Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hydrothermal treatment of nickeliferous laterite with ferric chloride solutions 1981

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

Item Metadata

Download

Media
UBC_1982_A6_7 M86.pdf
UBC_1982_A6_7 M86.pdf [ 7.62MB ]
Metadata
JSON: 1.0078720.json
JSON-LD: 1.0078720+ld.json
RDF/XML (Pretty): 1.0078720.xml
RDF/JSON: 1.0078720+rdf.json
Turtle: 1.0078720+rdf-turtle.txt
N-Triples: 1.0078720+rdf-ntriples.txt
Citation
1.0078720.ris

Full Text

HYDROTHERMAL TREATMENT OF NICKELIFEROUS LATERITE WITH FERRIC CHLORIDE SOLUTIONS by NORMAN DONALD HOLLINGSWORTH MUNROE B.Sc. (Chemistry, Physics), University of Dar Es-Salaam. 1973 M.Phil (Mineral Process Engineering), University of Leeds, 1977 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES The Department of Metallurgical Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1981 (c) Norman Donald Hollingsworth Munroe, 1981 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or pub l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Mf^LUJ(LCHOB L Nfe^HiNG- The University of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada' V6T 1W5 Date Mil. /VqV. 1981 ABSTRACT The extraction of nickel and cobalt, from nickeliferous l a t e r i t e , together with the hydrothermal p r e c i p i t a t i o n of hematite has been investigated. In order to emphasize the relevance and significance of this process, an appraisal is made of the state of the n i c k e l , cobalt and iron industries. A compilation of the annual production of the respective ores on the world market is included with an examination of the future uses and demand of nickel and cobalt. S o l u b i l i t y relationships for iron (III) compounds in aqueous solution are reviewed in terms of pH, solution com- position and temperature. The thermodynamic data used at elevated temperatures between 60°C (333°K) and 200°C (473°K) have been estimated by using the "Entropy correspondence p r i n c i p l e " method of Criss and Coble. A sample calculation is shown in Appendix A. The effects of (a) temperature; (b) f e r r i c chloride concentration, (c) hydrochloric acid concentration and (d) pulp density were studied in order to evaluate extraction conditions. Generally, metal extraction increased with temperature and f e r r i c chloride concentration. At 423°K, over 90 percent of the nickel was extracted with a f e r r i c i i chloride concentration greater than 1M. Since appreciable amounts of gangue dissolved under most conditions, thereby consuming acid, a discussion on the recovery of hydrochloric acid is presented. F i l t r a t i o n of the precipitated hematite has proved d i f - f i c u l t , because of the very fine nature of the p a r t i c l e s . An overview of the nucleation and growth of pa r t i c l e s in supersaturated solutions has therefore been included. This phenomenon is used to describe the phase changes which occurred during leaching experiments, and to propose an ap- proach by which coarser p a r t i c l e s might be achieved. i i i T A B L E OF CONTENTS Page A b s t r a c t 1 1 Table o f Contents i v L i s t o f Tab les 1 X L i s t of F igures x ' i l Acknowledgements . X 1 V Chapter 1 INTRODUCTION 1 1.1 General 1.2 Minera logy o f L a t e r i t e Ore Depos i ts 3 1.2.1 D e f i n i t i o n and Nature o f N i c k e l i f e r o u s L i m o n i t e s . 5 1.3 Economics, Market Survey and Mine Produc t ion o f N i c k e l . . 6 1.3.1 S ta te o f the N i cke l Indust ry 6 1.3.2 Marketed N i cke l Products 1 2 1.3.3 S ta te o f the N i cke l Market , 14 1.3.4 P r i c e s , . . . . 1 4 1.3.5 Resources - Reserves of N i cke l 19 1.3.6 Uses 1 9 1.4 Economics, Market Survey and Mine Produc t ion o f C o b a l t . . 21 1.4.1 S ta te o f the Coba l t Market and Indust ry 21 1.4.2 Resources - Reserves o f Coba l t 22 1 .4 .3 Produc t ion o f Coba l t 25 1.5 Iron Ore - Market Relevance to I ron Residue 26 1.5.1 The S ta te o f S tee l P roduc t ion 26 i v Chapter Page 1.5.2 The Role of the Iron Residue from Nickel-" •• Laterites in the Direct Reduction of Iron (DRI) Industry 26 1.5.3 Outlook for the Future Use of the Iron Residue .. 31 1.6 Review of Existing Methods of Nickel Extraction from 1 Laterites 32 1.6.1 Laterite Processing Methods - General 32 1.6.2 Pressure Acid Leaching 34 1.6.3 Matte Smelting 35 1.6.4 Ferronickel Production 36 1.6.5 Selective Reduction Roast/Ammoniacal Leach 36 2 A REVIEW OF PREVIOUS INVESTIGATIONS ON DIRECT ACID LEACHING.. 38 2.1 Direct Acid Leaching of Nickeliferous Laterites 38 2.1.1 Direct Leaching of Goethite 39 2.1.2 Mechanisms of Leaching of Goethite 43 2.1.3 Direct Leaching of Low Magnesium Limonites 46 2.2 Hydrothermal Precipitation in Solutions Containing Mainly Ferric Ions 47 2.2.1 Hydrolysis of Metal Species at Elevated Temperatures 47 2.2.2 Precipitation of Iron as Ferric Oxide 48 2.2.3 Hydrothermal Precipitation of Iron III Compounds. 53 2.2.4 Hematite-Goethite Relations in Acid Solutions ... 54 2.2.5 The Effect of pH on the Solubility of Iron III Compounds Without Complexing 54 v Chapter Page 2.2.6 Considering the Ferric Ion--Hematite Equilibrium. 55 2.2.7 Considering the Ferric Ion--Goethite Equilibrium. 58 2.2.8 Considering the Ferric Ion--Ferric Hydroxide Equilibrium 58 2.3 Thermodynamic Considerations Underlying the Hydros'i. thermal Treatment of Nickel Laterite with Ferric Chloride 69 2.3.1 The Effect of Chloride Complexing 70 2.3.2 Temperature/pH Consideration 70 3 EXPERIMENTAL 72 3.1 Mineralogical Investigation 72 3.1.1 Energy Dispersive X-ray Analysis via SEM . 