solution, nickel, cobalt and magnesium stay in the leach liquor, whereas the major portion of iron and aluminum are precipitated to form a barren residue. The precipitation reactions of iron and aluminum regenerate acid and therefore minimize the acid consumption. A typical Ni/Fe ratio of around 0.03 in laterite is increased by a factor of 200 to 300 times (Chou, 1977) in the leach liquor, resulting in a very high selectivity for nickel over iron. Pressure acid leaching of limonite is usually conducted at high temperatures (240~270°C) and high acidities (pH<l) (Boldt, 1967, Carlson, 1961, Chou, 1977). The recoveries of nickel and cobalt strongly depend on the temperature and the ratio of acid-to-ore. The leaching temperature 3 and acid addition required for high nickel and cobalt recoveries in excess of 90% vary widely from one ore body to another. There currently is no standard process for the acid pressure leaching of nickel laterites due to their complex mineralogy. Naturally, the metallurgical behavior of an ore is directly dependent on the mineralogical characteristics of that ore. Laterite ores vary widely in composition and often more than one nickel-bearing mineral is present. Hence the extraction of the nickel from laterites will depend on the reactivity of each nickel bearing mineral during acid leaching. Therefore, the importance of a complete mineralogical examination of the ore as part of a program on the development of a leach process cannot be overstressed. The objective of this research is to examine the mineralogy of a variety of laterite samples, investigate the dissolution behavior of the minerals making up the ores, and eventually understand the chemistry of laterite pressure acid leaching from a mineralogical point of view. 4 C H A P T E R 2 L I T E R A T U R E R E V I E W 2.1 Mineralogy of Nickel Laterite One of the major objectives of this research was to examine the mineralogy of nickel laterite ores. In this section the literature on mineralogy of laterites is reviewed, with the focus on: (1) the formation of the laterite profile, (2) the laterite profile feature, (3) common minerals existing in the laterites and (4) the nature of these minerals. 2.1.1 The Ore Sequence in Laterite Deposits Lateritic nickel ore is becoming increasingly important for the world's nickel supply. Its economic importance has already stimulated intensively geological, mineralogical and geochemical research. Numerous analyses of profiles through nickeliferous laterites and associated ultramafic rocks as well as general research on laterite ore formation have been published (Bernardelli, 1983, Burger, 1996, Esson, 1978, Haldeman, 1979, Lithgow, 1993, Monti, 1996, Ogura, 1979, Schellmann, 1983 , Troly, 1979, and Ogura, 1979) Nickel laterite deposits are derived from extensive tropical weathering of the basic rocks peridotite or serpentine and are usually found in tropical or subtropical regions such as New Caledonia, Indonesia, the Philippines, Australia, Brazil, Cuba and other tropical countries 5 (Golightly, 1979). A major constituent of peridotite rock is olivine, a silicate of magnesium and iron [(Fe,Mg)2Si04]. It often contains small amounts of nickel due to the fact that the ionic radii of Fe2+, Mg 2 + and Ni 2 + are very similar so that nickel can substitute into the olivine structure in place of either iron or magnesium. The nickeliferous peridotite typically contains about 0.2% Ni, 0.02% Co, 0.2% Cr and 10% Fe. Serpentine is an alteration produce of olivine, from which iron has been removed by weathering, leaving a hydrous magnesium silicate structure. Nickeliferous peridotites are attacked and gradually dissolved by surface waters that have been made acidic by dissolved carbon dioxide and organic acids derived from decomposing vegetable matter. The ground water, containing dissolved iron, nickel, magnesium and silica, percolates downwards. Iron is rapidly oxidized and is precipitated from the solution near the surface as a hydrated iron oxide goethite. Much of the cobalt and part of the dissolved nickel is co-precipitated with the iron in solid solution in the goethite lattice, resulting in an iron-rich, nickeliferous mineralization which is usually described as limonitic ore. The nickel remaining in the solution, together with magnesium and silica, is carried on down through the underlying basic rock. As it descends, the pH of ground water increases due to reactions with the bed rock and this causes the precipitation of hydrous nickel-magnesium silicate, known as garnierite (Ni,Mg)3Si205(OH)4. Nickel is also incorporated in the serpentinized silicates forming a sprolite ore through ion exchange reaction (Golightly, 1979 (1)): Mg3Si205(OH)4 + 3Ni2+ <-> Ni3Si205(OH)4 + 3Mg2+ (2-1) 6 Swchellman (Swchellman, 1983) suggested that nickel silicate ores were normally slightly weathered serpentines which had lost a considerable percentage of magnesium and were therefore able to take up nickel. In the areas of impeded water drainage allowing saturation conditions for nontronite or quartz formation, a smectite zone may form below the limonitic ore. The following simplified overall equation serves to illustrate the formation of goethite plus nontronite from olivine. 4Fe2Si04 + 8H++402 <-> Fe2Si4O,0(OH)2 + 6FeOOH (2-2) Olivine Nontronite Goethite A typical lateritic nickel deposit can usually be divided into following distinct zones, which occur at increasing depths from the surface: Limonite zone (L ore): This zone consists primarily of goethite a -FeOOH and may also contain maghemite and less frequently small quantities of hematite. Minerals such as spinel, magnetite or maghemite, primary talc, amphiboles and rarely chlorite may also persist in this zone. Other minor constituents are gibbsite, secondary quartz and manganese oxides. More rarely kaolinite, relic chromite and a few relics of serpentine are'commonly present in the more than 60 urn fraction. Nickeliferous limonite is generally quite uniform in chemical and mineralogical composition. 7 Goethite crystallizes as fine needles, laths or globular particles. Small amounts of Ni, Cr, Al and Si can be detected in goethite. As shown by synthesis of Ni goethite, its capacity to take up nickel is restricted to about 1.5% (Schellmann, 1983). Limonite ore contains low silica, low magnesia, typically 1.4% Ni, 0.15% Co and >40% Fe (Reid, 1996). Smectite -Quartz Zone (Nontronite zone): This is an intermediate layer largely consisting of soft smectite clays, usually nontronite, and hard crystalline quartz. Black Mn-oxide persists on fractures and as irregular accretions in this horizon. Goethite may be present as masses or as coatings on fracture (Burger, 1996 (1)). In the nontronite zone, some nickel substitutes for iron in goethite but the bulk of the nickel is associated with nontronite. The nontronite lattice has substaintial capacity for cation exchange and can grade over 2% nickel. The Mn-oxides readily adsorb metallic ions and over 5% grade for both Ni and Co are common in the microprobe analysis of Mn-oxides. The texture of parent rock is often well preserved. For a wide range of climatic and petrological condition this zone does not develop. Where the zone is developed, however, the maxima for both nickel and cobalt are achieved within the nontronite zone, for example, Bulong laterite profile (Burger (2), 1996). Saprolite Zone (Serpentine ore): This is a zone of somewhat altered bedrock in which most of the parent rock minerals are present and the original structures and textures are well preserved. Shape and mineralogical composition of saprolitic horizons are chiefly affected by the nature of the bed rock types, their degree of serpentinization and the regime of the soil water. Garnierite precipitates in veins and pockets within the ore can be found in some deposits. Saprolite in 8 unserpentinized peridotite in a well drained location contains a core of unaltered olivine-green harzburgite surrounded by saprolite rims, and fracture fillings of "garnierite" is important. In serpentinized profiles the saprolite zone consists of boulders of partly weathered bed rock. Saprolite zone contains high silica, high magnesia, typically 2.4% Ni, 0.05% Co, and < 15% Fe. Bulk grades of saprolite horizons may exceed 3% Ni over thickness of several meters in garnierite-rich saprolite. In nontronite rich profiles, however, bulk Ni rarely exceeds 1.5% (Burger, 1996(1)). Transition zone: There is normally a transition zone of a few meters thick at the base of limonite between the limonitic and the saprolitic zone. It is usually enriched in manganese, cobalt and nickel in iron-manganese oxide or poorly crystalline lithiophorite and cryptomelane (K2Mn8016). No discrete Ni minerals exist in this zone. Ionic Ni is incorporated within or adsorbed by a number of secondary oxide and silicate minerals. It is very heterogeneous, containing intermediate values of nickel, magnesium, iron and silica. The ore sequence described above is not found in all deposits. In some deposits the silicate ore zone is not present, (e.g. Conakry, Guenee and Skinda, India), while in others the nickel limonite zone has been eroded away, (e.g. Morro do Niquel, Brazil). The smectite zone forms under restricted drainage condition, or under a tropical wet-dry climate, (e.g. Murrin Murrin and Bulong, Australia). In the conditions of extremely intensive weathering, i.e. high precipitation and excellent drainage, the saprolite zone would be suppressed in favor of the transformation of 9 parent rock directly to limonite. Figure 1 shows the typical ore profiles of deposits formed under different climate conditions (Krause, 1997). HUMID EQUATORIAL CLIMATE DRYER EQUATORIAL CLIMATE FORMATION Iron Cap Iron Shot Umorite Overburden Limonite Ore Nontronite SaproGte with Boulders Peridotite Figure 2- 1. Effect of Climate on typical profiles of nickeliferous laterite orebodies. On the basis of a broad review of the literature, two conclusions can be reached: (1) in limonitic laterites nickel is associated with goethite, being located within the goethite lattice, and (2) in serpentinic laterites nickel substitutes for some of the magnesium in the silicate minerals (talc, chlorite) since both elements have similar ionic radii (Ni2+, 0.69A; Mg 2 +, 0.66A). 10 2.1.2 Nature of Minerals in Nickel Laterite Goethite - aFeO(OH): Goethite is a major constituent in limonitic ore. It is the most frequently occurring form of iron oxide in soils (Schwertmann, 1977). It has the greatest stability under most soil conditions. Goethite occurs in almost every soil type and climate region and is responsible for the yellowish-brown color of many soils. The basic structure unit of iron oxide is an octahedron in which Fe atom is surrounded by six O or by both O and OH ions. The O and OH ions form layers in either hexagonal or cubic closing packing. Goethite is in the former case. Goethite structure consists of double chains of Fe-O-OH octahedra extending along Z-axis as shown in Figure 2-2 (Schwertmann, 1988). Synthetic goethite is usually acicular with the Z-axis direction lying along the needle axis. This acicular morphology is also found in soils. a Goethite Figure 2- 2. The ideal model of goethite 11 The Fe3+ in the octahedral position in the goethite may be partially replaced by other trivalent metal cations of similar size, such as Al 3 + , Mn 3 + and Cr 3 +. Aluminum is the best known example for the substitution, although other metals such as Ni, Cr, Mn, Co, and Si may also substitute for Fe (Schwertmann, 1988). Hematite - Fe 20 3: Hematite consists of layers of Fe06-octahedra which are connected by edge and face sharing and stack perpendicular to the c direction as illustrated in Figure 2-3 (Schwertmann, 1988). Two thirds of the octahedra interstices are filled with Fe (III). It is the second most frequent soil iron oxide, but appears to be absent in soils recently formed under a humid temperature climate such as northern and mid-Europe and the northern part of American continent. Figure 2- 3. The ideal model for hematite. 12 Maghemite - Fe 2 0 3 : Maghemite is especially common in highly weathered soils of tropical and subtropical climates. The mineral is reddish-brown and ferromagnetic. The structure of maghemite is more variable than that of the other Fe oxides. Maghemite, like magnetite, generally has a spinel-like structure based on a unit cell containing 32 close-packed O. There are 21 1/3 Fe(III) ions distributed among the 8 tetrahedral and 16 octahedral sites; 2 2/3 sites per unit are vacant. The vacancies could occur in either octahedral sites, tetrahedral sites or both. Both cubic and tetrahedral structures may be possible depending on whether the vacancies are ordered or not. Random arrangement of vacancies give rise to the cubic structure, while the tetrahedral structure requires an ordered arrangement of vacancies (Eggleton, 1988). Lithiophorite - (Al , Li )Mn0 2 (OH) 2 : Lithiophorite has a layered structure with alternate sheet of Mn 4 + 0 6 and (Al, Li)(OH)6 octahedra. Lithium occupies one-third of the Al sites. Mn oxides in soils have high sorption capacity. Specific adsorption of heavy metal cations by Mn oxides occurs in the order Pb > Cu > Mn > Co > Zn > Ni and leads to the accumulation of relatively high concentrations of these ion in Mn oxides. The Co that is adsorbed by Mn oxides becomes more strongly bound with increasing time as it is oxidized and replaces Mn in the crystal structure (Gilkes, 1988). In many soils almost all of the Co is associated with the manganese minerals (McKenzie, 1977). Lithiophorite having an average of 10.0% Ni and 11.24% Co was found in the nickeliferous laterites of Sukinda (Das, 1995). Kaolinite - A I 2 Si 4 O, 0 (OH) 8 : Kaolinite is one of the most widespread clay minerals in soils. It is abundant in soils of warm moist climates and is a prominent constituent of oceanic sediments in 13 the equatorial belt. The composition of kaolinite corresponds closely to the formula, there being little or no atomic substitution. 'The ideal compositions are: A1 2 0 3 39.5, S i 0 2 46.5, H 2 0 14.0%. It has low cation and anion exchange capacity (Dixon, 1977). Figure 2-4 shows the structure of kaolinite (Klein, 1993). It consists of tetrahedral S i 2 0 5 sheet bonded to a gibbsite octahedral sheet. Kaolinite is an important industrial mineral, being used as a filler in paper and as an essential raw material in the manufacture of ceramics. Figure 2- 4. Diagrammatic sketch of kaolinite structure (Klein, 1993). Talc - Mg 3Si 4Oi 0(OH) 2: Talc is a secondary mineral formed by the alternation of magnesium silicates, such as olivine, pyroxenes and amphiboles, and may be found as pseudomorphs after these minerals. There is little variation in the chemical composition of most talc. Pure talc contains MgO 31.7, S i 0 2 63.5, H 2 0 4.8%. Small amounts of A l , Ti may substitute for Si and Fe may replace some of the Mg. Talc is composed of two Si- tetrahedral sheets and with a central 14 octahedral sheet in which all octahedral positions are occupied by Mg. The structure of talc is shown in Figure 2-5 (Klein, 1993). Figure 2- 5. Diagrammatic sketch of talc structure. Clinochlore - Clinochlore is a member of the chlorite group silicates. The structure of chlorite is composed of two sheets of talc (or pyrophyllite) separated by a brucite- (or gibbsite-) like octahedral sheet. The chemical composition of the octahedrally-coordinated cations that make up the two types of the octahedral sheets for the minerals in chlorite group is not the same even for the specific chlorite structure. More frequently either Al(OH)3 or Mg(OH)2 dominates the chemical compositions of the interlayer hydroxide sheet in most chlorites, but cations of Fe, both 2+ 3+ Fe and Fe , Mn, Cr, Cu, V, Ti, and Ni have been reported to occur as a part of chlorite structure (Barnhisel, 1977). The general formula of chlorite may be represented as follows: A5_ 6Z 4O 1 0(OH) 8 , where A = Al, Fe 2 +, Fe 3 +, Li, Mg, Mn, Ni, and Z = Al, Si, Fe3 +. 15 Spinel: The minerals of the spinel group are based on arrangement of oxygen in approximately cubic closest packing along (111) planes in the structure. The cations that are interstitial to the oxygen frame work are in octahedral and tetrahedral coordination polyhedral with oxygen. The general chemical formula of spinel group is XY 2 0 4 , where X and Y are various cations with variable valence. Several end members of spinel are: Spinel MgAl 20 4, Magnesiochromite MgCr 20 4, Chromite FeCr204, Galaxite MnAl 20 4, Magnetite Fe3+(Fe2+Fe3+)04, Hercynite FeAl 20 4 and Franklinite ZnFe204. The minerals of the spinel group show extensive solid solution between the various end-members. There are, for example, solid solutions of chromite and magnesiochromite, spinel and hercynite (Klein, 1993). 2.2 Nickel Laterite Ore Processing 2.2.1 General After the Second World War, Inco's Process Research Department tested nickel laterite ores of various compositions from ten different countries (Toomve, 1979). The experimental method employed for nickel extraction were either entirely pyrometallurgical, entirely hydrometallurgical or a combination of pyrometallurgical followed by hydrometallurgical steps. No single process was found suitable for all ore types. Because of the complex mineralogical characteristics of laterite, lateritic nickel ores differ substantially, even within the same ore body, in their amenability to extraction methods. So far four basic techniques have been used to treat nickel laterites (Canterford, 1975, Kerfoot, 1988, and Monhemius, 1987): smelting to ferronickel; smelting in the presence of a sulfur-containing compound to produce nickel matte; 16 sulfur acid leaching at high temperatures and pressures; and reduction roasting followed by ammonia leaching at atmospheric pressure. A list of plants treating laterite ore by both pyro and hydrometallurgical processes is shown in Table 2-1 (Reid, 1996). Table 2-1. Nickel laterite ore processing. Plant Country Product 1995 Production Ni (txlO3) Co (txlO3) SMELTERS SLN New Caledonia Ferronickel 42 -Nickel Matte 10 0.1 PT Inco Indonesia Nickel Matte 45 -Falcondo Dominican Rep Ferronickel 32 -Cerro Matoso Columbia Ferronickel 25 -Pamco Japan Ferronickel 39 -Sumitomo Japan Ferronickel 17 -Nippon Yakin Japan Ferronickel 13 -Larco Greece Ferronickel 9 Codemin Brazil Ferronickel 9 Aneka Tambang Indonesia Ferronickel 11 Glenbrook USA Ferronickel Closed Sub Total 266 0.1 AMMONIA L E A C H Queensland Nickel Australia Nickel Metal, Cobalt Sulfide 28 1.4 Nicaro Cuba Nickel Oxide 9 -Punta Gorda Cuba Nickel Oxide 13 -Los Camariocas Cuba Uncompleted - -Tocantins Brazil Ni-Co Carbonate 10 0.3 Sub Total 60 1.7 PRESSURE ACID L E A C H Moa Bay Cuba Ni-Co Sulfide 18 1.7 Grand Total 344 3.5 17 A number of new laterite projects are appearing in the South West Pacific region, the majority being based on hydrometallurgy, principal pressure acid leaching technology. Some of these are listed in Table 2-2 (Metheson, 1996). Additional projects are in various stages of development. Table 2-2. New laterite projects using pressure acid leaching in the South West Pacific region. Company Project ore body Production (tx 103) Ni Co Anaconda Murrin W Australia 45 3 Resolute/Samantha Bulong W Australia 15 1.35 Centaur Cawse W Australia 7.5 1.35 Calliope Gladstone NewCal/Indo Philippines 18 1.8 2.2.2 Ferronickel Smelting and Matte Smelting The rotary Kiln-electric furnace process for the smelting of garnierite ore to ferronickel was first introduced in the mid-1950s in New Caledonia (Boldt, 1967). This process was derived from iron ore smelting technology and since its successful introduction in New Caledonia it has become the predominant method for ferronickel production. In the approaches adopted for the processing of Ni laterite ores, various plants show substantial differences. However, there are four processing stages generally followed by all operations: drying, calcination reduction, smelting and refining (Basto, 1997, Daenuwy, 1997, Diaz, 1987, Kohga, 1997,Wiryokusumo, 1997, Uceda, 1997). 