72 3.1.2 X-ray Diffraction Analysis 73 3.2 Quantitative Chemical Analysis 73 3.3 Apparatus Design 74 3.4 Pressure Leaching Experiments 74 3.4.1 Experimental Procedure 74 3.4.2 Liquid/Solid Separation 78 3.5 Analytical Methods 79 3.5.1 Atomic Absorption Analysis 79 3.5.2 Determination of "Free Acid" 80 4 RESULTS 84 4.1 The Leaching of Nickel iferous Laterite in Aqueous Ferric Chloride 84 vi Chapter Page 4.1.1 Effect of Temperature 84 4.1.2 Effect of Concentration 87 4.1.3 Effect of Pulp Density 87 4.2 The Leaching of Nickel iferous Laterite in Ferric Chloride/Hydrochloric Acid Solutions 90 4.3 Leaching of Nickeliferous Laterite in the Presence of Ferrous Chloride 93 4.4 Acid Consumption During a Batch Leach (423°K) 94 4.4.1 The Effect of Hydrolysis on Acid Consumption 96 4.4.2 "Free Acid" Concentrations in Ferric Chloride 1 Solutions 99 4.5 The Determination of Leaching Equilibrium 99 4.6 Simulation of a Continuous Leaching Circuit 102 4.7 Morphology of the Iron Residue 103 4.7.1 Hematite from the Hydrolysis of Ferric Chloride.. 103 4.7.2 Iron Residue from the Hydrothermal Treatment of Nickeliferous Laterite with Ferric Chloride Solutions 107 4.7.3 Iron Residue from the Hydrothermal Treatment of 110 Nickeliferous Laterite with Ferric Chloride/ Hydrochloric Acid Solutions 110 4.8 Comparison of Nickeliferous Laterite and the Iron Residue by Electron Microanalysis 110 4.9 Comparison of Particles Found in Nickeliferous Laterite and the Iron Residue by Electron Microanalysis 114 vii Chapter Page 5 DISCUSSION 1 1 9 5.1 Review of Leaching Results H 9 5.2 The Significance of Acidity During Hydrolysis 120 5.3 Nature of the Iron Residue 121 5.4 Hydrothermal Precipitation of Hematite in Super- saturated Ferric Chloride Solutions 123 5.4.1 Homogeneous Nucleation 123 5.4.2 Growth of Nuclei 125 5.5 The Degree of Supersaturation for Hematite in a Goethite Saturated Solution 127 6 CONCLUSION 131 7 SUGGESTIONS FOR FUTURE WORK 133 7.1 Improvement in Experimental Procedure 133 REFERENCES 134 APPENDICES A Criss and Cobble Calculations 138 B X-ray Diffraction Results 1 5 1 vi i i LIST OF TABLES Table Page 1.1 Distribution of Major Elements in Nickel i f erous Li.monites ... 7 1.2 Principal World Nickel Producers 9 1.3 World Nickel Mine Production, 1977, and Capacity 1977, 1978.. and 1980 11 1.4 Commercial Forms of Primary Nickel 13 1.5 Identified World Nickel Resources 20 1.6 Western World Cobalt Market Consumption Shares 23 1.7 World Raw Steel Production by Region 27 1.8 Direct-reduced Iron Processes Applied Commercially up to the End of 1979 28 1.9 Growth of World DRI Capacity 30 2.1 Activation Energies for Direct Dissolution of Hematite and Goethite 42 2.2 Equilibrium Data for the Fe 3 + — F e ^ — H 2 0 System 57 2.3 The pH Corresponding to Various Activities of Ferric Ion in Equilibrium with Hematite 59 2.4 Equilibrium Data for the Fe3+—FeOOH—H20 Metastable System.. 60 2.5 The pH Corresponding to Various Activities of Ferric Ion in Equilibrium with Goethite 61 2.6 Equilibrium Data for the Fe 3 +—Fe(0H) 3—H 20 Metastable System 62 2.7 The pH Corresponding to Various Activities of Ferric Ions in Equilibrium with Ferric Hydroxide 63 ix Table Page 2.8 Theoretical Equilibrium pH at Various Temperatures and Activities for the F e 3 + — F e ^ — H 2 0 Metastable System 64 2.9 Theoretical Equilibrium pH at Various Temperatures and Activities for the F e 3 + — FeOOH—H2n Metastable System 65 3.1 Chemical Analysis of the Nickeliferous Laterite 75 3:2 pH values of Hydroxides in Equilibrium with their Metal Ions. 81 4.1 Metal Extraction from Nickel ferous Laterite with Aqueous FeCl y. 86 4.2 Leaching results in the Presence of "Free Acid" at 423°K . 92 4.3 Leaching Results in the Presence of FeCl 2 at a Pulp Density of 400 g/1 94 4.4 Acid Consumption During a Batch Leach at 423°K with a Pulp Density of 400 g/1 95 4.5 Comparison of "Free Acid" Concentration in Filtrates with Acid Concentration Due to the Hydrolysis of FeCl 3 97 4.6 Iron Precipitation from FeCl 3 Solutions at 423°K 100 4.7 Percentage Nickel Extraction from Nickeliferous Laterite in the Presence of High Concentrations of Nickel 102 4.8 Percentage Metal Extraction in a Simulated Continuous Leaching Circuit 104 4.9 Chemical Analysis of the Iron Residue for Each Stage 106 5.1 Comparison of Raw Materials used in the Iron and Steel Industry 122 A.l Best Values of a^, b^, a^., 3̂ . and Ŝ  (H ) at Several Temperatures 144 x Table Page A.2 Standard Free Energies, Entropies and Partial Molal Ionic Heat Capacities Adopted for Species Participating in 3+ Reactions Considered in the Fe —H^O System 145 A.3 Summary of Average Heat Capacities Over the Ranges of 298°K to Upper Limits of 333°, 373°, 423° and 473°K (Estimated by the Correspondence Principle) 149 A. 4 Summary of Computations of the Free Energy of Hydrolysis 3+ Reactions Considered in the Fe — H,,0 System 150 B. l X-ray Diffraction Patterns of Goethite Using K Radiation ... 152 a B.2 X-ray Diffraction Patterns of Hematite Using Ka Fe Radiation 153 B.3 X-ray Diffraction Patterns of Chromite Using K Fe a Radiation 154 xi LIST OF FIGURES Figure Page 1.1 Chemical composition of nickel laterite ore zones 4 1.2 Nickel producer inventories, 1967-1980 (Noncommunist World).. 1 5 1.3 Refined nickel production and producer deliveries, 1967-1980 (Noncommunist World), ^ 1.4 Noncommunist world nickel consumption / ^ 2.1 The structure of diaspore and goethite (after Bragg) - 41 2.2 Experimental temperature vs pH 2 g 8 diagram for the Fe 3 +—Fe 2 0 3 —H 2 0 system ; 4 9 2.3 Free energy versus temperature for iron (III) hydrolysis reactions 66 2.4 Influence of pH on the solubility of iron (III) hydroxides and oxide at 423°K 6 7 3+ 2.5 Temperature versus pH diagram for the Fe —Fe^O^—H^O system 68 3.1 The leaching apparatus " 76 3.2 The pressure f i l t r a t i o n , device ' 7 7 3.3 pH titration curves of metal chlorides in hydrochloric acid versus sodium hydroxide 82 4.1 Effect of temperature on nickel extraction from nickeliferous laterite 85 4 .2 Effect of FeCl 3 concentration on nickel extraction at 448°K. . 8 8 4.3 Effect of pulp density on nickel extraction with 1M FeCl 3 at 423°K 8 9 xi i Figure Page 4.4 Effect of "Free Acid" strength at several FeCl 3 concentra- tions on nickel extraction at 423°K, and a pulp density of 400 g/1 91 4.5 "Free Acid" concentration and percentage iron precipitation versus iron (III) concentration at 423°K 101 4.6 Metal extraction versus lixiviant composition at 423°K and a pulp density of 400 g/1 ' 105 4.7 Hematite particles precipitated from FeCI_3 at 423°K 108 4.8 Iron residue obtained from the hydrothermal treatment of nickeliferous laterite with FeCl 3 109 4.9 Iron residue obtained from the hydrothermal treatment of nickerliferous laterite with FeCl3/HCl / HI 4.10 Main nickel-bearing particles (goethite) 112 4.11 SEM x-ray analyser spectra for nickeliferous laterite and iron residue 113 4.12 Silicate particles after leaching in HC1/HN03 115 4.13 Magnetic particles found in nickeliferous laterite and the iron residue 115 4.14 SEM x-ray analyser spectra for silicate and magnetic particles — ,"''117 4.15 SEM x-ray analyser spectra for magnetic particles in the iron residue • 118 5,1 Free energy of formation of spherical embryos as a function of the diameter for a series of temperatures 126 xi i i ACKNOWLEDGEMENTS I am p a r t i c u l a r l y indebted to my Supervisor Dr. E. Peters for his supervision, keen interest and constructive c r i t i c i s m of this report. I sincerely thank him, not only - f o r his continuous guidance in matters of an academic nature but s p i r i t u a l as well. Special mention must be made of my wife, E l e i s e , who gave me moral support and insp i r a t i o n when d i f f i c u l t i e s arose; and believe me, there were many. Thanks are also extended to a l l members of the Department of .Metal 1urgical Engineering for their cooperation and unend- i n g a s s i s t a n c e . The author is grateful to the National Research Council of Canada for fin a n c i a l support in the form of a Research As si stantshi p. xiv 1 Chapter 1 INTRODUCTION 1 .1 General The majority of the world's known and anticipated reserves of nickel ores are l a t e r i t i c deposits. Nickeliferous limonites, which constitute one of three zones of a typical l a t e r i t e de- posit,are also future sources of cobalt, chromium and iron. Laterites are near-surface deposits of oxidized material formed by extreme weathering (or " l a t e r i z a t i o n " ) of parent ultramafic rock. In contrast, the majority of the world's present production of refined nickel is from primary sulphide deposits, which tend to extend deep into the earth. Although sulphide ores are r e l a t i v e l y easy to process (usually by a combination of ore-dressing and pyrometal1urgical techniques), mining costs are escalating and established mines are getting deeper and obtaining lower grade ore. Lat e r i t e s , however, are quite abundant and are much less expensive to mine. It is an established fact that processing costs for l a t e r i t e s are higher than sulphide ores, because of the 2 necessity of chemical and/or thermal treatment of the entire ore. However, improved processing technology, increased world demand for n i c k e l , and depleting sulphide reserves, indicate the future importance of l a t e r i t e ores. At present, this material is being treated for nickel and cobalt recovery at Nicaro and Moa Bay, Cuba,^ and a number of new i n s t a l l a t i o n s are projected for A u s t r a l i a , Indonesia, New Caledonia and the Phi 1ippine- Islands. Although the mineralogy and chemical composition of materials from these countries show s t r i k i n g s i m i l a r i t i e s , the methods used for the extraction of nickel and other desired metal values can be quite d i f - ferent. Those which are in commercial operation around the world include, pressure acid leaching, matte smelting, pro- duction of ferronickel in an el ec.tr i c.:: or blast furnace, and 2 selective reduction roast/ammoniacal leach. The object of this report is to investigate the chemistry of leaching nickeliferous l a t e r i t e with f e r r i c chloride solu- tions. In l i g h t of the fact that most of the nickel present in l a t e r i t e s occur in s o l i d solution with goethite, hydro- thermal treatment of the ore is a useful method of separating i t s metal values. Using appropriate techniques, nickel and cobalt can be recovered from the leach solution. Iron, on the other hand, is hydrolyzed to hematite and is referred to as the iron residue. 3 1.2 Mineralogy of Laterite Ore Deposits A l l l a t e r i t e s are formed by weathering of the parent rock in areas of abundant r a i n f a l l . Deposits are therefore found in tropical or subtropical regions, al though.'