18 Drying: All plants use dryers, mostly direct fired rotary dryers, for moisture elimination. Drying is usually conducted at about 250°C. The operation is controlled to a residual moisture in the ore of 15-20% to avoid excessive dusting during drying and subsequent flowsheet steps. Calcination and reduction: Elimination of chemical bound water and prereduction of the ore are generally conducted in rotary kilns which are operated counter-currently. The exception is Falconbridge's operation in the Dominican Republic where shaft furnaces are used for this purpose. Chemical bound water is typically released at about 400°C. As the ore moves toward the firing end of the kiln, temperature and reduction potential increase. Under the controlled reducing conditions used in rotary kilns, metallization of the ore starts at 500~600°C. When the ore reaches discharge end, its temperature has increased to 800~900°C and most of the NiO, but only a fraction of Fe oxide have been reduced to metal. Smelting: Electric furnace smelting is the route used for the production of ferronickel and matte from nickel lateritic ores. The liquidus temperatures of the metal'phases are strongly dependent on their S, C and Si content. At most smelting plants, the actual slag skimming temperature is -50-100°C above corresponding liquidus. Generally, the operating conditions are chosen to attain the superheat requirement for the metal phase. All FeNi producers require crude metal temperatures >1400°C to be able to carry out the refining step, since the liquidus of the refined FeNi is~1450°C. 19 Refining of ferronickel: Impurities such as S, Si, C and P must be removed from the crude metal in order to meet product specifications. Electric arc furnaces, oxygen blown converters and ladle furnace are used for refining. The products after refining contain 20~47%(Ni+Co) depending on the feed material and plant practice. Matte smelting is very similar to that used for ferronickel production, except that sulfur is added to form nickel and iron sulfides. There are only two plants producing nickel matte: SLN and P.T. Inco. In SLN, gypsum is added as a source of sulfur, whereas in P.T. Inco, elemental sulfur is added to the hot calcine before it is transferred to the electric furnace. Both companies use Peirce-Smith converters to upgrade furnace matte. The final converted mattes from the two plants are almost identical in composition: 77-78% Ni, 0.5-0.6% Fe and 21-22% S (Diaz et al 1988). Smelting currently accounts for 80% nickel production from laterites. The nickel recovery in smelters ranges from 85% to 95%, depending on the ore feed and furnace operating conditions. It is of significance that only the smelters that produce matte allow the opportunity for cobalt to be recovered. Nickel and cobalt are separated from matte in the matte refining operation. Most of the cobalt in the furnace ore feed reports to the ferronickel, which is not a problem to stainless steel producers, but the presence of cobalt can be a problem to producers of nickel containing specialty steels. 20 2.2.3 Ammonia Leaching (Caron Process) Professor Caron at Delft University in Holland developed the Reduction Roast - Ammonia Ammonium Carbonate Leach process in the 1920s. The so-called Caron process has been in operation at Nicaro, Cuba, since the 1940s. The process is applicable to high iron limonitic laterite ores and involves four main operations (Power, 1977): ore drying and grinding, reduction roasting, leaching with amoniacal ammonium carbonate solution and metal recovery from solution. Ore drying and grinding: Ore, which may be blended to give as constant a composition as possible, is first dried in direct-fired rotary kilns to about 2-3% moisture. Since the moisture content of the raw ore is usually in the range 30-50wt%, considerable energy is expended in the drying operation. Following drying the ore is milled to reduce the particle size to 74um or less. Reduction roasting: The reduction roasting step is critical in the process. The dried ore is reduction roasted in multihearth roasters in an atmosphere containing hydrogen and carbon dioxide from combustion of fuel oil in a deficiency of air. Nickel and cobalt are selectively reduced to the metallic form and thus rendered amenable to ammoniacal leaching, while iron remains as insoluble oxides and silicates. Although the objective is to cause selective reduction of nickel and cobalt, typically 10% of the iron content of the ore is also reduced and the product is an iron-nickel alloy rather than elemental nickel metal. The roasting reaction can be presented in idealized form as: 21 NiO + 2 Fe203 + 3 H 2 o FeNi + Fe304 + 3 H 20 (2-3) Leaching: The reduced ore discharged from the furnace is passed through coolers in a reducing atmosphere and cooled to 150-200°C, then is discharged into quench tanks containing ammoniacal ammonium carbonate leach solution. Leaching is carried out in a series of agitated tanks into which air is introduced in order to oxidize and dissolve the iron-nickel alloy. In spite of the high pH of the solution, which is about 10, hydrolysis of nickel and cobalt is prevented by their strong affinities for dissolved ammonia, which result in the formation of soluble nickel and cobalt complex ammine ions. Iron, on the other hand, although initially dissolving as ferrous ammine complexes, is rapidly oxidized to the ferric state, which hydrolyses and precipitates as ferric hydroxide, eventually leaving the process in the leach residue. The reactions are idealized in the following equations: Metal recovery: In the original Nicaro process no attempt was made to recover cobalt separately. Cobalt followed nickel through the process and reported in the final crude nickel oxide product. Increased demand and higher prices have made cobalt recovery economically attractive and all modern plants incorporate a cobalt-recovery step. The pregnant liquor overflow from the washing thickeners is first filtered to remove suspended solids and then hydrogen sulfide gas or FeNi + 0 2 + 8 NH 3 + 2 H 20 <-> Ni(NH3)62+ + Fe(NH3)22+ + 4 OH" (2-4) 4 Fe(NH3)22+ + 0 2 + 2 H 20 + 8 OH" <-> 4 Fe(OH)3 + 8 NH (2-5) 22 ammonium sulfide solution is mixed with the liquor, to cause precipitation of cobalt sulfide. The precipitated material is thickened, washed and dried and sold as a mixed sulfide byproduct. The cobalt-free solution then passes to the nickel precipitation stage. Here ammonia and carbon dioxide are removed from solution by steam heating in stripping stills, causing precipitation of basic nickel carbonate (BNC). The final part of the process, the treatment of the BNC filter cake to produce the ultimate nickel product, is where major differences in process design are found. In the original Nicaro process, the filter cake was calcined in a rotary furnace to drive off the final carbon dioxide and to form nickel oxide product. In the variation developed by Sherritt Gordon, which is in use at Surigao refinery in the Philippine, BNC filter cake is dissolved in ammonium sulfate solution. The nickel ammonium sulfate solution is then heated in horizontal autoclaves and treated with hydrogen under pressure to precipitate elemental nickel metal powder from solution. Theoretically, there is no consumption of the leaching reagents for Caron-type process. The leaching chemistry provides the advantage of selective leaching for nickel and cobalt and minimal corrosion problems in leaching systems. Energy requirement, however, for this process is high, and overall metal recovery is low: typically 75-80% recovery of nickel, 40-50% recovery of cobalt. 23 2.2.4 Pressure Acid Leaching (Moa Bay Process) The laterite plant at Moa Bay, Cuba, was the first, and remains the only, commercial plant to treat limonite ore using pressure acid leaching. Originally completed in 1959 by Freeport Sulfur, the plant is now jointly operated by Cubaniquel and Sherritt International Corp. The ore deposit in Moa typically contains 1.3% Ni, 0.12% Co, 0.55% Mg, 4.8% Al, 4.3% Si and 2% Cr. The major constituent of the ore is goethite which accounts for about 70% of the ore. Aluminum is present as gibbsite and accounts for 10% of the ore. Residual serpentine and quartz are the other main components. Nickel and cobalt are not present as discrete mineral phases but are associated with goethite and to a lesser extent in spinels. A schematic flowsheet of the process at Moa is presented in Figure 2-6 (Chalkley, 1996). The leaching circuit consists of five independent trains; each having four pachuca type vertical reactors connected in series. Ore thickener underflow pumps deliver slurry to a direct contact preheater, where the slurry is heated by low pressure (100 kPa) steam to between 70 and 80°C. The preheated slurry flows by gravity to one of two storage tanks. Slurry is pumped from the storage tank by a centrifugal pump to a positive displacement pump which deliver the slurry to a direct contact heater where the slurry, containing about 45% solids is heated to 246°C with high pressure (4500Kpa) steam. The preheated slurry flows from the preheater to the first reactor which is also fed with concentrated sulfuric acid through a titanium injection line. Agitation of the slurry in the autoclaves is achieved by injecting high-pressure steam into a central draft tube, which causes circulation of the slurry in the autoclaves as well as maintaining the required 25 reaction temperature. Slurry passes through the reactors by overflow pipes. Residence time in the autoclaves is 1-2 h, depending on the ore composition and the acid to ore ratio being used and this is typically about 0.22 kg acid/kg ore treated. The parameter which has the biggest impact on metal extraction is the acid addition, which in turn sets the free acid concentration in the leach discharge solution. Metal extraction increases with acid addition. Acid addition also affects the sedimentation characteristics of the leaching discharge slurry, with the sedimentation of the slurry generally improving at higher acid concentrations in the leach solution. Since the plant commenced operation, metal extraction averaged 94% and values in excess of 95% are now regularly obtained with good control of acid addition. The metal concentrations in the leach solution are largely influenced by the solid content of the ore slurry feeding the leaching reactor and the ore grade. The nickel concentration is generally in the range of 5.8 to 6.8 g/L and the free acid concentration is about 32 g/L (Chalkley, 1996). The leached slurry is passed through a heat exchanger and onto a flash tank to be let down to atmosphere pressure. A seven stage CCD circuit is used to separate the soluble nickel and cobalt from the leach residue. Solution from CCD wash circuit is pumped to the reduction and neutralization section. Prior to the neutralization of free acid, the raw liquor is treated with hydrogen sulfide gas to reduce Fe(III) to Fe(II) and Cr(VI) to Cr(III), and to precipitate copper as copper sulfide. Complete reduction of Cr(VI) to Cr(III) is achieved, while 50-75% of the Fe(III) is reduced. The copper concentration is controlled to less than 0.008 g/L. The neutralization 26 circuit consists of four agitated tanks. Limestone mud slurry at 40% solids, which analyzed about 90% CaC0 3 is added as the neutralization reagent. The pH of the neutralized liquor is maintained in the range of 2.1 to 2.3. The concentration of nickel plus cobalt in the precipitation feed solution is maintained below a maximum value of about 4.8 g/1 to prevent the tendency to form sulfide pellets in the autoclave. Nickel and cobalt are selectively precipitated from the neutral solution by hydrogen sulfide gas in the sulfide precipitation plant. Zinc also precipitates, but magnesium, manganese, iron and aluminum remain in solution. The precipitation reaction is carried out in horizontal agitated autoclaves which operate at 120°C and 1034kPa. Under these conditions 99% of the nickel and 98% of the cobalt precipitate as sulfides. In order to achieve high metal recoveries and to control the particle size of the precipitated sulfides, recycling of the precipitate is employed in the ratio of two parts recycle to one part new precipitate. After completion of the reaction, the resulting slurry is cooled and excess H2S is removed by flashing to atmospheric pressure. Thickeners are then used to wash and to concentrate the nickel-cobalt sulfide slurry to 65 per cent solids ready for shipment. A major feature of the acid leaching of limonite ores is the formation of scale in the reactor. The scale builds up to a thickness of 75 to 100 mm over a period of the three to four months. The leach train is shut down and the scale is removed by pneumatic hammers. Staged acid addition and improved agitation are claimed to reduce the rate of scale formation. Attempts to reduce 27 scale formation by increasing the diameter of the steam injection line and increasing the quantity of steam used for agitation have proven unsuccessful. 2.3 Chemistry of Pressure Acid Leaching 2.3.1 General Sulfuric acid leaching of nickeliferous limonite is performed at temperatures high enough to solubilize nickel and cobalt as aqueous sulfates while converting most of the iron in the ore and aluminum content to insoluble hematite and aluminum sulfate (Carlson, 1961). A brief and generalized account of the major reactions is provided here (Motteram, 1996, Taylor, 1996, Papangelakis, 1996). The dissolution of most divalent metals in sulfuric acid may be described as follows: Where M = Ni, Co, Cu, Mg, Zn, and Fe2+ Iron, as goethite, is extracted into solution as ferric iron and under the prevailing conditions, is largely precipitated as hematite. The transformation of goethite to hematite proceeds in two distinct reaction steps. Goethite is dissolved during the first step, MO + H 2S0 4 -> MS0 4 + H 20 (2-6) 2 FeOOH + 3 H 2S0 4 -> 2 Fe3+ + 3S042" + 4 H 20 (2-7) 28 Hydrolysis of ferric ions then proceeds in one more step. Simplistically, the iron hydrolysis reaction may be written as: 2 Fe3+ + 3 H 2 0 ^ Fe203 + 6 H + (2-8) Overall, the reaction can be written as: 2FeOOH Fe203 + 3 H 20 (2-9) The result of the above reactions is to minimize acid consumption, which is the prime reason for leaching at high temperature. Minor amounts of acid are consumed by the iron remaining in the solution. Aluminum, generally occurring in laterites as gibbsite or associated with goethite, is brought into solution by sulfuric acid. A significant portion of the aluminum in solution is also subsequently hydrolyzed and precipitated. Unlike iron, it is precipitated as a sulfate containing salt, such as hydronium alunite: 3A12(S04)3 + 14 H 20 -> 2 (H30)A13(S04)2(0H)6 + 5 H 2S0 4 (2-10) At temperature > 280°C and acid concentrations > 60 g/L H 2S0 4, the basic aluminum sulfate (the oxide formula A1 2 0 3 .2S0 3 «H 2 0) also forms (Davey, 1962): A l 3 + + S04 2' + H 20 -> A10HS04+ H (2-11) 29 In the presence of cations such as sodium derived from the borefield water and from decomposition of a portion of the clay components of the ores, precipitation of a portion of the aluminum as the sodium alunite may be expected: The extent of the precipitation of the aluminum and the composition of the precipitate are largely decided by the temperature, free acidity, and availability of cations such as sodium and potassium. Between pH values of 1 to 0.6 (measured at 25°C) in a pure (iron free) system, equilibrium concentrations of A l 3 + have been reported between 0.54 to 3.24 g/L at 250°C. Lower temperatures and higher acidities cause an exponential increase in solubility (Papangelakis, 1994). Although the hydrolysis reactions of aluminum sulfate release a portion of the acid, the aluminum is nevertheless a significant net consumer of acid. Manganese occurs mainly in the form of Mn(IV) in lateritic ores (as lithiophorite or asbolite). Since the leach liquor contains only Mn(II), the Mn(IV) is believed to react with any dissolved Fe(II) and with Cr(III) (Krause, 1997). 