-some 1 ateri te deposits are known to exist in more temperate zones. The nickel in the ultramafic rock usually occurs as an ion replace- ment in the s i l i c a t e l a t t i c e or as an associated sulphide. Surface waters containing carbon dioxide leach nickel out of the ultramafic rock, carrying i t downward. Reprecipitation of the metal deeper in the ore deposit may cause an enrichment of nickel in the range of 5 to 20 times the content of the parent rock. Magnesia and s i l i c a may be leached from the parent rock, leaving varying concentrations of n i c k e l , cobalt, chromium, iron and aluminium in the residual surface mantle. The end result of the " 1 a t e r i z a t i o n " process is the formation of three d i s t i n c t zones overlying each other and varying in chemical and mineralogical composition. Figure 1.1 depicts the variation in chemical composition and ore type , with 3 4 depth for a typical deposit. ' The three zones of a typical l a t e r i t e deposit are: the leached or canga zone, the iron oxide or limonitic zone, and the s a p r o l i t i c or magnesium-silicate-rich zone. Below the Increasing Metal Content Figure 1: Chemical composition of nickel laterite ore zones. 5 s a p r o l i t i c zone l i e s the remaining unaltered parent rock. Any of the three zones may be extremely thin or absent, while the thickness of individual zones may vary as much as 30-plus meters. Some deposits may even be composed e n t i r e l y of almost one zone. 1.2.1 Definition and Nature of Nickeliferous Limonites 5 The term nickeliferous limonite is used to desig- nate, the iron oxide rich zone of a l a t e r i t e deposit. It con- s i s t s of a cryptocrystal1ine mixture of goethite, with minor amounts of other hydrous oxides. Hematite, maghemite and chromite are accessory minerals commonly present, along with varying subordinate amounts of nickeliferous phyl1osi1icates of the montmori11onite and serpentine type. The presence of s i l i c a t e s has a s i g n i f i c a n t impact on the e f f i c i e n c y of the extractive procedures, even though they contain less than a tenth of the total n i c k e l . Most limonites contain at least 90 percent goethite. Three-quarters of the ore's nickel and half of i t s chromium content occur in s o l i d solution with the iron minerals. How- ever, more than 90 percent of the cobalt occurs in associa- tion with manganese oxides. The only s p e c i f i c nickel mineral detected in l a t e r i t e s 6 is g a r n i e r i t e . Investigation has shown the presence of up to an equivalent of 15 mol percent diaspore, and considerable amounts of adsorbed water. The l a t t e r is released upon heat- ing between 383°K and the temperature of dehydroxy1ation - 685°K. 6 Nickeliferous limonites analyze t y p i c a l l y in the range 0.8 - 1.5 percent nickel , 0.1 - 0.2 percent cobalt, 1.5 - 3.0 percent chromium and 45 - 50 percent iron. In addition to nickel and cobalt, these ores are obviously important as a source of chromium and iron. A l l four metals should therefore be considered as potential market products. Table 1.1 shows the d i s t r i b u t i o n of major elements in nickeliferous limonites. 1.3 Economics, Market Survey -and-Mine'Production of • Ni ckel 1.3.1 State of the Nickel Industry Salient s t a t i s t i c s have revealed how v i t a l nickel is to the iron and steel industry and the key role i t has played in the development of the current aerospace industry. Nickel's greatest value is in alloys with other elements, where i t adds strength and corrosion resistance over a wide range of temperatures. 7 Tab 1e 1.1 Distribution of Major Elements in Nickeliferous Limonites Content Percent Di s t r i buti on Percent NICKEL Goethi te a-FeOOH 0. 5 - 2 75 - 95 Manganese Oxi des 4 - 20 5 - 15 Maghemi te Y-Fe 20 3 0. 2 - 1 2 - 15 S i l i c a t e s 2 - 5 1 - 5 COBALT Manganese Oxi des 4 - 20 80 - 100 Maghemi te 0 - 0.3 0 - 2 0 CHROMIUM Goethite 1 5 0 - 7 0 Spinel 10 - 30 30 - 50 IRON Goethi te 50 90+ 8 Until 1870, nickel production was limited to small de- posits in China, Germany, Greece, I t a l y , Norway, Sweden, and the United States. 7 A nickel s i l i c a t e ore was discovered in New Caledonia in 1864 by Gamier, after whom the mineral garnierite was named. New Caledonia was the principal source of nickel from 1875 unt i l 1905, when Canada became the leading producer. The Sudbury area of Canada has remained the p r i n c i - pal source of nickel in the world. In 1977, the main nickel industries in market economy of the free world, were located in Canada, New Caledonia and Australia (see Table 1.2). International Nickel Company of Canada, Ltd. (INCO) operated 15 mines in Canada, with several on standby. The firm also operated ore concentrators, smelters and r e f i n e r i e s in Canada, as well as a nickel carbonyl refinery in Clydach, Wales, and a large integrated r o l l i n g mill at Huntington, West V i r g i n i a . Falconbridge Nickel Mines Ltd. operated mines, concentra- tors and smelters in Ontario and Manitoba. Smelted matte was shipped to the firm's refinery at Kristiansand, Norway. At this f a c i l i t y , cathode n i c k e l , nickel plating anodes and nickel sulphate were produced along with other associated metals. Sh e r r i t t Gordon Mines Ltd. presently operates a refinery Table 1.2 Principal World Nickel Producers Country Company Nickel Products Nickel metal and matte. Nickel oxide and mixed nickel-cobalt sulfides. Nickel-copper-cobalt matte. Nickel oxide sinter, soluble nickel oxide, nickel metal (cathode and pellets), ut i l i ty shot and pig. Nickel-copper matte. Nickel metal. Nickel oxide and sulphide. Dominican Republic. Ferronickel. Nickel metal, oxide, salts. tARCO-Socie'te' Miniere el Metallurgique de tarymna S.A. . . Ferronickel. Exploraciones y Explotaciones Hineras Izabel (Inco Ltd.). Nickel matte. Nickel matte. Ferronickel. Nickel metal. Nickel metal. Ferronickel. Ferronickel. Ferronickel. Ferronickel. Ferronickel. Ferronickel. Nickel oxide. Nickel oxide. Nickel oxide. Ferronickel shot, matte. Nickel metal. Nickel briquets, powder, nickel-cobalt sulfides.. Nickel metal. Nickel matte. Nickel metal and nickel-cobalt salts. Ferronickel. Nickel briquets and powder. U.S.S.R Nickel metal and matte. 10 at Fort Saskatchewan, Alberta, using imported nickel materials and high grade concentrates from INCO's Mine in Thompson Mani- toba, after the former company shut down i t s Lynn Lake mine in 1977. In countries with cen t r a l l y controlled economies, the U.S.S.R. produced 80 percent of a l l nickel output. The nickel l a t e r i t e mines in Cuba reportedly produced most of the rest. The principal nickel producer in New Caledonia in 1977 was Societe Metal 1urgique Le Nickel, S.A. (SLN). The company operated mines and a ferronickel smelter at Doniambo. In the past, SLN together with Le Syndicat Independant Des Mines have exported large tonnages of nickel ore to Japan. Western Mining Corporation Ltd. was the largest producer of nickel in Australia during 1977. In Western Australia the firm operated several mines and a concentrator at Kambalda, a smelter at Kalgoorlie and a refinery at Kwinana. I n s t a l l a - tion of a new high capacity shaft smelting furnace was com- pleted in late 1978. Since then, the company has exported nickel matte and concentrate to Japan. g Table 1.3 l i s t s the 1977 nickel production and projected capacities by country. As is shown, 25 percent of a l l nickel production was derived from nickel l a t e r i t e s whilst 80 percent Table 1.3 World Nickel Mine Production, 1977, and Capacity 1977, 1973, and 1980 (Thousands Tons) Production Capacity 1980 in 19/7 1977 1978 North America: 18 16 United States 13.7 18 Canada 259.4 275 275 275 Total 273.1 293 293 291 Central America and Caribbean Islands: Cuba 40.6 42 42 52 Dominican Republic 26.7 37 37 37 Guatemala -- -- 10 14 Total 67.3 79 89 103 Europe: 3.1 Poland 3 3 4 U.S.S.R. 158.0 170 170 210 Other! 21.4 50 50 70 Total 182.5 223 223 284 Oceania: 115 Australia 94.5 95 95 New Caledonia 120.2 150 150 170 Total 214.7 ' 245 245 285 Asia: 29 70 Indonesia 15.4 29 Philippines 16.5 35 35 55 Total 31.9 64 64 125 Africa: 20 Botswana 13.3 20 20 Rhodesia, Southern 17.6 17 17 22 South Africa, Republic of 25.4 25 25 28 Total 56.3 62 62 70 Other 25.2 25 25 752 World Total 851.0 991 1.001 1,233 Western Europe, principally Greece and Yugoslavia. Host expansion will occur in Central and South America. 12 of world resources, exclusive of seabed nodules, are contained in l a t e r i t e s . 1.3.2 Marketed Ni ckel Products Primary nickel is marketed in the form of nickel cathodes, powder, briquets, rondelles, p e l l e t s , ingots, metal shot, nickel oxide sinter and f e r r o n i c k e l . The chemical com- positions of commercially produced primary nickel forms are Q given in Table 1.4. Commercial nickel in these forms is normally more than 99.5 percent pure except for ferronickel and nickel oxide s i n t e r . The ferronickel produced in the United States contains 40 to 50 percent nickel and is sold in 50-pound pigs, whereas that produced in other countries contains 20 to 38 percent n i c k e l . Nickel oxide sinter contains 76 percent ni c k e l . A new product, incomet, introduced in 1974, contains 94 to 96 percent nickel and has replaced the 90 percent nickel oxide sinter introduced in Canada, but now produced only by Queens- land nickel proprietary in Au s t r a l i a . The a v a i l a b i l i t y of carbonyl pellets was greatly i n - creased with the start up of INCO's Canadian Copper C l i f f Nickel Refinery in 1974. Demand for this form of nickel has increased due to i t s high purity (99. 97 percent), ease of handling and storage. Table 1 .4 Commercial Forms of Primary Nickel Composition, percent Ni C Cu Fe S Co 0 Si Cr Pure unwrought nickel: Cathode 39:9 0.01 0.005 0.002 0.001 . . . ~ •• - Pellets 99,97 .01 .0001 .0015 .0003 0.00005 — . . . Powder 99.74 .1 . . . .01 .001 — 0.15 . — Briquets 99.9 . .01 .001 .002 .0035 .03 — — " - - Rondelles 99,25 ..022 .046 .087 .004 .37 .042 . . . Ferronickel^ 2 20-50 1.5-1.8 — Balance .3 (2) . . . 1.8-4 1.2-1.8 Nickel oxide 76,0 — .75 .3 .006 1.0 Balance . . . . . . Nickel s a l t s : 3 Nickel chloride 24,70 . . . — — — — — Nickel nitrate 20,19 — . . . . . . . . . . . . . . . . . . Nickel sulfate 20.90 Ranges used to denote variable grades produced. Cobalt (1 to 2 percent) included with nickel. 'Theoretical nickel content. 14 1.3.3 State of the Nickel Market Sharp swings in.nickel d e l i v e r i e s and refined pro- duction have occurred since 1967. A s i g n i f i c a n t buildup of producer inventories during 1975-1977 was followed by a steep decline in 1978-1979. However, during 1981 a modest increase in producer inventories is anticipated. The short and drastic swings in consumption patterns are i l l u s t r a t e d in the following charts. See Figures 1.2, 1.3 g and 1.4. Recent market studies suggest that the h i s t o r i c a l growth in nickel consumption of 6 percent per year began to weaken in 1975-1976. It is now estimated that 4 percent per year is a reasonable yardstick for growth expectations through 1985 and possibly in 1990. The 4 percent trend assumes that the rate of capital spending w i l l be stimulated, and that i n - creased u t i l i z a t i o n of nickel w i l l occur in energy programs associated with coal processing, o i l and gas d r i l l i n g and the development of the nickel-zinc battery. 