3A12(S04)3 + Na2S04 + 12 H 20 -> 2 NaAl3(S04)2(OH)6 + 6 H 2S0 4 (2-12) 3 Mn0 2 + 2 Cr 3 + + 2 H 2 O o 3Mn2+ + 2 HCr04" + 2 H -+ (2-13) Some of the metals are present as silicates and also react with acid. The overall reaction of the silicates may be represented by the equation below: 30 MO.Si0 2 + H 2S0 4 -» MS0 4 + Si02 + H 20 (2-14) In this case, the corresponding silica may be considered to dissolve, followed by polymerisation and precipitation, such that the net extraction of silica is low. 2.3.2 Dissolution of Goethite and Precipitation of Hematite Goethite is a major constituent of the limonite, its dissolution strongly affects the rate and mechanism of nickel dissolution because nickel is associated with goethite. The transformation of goethite into hematite is a key reaction in the leaching of nickeliferous limonite ore with sulfuric acid at high temperature. Warren and Devuyst (Warren, 1973) reviewed metal-oxide dissolution in acidic media, and showed that the adsorption and desorption of protons and cations were key factors. A reaction mechanism explaining dissolution as a function of adsorption and desorption of species at the oxide-liquid interface was proposed. Si02 + 2H 2 0-> Si(OH)4 (2-15) n Si(OH)4 -> (Si02)n + n H 20 (2-16) Surface hydroxylation s |-FeO + H 20 <-> s |-Fe(OH) '2 Surface protonation s |-Fe(OH)2 + H 3 0 + <-» s |-FeOH+ + 2 H 20 31 Anion adsorption s|-FeOH++X- s|-FeOH+X' Desorption s f-FeOHX-> s| + FeOHX, The proposed mechanism suggested that the ability of the oxide to adsorb water and protons is critical to a dissolution process. The aqueous chemistry of iron precipitation is complex as indicated in Dutrizac detailed review of the chemistry of iron precipitation by ferric ion hydrolysis (Dutrizac, 1980). The classic study of Posnjak and Merwin (Posnjask, 1922) for the hydrolysis of iron from sulfate solution from 50-200°C determined that in the Fe203-S03-H20 system, the only stable solid phases were Fe203, Fe203.2S03 and Fe 20 3«3SO, at 200°C. Goethite is stable at room temperature but dehydrated to hematite above 130°C. In fact, for the following reaction: the change in free energy is negative, which suggests that goethite is not stable thermodynamically. Umetsu et al (Umetsu, 1977) studied the effect of aqueous nonferrous metals such as Zn and Cu on the hematite formation in the Fe203-S03-H20 system. They demonstrated that the limit of concentration of free acid for hematite formation increased in the presence of the nonferrous metals. At 200°C, addition of 50g/L Zn allows formation of hematite at up to 80g/L H 2S0 4, whilst in the absence of Zn, hematite formation ceases at 55g/L H 2S0 4. It is assumed that the 2 FeO(OH) -> Fe203 + H 20 (2-17) 32 presence of nickel sulfate will influence the precipitation of iron in a manner similar to that of zinc (Queneau, 1986). Queneau reported that as with zinc and nickel, aqueous magnesium allows the formation of hematite at higher free acid concentrations. The result, at a given free acid concentration, is less sulfur loss to the leaching residue and lower iron content in the product liquor. Tindall and Muir (Tindall, 1996) tested the effect of background salts on the transformation of goethite to hematite at 230°C and 250°C. While the addition of Na2S04 and MgS04 enhanced the transformation rate, the addition of A12(S04)3 and Cr2(S04)3 completely stopped the reaction without significantly affecting the concentrations of acid or iron in solution. The effect of MgS04 is not clear, but it is believed that in the presence of Na2S04, transformation takes place via the formation of a more stable sodium jarosite intermediate, and that A l 3 + and Cr 3 + are adsorbed strongly onto the iron oxide surface and inhibit the protonation and therefore the transformation. It is of great significance that even trace quantities of A l 3 + or Cr3+inhibit the reaction consistent with a surface adsorption effect. 2.4 Leaching Kinetics and Nickel Extraction during Pressure Acid Leaching Leaching kinetics and nickel extraction from various laterite deposits have been widely studied. Generally, the effect of several key parameters, such as leaching temperature, acid-to-ore ratio, pulp density and reaction time on the leaching kinetics and nickel extraction during pressure acid leaching was researched. 33 Carlson (Carlson, 1961) first reported the kinetics of the pressure leaching, demonstrating the importance of acid/ore ratio and temperature. The remarkable feature of their work was the shape of the nickel extraction curves. Most of the leaching was completed in the first few minutes, with little extraction occurring during the remaining hour. Chou et al (Chou, 1977) examined the effect of temperature, acid/ore ratio, percent solid, and particle size on sulfuric acid pressure leaching of nickeliferous limonite samples from south New Caledonia. It was found that neither the rate nor the extraction of nickel was affected by the ore grinding when leaching at 250°C, 0.24 acid/ore ratio. The temperature dependence of the nickel extraction rate between 250 to 275°C was pronounced during the first 10 minutes, but became almost temperature independent after 15 minutes; the 225°C and 300°C leaching temperatures resulted in significantly lower nickel extraction than did the 250 to 275°C intermediate range since lower temperature did not provide the driving force needed for rapid nickel extraction, and very high temperature resulted in coprecipitation of nickel with the solid hydrolysis products. Motteram (Motteram 1996) examined the sulfuric acid pressure leaching response of the composite, a projected average feed, from Murrin Murrin laterite deposit. For the extraction of 95% or more of the nickel and a corresponding 93% of cobalt within 60 to 90 minutes, a terminal free acid concentration of about 40 g/1 is required for leaching tests conducted between 250 and 260°C, with pulp density 40%. In this research four special samples including an overall composite representative of the majority of the resource, a limonitic composite containing higher Co, Fe and Al and lower Mg than average, a magnesitic composite, with a higher average 34 magnesium content, and a high silica composite (clay) were leached in batch. Kinetic nickel extraction data are summarized in Table 2-3. Table 2- 3. Kinetic nickel extraction data of Murrin Murrin laterite samples. Ore % solid Kg/t H 2S0 4 % Nickel Extraction g/1 H 2S0 4 120 min 30 min 60 min 90 min 120 min Overall 33.8 369 81.9 90.7 94.4 95.5 48.0 Limonitic 34.8 297 76.7 88.9 93.0 94.2 37.5 Magnesitic 29.9 442 86.1 94.1 97.0 97.1 42.8 Clay 33.5 378 87.2 95.9 97.2 97.7 55.8 With leaching conditions at 250°C, 30 to 35% solid, acid addition targeting for 45 g/1 in the products liquors, extraction of about 95% Ni was achieved with three of the feeds within 90 minutes, with the limonitic composite lagging behind slightly, possibly because of a lower acidity in the product liquors. Continuous pressure leaching tests were also carried out on the four samples. The results showed the significant effect of acid addition and leach temperature on the leaching kinetics. For the overall composite, at 250°C and with an acid addition about 36 kg/t, extractions were 92.4% Ni and 90.7% Co after 90 minutes, and 95.1% Ni and 93.1%Co after 120 minutes. An increase in the leaching temperature to 255°C at a comparable acid addition, or an increase in acid addition by about 25kg/t at 250°C both reduced the retention time to 90 minutes for extraction of 95% Ni and 93% Co. For limonitic composite, at 250°C and with an acid addition of 292 kg/t, extractions were 92.4% Ni and 90.7% Co within 120 minutes. For the magnesitic composite, 35 extractions of 95% Ni and 90% Co were attained after 120 minutes at 250°C and with a acid addition of 440 kg/t, an increase in the acid addition by 40 kg/t improved the extraction to 96% Ni and 94% Co. For the high silica composite, at 255°C and 90 minute retention time, with an acid addition of 348kg/t, the extractions yielded 96.5% Ni and 92.3% Co after 90 minutes. Hellsten (Hellsten, 1996) reported the sulfuric pressure leaching results of Cawse laterite deposit. Bench scale test work indicated that pressure leaching with sulphuric acid at 250°C gave nickel and cobalt extraction of at least 95% for a retention time of 75 minutes. Acid consumptions for all ore types were low ranging from 240 kg/t for the siliceous cobalt to 400 kg/t for the nontronite mineralization. Acid consumption for a representative blend of up-grade types averaged 375 kg/t. Calliope Metals Pty Ltd. is studying the feasibility of building and operating a laterite leaching plant to process limonite and limonite/saprolite transition ores from New Caledonia, using pressure acid leaching (Faris, 1997). Batch tests had been conducted on limonite samples. In all cases, nickel and cobalt extractions in excess of 95% and 93%, respectively, were achieved at temperature of 255°C or less, with one hour or less retention time. Optimum acid consumption, which varied between 210 and 280 acid kg/t ore, was governed primarily by the magnesium and aluminum contents of the ores. Krause et al investigated the nickel extractions obtained during batch-wise leaching of different laterite ores from Indonesia and New Caledonia (Krause, 1997). They concluded that the 36 optimum sulfuric acid addition required for obtaining high extraction of nickel (-95%) and cobalt (>90%) depended mainly on the Mg, Al, and FeC03 (siderite) contents of the ore. They also suggested that for a leaching temperature of 240~270°C, the following equation can be used to estimate the sulfuric acid requirement: wt% H 2S0 4 = 4 + 6(wt% Mg) + 2.4(wt%Al-0.8) + 3(wt%Ni+Co+Mn) + 4wt%C02 (2-1) Where wt% H 2S0 4 denotes the mass of (100%) H 2S0 4 required on a dry ore basis, and Wt% Mg, Al, etc, stand for the element analysis of the ore. According to Krause et al, equation (2-1) has proved to be reliable for a large variety of different lateritic ores, provided that a sufficiently long retention time was allowed for leaching. They also demonstrated that increasing acid addition at a given temperature by about 15%, decreased the retention time for leaching by about 20%. In summary, leaching kinetics and the nickel extraction from laterites mainly depend on the temperature and the acid addition. Leaching at 250°C is favorable for most of the laterite leachings. Acid addition for leaching different laterites shows significant variation from deposits to deposits as well as among the samples from the same deposit. High Ni and Co extractions can be obtained, provided that the sufficient quantity of acid is added and a sufficiently long residence time is provided at temperature above 220°C. The pulp density does not have a very 37 pronounced effect on the extent of Ni and Co extractions and on the leaching kinetics in the range of pulp densities of interest to commercial operations (25-40% solids). 2.5 Leach Residue Sobol (Sobol and Kukoev, 1972) investigated the leaching residue of the Moa Bay laterite ore with sulfuric acid. It was found that nickel and cobalt losses with leaching tailings had a twofold character: with primary unleached minerals and secondary connected with the formation of products following hydrolysis of sulfates of aluminum and other component. It was concluded that from the initial leaching period, the particles of iron oxides began to be covered by a film made up of the basic sulfates and chromates of trivalent iron and aluminum: hydronium-alunite (H30)A13(S04)2(0H)6, hydronium-jarosite (H30) Fe3(S04)2(OH)6, hydronium chromate alunite (H30)Al3(Cr04)2(OH)6, and hydronium chromate jarosite (H30) Fe3(Cr04)2(OH)6. The basic sulfate of the scale is multi-component solid solution with varying composition, in which part of the hydronium ions (H 30+ ) is replaced by cations Ni 2 + , Co2 +, Mn 2 + etc., while part of the hydroxyl ions is replaced by anions of silica acids. The appearance of this film produced secondary losses of nickel and cobalt with tailing; Ni 2 + , Co2 +, and Mn 2 + coprecipitate with hydronium and other basic sulfates with an approximately direction proportional concentration in the pregnant liquor. Co-precipitation with hematite was negligible. 38 2.6 The Scope of the Present Study Pressure acid leaching studies have been carried out on various laterite deposits. Previous research on nickel extraction from laterite ore and leach kinetics of pressure acid leaching of laterite showed that high N i and Co extractions can be obtained, provided that the sufficient quantity of acid is added and a sufficiently long residence time is provided at temperature above 220°C. The acid addition for the high nickel and cobalt extraction and fast leach kinetics, however, showed significant variation among deposits treated using pressure acid leaching. It is generally accepted that the ore compositions, especially the contents of M g and A l in the ore, are the key factors for the acid addition calculation, but the information about the effect of mineralogy is scarce. Laterites are complex mixtures of various minerals. The extractions of the nickel and cobalt directly rely on the compositions and mineralogy of the ore as well as the reactivity of their constituents. Hence, the scope of the present research was to firstly examine minerals existing in the laterite ore samples and nickel associations with various constitutional minerals. Interruptive autoclave leaching experiments were conducted with acid addition calculated from Falconbridge spreadsheet, followed by mineralogical examination of leach residue to investigate the dissolution behavior of laterite constituents under pressure acid leaching, and eventually understand the pressure acid leaching chemistry and provide fundamental information for process optimization. 39 CHAPTER 3 EXPERIMENTAL PROCEDURES The experimental program developed for the present study can be divided into three phases: 1. Mineralogical examination of 8 laterite samples provided by Falconbridge. 2. Pressure acid leaching experiments to generate leach residue. 3. Residue characterization to understand the chemistry during pressure acid leaching. The examination of mineralogy of ore samples began with X-ray diffraction analysis to identify the possible mineral phases. Secondly, EDX analysis was employed to confirm the minerals actually existing in the ore. For residue characterization, only X-ray diffraction was utilized. A Parr autoclave was used to carry out pressure acid leaching experiments. 3.1 Sample Descriptions and Head Assay Experimental Procedure 3.1.1 Sample Descriptions All samples were provided by Falconbridge Limited. Ouaco LD-95-75 and Poum 8 are two samples from two laterite ore bodies in New Caledonia. A sample named Ivory Coast blend is the -28 mesh fraction of two drill hole's mixture from Ivory Coast ore body. LD-97-13, LD-97-18, LD-97-24, LD-97-28 and LD-97-32 are five drill hole samples from Ivory Coast ore body. The five samples were wet screened through -65 Tyler mesh and the over sized fraction was 40 ground to pass the same screen. These samples were then dried at 105°C to remove free moisture before shipment. 3.2.2 Head Assay Experimental Procedure Ore samples were submitted to two local labs, Chemex and International Plasma Lab (IPL) for their head assays. Fusion and multi-acids methods were applied to decompose the solid samples. Ouaco LD-95-75, Poum 8 and Ivory Coast blend were fused using Na202, followed by HC1 digestion at IPL. The three samples were also digested by the aqua regia method and triple acids, namely, hydrochloric, nitric and perchloric acid at Chemex. Fusion and triple acids were used to decompose samples LD-97-13, LD-97-18, LD-97-24, LD-97-28 and LD-97-32 at IPL. All solution samples from the ore digestions were analyzed by ICP for metals Ni, Co, Fe, Mn, Al, Mg, Cr and Si. 3.2 Pressure Acid Leaching Experimental Program 3.2.1 Pressure Acid Leaching Apparatus and Procedures Figure 3-1 shows the autoclave setup used for the pressure acid leaching experiments. The tests were carried out on a 1 L scale in a standard 2 L Parr titanium autoclave equipped with internally mounted cooling coil. The autoclave can be quenched from temperature 260°C to room temperature within 5 minutes. Agitation was provided by two 58 mm diameter 45° pitched blade impellers. The stirring speed was maintained at 625 rpm to provide uniform mixing for all the 41 experiments. At this speed, approximately 60 minutes was required to achieve the temperature of240-260°C. Sulfuric acid was used to leach all the laterite samples. For the interruptive leaching, concentrated acid (-96%) was injected through a tube extending about 25 cm from the base of the autoclave head under 50 psi nitrogen overpressure. Figure 3-1. The autoclave setup for pressure acid leaching experiments. The leaching tests were performed according to the following procedures: 1. Ore samples were dried overnight at 50°C. Acid was premixed with ore slurry in the titanium liner when leaching tests were performed at 220-240°C. For the 250 and 260°C leaching test, ore was mixed with DI water in a titanium liner. 42 2. The pulp was brought to the desired operating temperature within 60 minutes. Acid was injected at 244°C and 252°C for leaching temperature at 250°C and 260°C respectively due to the temperature increase after acid injection. For the set point temperature of 240°C, autoclave temperature was controlled in the range of 238°C ~240°C; for set point temperatures 250°C and 260°C, the autoclave temperatures were controlled in the range of 246°C ~ 252°C and 255°C ~ 261°C respectively. Acid addition was calculated from a spreadsheet provided by Falconbridge Limited as described in Appendix II. 3. Autoclave was quenched to cool as quickly as possible. A suction filtration apparatus was used for solid liquid separation with Whatman #1 filter paper. 4. Approximately half of the filtrate volume of warm (80°C) DI water was used for displacement washing of filter cake. Filter cake was repulped with about 2X volume DI water. The washing solution and solid was filtered via suction filtration apparatus with Whatman #1 filter paper. 5. Leaching residue was dried in a vacuum oven for overnight and the filtrate, wash solution and residue samples were prepared for analysis. 3.2.2 Analytical Methods and Calculation of Metal Extracitons Residues and solution samples were analyzed at IPL. Filtrates and washing solutions were routinely analyzed for Ni, Co, Fe, Mn, Al, Mg, Cr and Si by ICP. Leaching residues were fused by peroxide and dissolved by hydrochloric acid before subsequent ICP analysis also for Ni, Co, Fe, Mn, Al, Mg, Cr and Si. Because of the interfering effect of Na, the detection limit for Mg in 43 the residue was >500 ppm. Except for the LD-7-13 residue, Mg could not be detected in residues obtained from the leach of the other samples. A metal mass balance was conducted for every leach test to ensure the calculated nickel and cobalt extractions were reasonably accurate. The percent metal extraction was based on the solution and residue assays: Percent metal extraction = ; M e J J / K S , • 100% (3-1) "Me + "Me W^l: weight of the metal in solution. W^f '• weight of the metal in residue. 3.3 Determination of Free Acid Concentration The laterite leach liquor is a complicated mixture of metal sulfates. Analysis of aqueous metal-containing solutions for acid is complicated by the propensity of many metal ions to hydrolyze and produce acid. Standard titration methods are generally obviated since hydrolysis in many cases may be extensive at pH values well below 7. The higher the hydrolyzable metal ion concentration is relative to the acid concentration, the more serious the problem. Measurement of pH in strongly acidic-high metal concentration solutions is complicated by the high ionic strength of the solution and possible non-linear response of the pH electrode. In addition, in a 44 high sulfate medium, sulfuric acid may not be fully dissociated. Calibration of the electrode response with solutions containing precisely known concentrations of the various metal salts may be workable, however it is often the case that the composition of samples is neither fully known, nor constant. One possible approach to the problem is to use the method of standard addition. The pH electrode is used like any other ion selective electrode. The response is recorded in mV rather than in pH units. The electrode is calibrated with standards containing known acid concentrations (eg H2S04) in a high ionic strength solution. Only the "slope" (eg mV/g/L H2S04) of the electrode is required. The sample of interest is then diluted into the same high ionic strength medium. The electrode response of a known volume of sample is recorded. Next a small volume of known acid is added. The addition should not significantly change the ionic strength. The electrode response is recorded again. If desired, further additions of standard acid may be made. The concentration of acid may be calculated based on the Nernst equation as follows: E 0 = E* + m log C( o (3-2) E, = E* + mlog[(C0V0 + C sV s)/(V0 + Vs)] (3-3) E 0 : mV reading of the sample E*: constant 45 E,: mV reading of sample plus spike of standard acid m : electrode slope (from calibration) in mV/concentration units C 0: unknown acid concentration in diluted sample V 0 : sample volume (mL) C s: standard acid concentration (units must be the same as those used for the calibration) V s: spike volume (mL) Subtracting equation 3-2 from 3-3: E, - E 0 = m log [(C0V0 + CSVS)/((V0 + VS)C0)] (3-4) Rearranging this gives: (3-5) [{(V0 + V s )10« E , - E 0 ) / m ) } - V0] A high quality electrode is essential since a deviation of + 1 mV can easily lead to errors on the order of 10%. A pH meter with a resolution of 0.1 mV is desirable. Enough time is required to acquire a stable reading. This may take 3-5 minutes. Spike acid amounts should be in the range of 50-200% of that in the original sample. The mV readings for the diluted sample and after additions of spikes must fall within the calibration range. Temperature of the samples and standards should be within 0.5°C of each other. Reagents and standards include the following: 46 MgS04 7H20 (certified ACS grade or better) 0.01, 0.075 and 0.2 N H 2S0 4 standards in 2 M MgS0 47H 20 (493 g/L) 0. 650 N H 2S0 4 in 2 M MgS04 7H20 (493 g/L). The presence of high MgS04 7H20 in samples, standards and spikes provides a high and relatively constant background sulfate concentration. This facilitates the high ionic strength required and mitigates the interfering effects of relatively low, but variable sulfate levels. MgS04 7H20 is a convenient choice since Na+ and K + salts of sulfate are not soluble enough. The +2 charge on Mg + 2 also provides a higher ionic strength. The samples are diluted so that they are within the calibration range. This also lowers the sample sulfate concentration to a value that is small relative to the background sulfate. The acid concentration in the spikes must be high enough to allow addition of a fairly small, but practical volume. However, it cannot be too high since this would result in a substantial volume of mixing error. The procedure for the free acid measurement is as follows: 1. Prepare standard samples that contain 0.01, 0.075 and 0.2 N H 2S0 4in 2 M MgS0 47H 20. 2. Dilute the filtrate and washing solution so that the solutions contain 2 M MgS04 7H20 and 0.74 - 3.2g/L H 2S0 4. 3. prepare a water bath at 25 °C. 4. Calibrate the pH probe and meter with suitable standards to get the slope m. 5. Pipette 50 mL of a diluted sample into a clean dry beaker containing a stir bar, immerse the clean and dry pH probe in the solution and record the mV reading after it has stabilized. 47 6. Add a known volume of the standard acid (0.650N H 2S0 4 in 2M MgS04 7H20), record the mV reading. 7. Calculate the acid concentration of the sample from equation (3-5) and the sample dilution factor. The idea for this method is from a chemical analysis text book (Harris, 1991). The procedure described above was developed in this laboratory (Wassink, 1998). Based on the analysis of a known concentration standard solution sample, an accuracy of around 96% can be obtained by this method. 3.4 Methods Used for Mineralogical Investigation 3.4.1 Removal of Iron Oxide The major constituents of the limonite ores and leach residues are goethite and hematite. It is very difficult to identify the minor phases existing in ore samples and leaching residues by X-ray diffraction because of the inherent detection limitation. It is also difficult to examine the chemistry of the particles existing in the minor phase without removing goethite from ore samples or hematite from leach residues. Selective dissolution by Dithionite-citrate-bicarbonate solution, partial dissolution by hydrochloric acid, magnetic separation and sieving were therefore applied to concentrate the minor phases. Selective Dissolution by Dithionite-Citrate-Bicarbonate: Iron oxides are reductively dissolved 48 if the redox potential of the extractant is sufficiently low. Dithionite, a powerful reductant, reduces all Fe(III) oxides to Fe(II) at pH value below 9-10. However, if the pH goes below 6.5, FeS and elemental sulfur may precipitate. The optimal pH is 7 - 8 and this is achieved by buffering the solution with sodium bicarbonate. The dithionite quickly loses its reducing abilities and therefore a complexing agent such as Na-citrate must be introduced to the system to keep the ferrous iron in solution. This method dissolves Fe-oxides of different crystal quality with almost no destructive effect on Fe-silicate clay minerals. This technique was proposed by Mehra and Jackson (Mehra, 1960) for estimating the "free Fe oxide" in soils and clays and modified by Anita Lam for removing goethite from laterites and hematite from leaching residues. The experiment procedure consists of the following steps: 1. Prepare a 0.3 M Na-citrate solution and a separate 1 M NaHC03 (sodium bicarbonate) solution. 2. Make sure a water bath is ready at 80°C. 3. To 0.5 g of ore, add 40 mL of 0.3 M Na-citrate solution and 5 mL of 1 M NaHC0 3 solution. 4. Bring the mixture to 80°C and then add 1 g of solid Na2S204 (sodium dithionite). 5. Stir the mixture continuously for one minute and then occasionally for the next fourteen minutes. The total reaction time is fifteen minutes. 6. To promote flocculation (for easier solid-liquid separation), add 20 mL of saturated NaCl solution followed by 20 mL of acetone. 7. Filter the mixture in a vacuum flask with Whatman #5 filter paper followed by a wash with a 49 small amount of the 0.3 M Na-citrate solution. 8. The solutions are analyzed for metals. The residues dried in vacuum dryer at 80°C are used for X-ray diffraction and SEM-EDX studies. Partial Dissolution by Hydrochloric Acid: Because of the different dissolution rates of the minerals in the laterites, concentrated hydrochloric acid (37%w/w) was used to partially dissolve the minerals and the leaching residue of the Ouaco ore. The mixture was boiled for about 20 minutes to the leach acid soluble species and to concentrate the minor mineral constituents in the digestion residue. Sieving: The samples were screened using Tyler sieves #140 (106u.m aperture) and #400 (38pm). Because goethite and hematite showed very fine particle size (a few microns), major portions of these phases reported to the -400 mesh fraction. Minor phases such as manganese oxide, gibbsite and some silicates were therefore retained in the +400 mesh fraction. Magnetic Separation: A magnet was used to separate the magnetic particles from bulk sample. The chemistry of the magnetic particles was studied using SEM-EDX. 3.4.2 Calculations of Metal Concentrations in DCB-treatment -soluble Fraction Concentrations of various metals in the DCB-treatment-soluble fraction were calculated from solution sample assays according to the following equation: Cm = CV/(W-Wr).100% (3-6) 50 Cm: concentration of metal in DCB-treatment soluble materials. V: solution volume. C: solution assay. W: initial sample weight treated by DCB. Wr: residue weight after DCB-treatment. 3.4.3 Electron Microanalysis (SEM-EDX). The scanning electron microscope (SEM) was utilized to study the occurrence of various particles found in the ores and leaching residues. In order to get high resolution, samples were coated with Au-40% Pd alloys using sputter coater, operating at conditions of 700V, 25 mA for 4 minutes. A Kevex 8000 energy dispersive X-ray spectrometer (EDX) was employed to investigate the nickel association with constitutional minerals in laterites and the chemistry of the particles existing in the ore by semi-quantitatively analyzing their compositions. For the EDX analysis, powder particles were mounted on a stand with double stick and coated with carbon. EDX- spectra were collected with the following conditions: accelerating voltage 20KV, live time 200 seconds (30% dead time), time constant 12 u, sec. 51 3.4.4 X - Ray Diffraction Analysis X-ray diffraction patterns of ore samples, leaching residues, and the concentrated samples were obtained with a Siemens D5000 X-ray diffractometer using a copper target at the test conditions of 40 kV, 30 mA. For all ore bulk samples and leaching residue bulk samples, a program with a step scan of 0.01° 20 , a counting time of 1 second per step, and a scanning range from 3 to 60° 29 was used. For concentrated samples, a program with a step scan of 0.04° 29 , a counting time of 2 second per step, and a scanning range from 3 to 60° 29 was used. The peaks were matched by the computer data files to identify the various minerals. Background subtraction and data smoothing were applied on all spectrums reported here. Because of the limitations of the X-ray diffraction, peaks of minerals occupying less than 5% of the mixture may not appear in the scan. In the present study, the minerals which can be identified by the X-ray diffraction pattern of the bulk sample are reported as major constituents of the sample. CHAPTER 4 52 RESULTS AND DISCUSSION 4.1 Head Assays and Selective Dissolution Results of Ore Samples 4.1.1 Head Assays Ouaco LD-95-75, Poum 8 and Ivory Coast blend ore samples were air dried and digested using the fusion method at IPL and three acid and aqua regia methods at the Chemex lab. Five drill samples LD-97-13, LD-97-18, LD-97-24, LD-97-28 and LD-97-32 were dried at 100°C and decomposed at IPL using fusion and triple acids method. Metal concentrations were determined by ICP. The chemical assays of the ore samples from IPL and Chemex lab, together with the assays from Falconbridge are tabulated in Table 4-1. There are marked differences between the assays reported by Falconbridge, Chemex and IPL. Results showed differences even when following the same digestion method. For all the samples digested using fusion method, the assays for metals Ni, Fe, Mn and Cr from Falconbridge were higher than that from IPL. The samples were not completely digested by the aqua regia method resulting in low metal assays. Fusion with Na 20 2 caused interferences in subsequent analysis. The presence of the Na in the sample influenced the metal assays by ICP analysis. Magnesium was particularly sensitive to the effect of sodium. 53 Table 4- 1. Head assays for laterite samples used in the present study. Sample name Digestion method (company) Ni (%) Co (%) Fe (%) Mn (%) Al (%) Mg (%) Cr (%) Si (%) Ouaco fusion(Falconbridge) fusion (IPL) 3 acid (Chemex) aqua regia (Chemex) 1.5 0.26 50 1.59 2.42 0.38 2.74 1.13 1.4 0.26 43 1.5 2.15 0.26 2.0 0.87 1.39 0.19 >30 1.32 1.85 0.20 1.19 n/a 1.33 0.23 >30 1.68 1.83 0.14 0.88 n/a Poum fusion(Falconbridge) fusion (IPL) 3 acid (Chemex) aqua regia (Chemex) 1.31 0.04 50 0.46 3.02 0.60 2.70 1.44 1.1 0.05 44 0.41 2.92 0.34 2.2 0.91 1.18 0.03 >30 0.39 2.20 0.20 0.93 n/a 1.09 0.03 >30 0.40 2.12 0.14 0.81 n/a Ivory Coast blend fusion(Falconbridge) fusion (IPL) 3 acid (Chemex) aqua regia (Chemex) 1.27 0.12 55.1 1.01 1.16 0.36 0.52 high 1.2 0.13 44 0.74 1.33 0.29 0.60 2.8 1.25 0.14 >30 0.79 1.30 0.40 0.60 n/a 1.17 0.13 >30 0.80 0.099 0.35 0.52 n/a LD-97-13 fusion (Falconbridge) fusion (IPL) 3 acid (IPL) 1.12 0.02 20.5 0.32 0.70 7.24 0.79 23 1.10 0.042 18 0.295 0.72 7.25 0.63 19 1.2 0.04 20 0.28 0.44 7.58 0.37 n/a LD-97-18 fusion (Falconbridge) fusion (IPL) 3 acid (IPL) 1.08 0.08 39.2 0.53 4.2 0.25 0.81 9 1.00 0.093 35 0.50 3.97 0.052 0.69 7.3 1.2 0.092 39 0.46 4.07 0.24 0.64 n/a LD-97-24 fusion (Falconbridge) fusion (IPL) 3 acid (IPL) 1.11 0.00 53.2 1.25 1.31 0.48 0.54 0.99 1.00 0.011 51 1.10 1.26 0.20 0.41 1.3 1.20 0.012 52 0.80 1.18 0.45 0.46 n/a LD-97-28 fusion (Falconbridge) fusion (IPL) 3 acid (IPL) 1.20 0.19 54.6 1.17 1.96 0.34 0.90 2.00 1.10 0.184 50 0.99 1.75 0.067 0.71 1.9 1.20 0.178 51 0.99 1.67 0.30 0.79 n/a LD-97-32 fusion (Falconbridge) fusion (IPL) 3 acid (IPL) 1.39 0.12 49 0.75 2.78 0.44 0.76 4.08 1.30 0.122 46 0.66 2.5 0.20 0.60 3.3 1.50 0.127 47 0.76 2.59 0.39 0.65 n/a 54 In general, laterite samples from Ivory Coast have a higher silicon content than the samples from New Caledonia whereas New Caledonia laterites have a higher aluminum content than Ivory Coast laterites except two drill samples LD-97-18 and LD-97-32. The concentrations of Ni, Co, Mn and Cr are highest in Ouaco LD-95-75 sample. Sample LD-97-13 is very high in magnesium value of about 7.5% Mg. However, the magnesium content of the rest of the samples is relatively low compared to the Moa Bay laterite which is assayed at about 0.55% Mg (Chalkley,1997). 4.1.2 Selective Dissolution Results for Ore Samples Five drill ore samples LD-97-13, LD-97-18, LD-97-24 LD-97-28 and LD-97-32 were dissolved by the dithionite-citrate-bicarbonate (DCB) method to selectively remove the iron oxides. In the laterite samples, lithiophorite is contained in a small amount and possibly dissolved during DCB treatment. The DCB results for one treatment are presented in Table 4-2. DCB treated solutions were analyzed by ICP for their metal concentrations. The compositions of the DCB dissolvable fraction calculated from solution assays according to equation (3-6) are listed in Table 4-3. The DCB treated residues were physically examined by XRD (section 4.1). It can be seen that goethite was partially dissolved by DCB in one treatment (except in LD-97-13) and that maghemite persisted in all DCB treated residues. The DCB dissolvable fraction predominantly contains iron oxides goethite and maghemite. According to the EDX study of maghemite (Table 4-5 and Table 4-8), metal substitutions are present to a lesser extent (except nickel), indicating 55 that metals in the DCB treated filtrates are mostly present in goethite. Table 4-3 indicated that metals Ni, Co, Mn, Cr and Al substituted for Fe in goethite matrix in all samples. Table 4- 2. DCB-treatment results. Sample Initial Residue DCB dissolvable weight (g) weight (g) fraction % LD-97-13 1.02 0.77 26.5 LD-97-18 1.00 0.38 62 LD-97-24 1.00 0.08 92 LD-97-28 1.02 0.23 77.5 LD-97-32 1.00 0.27 73 Table 4- 3. The metal concentrations in DCB-dissolvable fraction in the laterite ores*. Sample Ni Co Fe Mn Al Mg Cr Si % % % % % % % % LD-97-13 2.31 0.12 48.5 0.83 0.33 0.71 0.39 2.83 LD-97-18 1.4 0.13 51.9 0.73 0.81 0.15 0.51 0.15 LD-97-24 1.10 0.01 49.8 1.19 0.99 0.34 0.43 0.81 LD-97-28 1.07 0.15 45.4 0.96 0.77 0.09 0.44 0.48 LD-97-32 1.5 0.14 54.8 0.81 0.96 0.67 0.50 0.12 * calculated from filtrates assays 4.2 Mineralogical Examination Results 56 4.2.1 Mineralogy of Ouaco LD-95-75 Figure 4-1 shows the X-ray diffraction pattern for Ouaco LD-95-75 bulk sample. Figure 4-2 illustrates X-ray diffraction pattern for Ouaco LD-95-75 residue which was obtained from partial dissolution by hydrochloric acid. Diffraction peaks of minerals goethite and maghemite were identified in the pattern for Ouaco LD-95-75 bulk sample, indicating that the main minerals in Ouaco ore are goethite and maghemite. Minor phases such as spinel [Mg(Al, Fe)204], lithiophorite, quartz, and gibbsite were identified from the XRD pattern of partially dissolved residue (Figure 4-2). The results also indicated that spinel, lithiophorite, quartz and gibbsite are more resistant to the hydrochloric acid than goethite and maghemite. Needle Mn-oxide particles were observed under the SEM. SEM-EDX results indicated that nickel in Ouaco LD-95-75 was mostly incorporated into goethite. Nickel and cobalt showed an association with lithiophorite. A portion of the aluminum was present as gibbsite. Table 4-3 summaries the EDX analyses for the minor phase particles of Ouaco LD-95-75 sample. Figure 4-3 shows the secondary electron image of particles present in the Ouaco LD-95-75. Figure 4-3a represents the agglomerated goethite particles at 20,000x magnification. Goethite demonstrates a thin needle shape morphology with an approximate diameter 0.2 pm. All minor phase particles found in this sample were coarser than the goethite particles. 57 a o a> GO 6 S 00-801; in CA GJ in C/3 <a (D r—t CJ) Q CD CN • IT) ON I Q o o a 3 o <^  o CD "cL, S cd 3 O CD bp 00' 0 58 00 U-i o c o CO u <u l-l bO m Q CD CN O co •3 •-H CO &, s O CD C ' 3 -4—» X) o cu 3 12 'co I) t-i >-, • £ CD •4-* •4—* ca On c o 12 "o CS O o o t-l T3 >, m 1"-ON Q •-J o o o o cd cfci <4H C3 CD i s E g J 5 00 JL _ J 00' 0 £3 " < - f X i (U t>0 Figure 4-3. Secondary electron image of particles in Ouaco LD-95-75. (a: goethite; b: lithiophorite; c: Mn-oxide; d: spinel). 61 Table 4- 4. Minor phase compositions in Ouaco Ore. (as measured by EDX analysis) wt% NiO CoO A 1 A Si02 Cr 20 3 Mn0 2 FeO K 2 0 MgO Lithiophorite (Fig.4-3b) 10.85 15.19 9,39 64.58 4.01 11.07 15.45 8.73 59.82 3.21 9.13 12.81 10.72 64.12 Mn-oxide (Fig. 4-3c) 4.10 7.83 85.74 2.78 5.54 10.63 81.40 2.46 Spinel (Fig. 4-3d) 34.74 5.53 29.68 2.95 7.26 19.84 36.92 2.82 33.77 7.63 18.86 33.29 5.45 30.25 5.99 6.94 18.08 4.2.2 Mineralogy of Poum 8 Figures 4-4 and 4-5 illustrate the X-ray diffraction patterns for Poum 8 ore bulk sample and +50 mesh fraction of Poum ore respectively. Only goethite was detected in the bulk sample XRD pattern, indicating that the major mineral constituent of Poum 8 was goethite. Spinel, lithiophorite and gibbsite were identified in the +50 mesh fraction of Poum ore by XRD and SEM-EDX. The Mn-Al oxides having a similar composition to lithiophorite (Mitchell, 1967, Das, 1995) were found in the +50 mesh fraction by SEM-EDX. Al-Mn oxide particles with relatively a high aluminum content and a low nickel content were also present in the +50 mesh fraction as determined by SEM-EDX. There is a possibility that the high Al content Mn-oxide is a (Mn, Al)-spinel known as galaxite (MnAl204). Ni and Co are probably incorporated into galaxite. According to Blesa (Blesa, 1994), Co and Ni was found to be in spinel structures, but 63 3 o PH U-I O c H—» o CvJ <H-H J=! on <D 0> h H Q CD CN O + t H OH c o oo •c s ° u •= •5 «> • - E i f Its • a o S o. a CO ,4: J <5r "•H O ca , h n <H-I C3 h H I 1) h H 8 0 " b b Z 00' 0 64 galaxite was not confirmed to be in the laterites. The +50 mesh spinel had the same composition as found in the Ouaco LD-95-75 sample. The particles containing high Cr and Fe and low Mg was also found in +50 mesh fraction of Poum ore. These particles probably also belong to the spinel group. It is difficult to distinguish spinels by XRD since they have similar XRD patterns. As seen in Ouaco LD-97-75 sample, Al was found as gibbsite in Poum ore. Table 4-5 shows EDX results for the high Cr-Fe spinel, lithiophorite, and (Mn, Al)-oxide particles in +50 mesh fraction of Poum ore. Figure 4-6 shows the gibbsite EDX spectrum in Poum ore. Figure 4-7 shows the goethite particles in Poum ore sample. Table 4- 5. Compositions of particles in +50 mesh fraction of Poum ore. (as measured by EDX analysis) wt% NiO CoO A1 20 3 S0 3 C r 2 0 3 M n 0 2 FeO MgO Spinel 8.67 61.77 22.42 7.14 14.94 57.00 18.75 9.32 10.92 58.19 18.49 12.40 Lithiophorite 17.17 11.95 18.27 52.60 17.42 11.47 20.11 51.00 18.82 10.82 19.07 51.29 Mn-Al oxide (Galaxite ?) 