1.3.4 Prices In 1980, INCO introduced a temporary 6 percent across-the-board discount from published prices for competi- tive reasons. Further price erosion is not expected for the following reasons: 15 Figure 1.2 Nickel producer inventories, 1967-1980 (Noncommunist World). 16 Refined nickel production and producer deliveries, 1967-80 (Noncommunist World) 1.400 1,200 1.000 1966 1968 1970 1972 1974 Year end 1976 1978 1980 - Refined nickel production 1967-79 Refined nickel producer deliveries 1S67-79 — - — Estimate Estimate Figure 1.3 Refined nickel production and producer deliveries, 1967-80 (Noncommunist World). 17 F igure 1.4 Noncommunist World n i c k e l consumpt ion. t 18 1) Producer inventories w i l l be maintained at more reason- able levels and consumer inventories now stand at a low 1evel . 2) Production cutbacks have been i n i t i a t e d by the major producers in order to keep inventories in lin e with demand. 3) Production costs have continued to escalate sharply. Nickel products derived from l a t e r i t e ores (which account for about 45 percent of non-communist world production) are under special cost pressure. At cer- tain l a t e r i t e based f a c i l i t i e s , energy costs are 50- 60 percent of the total production cost. 4) Under present economic circumstances, the published price, of $3.45 (U.S.) per pound does not y i e l d an adequate return on investment. Over the longer term however, nickel markets face an impressive building of latent demand in the energy and trans- portation sectors. Normal levels of nickel consumption are anticipated through 1982, by which time, some of the currently idle standby capacity may be required. In other words, the current market weakness has not yet run i t s course, but o p t i - mism and confidence are growing, now that stronger demand l i e s ahead. 19 1.3.5 Resources - Reserves of Nickel Total i d e n t i f i e d resources of nickel have been estimated at 230 m i l l i o n tonnes, whilst world nickel reserves g were 60 m i l l i o n tonnes. A report from the U.S. Geological Survey^ indicated that the combined sulphide and l a t e r i t e world resource was approxi- mately 7 b i l l i o n tonnes averaging about 1 percent ni c k e l . This figure appears to be very modest, because counting only deposits in countries bordering on the Caribbean and Islands of the Western P a c i f i c , the tonnage of ni ckel i ferous l a t e r i t e s ana-: lysing more than 0.8 percent nickel and averaging about 0.1 per- cent cobalt, amounts to at least 10 b i l l i o n tonnes. 1.3.6 Uses In 1979, pure unwrought nickel constituted 68 per- cent of the total primary nickel market; f e r r o n i c k e l , 20 per- 8 cent; and nickel oxide s i n t e r , 11 percent. The pure forms were u t i l i z e d mainly in the production of nickel wrought products, high-nickel heat and corrosion-resistant a l l o y s , copper-base alloys and in e l e c t r o p l a t i n g , whereas ferronickel and the oxide sinter were used largely in the production of stainless and alloy steels. 20 Table 1.5 Identified World Nickel Resources (Thousand Tons) Reserves Other , Resources Total Resources North America: United States Canada 200 8,700 14,900 12,500 15,100 21,200 Total Africa 8,900 2,300 27,400 6,700 36,300 9,000 Central America and Caribbean: Islands: Cuba Dominican Republic Guatemala Puerto Rico 3,400 1,100 300 14,200 100 900 900 17,600 1,200 1,200 900 Total Europe: U.S.S.R. 4,800 8,100 16,100 13,200 20,900 21,300 Oceania: Australia Indonesia New Caledonia Philippines 5,600 7,800 15,000 5,700 3,200 55,000 31,000 10,600 8,800 62,800 46,000 16,300 Total 34,100 99,800 133,900 South America: Brazil Colombia Venezuela 460 900 3,640 600 700 4,100 1,500 700 Total 1,360 4,940 6,300 2 World total (rounded) 60,000 168,000 228,000 3 World total,seabed nodules — 760,000 760,000 Derived in consultation with U.S. Geological Survey and revised March 1978. 2 Excludes nickel associated with seabed manganese nodules. 3 Based on Holser, 1975 paramarginal resource estimate of 76 billion dry short tons of sea nodules and the NMAB-323, 1976 estimate of average nodule composition of 1 percent copper, 1 percent nickel, 24 percent manganese, and 0.35 percent cobalt. 21 The approximate current use pattern of nickel is shown below: Percentage of Uses A l l Nickel Stainless steels 41 Nickel plate 16 High nickel alloys (monels, coinage) 12 Nimonic alloys (60% Ni), Cr, Fe 10 Iron and s t e e l castings 9 Copper and brass products 3 Others (batteries, magnets, catalysts) 9 100 1.4 Economics, Market Survey and Mine Production of Cobalt 1.4.1 State of the Cobalt Market and Industry World cobalt production from primary sources in 1980 remained unchanged from the previous year at 26,000 ton- nes. Secondary production increased to 1,000 tonnes, up 20 percent. However, real consumption f e l l as much as 15 per- cent bel ow the 1 979 l e v e l . Moreover, as a result of destock4 ing, the apparent consumption f e l l by as much as 35 percent. It i s generally f e l t that the combined effects of lower con- sumption, increased substitution and unprecedented destock- ing reversed the world cobalt demand equation for 1980. 