3.70 6.49 47.78 36.53 2.61 11.91 45.59 39.89 11.06 46.12 37.79 5.03 4.2.3 Mineralogy of Ivory Coast Blend Figure 4-8 and Figure 4-9 show the XRD patterns for the Ivory Coast ore bulk sample and +400 mesh fraction sample respectively. Ivory Coast ore mainly consisted of goethite, maghemite and 65 ft! 0.000 Range- 10.230 keV 9.63 Integral 0 2E Figure 4-6. EDX spectrum of gibbsite in Poum ore sample. Figure 4- 7. Secondary electron image of goethite particles in Poum ore. 66 CO CD CD h H CO CD Q CD CN T 3 CD oo cd O u o > I—H <H-I o c _o *H—» o cc! CO CD O O + h H •a CD s OH C %-> O CCj h H >< ON CD h H bp 00' 0 68 quartz. Lithiophorite and clinochlore were identified by XRD in +400 mesh fraction. Nickel showed an association with maghemite, clinochlore, and Al-Mn oxide. Cobalt showed an association with Al-Mn oxide. In the Al-Mn oxide, Ni and Co compositions showed significant variation as shown in Figure 4-10. The NiO content increased as the A1203 content decreased. The precise nature of the Ni and Co containing Al-Mn oxide particles is not known from present study. The composition obtained by EDX showed significant difference from the composition of lithiophorite reported in the literature and there were no extra peaks present in the XRD pattern for those particles. There are two possibilities producing these results: (1) The particles contribute X-ray counts to the detection of the lithiophorite structure, and therefore the particles are a lithiophorite phase. However, the composition data reported in the literature is not enough for the lithiophorite identification; (2) the particles are amorphous and hence can not be identified by XRD. Figure 4-11 shows a secondary electron image of goethite particles in Ivory Coast blend. The photomicrograph of well crystallized goethite showed very fine needles with approximately 60 nm diameter and 400 nm length. 4.2.3 Mineralogy of LD-97-13 Figures 4-12 and 4-13 are the XRD patterns for the LD-97-13 bulk sample and the DCB treated sample. Goethite, talc, clinochlore, and quartz were identified by bulk sample XRD. Goethite peaks were not found in the DCB treated residue diffraction pattern, suggesting the complete A1 2 0 3 % gure 4- 10. Relationship between A1 2 0 3 and NiO content in A l - M n oxide in Ivory Coast blend sample. ure 4-11. Secondary electron image of goethite particles in Ivory Coast blend. 70 72 dissolution of goethite. Of 1.02 g of ore sample dissolved by DCB, 0.77 g reported to residue fraction. This indicated that goethite accounts for less than 23% of the ore and silicates such as talc, clinochlore, and quartz were present in significant amounts in LD-97-13. Two typical clinochlore EDX spectra were obtained indicating that there may be two distinct clinochlores present in LD-97-13. However, it is difficult to be distinguished by XRD since they have quite similar diffraction patterns. XRD peaks of clinochlore in LD-97-13 were broad suggesting that both clinochlores contribute to the peaks. The Al-Mg clinochlore particles in the +140 mesh fraction appeared silver colored to the eye, but had various shades of light green under the optical microscope. The head assay for Mg in LD-97-13 was about 7%, while the Mg content in Fe-oxide fraction was only 0.71% as listed in Table 4-3. This indicated that a significant portion of Mg was present as silicates such as clinochlore and talc. Talc shows very little variation in chemical composition. A small amount of iron and a minor amount of nickel (NiO 0.2-0.7%) in some of the talc particles were detected by EDX. Nickel showed an association with clinochlore and magnetic particles. The nickel content detected in magnetic particles ranged from 0 (below detection) up to around 9%. The magnetic fraction exhibited very fine particle size and some had a high silica content. The high Si content may have resulted from the agglomeration of maghemite and some other silicates, which was observed under SEM. Nickel content in Fe-Mg chlorite varied from 0 (below detection) to around 4%. A trace amount of Ni were detected in Al-Mg clinochlore. Typical compositions of particles existing in LD-97-13 were summarized in Table 4-6. 73 Table 4- 6. Particle compositions in LD-97-13. (as measured by EDX analysis) wt% NiO A I A Si02 Cr 20 3 MnO Fe 20 3 MgO Talc 69.35 1.59 29.06 Mg3Si4O l 0(OH)2 67.45 1.92 30.67 68.09 1.16 30.75 Magnetic 3.25 19.39 69.51 7.84 particles 4.45 20.71 2.32 72.53 4.24 2.21 4.65 88.9 Clinochlore 2.17 7.18 38.93 42.63 9.09 (Mg, Fe, Al)6(Si,Al)4O10(OH)8 2.55 37.41 37.47 22.68 2.27 38.03 47.76 11.08 Clinochlore 0.31 28.98 37.25 2.38 2.23 28.85 (Mg, Al)6(Si, Al) 4O1 0(OH)8 0.39 28.82 37.02 2.52 2.58 28.67 0.21 27.97 38.29 1.99 1.88 29.65 4.2.4 Mineralogy of LD-97-18 Figure 4-14 and Figure 4-15 illustrate the X-ray diffraction patterns for the LD-97-18 bulk sample and DCB treated sample. Goethite, quartz, kaolinite, and maghemite were identified in both bulk sample and DCB treated sample by XRD. In +400 mesh fraction Ni and Co containing manganese oxide particles were found by SEM-EDX. Typical compositions of kaolinite particles and manganese oxide are summarized in Table 4-4 and Table 4-8. Fe and Na 5 ^ o a a . S I E 0) N j - . O ii * g> ea o 3 5 a a 2 id d a 2 hfi) 74 00'Z0b S p u r i a ; } 00 ' 0 75 76 were detected in kaolinite particles. The Al concentration in the LD-97-18 ore sample was approximate 4%. According to Table 4-3, Al content of Fe oxide in LD-97-18 was 0.81%. These results suggest that most of the Al in LD-97-18 is contained in kaolinite. Table 4- 7. Compositions of kaolinite in LD-97-18. (as measured by EDX analysis) wt% A1203 Si02 Cr 20 3 Fe 20 3 Na20 Kaolinite Al2Si205(OH)4 35.28 41.22 13.30 10.20 36.02 41.44 1.61 8.43 9.32 35.95 40.8 11.43 11.82 Table 4- 8. Compositions of Mn-oxide in LD-97-18. (as measured by EDX analysis) wt% NiO CoO Mn0 2 A1203 Fe203 MgO Ti0 2 Si02 BaO Mn-oxide 10.6 7.68 41.99 27.15 13.89 4.86 7.68 11.20 4.93 23.08 20.06 15.38 11.46 5.30 3.52 47.09 16.14 10.84 4.71 12.41 4.2.5 Mineralogy of LD-97-24, LD-97-28 and LD-97-32 Figure 4-16 and Figure 4-17 are the X-ray diffraction patterns for the LD-97-24 bulk sample and the DCB treated sample. Only goethite peaks were detected in the XRD pattern of bulk sample. 77 00*0Zi S^UHOQ 00" 0 78 00" LSll s:j.unoo 00'0 79 Hematite was identified by XRD in the DCB treated sample. As illustrated in Table 4-2, 92% of ore sample was selectively dissolved during one DCB treatment, indicating that LD-97-24 consists of relatively pure goethite and that very small amounts of minor minerals exist in LD-97-24. The goethite peaks after DCB treatment had very high counts, which was probably due to the goethite particles left in the DCB treated residue having higher crystallinity than goethite in the bulk sample. Of 30 g of ore sample that was wet screened with a 400 mesh sieve, only 0.25 g reported to the +400 mesh fraction, indicating that the particles in the ore were very fine. Figure 4-18 and figure 4-19 are the X-ray diffraction patterns for the LD-97-28 bulk sample and the DCB treated sample. Goethite, maghemite and clinochlore were found by the X-ray diffraction pattern of the LD-97-28 bulk sample. Minor phases such as talc and kaolinite were identified by the X-ray diffraction pattern of the DCB treated sample. Al-Mn oxide particles were found in the +400 mesh fraction by SEM-EDX. EDX analysis showed that talc had similar composition to that found in LD-97-13 sample (Table 4-6). Kaolinite had similar composition as that found in LD-97-18 (Table 4-7). Major components of the clinochlore in LD-97-28 were Al, Mg and Si. It had similar compositions and the same appearance as the Al-Mg clinochlore found in LD-97-13 (Table 4-6). Nickel showed an association with maghemite, Al-Mn oxide and some of the clinochlore in trace amounts (NiO 0.3% in average). Typical compositions of the Ni-containing Al-Mn oxide and magnetic particles in LD-97-28 are listed in Tables 4-9. Figure 4-20 and Figure 4-21 display X-ray diffraction patterns for LD-97-32 sample. XRD and SEM studies showed that the LD-97-32 sample mainly consisted of goethite, quartz, maghemite, 80 . CD i co 3 oo C N i r-0 \ l Q ,o 0) . O H o .2 -*—» o cd oo CD =3 00*083 00' 0 81 OO <D <D 00 <u Q CD ccj co OO CN i C--O N I Q i-J -d u ta « U Q O H a o %-» o t-i X ON 3 00 00' 0 ! 82 CO CD CD i-i b O CD C D C N CL, CCJ 3 CN • Q c2 6 CD 1 a _o %-» o cd o CN CD b p 00•see 00' 0 83 o ca ca H « E (U N ca 3 o C N c n ON I Q ts CD t-i • H CQ U Q >~ CD s a, c o o e g (-1 >< C N i CD 00' 0 84 and talc and to a lesser extent kaolinite. Ni and Co containing Mn-oxide particles and chromite were observed under the SEM. SEM-EDX results showed that the magnetic particles and talc had similar compositions as found in the LD-97-13 sample, kaolinite had similar composition to that detected in the LD-97-28 sample. Table 4- 9. Compositions of Al-Mn oxide and magnetic particles in LD-97-28 (as measured by SEM-EDX) wt% NiO CoO A 1 A Si02 Mn0 2 Fe 20 3 Al-Mn oxide 25.11 6.70 11.95 44.33 11.91 22.83 6.09 10.86 49.39 10.83 1.92 5.75 40.25 12.02 19.30 20.76 Magnetic particles 3.76 96.24 4.47 95.53 4.16 95.84 4.2.6 Summary of Mineralogical Examinations Table 4-10 summarized the mineralogical examination results for Ouaco LD-95-75, poum 8, Ivory Coast blend, LD-97-13, LD-97-18, LD-97-24, LD-97-28 and LD-97-32. The minerals identified by XRD in bulk sample, in concentrated sample and the nickel bearing minerals found by EDX analysis are listed in this table. Results have shown that there are significant mineralogical variations among the laterite samples. Goethite, maghemite and quartz were common minerals in all laterite samples. Al found as gibbsite was present in the two New 85 Table 4- 10. Summary for mineralogical examination results Sample Minerals Ouaco LD-95-75 Bulk sample Goethite and maghemite. Concentrated sample Spinel, lithiophorite, quartz and Mn-oxide. Ni bearing mineral Goethite, maghemite and lithiophorite Poum 8 Bulk sample Goethite. + 50 mesh fraction Maghemite, spinel, lithiophorite, quartz and gibbsite. Ni bearing mineral Goethite, maghemite and lithiophorite. Ivory Coast blend Bulk sample Goethite, maghemite and quartz. +400 mesh fraction Goethite, clinochlore, lithiophorite and quartz. Ni bearing mineral Goethite, clinochlore, maghemite and lithiophorite. LD-97-13 Bulk sample Goethite, maghemite, quartz, clinochlore and talc. DCB-treated sample Maghemite, quartz, clinochlore and talc. Ni bearing mineral Goethite, maghemite, clinochlore and talc. LD-97-18 Bulk sample Goethite, maghemite, quartz and kaolinite. DCB-treated sample Goethite, maghemite, quartz and kaolinite. Ni bearing mineral Goethite, maghemite and Al-Fe-Mn-oxide (+400 mesh fraction). LD-97-24 Bulk sample Goethite. DCB-treated sample Goethite and hematite. Ni bearing mineral Goethite. LD-97-28 Bulk sample Goethite, maghemite, clinochlore and talc. DCB-treated sample Goethite, maghemite, clinochlore and talc. Ni bearing mineral Goethite, maghemite, clinochlore, and Al-Fe-Mn oxide (+400 mesh). LD-97-32 Bulk sample Goethite, maghemite, quartz, talc and kaolinite. DCB-treated sample Goethite, maghemite, quartz, talc and kaolinite. Ni bearing mineral Goethite, maghemite, talc and Al-Fe-Mn-oxide (+400 mesh fraction). 86 Caledonia laterite samples Ouaco LD-95-75 and Poum 8. Silicates such as talc, kaolinite and clinochlore were found in Ivory Coast laterites and the silicate minerals existing in the Ivory Coast laterites vary among the samples. Goethite in LD-97-13 was less than 23%. A significant portion of Mg was found as talc and clinochlore in LD-97-13. In LD-97-18, a large fraction of the Al was present as kaolinite. 92% of the LD-97-24 sample was goethite. The LD-97-28 sample contained major minerals goethite, clinochlore and maghemite. Talc and quartz were present as minor phases. LD-97-32 was composed of goethite, talc, goethite, quartz and minor phase kaolinite. SEM-EDX results indicated that the major mineral host for nickel in these laterite samples was goethite. Nickel was also associated with maghemite, lithiophorite and showed an association with various silicates to different extents in Ivory Coast laterites. High nickel and cobalt containing Al-Fe-Mn-oxide particles were normally found in +400 mesh fraction of Ivory Coast laterite samples by SEM-EDX. 4.3 Pressure Acid Leaching Experimental Results 4.3.1 Interruptive Leaching of Ivory Coast Blend Four interruptive autoclave leachings were performed on the Ivory Coast blend sample with leaching leach times of 5, 10, 15 and 30 minutes. Leaching experiments were performed at 260°C, at a pulp density 30% solids and an acid to ore ratio of 0.18. Metal concentrations in the 87 leaching liquor and residues are summarized as mass balance in appendix III. Metal extractions were calculated and are tabulated in Table 4-11. Mg extraction was not calculated since its content in the residue was lower than the limitation (500ppm) of ICP detection, which suggested a high magnesium extraction. Table 4-11. Metal extractions for pressure acid leaching experiments of the Ivory Coast blend. (T = 260°C, A/O = 0.18, pulp density = 30% solids) Time Ni Co Fe Mn Al Cr Si Total S Free acid in min. % % % % % % % % filtrates (g/1) 5 78 88 0.76 84 41 2.46 2.56 1.18 25.33 10 85 89 0.76 87 36 1.78 2.50 1.15 24.73 15 97 97 0.61 94 33 0.83 2.46 1.06 20.5 30 97 96 0.77 94 23 0.70 2.05 1.17 26.88 As listed in Table 4-11, the extractions of Ni, Co and Mn increased with leach time. 78% Ni and 88%) Co were leached out during the first 5 minutes of leaching. Extractions of the Ni and Co reached 97% after only 15 minutes of leaching. After 15 minutes, no further metal extractions occurred. The extraction of Al decreased with leach time and the decrease was significant during the leach period of 15 - 30 minutes. The extraction of Cr also decreased with leaching time and had very low value. The iron extraction was very low only of about 0.7%. Figure 4-22 demonstrates the acid consumption with leaching time. The acid consumption shows a maximum peak after 15 minutes of leaching, which probably was due to the dissolution 88 Figure 4- 22. The acid consumption during interruptive pressure acid leaching. of maghemite. It can be assumed that maghemite dissolved slowly during the first 10 minute leaching period. However, the dissolution rate increased after 10 minutes of leaching, therefore increases the acid consumption. As listed in Table 4-11, cobalt extraction was higher than nickel extraction after 10 minutes of the leaching, which also illustrates the effect of the slower dissolution rate of the maghemite than goethite. In Ivory Coast blend sample, Co showed an association only with major mineral goethite, while Ni showed an association with two major constituents goethite and maghemite. Obviously the undissolved maghemite kept nickel in the leach residue and resulted in lower nickel extraction. The fact that the initial rate of cobalt extraction was always higher than that of nickel was also reported by Chou (Chou, 1977). Unfortunately, there is no quantitative results for the mineralogical concentration to confirm the statement. The extraction of iron (or the total iron left in the leaching solution) was lowest after 15 minutes of leaching may be due to the maximum acid consumption which produced a lowest 89 free acid concentration in the leaching liquor. This result suggested that iron left in the leach liquor were strongly dependent on the terminal acidity. The maximum acid consumption also resulted in lowest sulfur in the leach residue. The acid consumption decreased after 15 minutes of leaching was considered to be the precipitation of hematite and alunite. Figure 4-23 shows XRD patterns for leaching residue bulk samples of Ivory Coast blend after interruptive autoclave leaching. Figure 4-24 shows XRD patterns for DCB treated leaching residue samples. Two figures demonstrate a qualitative mineralogical change during pressure acid leaching of Ivory Coast blend with leaching time. Heating of the slurry up to 260°C seemed to have no effect on the mineralogy. XRD analysis of solids taken before acid injection showed a similar pattern to that of the unheated laterite. Goethite peaks appeared in the XRD spectrum for the bulk sample residues after 5 and 10 minutes of leaching. However, the peaks for goethite were not identified in the diffraction pattern for the bulk sample residue after 15 minutes of leaching, indicating that goethite fully dissolved within 15 minutes. Diffraction peaks for maghemite also disappeared from the XRD spectrum for the bulk sample residue after 15 minutes of leaching and few magnetic particles were collected with a magnet in this residue. This suggested the complete dissolution of maghemite after 15 minutes of leaching. Figure 4-24 (c) showed that the alunite was a major constituent in the DCB treated leaching residue after 30 minutes of leaching. The XRD results and Al extraction data accordingly suggested that the amounts of the alunite precipitates increased rapidly after 15 minutes of leaching. Minor phase clinochlore persists in the leach residues as indicated by the XRD patterns of DCB treated residues. 9 0 91 92 The results for Ni extraction and XRD results for the leach residues demonstrated the effect of goethite and maghemite dissolution on the Ni extraction. As goethite and maghemite were fully dissolved during leaching, Ni extraction was near complete, indicating that the nickel extraction was governed by goethite and maghemite dissolution. This conclusion also reflected that over 97% of the Ni in the Ivory Coast blend sample was contained in the goethite and maghemite. To investigate the effect of acid addition on the leaching kinetics, 30% pulp density of Ivory Coast blend slurry was also leached at 260°C and acid/ore ratio of 0.21 for 5 minutes. The metal extractions were: 89% Ni, 92% Co, 0.86% Fe, 90% Mn, 50% Al and 2.65% Cr and the total sulfur in the residue was 1.38%. Metal extractions were significantly influenced by the increase of the acid addition. 5 minutes of leaching at acid/ore ratio of 0.21 gave higher Ni and Co extractions than 10 minutes of leaching at acid/ore ratio 0.18. The change of acid/ore ratio from 0.18 up to 0.21 also left higher Fe and Al contents in the pregnant liquor and higher sulfur in the leach residue. SEM was employed to characterize the residues produced during the leaching. Figure 4-25 displays the evolution of the residue during the leaching process. As shown, the residue solids after 5 minutes of leaching composed of goethite needles and very fine hematite particles in different sizes. After 15 minutes of leaching, hematite particles size became uniform. Figure 4-25 also illustrates the growth of hematite particle size with the leaching time. (b) Figure 4-25. The secondary electron images of leaching residues produced after pressure acid leaching of Ivory Coast blend with different leaching time. (a):5 minutes, (b): 10 minutes; ( c ): 15 minutes; (d): 30 minutes. 95 4.3.2 Leaching of LD-97-13 Two autoclave leaching experiments were carried out on sample LD-97-13 for 30 and 60 minutes at temperature 250°C, an acid/ore ratio of 0.38 and 17% pulp density. Metal extractions after leaching are tabulated in Table 4- 12. Table 4- 12. Metal extractions after pressure acid leaching of LD-97-13 (T = 250°C, acid/ore ratio = 0.38, pulp density = 17% solids). Time Ni Co Fe Mn Al Cr Mg Si Total S Free acid in min. % % % % % % % % % filtrates (g/1) 30 94 93 6.1 97 61 3.8 32 0.83 0.37 40.05 60 92 89 5.7 98 53 5.14 34 0.80 0.45 42.58 Experimental results show that reaction time has an insignificant effect on the metal extractions from LD-97-13 (except Al extraction which decreased with leaching time). Both leaching experiments gave lower than 95% of Ni and Co extractions. The extraction of Mg and the total sulfur in the residue were extremely low, whereas the extractions of Fe and Al arid free acid concentration were relatively high. The high iron extraction was probably caused by the high terminal acidity in the leaching solution and low slurry pulp density. The effect of the acidity and pulp density on the iron extraction was discussed in section 4.4. The acid consumptions were about 187.4kg/t after 30 minutes of leaching and 197.6kg/t after 60 minutes of leaching, indicating'that the dissolution reactions continued after 30 minutes of leaching and after the full dissolution of goethite. 96 Figure 4-26 shows the XRD spectrums for the bulk residues of LD-97-13 after 30 and 60 minutes of leaching, and Figure 4-27 for the DCB-treated residues. Goethite and maghemite peaks were not found in the spectrums for residue of 30 minutes of leach, indicating goethite and maghemite completely dissolved after 30 minutes of leaching. Clinochlore peaks clearly appeared in the XRD patterns for residues after 30 minutes of leaching. Those peaks were not found in the XRD patterns for residue after 60 minutes of leaching, suggesting that clinochlore was dissolved during leaching with slower dissolution rate than that of the goethite and maghemite at present leaching conditions. Talc peaks were found in all of the XRD patterns for the residues, showing its resistance to the pressure acid leaching. As described in section 4.2.3, most of the magnesium was present as talc in the LD-97-13. undissolved talc kept Mg in the residue, which resulted in low Mg extraction and high terminal acidity in the leaching solution. Small amounts of nickel were incorporated into talc matrix, therefore, the undissolved talc also resulted in low nickel extraction. Alunite peaks were not found in XRD spectrums was probably due to its small amounts and relatively low counts. However, the decrease of the Al extraction and the increase of the sulfur in the leaching residue with time indicated the precipitation of aluminum sulfate. 97 98 99 4.3.3 Leaching of LD-97-24 and LD-97-28 Samples LD-97-24 and LD-97-28 were leached at 250°C for 30 and 60 minutes. The other leaching parameters and metal extractions are listed in Table 4-13. It is obvious that leaching time has a positive influence on Ni and Co extraction and that 97% Ni extraction can be attained within 1 hour of leaching at present leach conditions for both samples. During the leaching of both samples, iron extraction was very low and decreased with time. Al extraction also decreased with time and sulfur content in the residue increased with time. Mn extraction was relatively low, which was probably due to the co-precipitation of the Mn with hematite as illustrated by the selective dissolution results (see section 4.2.5) or undissolved lithiophorite. Cr extraction obtained from the leach of these two samples was relatively high compared to that from the leaching of the other samples. Table 4-13. Leaching results for LD-97-24 and LD-97-28. Conditions Metal Extractions Sample Time A/O Slurry* Ni Co Fe Mn Al Cr Si Total S Filtrate free Min. ratio Wt% % % % % % % % % acid (g/1) LD-97-24 30 0.18 39 81 81 0.41 32 50 42 10 1.05 27.15 60 0.18 39 97 86 0.27 31 45 45 7.28 1.24 26.79 LD-97-28 30 0.24 24 93 90 0.47 60 54 29 6.66 1.14 29.27 60 • 0.24 24 97 96 0.31 65 50 28 6.64 1.25 29.17 * Slurry = pulp density. Figure 4-28 illustrates the XRD patterns for leach residue of LD-97-24. Goethite peaks was found in the XRD spectrum for the residue after 1 hour of leaching. Figures 4-29 shows the X-100 101 ray diffraction patterns for the LD-97-28 bulk sample residues after 30 and 60 minutes of leaching, and Figure 4-30 for the DCB-treated residue samples. Goethite, maghemite and clinochlore, the major components of this ore, were present in the residue after 30 minutes of leaching as illustrated by XRD results. However, goethite and maghemite were not identified in the XRD spectrums for the residue after 1 hour of leaching. Clinochlore persisted in the residue after 1 hour of leaching. As demonstrated in the leaching results of the LD-97-13, clinochlore dissolved after 60 minutes of the leaching of LD-97-13. The different dissolution behavior of clinochlore in the leaching of LD-97-13 and LD-97-28 was probably due to the different terminal acidity in the leaching solution. Alunite was found in the DCB treated leach residues obtained from both 30 and 60 minutes of leaching of LD-97-28. 4.3.4 Leaching of LD-97-18 and LD-97-32 Autoclave leaching experiments were conducted on samples LD-97-18 and LD-97-32 at 250°C for 30 minutes. Table 4-14 summarized leaching conditions and leaching results for LD-97-18 and LD-97-32. Table 4- 14. Leaching conditions and results for LD-97-18 and LD-97-32 (Temp.= 250°C, time = 30 minutes). Sample A/O Slurry Ni Co Fe Mn Al Cr Si Total S Filtrate free ratio Wt% % % % % % % % % acid (g/1) LD-97-18 0.39 17 98 98 2.32 94 56 18 3.0 1.42 32.25 LD-97-32 0.30 20 98 98 0.80 83 60 34 4.2 1.08 32.65 102 103 00 ' f rSbE s^unoo 0 0 - 0 104 The results showed that at 250°C, an acid to ore ratio of 0.39 for LD-97-18 and 0.30 for LD-97-32 enabled 98% nickel and cobalt extraction within 30 minutes. Fe extraction from the leaching of LD-97-18 was higher than that from LD-97-32, which was probably due to the pulp density differences as subsequent discussion (section 4-4). Figure 4- 31 shows the XRD pattern for leach residue bulk sample of LD-97-18 and Figure 4- 32 for the DCB-treated residue. The leaching residue of LD-97-18 predominantly consisted of hematite and quartz. Talc, kaolinite and alunite were found in the DCB treated leaching residue as illustrated in Figure 4-32. The leaching residue of LD-97-32 mainly consisted of hematite, talc, kaolinite (Figure 4-33). Quartz and alunite was identified in the DCB-treated leaching residue of LD-97-32 (Figure 4-34). Goethite and maghemite peaks were not identified in the XRD spectrums for residues obtained from 30 minutes of leaching of LD-97-18 and LD-97-32, suggesting that goethite and maghemite completely dissolved during 30 minute leaching of both samples. 4.3.5 Selective Dissolution Results of Leach Residues The leach residues were also treated with DCB technique to remove iron oxide and therefore to concentrate the minor phases in the leach residue. The DCB results for one treatment and the compositions of the DCB dissolvable fraction are listed in Table 4-15. For the leaching residues in which goethite and maghemite were not present, the DCB dissolvable fraction mostly consisted of hematite. As seen from Table 4-15, nickel and cobalt contents in DCB-soluble 105 co CD a CD Q C D C N OO i O N Q O J D =3 X> CD rs <u i-< cc g o C3 CD •s a _o %-> o ccj CO c n i CD 2? 00'EBfrt 00' 0 106 co CD U i-i 00 CD Q C D C N ON Q i—l O CD 3 ."2 'co CD l -H 00 c IS o CS J D T3 CD •a CD +-» CQ O Q s-i CD cd &, C .2 O cd cd C N CD l-i 3 00 00*80BS 00' 0 108 a C3 a 1/3 "fN • g o 3 «• m O E ^ E a I ..J in (3D m r^ -O N I Q o CD rs 0) u, 00 a 13 o ca J D ~o CD ta CD CQ U Q hH ca P H a o O ca hH r o i <u hH 00 00' 0 109 material (mostly hematite) are very low indicating that very small amounts of nickel and cobalt co-precipitated with iron into hematite, which is consistence with the research of Sobol (Sobol, 1971). The content of Al, Cr, and Si in the DCB filtrates was relatively high. Table 4-15. The compositions of DCB-dissolvable fraction in the leach residues*. Sample initial residue Ni Co Fe Mn Al Mg Cr Si name weight g weight g % % % % % % % % LD-97-13 1.04 0.69 0.065 0.001 55.30 0.01 0.06 0.010 0.68 2.56 LD-97-18 1.08 0.36 0.048 0.002 57.03 0.06 0.19 0 0.65 0.33 LD-97-28 2.00 0.23 0.07 0.01 64.95 0.46 0.30 0 0.65 0.32 LD-97-32 1.00 0.18 0.033 0.005 60.18 0.23 0.15 0.004 0.46 0.36 Calculated from DCB-treatment solution assays 4.4 Morphology of the Hematite The characteristics of hematite particles produced after 1 hour of pressure acid leaching of Ouaco, Poum and Ivory Coast blend were studied under SEM. Table 4-16 lists the leaching conditions for generating the residues. Figure 4-33 shows the morphology of the hematite particles. Table 4- 16. Sulfuric acid leaching conditions generating residues. Ore type Temperature °C Pulp density (W/W) Acid/ore (W/W) Leach time (min) Ouaco 240 30 0.24 60 Poum 240 & 260 30 0.30 60 Ivory Coast 240 29 0.18 60 1 1 0 I l l (c) (d) Figure 4-35. The secondary electron image of the hematite particles obtained from pressure acid leaching of various laterite samples, (a): Ivory Coast blend; (b) Ouaco LD-95-75; ( c ):Poum 8 at 240°C; (d): Poum 8 at 260°C. 112 The hematite particles are almost spherical and the size of the particles obtained from the leaching of three samples showed slight differences. The biggest particles were formed from the leaching of Ouaco ore, the smallest of Ivory Coast ore. This may result from the size differences of the goethite particles in the ore and the nucleation of hematite on goethite surface. Goethite needles in the Ivory Coast blend have finest size among those three samples (see section 4.1). The hematite nucleates on the surface of the goethite. The finer the goethite particles, the more the nucleation site, the smaller the hematite particle size. Temperature showed insignificant effect on the hematite particle size formed from leaching Poum ore sample. 4.5 Discussion of Results 4.4.1 Metal Extractions During Pressure Acid Leaching Autoclave leaching results showed that over 95% of nickel and cobalt extractions were achieved from leaching the Ivory Coast laterite samples (except for LD-97-13) at temperatures of 250 ~260°C, with acid addition and slurry pulp density calculated from the Falconbridge spreadsheet (Appendix II) and sufficiently long leaching time provided. Ni and Co extraction kinetics showed a strong dependence on the dissolution rate of the goethite and maghemite. The fast goethite and maghemite dissolution rate provided rapid Ni and Co extraction kinetics. When goethite and maghemite in the Ivory Coast blend sample were fully dissolved within 15 minutes as illustrated by XRD results, 97% Ni extraction was attained after only 15 minutes of leaching. However, more than 30 minutes was need to achieve over 95% of nickel extraction from leaching samples LD-97-28 and LD-97-24 since the dissolution of goethite and maghemite was 113 not complete after 30 minutes of leaching. It is obvious that the factors involved in increasing the goethite and maghemite dissolution rate also increase nickel extraction kinetics. The results also indicate that most of nickel in the laterite ore samples are associated with goethite and maghemite. Relatively low Ni extractions were obtained from leaching LD-97-13 as a result of the undissolved talc (a major portion of the sample). Present results also show that Fe extraction (or total iron left in the solution) was affected by three factors: terminal acidity, leach time and pulp density. Table 4-17 summarizes the Fe extractions obtained from leaching five drill samples at 250°C. It clearly illustrates the effect of these three factors on the Fe extraction. Table 4- 17. Comparison of iron extraction obtained from the leaching of five drill samples. Sample Leach time (min) Pulp density (w/w) % Free acid in filtrate (g/1) Fe extraction % LD-97-24 30 39 27.15 0.41 60 39 26.19 0.27 LD-97-28 30 24 29.27 0.47 60 24 29.17 0.31 LD-97-32 30 20 32.65 0.74 LD-97-18 30 17 32.25 2.3 LD-97-13 30 17 40.05 6.1 60 17 42.58 5.7 114 As seen in Table 4-17, lower terminal acidity, longer leaching time and denser pulp slurry result in lower Fe extractions (less amount of iron in the solution). Higher acidity contributes to a higher iron solubility in solution and therefore a higher Fe extraction during leaching. Longer leaching times allow further iron precipitation and therefore resulted in a lower iron concentration in the leaching solution. The higher Fe concentrations in the leaching solutions obtained from leaching at lower pulp density were probably caused by the hematite nucleation on crystal fragments. The low density slurry provided less opportunities for nucleation and therefore a lower hematite precipitation rate and higher Fe concentration in the leaching solution. As described in section 4.4, finer mineral particles produced finer hematite particles. This supports the conclusion that hematite nucleates on the solid surface. The heterogeneous nucleation of hematite during pressure acid leaching was also described by Papangelakis (Papangelakis, 1996) and Chou (Chou, 1977). Al extractions were in the range of 50-60% when five drill samples were leached at 250°C and showed dependence on the free acid concentration and leaching time. Higher terminal acidity provided a higher Al solubility and therefore higher Al extractions during leaching. The fact that Al extraction decreased with leach time (Table 4-12 and Table 4-13) was probably due to the increasing amount of the alunite precipitates. The Al extraction was only about 23% when leaching experiment was carried out at 260°C for 30 minutes (Table 4-11), reflecting the effect of temperature on the reprecipitation of Al. 115 The extraction of Mn varied significantly from 30% in the leach of LD-97-24 (Table 4-13) to 98%o in the leach of LD-97-13 (Table 4-14) and also showed dependence on the terminal acidity. The results showed that lower Mn extraction from leaching samples LD-97-28 and LD-97-32 corresponded to a higher Mn content in the DCB dissolvable fraction of their leaching residues. This probably resulted from the presence of lithiophorite in the leaching residue and/or manganese co-precipitate with iron into hematite during leaching. The extraction of Cr obtained from leaching the five drill samples changed from about 5% to about 45%. In general, the high Mn extractions after leaching resulted in lower Cr extraction, which was probably due to the oxidation of Cr(III) by Mn(IV) according to the chemical reaction (2-12) and metal chromate precipitation from the leaching solution. The existence of the metal chromate in the leaching residue was reported by Sobol (Sobol, 1971). However, it was not identified in the present leaching residue study and therefore we can not confirm this conclusion. As illustrated in Table 4-15, the Cr content in the DCB dissolvable fraction of residues obtained from leaching four drill samples is very high, indicating that Cr Co-precipitated with Fe as hematite during leaching. The content of Mg in the leaching residue was lower than detection limit (500ppm) (Except in the leach residue of the LD-97-13 sample). According to the estimation from the head assays, the extraction of the magnesium was 100%). The low Mg extraction from leaching LD-97-13 was due to the undissolved talc. 4.4.2 Dissolution of Various Minerals during Pressure Acid Leaching Goethite is a major Ni containing mineral in all laterite samples used in this study. Its dissolution behavior significantly affected the nickel extraction. Experimental results showed 116 that the goethite dissolution rate mainly depended on the temperature and free acid concentration in the leaching solution. Higher temperature and higher terminal acidity provided faster goehtite dissolution rate as illustrated in Table 4-18. Maghemite demonstrated a similar dissolution behavior to that of goethite. Table 4- 18. Comparison of the goethite dissolution kinetics at various acid concentrations. Sample Temperature °C Free acid (g/1) Time* Ivory Coast blend 260 20.5 within 15 minutes LD-97-24 250 26.19 more than 1 hour LD-97-28 250 29.17 within 1 hour LD-97-32 250 32.65 within 30 minutes LD-97-18 250 32.25 within 30 minutes LD-97-13 250 42.58 within 30 minutes * Leaching time required for complete goethite dissolution. Talc was found as a major phase in the sample LD-97-13 and a minor phase in the samples LD-97-28 and LD-97-32. It persisted in the relevant leaching residues, reflecting its resistance to the pressure acid leaching. The large amounts of talc in LD-97-13 resulted in less than 95% of nickel extraction because of its association with nickel. Kaolinite was found as a major phase in the sample LD-97-18 and a minor phase in the sample LD-97-32. It was also present in the corresponding leaching residues, showing its refractory character during pressure acid leaching. Clinochlore was found in the samples LD-97-13 and LD-97-28. It was detected in the residue obtained from leaching LD-97-28 at 250°C for 1 hour. However, it was dissolved after 1 hour of 117 leaching of the LD-97-13 sample. These differences were probably produced by the different terminal acidity in the leaching solution, suggesting that the free acid concentration significantly affected the dissolution rate of the clinochlore. The reactivity of clinochlore and goethite during pressure acid leaching showed differences with that of atmospheric leaching. Canterford (Canterford, 1978) reported that nickel was more readily extracted from nickel-bearing clays such as chlorite than it was from goethite during the atmospheric leaching. However, the present results showed that the dissolution rate of goethite was higher than that of the clinochlore during pressure acid leaching. 