22 Cobalt's use in high performance services p r a c t i c a l l y guarantees i t s continued growth in demand, regardless of high prices. There are two areas in which cobalt seems to be r e l a - t i v e l y immune to substitution in the time horizon of ten to twenty years. These are super alloys and catalysts. Super alloys are primarily used in three major growth areas: 1) Turbines and j e t engines. 2) Surgical implants. 3) Oil f i e l d goods. In applications where the high temperature performance is c r u c i a l , properties imparted by cobalt are so valuable that the price is less important than long-term a v a i l a b i l i t y . The second most important metallurgical use of cobalt is in magnets. Permanent magnets of the alnico variety, which exhibit excellent magnetic properties, have t r a d i t i o n a l l y been a large consumer of cobalt. Recently, however, this demand ' for the metal has been declining due to the advent of hard f e r r i t e s as a substitute. The approximate current use pat- tern of cobalt, i s shown in Table 1 . 6 . 1.4.2 Resources - Reserves of Cobalt Most cobalt resources are only available as by- products of mining for more abundant elements. Only the 23 Table 1.6 Western World Cobalt Market Consumption Shares 9 (Percentage) 1970 1980 19901 Magnetic Alloys 20 16 17 Cemented Carbides 5 7 8 Superalloys and Other Alloys 45 47 40 Ceramics and Enamel - 12 10 8 Chemicals 18 20 27 Estimated 24 Moroccon Bouazzer deposit is of a high enough grade to neces- sitate production of cobalt as the principal metal. Much of the world's i d e n t i f i e d resources are in the form of l a t e r i t i c nickel ores in t r o p i c a l regions, such as the Philippines, Indonesia and New Caledonia. Australian pro- duction from l a t e r i t e s has become an increasingly s i g n i f i c a n t source as has the Cuban. Nevertheless, most cobalt currently comes from sulphide and oxide deposits in Zaire, Zambia, Finland and Canada. Reserves It has been estimated that the world's continental re- sources amount to some 9 mi l l i o n tonnes of cobalt. An e s t i - mation by country is outlined as follows: Zaire 3,090,000 tonnes Zambia 1,704,000 tonnes Cuba 1 ,049,000 tonnes USA 764,000 tonnes Canada 249,000 tonnes Of these resources, the estimated reserves are: Zaire 1,520,000 tonnes 42% Zambia 699,000 tonnes 19% Canada 20,000 tonnes 0.5% from a total reserve of 3 ,618,000 tonnes. To this can be added a further 1 2 mi 11ion tonnes resource hoistable from the sea bed. 25 Thus, i t can be seen that Zaire and Zambia possess 53 percent of.the known, land resource and 51 percent of the proven reserves. 1.4.3 Production of Cobalt The significance of these reserves has been trans- lated into production and the contribution of Central A f r i c a is as follows: During the period 1925-1975 Zaire produced 947,500 tonnes 1933-1975 Zambia produced 51,993 tonnes and more recently: Zaire Zambia Canada World (tonnes) (tonnes) (tonnes) (tonnes ) 1 976 10,700 1 ,700 1 ,900 17,400 1977 10,200 1 ,700 1 ,400 17,435 1 978 12,300 1 ,800 1 ,450 20,090 1979 12,000 3,000 1 ,600 21,800 1980 13,400 3,000 - 23,214 Hence, the Central A.fri can source is established as the producer of some 70 percent of the world's cobalt. 26 1.5 Iron Ore - Market Relevance to Iron Residue 1.5.1 The State of Steel Production World steel production, which had been advancing annually since 1977, was projected by the International Iron and Steel Institute (IISI), to f a l l by 4 percent in 1980 to 717.7 m i l l i o n tonnes. This was primarily due to a state of recession in the iron and steel industry in North America and Europe. However, decline is not the whole picture, because there is an anticipated increase in steel output of about 2.2 percent in eastern bloc countries, and 3.6 percent in develop- ing countries. It was expected that three countries would record s i g n i - ficant production increases in 1980. Italy up by about 9.8 percent, Brazil up by about 10.7 percent and South Korea up by 13.1 percent over the 1979 figures. 1.5.2 The Role of the Iron Residue from Nickel-Laterites in the Direct Reduction of Iron (DRI) Industry Despite a rather uncertain start with regards to scaled up commercial operations, 12 direct reduction iron processes had been applied commercially throughout the world by the end of 1979 (see Table 1.8). 27 Table 1.7 World Raw Steel Production by Region (Million Tonnes) 1977 1978 1979 Projected 1980 Percentage Change 1979-1980 World 7088 712.5 747.4 717 - 4.0 U.S. 132.2 123.8 123.3 100.8 -18.2 Japan 117.1 102.1 111.7 111 .8 0 E.E.C. 155.6 132.4 140.0 128.4 - 8.2 Developing Countries 31.1 46.6 55.6 57.6 + 3.6 Eastern Europe 185.1 211.9 209.4 214.4 + 2.2 Source: International Iron & Steel Institute. Table 1.8 Direct-reduced Iron Processes Applied Commercially at Year-end 1979 Process Type and Equipment Reductant Product Hyl Batch Gas Sponge iron, lump, or pellets. Retort Vertical Shaft Gas Sponge iron or lump. Hoganas Saggers in tunnel kiln Coal or coke Sponge iron or metal powder. Shaft Gas Lump or pellets. Shaft Gas Lump or pellets. High-iron briquets(HIB) Fluid bed Gas Iron briquets. SL/RN Rotary kiln Coal or coke Lump, pellets, fines, or briquets. Kinglor-Metor Rotary kiln Coal or coke Lump or pellets. Esso (FIOR) Fluid bed Gas Briquets Purofer Shaft Gas Lump or pellets. Krupp Rotary kiln Coal Sponge iron. -Chalmers (ACCAR) Rotary kiln Coal/oil/gas Sponge iron. Nippon Steel 1 Rotary kiln Coal or coke