4.4.3 Terminal Acid Concentration after Leaching For all the leaching experiments carried out in this study, acid addition was calculated from a spreadsheet provided by Falconbridge Limited. The calculation program was based on the ore compositions and typical metal extractions with targeting free acid concentration of 35g/l for every test. However, the free acid concentration in the leaching liquors obtained from leaching five drill samples showed marked differences, which was probably caused by, (1) the different metal extractions achieved from leaching these five samples. As reported in the section 4.2, the extractions of Fe, Al, Cr, Mn obtained from leaching the five drill samples varied in a wide range. The typical metal extractions assumed for acid addition calculation were not followed during actual leaching; (2) the different mineralogy of the ore samples. Ore samples consisting of talc and kaolinite produced leaching solutions with high terminal acidity (Table 4-12 and Table 4-14). Talc and kaolinite showed resistance to pressure acid leaching. Therefore, a 118 portion of Mg and Al contained in these silicates did not consume acid during leaching, eventually resulted in high free acid concentration in the leaching liquor. The mineralogy varied among the five drill samples. In the present study, the mineralogy of the ore showed an insignificant impact on nickel extraction (except for LD-97-13), which was probably due to the fact that most of the nickel was associated with goethite and maghemite. However, the presence of the refractory minerals in the ore samples significantly affected the terminal acidity in the leaching solution and resulted in different goethite dissolution rate which affected the nickel extraction kinetics. For a successful acid leaching, one of the key parameter is rapid nickel and cobalt extraction with minimal acid consumption. Present studies showed that for minimizing acid consumption and obtaining fast nickel extraction, the mineralogy of the ore should be taken into account while calculating acid addition. 119 C H A P T E R 5 CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions Following are the principal conclusions reached from the study of pressure acid leaching of a variety of laterite samples and investigations into the mineralogy of both those laterite ores and their respective leach residues after pressure acid leaching with sulfuric acid. 1. Goethite and maghemite were found to be two predominant constituents in all the laterite samples. Most of the nickel was contained in the goethite and maghemite. Mn-oxide was commonly found in the laterite samples as a minor phase and a nickel and cobalt hosting mineral. Silicates such as quartz, talc, kaolinite and clinochlore were found in the Ivory Coast laterite samples. The silicate species present in Ivory Coast laterite samples and their proportions varied among the five drill samples. Nickel showed an association with talc and clinochlore to various extents. 2. Over 95% of nickel and cobalt extractions were achieved from leaching Ivory Coast laterite samples (except for LD-97-13) at the leaching conditions of temperatures of 250 ~260°C, an acid addition and slurry pulp density being calculated from the Falconbridge spreadsheet and providing long enough leaching time. Ni and Co extraction kinetics strongly depended on the goethite and maghemite dissolution rate. Faster goethite and maghemite dissolution rate 120 provided rapid Ni extraction kinetics. The relatively low Ni extraction obtained from leaching LD-97-13 resulted from the large portion of undissolved talc. Very small amounts of Ni and Co coprecipitated with Fe as hematite. 3. The extraction of Fe during pressure acid leaching was affected by terminal acidity, leach time and pulp density. Lower terminal acidity, longer leaching time and increased slurry pulp density resulted in lower iron extractions. The effect of pulp density on Fe extraction was thought to be the precipitation of hematite particles on the solids surface. The extraction of Al showed dependence on leaching temperature and time. Lower Al extraction was obtained at higher temperatures and longer leaching times as a results of alunite precipitation. The extractions of Cr and Mn from leaching the five drill samples varied widely. 4. Goethite dissolution rate showed a strong dependence on the temperature and terminal acid concentration. High temperature and high terminal acidity favored rapid goethite dissolution. Goethite was completely dissolved from Ivory Coast blend sample within 15 minutes for leaching test conducted at 260°C and a free acid concentration of 20.5g/l. It was also fully dissolved from samples LD-97-13, LD-97-18, and LD-97-32 within 30 minutes at 250°C and free acid concentrations > 29g/l. However, goethite was detected in the leaching residue of LD-97-24 after 1 hour of leaching when leaching test was conducted at 250°C and a free acid concentration of about 27g/l. The leaching behavior of the maghemite was similar to that of the goethite. 121 5. Talc, kaolinite and quartz present in the ore samples were found in the corresponding residues, showing their resistance to the pressure acid leaching. The dissolution rate of the clinochlore exhibited dependence on terminal acidity. It was fully dissolved from LD-97-13 after 1 hour of leaching and a free acid concentration of about 40g/l. However, it was found in the leaching residue of LD-97-28 after 1 hour of leaching and a free acid concentration of about 29g/l. 6. Leaching results illustrated that the terminal acidity was significantly affected by the mineralogy of the ore. The samples consisting of talc and kaolinte provided leaching liquors containing high free acid concentration. 7. Mineralogy is the primary consideration when trying to characterize the leaching behavior of the new ores. 5.2 Recommendations for Further Work 1. The focus of this thesis was a fundamental investigation of the pressure acid leach behavior of the minerals contained in laterite ores. The results presented in this thesis have only qualitatively revealed the mineralogy of the laterite ore samples and residues as well as some basic leaching characteristics. The quantitative analysis of the concentrations of various laterite constituents in the ores and residues should be performed in an attempt to gain a comprehensive understanding of leaching behavior of various minerals. For example, the 122 dissolution rate of maghemite during first 10 minutes; the dissolution rate of talc, kaolinite and clinochlore during pressure acid leaching. 2. During the SEM-EDX experiments, it was found that the compositions of the naturally occurring minerals varied from one particles to another. Hence it is difficult to calculate the nickel distribution among constitutional minerals according to only the mineralogical concentrations. Sequential dissolution may be a choice for investigating the nickel distribution among the laterite constituents. DCB technique is selective for the dissolution of Fe oxide, however it possibly dissolves lithiophorite as well. Therefore the selective dissolution of lithiophorite is recommended prior to the DCB treatment. This would provide the information about nickel distributions among silicates, Fe oxide and other metal oxides. 3. Present results illustrate that lower pulp density slurry results in a higher Fe extraction. Further tests are needed to confirm this conclusion. This would offer significant information for the maximum iron injection during pressure acid leaching. 4. The leaching results also demonstrate that the metal contents in the leaching liquor varied among samples. The relationship between those metal contents in the solution is not fully understood from this study. Also, the extractions of Cr and Mn varied widely through the experimentation, and therefore a fundamental study about leaching solution is recommended for an accurate acid addition calculation. 123 Results obtained from leaching the five drill samples suggests that in order to achieve high nickel extraction and faster leaching kinetics with minimal acid consumption, the mineralogy of the ore should be considered for an accurate acid addition calculation. 124 BIBLIOGRAPHY Barnhisel, R.I., 1977, "Chlorites and hydroxyl interlayer vermiculite and smectite", Minerals in Soil Environments, (ed. By Dixon J.B and Weed S.B.), Published by Soil science society of America, pp. 331. Basto, E., Duran, J., and Cano, G., 1997, "Cerro Matoso - A process update", Proceedings of the Nickel-Cobalt 97 International Symposium, August Sudbury Ontario, Vol. Ill, pp.207-215. Bernardelli, A., Melfi, A.J., Olivenira, S.M.B., and Trescases, J.J., 1983, "The Carajas nickel deposit", Proceedings of the II International Seminar on Lateritisation Processes Held in Sao Paulo, Brazil, Pp. 107-118. Biernat, R. J., and Robins, R. G., 1972, "High-temperature potential/pH diagrams for the iron-water and iron-water-sulphur systems", Electrohcimica Acta, Vol. 17, pp. 1261 -1283. Blesa, M. A., Morando, P. J., and Regazzoni, A. E., 1994, Chemical Dissolution of Metal Oxide, Published by CRC Press. Boldt, J.R., 1967, Wining of Nickel: Its Geology, Mining and Extractive Metallurgy, Longmans, Toronton. Buger, P.A. (1), 1996, "Origins and characteristics of lateritic nickel deposits", Nickel 96, AusIMM, Kalgoolie, pp. 179-183. Burger, P.A.(2), 1996, "The Bulong, West Australia, Ni/Co laterite deposits-A case study", Nickel 96, AusIMM, Kalgoolie, pp. 37-41. Canterford, J.H., 1975, "The treatment of nickeliferous laterites", Minerals Sci. Engng, Vol. 7, no.l, January Carlson, E.T., and Simons, C.S., 1961, Extractive Metallurgy of Copper and Cobalt, Interscience publishers, New York, N.Y., 1961. Chalkley, M. E., Balan, R., Kranz, U.U.,and Sanchez, R., 1996, "The pressure acid leach process for nickel cobalt laterite: A review of operations at Moa nickel S.A.", Nickel/Cobalt Pressure Acid Leaching & Hydrometallurgy Forum, Perth, West Australia. Chalkley, M. E., and Toirac, I.L., 1997, "The acid pressure leaching process for nickel and cobalt laterite. Part I: Review of operations at Moa", Hydrometallurgy and Refining of Nickel and Cobalt, 27th Annual Hydrometallurgical Meeting of CIM. (Ed. By Cooper, W. C and Mihayalov, I), pp. 341-353 Chou, E.C., Queneau, P.B., and Rickard, R.S., 1977, "Sulfuric acid pressure leaching of nickeliferous limonite", Metallurgical Transaction B, pp. 547-554. 125 Daenuwy, A., and Dalvi, A.D., 1997, "Development of reduction kiln design and operation at P.T. Inco", Proceedings of the Nickel-Cobalt 97 International Symposium, August Sudbury Ontario, pp. 93-113. Das, S.K., and Sahoo, R.K., 1995, "Lithiophorite and magnetite in the nickeliferous laterites of Kansa, Orissa", Indian Journal of Earth Science, Vol. 23, pp. 20-23 Davey, P.T., and Scott, T.R., 1962, "The hydrolysis of aluminum sulfate solution at elevated temperature", Australia Journal of Science, Vol. 13, No. 4, pp. 229-241. Diaz, C M . , Landolt, C.A., Vahed, A., Warner, A.E.M., and Talor, J. C , 1988, "A review of nickel pyrometallurgical operations", Extractive Metallurgy of Nickel & Cobalt, Proceedings of 117th TMS Annual Meeting, (ed. By Taroler, G.P and Landolt C.A.), pp. 211-239. Dixon, J.B., 1977, "Kaolinite and serptine group minerals", Minerals in Soil Environments, (ed. By Dixon J.B and Weed S.B.), Published by soil science society of America, pp. 357 Dutrizac, J.E., 1980, Lead-Zinc-Tin 80 (ed. by J.M.Cigan, T.S.Mackey and T.J.O'Keefe), TMS-AIME, Warrendale, PA, pp. 532. Esson, J., 1978, " Chemistry and mineralogy of lateritic nickel deposit at Liberdade", Brazil, Trans. Instn Min. Metall. (section B) May, B53-59. Eggleton, R.A., Schulze, D. G., and Stucki, J. W., 1988, "Introduction to crystal structures of iron-containing minerals", Iron in Soils and Clay minerals, (ed. By Stuck, J.W., Goodman B.A and Schwertmann U), ppl41 Faris, M.D., and Collins, M.J., 1997, "The Calliope project: Pressure acid leaching of nickel laterite ores from New Caledonia", Hydrometallurgy and Refining of Nickel and Cobalt, 27th Annual Hydrometallurgical Meeting of CIM, (ed. By Cooper, W. C and Mihayalov, I), pp. 409-424. Gilkes, R.J., and McKenzie, R.M., 1988, "Goechemistry of manganese in soil", Manganese in Soils and Plants, (ed. Graham R.D., Hannam, R. J. and Uren N. C). Published by Kluwer Academic Publisher, pp. 33. Golightly, J.P., 1979, "Geology of Soroako nickeliferous laterite deposits.", International Laterite Symposium, New Orleans, Louisiana, February 19 to 21, pp.38-56. Golightly, J.P., 1979, "Nickeliferous laterite: A general description", International Laterite Symposium, New Orleans, Louisiana, February 19 to 21, pp.3-23. Haigh, C.J., 1967, Proc. Aust. Inst. Min. Met., 223, 49. Haldeman, E.G., 1979, "Geology of lateritic nickel deposits Dominican Republic" International Laterite Symposium, New Orleans, Louisiana, February 19 to 21, pp. 57-84. 126 Hellsten, K.J., 1996, "Cawse nickel laterite deposit", In Nickel /Cobalt pressure acid leaching & hydrometallurgy forum, May 13-14, Perth, Western Australia. Harris, D.C., 1991, Quantitative Chemical Analysis, W. H. Freeman and Co., 3rd edn. pp. 377-378. Kerfoot, D. G. E. and Weir, D. R., 1988, "The hydro and electrometallurgy of nickel and cobalt", Extractive Metallurgy of Nickel and Cobalt 117* TMS Meeting, pp.241-265. Klein, C , and Hurlbut, C.S., 1993, Manual of Mineralogy, Published by JOHN WILEY&SON, INC Kohga, T., Yamagiwa, M., Kubo, N., and Akada, A., 1997, "Recent ferronickel production at Hyuga smelter", Proceedings of the Nickel-Cobalt 97 International Symposium, August Sudbury Ontario, Vol. Ill, pp.217-228. Krause, E., Singhal, A., and Blakey, B. C , 1997, "Sulfuric acid leaching of nickelferous laterite", Hydrometallurgy and Refining of Nickel and Cobalt, 27lh Annual Hydrometallurgical Meeting of CIM, (ed. By Cooper, W. C and Mihayalov, I), pp. 441. Lithgow, E.W.,1993, "Nickel laterites of central Dominican Republic", International Symposium Extractive Metallurgy of Copper, Nickel and Cobalt, Volume I, pp.403-417. Mahteson, P.J., 1996, "The development of the nickel/cobalt laterite in the South West Pacific", Nickel 96, AusIMM, Kalgoolie, pp. 59-60. McKenzie, R. M., and Osmond, G., "Manganese oxide and hydroxides", Minerals in Soil Environments, (ed. By Dixon J.B and Weed S.B.), Published by Soil science society of America, pp. 181. Mehra, 0,P., and M.L.Jackson., 1960, "Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate" Clays and Clay Minerals, (ed. By. Swineford A.), Proc. 7th Natl. Conf., Washington D.C., 1958, Pergamon Press, New York, pp.317-327. Mitchell, R.S., and Meintzer, R.E., 1967, "Lithiophorite from Charlottesville, Virginia", The American Mineralogist, Vol. 52, Sept.-Oct, pp. 1545-1549. Monti, R., and Fazakerley, V.W., 1996, "The Murrin Murrin nickel project", Nickel 96, AusIMM, Kalgoolie, pp. 191-195. Motteram, G., Berezowsky, R., and Raudsepp, R., 1996, "Murrin Murrin nickel/cobalt project-project development overview", Nickel /Cobalt Pressure Acid Leaching & Hydrometallurgy Forum, May 13-14, Perth, Western Australia. 127 Ogura, Y., Keiichikoide, K., and Shimosaka, K., 1979, "Geochemistry and mineralogy of nickel oxide ores in the Southwestern Pacific Area", Proceedings of the International Seminar on Lateritisation Process, India, Dec, pp. 58-67. Papangelakis, V.G., Blakey, B.C., and Kambossos, J., 1994, "Behavior of aluminum during direct acid leaching of limonitic laterites", Impurity Control. Disposal Hydrometall. Process., 24,h Annu. Hydrometall. Meeting, pp. 315-325. Papangelakis, V.G., Georgiou, D., and D.H.Rubisov., 1996, "Control of iron during the sulfuric acid pressure leaching of limonitic laterites", Second Intenational Symposium on Iron Control in Hydrometallurgy, (ed. by J.E. Dutrizac and G.B. Harris), pp.263-274. Posnjak, E and Merwin, H.E., 1922, Amer. Chem. Soc, 44, 1965 Power, L.F., and Geiger, G.H., 1977, "The application of the reduction roast-amoniacal ammonium carbonate leaching to laterite", Mineral Science Engineering, 9, pp. 32-51. Queneau, P. B.,and Weir, R., 1986, "Control of iron during hydrometallurgical processing of nicekliferous laterite ores", Iron Control in Hydrometallurgy, (ed. By Dutrizac, J.E and Monhemius, A. J.), pp76-105 Reid, J.G., 1996, "Laterite ores-nickel and cobalt resource for the future", Nickel 96, AusIMM, Kalgoolie, pp. 11-16. Schellmann, W., 1983, "Geochemical principles of lateritic nickel ore formation", Proceedings of the II International Seminar on Lateritisation Processes Held in Sao Paulo, Brazil, July 4-12, pp. 119-135. Schwertmann, U and Talor, R.M., 1977, " Iron oxides", Minerals in Soils Environment, (ed. By Dixon , J.B and Weed S.B.), pp. 145. Schwertmann, U., 1988, " Some properties of soil and synthetic iron oxides", Iron in Soils and Clay Minerals, (ed. By Stucki, J.W., Goodman, B. A and Schwertmann, U.), pp. 203 Sobol, S.I., and Kukoev., 1972, "Structure of tailing obtained following the sulfuric acid leaching of oxidized nickel ores", Tsvetnye Metally, vol. 13, pp. 13-14. Taylor, A., 1996, "Engineering issues & economics of pressure leaching of laterite", Nickel /Cobalt Pressure Acid Leaching & Hydrometallurgy Forum, May 13-14, Perth, Western Australia. Tindall, G.P., and Muir, D.M., 1996, "Transformation of iron Oxide in Nickel Laterite, Second International Symposium on Iron Control in Hydrometallurgy", (Ed. by J.E. Dutrizac and G.B.Harris), pp. 249-262. 128 Toomver, T.T., 1979, "Development of InCo' selective reduction smelting process for nickel laterite ores", Proceedings of the International Seminar on Lateritisation Process, India, pp 252-271. Troly, G., Esterle, M., Pelletierm B., and Reibell, W., 1979, "Nickel deposits in New Galedonia-some factors influencing their formation", Proceedings of the International Seminar on Lateritisation Process, India, pp 85-119. Umetsu, Y., Tozawa, K and Sasaki, K., 1977, Canadian Metallurgy Quarterly, No. 16, pp. I l l Warren, LH., and Devuyst, E., 1973, "Leaching of metal oxides.", 1973, The International Symposium on Hydrometallurgy, (ed. by D.I.Evans and R.S.Shoemaker), AIME, Chicago , 229-265 Wassink, B., 1998, Personal communication, Feb. Wiryokusumo, Y., Loebis, A.S., and Miraza, T., 1997, "Pomalaa ferronickel smelting plant of PT Aneka Tambang", Proceedings of the Nickel-Cobalt 97 International Symposium, August Sudbury Ontario, Vol. Ill, pp. 175-205. Zelazny, L.U., and Calhoun, F.G., 1977, "Palygorskite, (Attapulgite), sepiolite, talc, pyrophyllite and zeolite", Minerals in Soil Environments, (ed. By Dixon J.B and Weed S.B.), Published by Soil science society of America, pp.435 129 A P P E N D I X I F O R M U L A S O F M I N E R A L I N L A T E R I T E S Goethite, maghemite, gibbsite, quartz, talc, kaolinite, clinochlore and some of the spinel group minerals were found in the laterite samples investigated in this study. Their formulas are listed in Table A- l . Table A-1. Empirical formulas of minerals found in laterites used in this study Mineral Formula Oxide group: Goethite FeO(OH) Maghemite Fe203 Gibbsite Al(OH)3 Lithiophorite (Al, Li)Mn4+02(OH)2 Aluminian chromite Fe(Al, Cr)204 Spinel Mg(Fe, A1)204 Galaxite MnAl 20 4 Silicates group: Clinochlore (Mg, Al)6(Si, A1)A0(OH)8 (Mg, Fe, Al)6(Si, A1)A0(OH)8 Kaolinite Al2Si205(OH)4 Quartz Si02 Talc Mg3Si4O10(OH)2 130 APPENDIX II CALCULATION OF ACID ADDITION For all autoclave leaching experiments, acid/ore ratio and pulp density were calculated from a spreadsheet provided by Falconbridge Ltd. This spreadsheet was based on the following fixed parameters: Total slurry volume = 11 Free acid concentration = (Wa Ca-At)/V, = 3 5 g/1 Initial anion concentration = WaCaMc/Ma(Va+Vw) = 75.5 g/1 Acid consumption during the leaching was calculated with the following assumptions: The final iron concentration = 2.5 g/1; Extractions of metals: Ni 96%, Co 92%, Mn 40%, Al 60%, Mg 40%, Cr 20%. The total acid consumption was calculated from following equation: A, = 2.5V,3/2Ma/AFe +W0( 0.96CNi/ANi + 0.92CCo/ACo + 0.4CM n/AM n + 0.6CA1/AA1 + 0.4CM l/AM R + 0.2CCl.3/2ACr) (A-l) 131 The amount of acid needed for leaching was given by equation (A-2): Wa = (35^+^/0, (A-2) Wa - total weight of acid C a - concentration of acid = 96% A, - acid consumed during reaction M c - molecular weight of acid = 98.0775 g/mol M a - molecular weight of anion = 96.0616 g/L V a - volume of acid V w - volume of water V, - liquid volume = V a + V w W 0 = ore weight, C N j C C o C M n C A I C M g , C C r are the concentrations of metals in the ore respectively, A n i , A c o , A m n , A a l , A m g , ACrare the atomic weight of metals respectively. 132 APPENDIX III MASS BALANCE SHEETS Interruptive autoclave leaching experiments were carried out on Ivory Coast laterite samples LD-97-13, LD-97-18, LD-97-24, LD-97-28, LD-97-32 and Ivory Coast blend. Appendix III summarized leaching conditions, the assays of filtrate, washing solution and leaching residue obtained by ICP analysis and free acid concentrations measured according to the procedures described in Section 3.3 for every test. The calculated assays and the metal extractions calculated from equation (3-1) are also illustrated in the mass balance sheets. It should be mentioned that the solid % represent the percentage of solid in the total solution and that the calculated assays are obtained from the assays of filtrate, washing solution and leaching residue according to following equation: W, + Wv. + W. W.. C„, = - ^ — ^ (A-3) Cm: Calculated metal assay. W} : Weight of metal in filtrate. Ww : Weight of metal in washing solution. Wr: Weight of metal in residue. W : Weight of ore. ore ° 133 o oo § o ^ ^ ^ C N o f * e o -a a o U bxi a 1/1 co -4-» fi fi CD 9} U pa CA l-i t—1 U \ -S C^  O 4> CD _ hJ H i—1 to G O 0 s -T3 fi CD 5 H—» OO cd O u >-> o oo >-. o > - ^ oo m o cn oo f -co 3 a HS O co o •5 ^ > S O o s Q as OD O1 ao CN OS i n 00 OS VO o VO OS oo - o d -CN i n m O CN CN o 00 CN r n CN O © CN O T i -n's vo VO vo m CN © T f CN T T OS T f oo 00 m t-~ © — d CN 177.289 0.853 0.512 178.654 SO p CN O CO OS i n i n CN i n © © d d Tt-o p VO i n CN OS cn CN OS OS CN —~* © o T T VO r~ r n i n m p r n © d VO VO VO vo o © o d d d I/-} r~ o CN T T O CN o CN d m CN O m 00 o CN © CN d 52.000 1.618 0.169 OS p CN m i n o © © d Tf-OS CN oo CN o © T f d 340.94 CN >n 3025 |Output: iResidue (g, %) iFiltrate (mL, g/L) Wash (mL, g/L) |Total (g) n/a | high 2.800 3.122 2.56 | o o VO o CN in o o vq VO VO Tr d d d d CN o © r n o VO o o r n m m 1.17 ' ' ' ' T f o OS o p © oo TT •^-* d d d 00 o m o o o o o Os 00 VO A in m VO T f d o o CN T i -en 00 r n 87.85 d d d d 87.85 o in CN o CN o o CN 00 7.69 N ? 0S-C3 C3 o s ^ o 00 cn C N T t fi o ~ O 00 CO •s 00 c oo iH ^ '3 < CD hH P H "3 fi s25 c _o *H—» O H 3 fi OO fi O CO ~o ccj O H 134 S 2 CN 0 CN S B a .© s © U ex e CO C « CU s- t—1 • o T3 "o cu u C C3 « £ S O ^ C U CU CU ~— -J H i—1 T3 C 5 H—» 03 O U >-, o-i oo > oo r-i ^- i M m oo r--00 in < < 00 fi 00 X! 00 5 O cu CD CU J-O O H Cd O N C/3 11.407 | 0.193 j 0.099 j 11.700 | CN O O l O l © i n © 00 i n © CN © d o i OO CN r o o © 00 i n © i n O l o i © 00 r o 00 00 vo vo oq O l r o O N d — 1 © o i 174.460 698 0 0.476 175.805 © V O © O l O l r o vo V O •"3-i n d d © d 00 i n r -O N 00 r o r o O N © © © o i i n © © r o r o r o O l V O © r n d d O l © V O oo r o © O N o © d © © r o 00 o r o vo O l r o d d © O N i n © o i © 52.000 1.498 0.299 oo o V O </-) i n r o © © © © vo CN O l © vo oo © O N d d 335.5 o 00 i n 1590 Output: Residue (g, %) Filtrate (mL, g/L) Wash (mL, g/L) Total (g) n/a | high 2.800 | 3.068 | 2.50 | o © V O © O l i n © © vo o i n oo r-; d © d d © © r o © V O © © r o i n 5.80 '—' '—1 r o © O N © © © o -O N vo 6.91 © — © © oo © r o © © © © © o -© V O A i n m vd d © © o l r o r o 89.01 © d © d 89.01 © m O l o O l O © O l r o 4.86 CO C/1 CO ed >. m cd CO < As -a 00 in < -a 'C x> cd m .a. c m < o 3 o 5 Fa CU c/l O N ssa ad < u -a X CO "5 cd m m CO < O 00 oo o c o w 00 CD fi w CO cd 12 CD CD cd C fc 00 o 3 in cd cd "cd O H 135 S 2 ~ 1 CN - o gj e o s © a CU CU X! V) <U u s oa CA 00 CL) H—» fi £ CL) ^ 3 O H -fi • o CCS . „ - o U (j I) N N J H J <d 00 T 3 O GO fi CL) s H—» OO CCJ O u oo cn cn oo o > 00 -fi 00 w t3 • i-H o _o oo fi ^ -fi s a C J CL) £ o ccS (/) 0 S ON 0 0 u 11.407 | 0.193 | 0.099 | 11.700 | CN O CN CN o i n o 00 i n p CN d d CN OO CN r~ cn o p 00 O i n CN CN — d T t T t oo m 00 00 v q VO 00 CN cn Ov d ~~ d CN 174.460 0.869 0.476 175.805 o VO p CN CN cn T t VO VO T t i n d d d d 00 i n Ov oo cn cn T t Ov o o d CN —™ i n o o T t m cn cn CN VO o cn d d CN o VO oo cn o Ov o o d d d cn 00 o cn VO CN cn d " d T t o OV T t i n d CN d 52.000 1.498 0.299 00 o VO m m cn o d d d vo CN CN o VO oo o OV d T t d 335.5 o 00 m 1590 |0utput: iResidue (g, %) JFiltrate (mL, g/L) |Wash (mL,g/L) |Total (g) n/a | high 2.800 3.068 2 50 | o o VO o CN i n o o v q o T t i n 00 d d d d o o cn o vo o o cn m 5.80 — ' — 1 cn o Ov o p o T t OV VO t— Ov VO d d d 00 o cn o o o o o o VO A i n m T t T t vo T t d o T t o CN T t m cn T t 89.01 d d d d 89.01 o m CN o t— CN o o CN T t rn 4.86 CO CO a CO CO < As cu ao CO < OJ 'C CO .D. on CO < Fa cu N ? -<? ssa -a CJ3 < ra -o X u +H CO 3 B3 u CO CO "ra < U 00 00 o T * fi o ° w 00 "ca -fi hH 0O - f i CyJ P H ^ '3 < <L> CJ hH P -H "3 fi 00 c o fi oo oo ca 13 '3 ca la H—* o 136 ™ ~ © gj 0» cu xn CU S tt an C O 13 S O U WD a u CU -J 1/2 a cu 3 £ 2 H u rt fi rt fi rt h J < -a fi cu s - r t CO CCS O u >> o cu CO T 3 Q £ r r t O O O i r ! O O co r o O O OA OD 12 < _2 w -fi .2? cu s .. l-l rt rt-£ O O O s o a u < rt P 3 O O 00 9.839 | 0.179 | 0.069 j 10.088 | rn rn o so o o ON r i d d r i SO CN 00 r f p rr o r f r f r f r i r i d r f r i so r f r-00 m so oo d — d r i 161.874 0.709 0.290 162.873 00 o 00 SO rn O r f r f rn in d d d d so so r-i r f r i r f rn p d rn 1 1 in o o o 00 r i SO r f o rn d d m 00 S O o n p rf o p d d d o Os 00 in 00 in OS SO r i d — d m o o Os as in r i in d r i d o o © oo o rn Os in — d SO O O in r~ in O S Os o d d d <n >n © rn in rn r f Os d in d 317.4 o r f SO 1500 a. Residue (g, %) :rate (mL, g/L) ish (mL. g/L) s Residue (g, %) IZ ca o H high 2.800 2.654 | o o sq o n in o o so SO in d d d d o o rn o so o o rn •— 1 ' — -~ • — 1 o OS o p O r f rn m d — d d o m o o o o o so in 00 A i n i n r f ' r f r i r f o r f o n rr rn r f d d d d o m r i o r~ CN o o r i O s rn ~ u--o CO CD X o ON r t r-- r o o CN rt^ CN >n ~ 0 D *—' G O '-rt j=S O C/3 OD cu fi rt ts — CO rt _o CJ cs W < OJ cu rt. " r t fi r-c co C o rt "rt o i—i 137 § a 2 ^ <N ^ O CW CJ s « 5B 1/3 a o a © U bD G -= « -c/l CD C CD CD hi fi hi-fi a % o ^ CD CD ~ £ H h J < c/l T3 T3 fi CD s c/l Cd O U o vo vo © in OO ro ro oo S a CD 00 A^ s CD 3B Ak ^ C O 0 s fi -fi O H 5 o CD CD UH Q Ov o V c/3 10.464 | 0.161 | 0.059 10.683 | r-m CN OV o o r -o o c n i n CN CN d d CN r o vo CN Ov T t VO m vo m i n t— CN m' d d T t Ov t— CN 00 Ov m Ov Ov m OV d d CN 163.500 0.802 0.501 164.803 m CN O CN i n cn Ov VO VO i n d d d d OV vo T t OV CN oo t— vq o T t d c n i n o o CN LLZ CN m o cn © d T t 00 VO m o T t o o d d d oo Ov OV o CN Ov Ov d — d i n i n o O T t i n i n d r n d 50.000 1.383 0.275 r-o o r— o VO T t o d d d CN i n p 00 I '-ve OV Ov d t n d 1— CN cn O oo i n 1825 |0utput: |Residue (g, %) |Filtrate (mL, g/L) Wash (mL, g/L) |Total (g) -5 .£P O o 00 r-o oo in p 15 CN CN CN o o o CN o CN o Ov VO in vq in d d d d o o o o m oo o vo o CN vq cn cn cn ~~ '—1 — CN o o o CN _ OV T t r— OV r— p C— t— m d d d Ov o o o o o o Ov m o rn t-; A l/-i T t rn d in T t T t o o T t OV in T t CN m T t Ov in d d d d ov o o o o CN in r- o in CN CN CN rn VO OV 00 o CN 00 T t Ov VO T t T t CN T t assays | Assays -a CD aci idg C/l >, J D h. Assa 4-* on Assa _o —i s cd 0. -<? 0 S >, ssa ad < CD -a X <D v: JS >> 3 ssa _o ssa cd < U fi O ^ O W) w w 00 CD fi tt! - f i S H c/l -fi CCS '5 < CD CD UH P H 13 c P H •3B c _o '-fi O H S fi C/l c o CD CD ccj 13 H-> O H © ^ m CN ON o 2 CU -C C/3 CU s "3 C 3 en e o o U OJD a u « CU -3 c s % cu cu C/3 -a 6 £ r--O N r t O N A r-' ci U ON ON H-, r-l 00 O N r t oo S a cu M 00 rt OH ^ ^ NO cu 5 O O Q £ 39.6751 0.282 | 0.056 | 40.013 | rt U 1.109 0.035 0.009 1.153 Mg 9.092 3.428-0.780 13.300 Weights (g) 0.436 0.539 0.130 1.106 Weights (g) Mn 0.016 0.461 0.109 0.585 CJ fe 36.225 1.893 0.459 38.577 o U 0.005 0.058 0.015 0.078 iz 0.125 1.702 0.394 2.221 | 23.000 0.368 0.030 rt u 0.643 0.045 0.005 5 /o or g/L) Mg 5.271 4.481 0.411 5 /o or g/L) < 0.253 0.705 0.069 Assay (' i 0.009 0.603 0.057 CJ fe 21.000 2.474 0.242 o U 0.003 0.076 0.008 iz 0.073 2.225 0.208 Quantity g or mL 172.5 765.0 1895.0 1 Description |0utput: iResidue (g, %) iFiltrate (mL, g/L) |Wash (mL, g/L) |Total (g) © © © rt © © •> © © rn ON © CN ^ CN NO O ON r f N O O N CN 0 0 NO in © ' © © © o O N 0 0 0 0 r f r f cn I T ) CN CN r-r-' r~" v d © © CN © r f © CN NO r f r- in © © © ' d © © •n r~ 0 0 CN ON O N CN cn CN CN © © © © © © © CN © © © r f © © in © ' d o d O N CN CN © © CN © r f CN r f o © © © © © ' © ' © © © © © in © CN © CN CN • r t ^H r^ —1 T3 ta tu a 0 S 138 © © >n in NO ON © 0 0 NO d cn cn r f a o fe i £ 12 < OJ C3 w rt o 1 cn C o o a cs o H 8 s 5 °-CN "° o * ° o a o U e C/3 CD -*-» 3 C CU CU CU U e JS "S oa CA ce £ CD a s S H O H -g E ca CU CD CD J H J CO T 3 u C S 6 £ O N I Q 3 Q. CD CD r t OO K C O C I O N O N I D r - H oo r-3 < o 4J 00 3 - C ^ - C 00 U O H ^ ^ Q N V O flOOQS 53 35.1961 0.242 | 0.052 | 35.4891 Cr 0.882 0.037 0.011 0.930 ao 8.967 3.829 0.803 13.598 l-H o ND ON CO UO 0 0 CN 0 0 NO CN •+-. •a © © © Wei Mn 0.013 0.461 0.101 0.575 Fe 31.844 1.580 0.355 33.779 Co 0 0 o © t — uo © t © ON © Co d © © © % CO NO T t NO ON UO CO 0 0 r-^ H d © CN 33 21.000 0.348 0.020 Cr 0.526 0.053 0.004 ao 2 5.350 5.509 0.316 i UH o 0.359 0.775 0.050 Assay (' | 0 0 © © CO NO NO © T t © Assay (' © © © Fe 19.000 2.273 0.140 Co 0.005 0.082 0.005 iz CO © r H ON NO CO r H T t d CN © Quantity g or mL 167.6 695.0 2540.0 1 Description |Output: iResidue (g, %) iFiltrate (mL,g/L) |Wash (mL,g/L) |Total (g) O © 0 0 Cd © © r- CO © © OV 0 0 B CO OV © CN r H NO © Ov [— NO OV CN CO VO © © © © UO © Ov OV VO 0 0 T t T t 0 0 © UO CN CN 0 0 r-' r~ VO CO © © CN CN 0 0 T t © CN T t UO T t f- r- VO CN © © ' © ' © ' uo © © UO ,_, CN 0 0 CN OV OV r -CN CO CN CN r— © © © © ' Ov © © © CN © o © - H CO © uo © r H r -© © od r-' uo' CN CN —* © © CN © T t T t CN T t © T t © © © o Ov © © © © 0 0 © © O T t UO © CN o © © CN ^ H r-H -—1 CN ON 139 N ? 0 S N ? C5> XS cd CJ X CA >> cd N ? 0 0 U O CN T t "Sb © VO 1 c .2 _3 "3 c/l ao c 2 © © Ov r o X cd E ^ T3 'a < CJ OJ UH P H "cd C E 3 e o I 3 C/l S o o u • C3 o H 140 8 o £ S CN M o ^ e o •o s o U 3 C CU O) -E 03 CU u s -2 "3 CO sfi iri u rt e o C cd CU CD CU 6X) e SS Q ^ o\ r i - r-Q ° -fc O N O N m fc| oo r-s c-e •3 <: £ 3 J so « o-CD - 3 * C/3 ^ *•? Q ON DO CU CU rt O tn U 14.127 0.339 0.101 14.567 -* O N O O N m © CO ro . — ' © © -1 in p T T O N in O N 00 CN © O N ro CN d N O r f N O o m N O CN CN © m © ©' © © 73.252 1.340 0.402 74.994 •ri o o © in N O r f © © CN © © © ' © CN Tl © in ro CN © in O N 00 CN © ' ^ © CN © © CN in r f N O in © 00 © © 00 CN N O ro CN r o ro © d © © © T, © r f O N CN © r o © ro © © CN © r f CN ©' © ' 42.000 1.787 0.225 CO © © o © CN N O CN © © © ©' © CO © r o r o »—i 00 CN © ' CN © 174.4 750.0 1785.0 ^? i i |Output: [Residue |Filtrate ( Wash (i |Total (g; © © CN © © r o © ro in O N r-' r-' © © NO 00 i—1 r f 00 oo 00 NO NO NO © © © © © CN in O N © NO NO CN © O N in r f r f ro ro' © CN r o r o NO © r f in r f in T, © © © © ' © © © r -© © © r -CN © © r -O N O N m ' 00 r o CO ro r o © CN ro r f 00 ON ON O © © © © © © © © © © ro oo © © 00 © CN © rt< "* N ? T-) CO CJ in CO fiN 3 3 c ' • r t • • J2 o cu — "cd J3 s— cn a cd E £ -6 '3 < cu CU u . fc 13 c o NO N£9 c _o '-rt Q. £ 3 C/5 o o o cd -rt o H 141 - S 2- * tN o CN CU CU -C cn CU CU c js 13 03 Oft in s © e o U Of) c u CU -co -»-» C 2 H £ 8 O CD CD -CO T3 o 00 T t CN o 00 I/O r- CO o CO 00 3 • • 'oo - f i v — ' 00 UH a CD £ O U, H H O Q 3 "2 '3 <: CD "C 3 -fi _ O H cd c5 NO ON 1 3 u c/} T t NO ON NO CN CO T t T t © T t © Os Os © o r o uo 00 .—1 © o CO © © T t © ' ^ H 1—I o ON 00 uo T t NO NO r o uo © T t UO © NO T t © r H T t CN T t 1.88 © © © ' © © © © T t ON © T t NO UO 00 UO Os UO © NO 00 © r o CN 00 Os uo CN 00 © CO CO T t © ' CN r H © T t r~ l UO T t ON 00 © CO © T t CO r-CN © uo CN 00 Os 00 © © r H as T t T t CN CN CN 1—1 © ' Tt © ~* i — ' CO 181.665 0.541 0.205 182.411 53.200 52.000 51.000 49.075 0.41 ON © © OS CN © r H r H © 00 T t © © o o CN ^ H O r H ^ H © r o © 1.51 © © © © © © © ©' 00 NO uo r ~ Os NO T t r o 00 NO UO o © © © CN © © © ON © 1.36 © ' CN © T t 00 © © CN © UO CN UO © ~* © © as NO CN CN T t r-UO © © © ' © ' ON CN t— © © ON as NO CN © ' CN © ' NO 00 © NO f -UO © © 55.000 0.846 0.093 CO © © uo T t © U0 © © © © © cn Os CN CN 00 uo 00 NO r o Assa; Assay © CO © ' '5 CJ bn 330.3 640.0 2210.0 Multiple A Falconbrid, IPL Assays |0utput: |Residue (g, %) JFiltrate (mL, g/L) |Wash (mL, g/L) |Total (g) Assays Head % | Calculated Assays [Extraction: ^ NO CN 00 CN 00 c o o cn 00 e CD _ 15 'JG UH co - f i rt E -a < CD CD u-P H c E ON T t T t c OH e 3 co C o o o H-» O H 142 o uo CN O 2 ON VO 0 - CN cn C O " O e o U C « <U H-J 3 C CU CU t/5 O) = PQ % s e ° S cd cu cu cu J H J OX) e C J CS Q £ rr CN i C--ON Q fc rT o CN od m o m 00 +-> /3o 00 I-H o 3 O H cd z% CU CU a 5 O O Q N : VD ON o 0S-u ro CN ro T, CN 00 r-o rt rt in rt-' © ' © ' rf ro V D 00 © rt in 00 in CN N O in © © ' © ' O N rf r~ ro in r-rt r-O N O N CN © ' rf' © © N O © CN O N CN © ro ro © ' rf 184.737 0.382 0.111 185.231 r--© © rf ro © © © CN in © © © ' © ' © ' CN rt in CN CN in 00 00 CN m CN © ' CO © ' rt © © CO t-~ in ro in ro © © ' © ' V D V D CN ro N O r- © © ' © ' © OO rt 00 rf © in CN CN © ' CN © ' V0 CN O N © <N in © ro © ' © ' 57.000 0.540 0.049 CN © © 00 rt © rf © © © ' © ' © rt rt © in in in rt O N ro © ' rt © 324.1 708.0 2247.0 i s N£9 i | Output: [Residue [Filtrate ( |\Vash (i |Total (g; © © © O N CS © CN O N ro CN © ' H-H ' — l © © © O N rf N O m rt rt rt © ' © ' © ' © ' © CN O N CN -—< 00 in rt ro '—_ CN ro \—I \—J © 00 © ro in O N © CN 00 © ' © © © © © o © rf CN © © r-ro CN O N in in in rf © CN ,_, rt © i—< H-H © © © © © ' © ' © ' © ' © © © CN i-H © © rf CN © NT O N T3 ca CJ a cn ca N ? flN ON v b CN I T N CN CN c 'H—» • • ,2 w 00 CU C "Ss 12 I— c n H ccj -6 < cu CU kH fc I d c fc V D m rt 00 & _o O H 2 3 c n C o o o cd I d H-» o H 143 cu CU JS C/3 CU CU S3 i2 13 PQ C/5 C/l C3 e o • M M "O C o u CM) e 'Ja w «« CD H—» 3 Ct) s £ H & - f i E % CD CD H —1 C/l oo CN r-O N O N T t Q t N "1 H O N NO CN J N 00 | N 00 -fi 00 3 12 'o < CD oo fi ^ — ^ s CD CD -fi N O £ O O Q ch 1 ND CN Tj-© c n ND © CN OS T t © © © ca "5 © © Os CN ND 00 i n © © i n CN ' " H r-C -T t ON 00 T t T t © -H © © © OS © Os Os © CN © ~* © © CN © © © © NO CN T) © T t T t 00 T t © 00 T t © NO ON c n r ~ NO i n 00 i n 00 CN CN © >n r~> OS T f CN © 00 © —H c n © T t o NO OS Os ON OS i n NO © i-H © cn © © ' 145.692 0.578 0.113 146.382 54.600 51.000 50.000 49.638 o NO o OS c n T t NO 00 © i n 00 i n © OS r -T t 00 ON OS © ' © © © © © ' © © OS Th CN © ND NO NO CN T t cn T t © © CN © © CN © © i n NO © CN © c n 0006 4255 0253 © © 5474 6842 0430 © © © 9361 4546 1995 © co © 4631 2100 1278 © CN © 54.0000 0.8080 0.0464 0224 6136 0356 © © © CA ^ . 0922 7197 2166 : Assa Assay © co © CJ an 269.8 715.0 2428.0 Multiple A Falconbrid, IPL Assays i s [Residue (g, %) Initiate (mL, g/L) |\Vash (mL, g/L) |Total (g) Assays Head % Calculated Assays N O CN O N CN m oo 3 c o '-fi • • -3 w 60 CD fi oa - c u m _fi rt E £ "H '3 < CD CD v. P H "3 c E m CN T t T t 3 c o ' - H * O H E 3 c/l c o o o CS H-» o H 144 cn c o cn CD +-* 3 c a U £ CD w •§ .§ S 2 H « s y CU CU C/3 CU CU c iS 13 (*) t/3 a s * rt CJ ^ S fc J ! O 11 II) U N J H J < cn rs "o 00 C N i O N O N Q r r C N " i ON NO C N C N 00 t— 3 . . 5 o 3 "2 < £ 3 I 00 rt o, •53 B 3 cd r /3 00 CD U, HRT O Q NO ON os •r. so CN in oo © so CN r t © © in o r t OS rtH r t CN r t OS rt © ' © ' ^ OS CN 00 © CN in © so Os cn so CN CN ©'• in r~-CN © CN SO r-in r t r t © CN ~* © r n 142.340 0.336 0.100 142.776 r f CN O r t in r t in CN CN © SO © © ' © © in OS © © r t SO r t r-CN 00 r t © ' CN © ' m © © Os 00 SO cn © r t © rt d © CN r t ir, CN 00 in SO in © © ' © © ' cn Os © CN SO © m 00 CN cn © m m r t © in CN r t CN © CN © 55.000 0.467 0.047 OS © © o m so ON in © © © © m © SO so OS r t cn © m © ' 258.8 720.0 2137.0 N ? i i |0utput: |Residue |Filtrate ( |\Vash (i |Total (g; © © SO © cd © 00 © Os CN © © OS 00 © OS © m, OS so © © ' © © ' © cn CN so r~ m OS so OS H ^ ^ © so SO r- Os Os 00 ~ H OS Os © rtH © ' © ' — 1 © © © m. © © © i H SO © © r t r t © ' 00 •n in m r t © r- r t r t os 00 © CN © © © © © © © © o © 00 CN CN rt< rtH 1 - 1 *rt s P 0 s T 5 CO o X Os C N C N o — o w 00 CD fi "cd J 3 I— cn ^ «j E £ "H CD CD u rt-"c3 c E m r-C N r j -3 G O '-t-> OH S 3 cn C o o c j cd "cd H—> o H 145 CU CU JS CU CU C j s 73 PQ in in cs M l cs o •o e c U 01 s -CD H-> c S CD 3 S •fi -5 £ H U r -o. -C £ cd CU CD CD J H J CD Q 5 CN co ov i Q vo r t _ T t T t r -H OV T t r-C N oo 00 3 < 3 a CD -fi CD os t J W VIS _ CN Os t~~ VO r -CN VO © tn en t> © ' © ' 00 00 CN 00 Os est en T t OS © wo CN © ' © ' © ' ^ © © m CN © CN CN CN l > CN CN © ' wo vo r-CN r-en © © 00 CN en OS >n © ' © 110.722 0.714 0.184 111.620 00 © © 00 T t CN t— VO © CN CN en © ' © ' © ' © ' m © t— © VO © wo VO c n © CN © CO © © 00 en 00 en OV CN © en © © Tt" o T t wo T t © T t © © ' © © CN CN r-H en •n r-00 © en ^ H cn © ' W0 en © T t T t © CN i -H © r H © 54.000 0.992 0.079 T t © © T t T t en Ov CN © © © © 00 CN © CN vq © © CO © ' en © ' 205.0 720.0 2340.0 N? i i •$$ i |0utput: |Residue |Filtrate ( |\Vash (i |Total (g] © © r- CN 00 © VO T t © =i CO r o © T t c n r o © ' © CN r~ 00 ^ H VO m Ov i -H 00 vo in wo m © ' © © © en © © 00 oo 00 © VO 00 r- m uo r o Ov CN CN CN CN in © 00 r H OV OV in in VO m vq f - VO vo CN © © ' © ' © ' 00 © © © o © © © © © © © © CN 00 Ov VO VO © ' T t T t T t T t © r- CN en m CN CN CN cn v q r-H r H r H r H r—'• © © © © Os © © © en r H Os © © Os r o en «o r o en oo' Ov NT OV T 3 CO o X ca N? n V VO £ ° C N C O L N 00 c o i-> O w ao CD fi "S -S I— en 4fi rt E £ «< CD CD UH P H c E 00 T t c *H-» O. a fi en C o CD O 13 H—» o H 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0078581/manifest

Comment

Related Items