Pellets. Sumitomo1 Rotary kiln Coal or coke Pellets. Kawasaki1 Rotary kiln Coke breeze Pellets. designed to produce sponge iron from waste materials accumulated at integrated steel plants. oo 29 The importance of direct-reduced iron (DRI) in the world's steel industry is reaching new heights, as demonstrated by the rapid growth of plant capacities in the past 10 years (see Table 1.9). During the 11-year period 1970-1980, DRI capacity has increased over 970 percent to 20.46 mi l l i o n tonnes per year. It is anticipated that new plants currently under con- struction should take capacity to over 31.0 m i l l i o n tonnes by 1 982. The U.S.S.R. remains the worl d 1 s .leader :in steel production with an estimated output of 152 mi l l i o n tonnes in 1980, which was up 1.9 percent above the 1979 figure. Japan showed an output of 111.8 m i l l i o n tonnes in 1980, which was almost i d e n t i - cal to the 1979 l e v e l . However, the f i n a l figures for the U.S., the world's t h i r d largest steel producer, revealed a decline of approximately 18.2 percent in 1980 output. The greatest concentration of DRI production is located in developing countries, where e l e c t r i c - a r c steel making, with i t s low investment cost, is very popular. DRI is attractive in such nations because they do not have good sources of scrap for supplying iron units to their e l e c t r i c furnaces. There is also a growing interest in DRI among developed countries as a means of d i l u t i n g the undesirable residual elements which con- taminate their scrap supplies. F i n a l l y , a concern of many steel producers in countries, such as the U.S., is that the Table 1 .9 Growth of World DRI capacity (Tonnes) Year Capacity 1970 1 ,906,000 1971 2,754,000 1972 3,324,000 1973 4,650,000 1974 5,775,000 1975 6,650,000 1976 7,810,000 1977 10,775,000 1978 14,325,000 1979 17,625,000 1980 20,460,000 31 rapidly growing number of e l e c t r i c - a r c furnaces w i l l raise the demand and price of scrap beyond economical l i m i t s . A l l of these factors contribute: to the high level of interest in DRI. The DRI process can be c l a s s i f i e d into those that require high-grade iron ore or agglomerates with s t r i c t limitations on impurities, or those that can reject the impurities to produce a high-grade concentrated product. It is anticipated that in.the.former case.treated iron, residue could meet acceptable specifications with regard to n i c k e l , chromium, and alumina for u t i l i z a t i o n in a DRI process. 1.5.3 Outlook for the Future Use of the Iron Residue The future u t i l i z a t i o n of chemi cal 1 y treated i ron residue from nickel l a t e r i t e s in DRI processes appearspromising for the following reasons: 1) The DRI processes should continue to r i s e in popularity throughout the world as increased use is made of e l e c t r i c - arc furnaces. 2) The greatest concentration of DRI production w i l l be located in developing countries, where e l e c t r i c - a r c steel making, with i t s low investment cost, is very popular. 3) The U.S. and other developed nations have embarked on a policy of helping developing countries help themselves. 32 4) The largest known t e r r e s t i a l reserves of nicki1iferous l a t e r i t e s occur in developing countries and in some instances close to existing Bauxite Industries. 5) By-product caustic from the Bayer process can be em- ployed for chromium extraction from the leached ore. 1.6 Review of Existing Methods of Nickel Extraction From Lateri tes 1.6.1 Laterite Processing Methods - General There are several commercial alternatives that exist for the extraction of nickel from nickel l a t e r i t e ores. The choice of the extraction process employed, however depends on several factors such as: 1) The physical and chemical nature of the ore body. 2) The geographical location, which is d i r e c t l y related to supply arid cost of necessary raw materials and energy. 3) The marketability of the end products, may depend on geographical location. Several processes for the recovery of nickel and cobalt have been described, a l l of which, however, show some dis- advantages. Typical examples of these processes are: s'el.ec- 13 14 tive extraction with d i l u t e acids, ' selective 33 conversion of nickelous and cobaltous oxides with gaseous hydrogen c h l o r i d e 1 5 ' 1 6 ' 1 7 and selective reduction of the above-mentioned oxides with carbon monoxide/carbon dioxide mixtures, f o l 1 owed by extraction with ammonia-ammonium carbon- 1 ft ate solutions. 1 9 Van Nes and Heertjes l a t e r described a method of i n - creasing the ratio of the amounts of nickel and cobalt with respect to iron. This was realized by converting nickel oxide and cobalt oxide, almost s e l e c t i v e l y and completely, into the corresponding chlorides. A step-wise process was used in which a gaseous mixture of steam and hydrogen chloride was passed through a f l u i d i z e d bed of the granulated ore, under proper conditions of temperature and concentration. The chlorides were recovered by extracting the chloridized ore with hot water. Roorda and Queneau, 1 1 more recently proposed a process, based on pyrometal1urgical selective reduction of n i c k e l i f e r - ous limonites, followed by aqueous chlorination in seawater. Though conducted only on a laboratory scale, process variables were kept within s p e c i f i c constraints, which could be imposed i n d u s t r i a l l y . The methods employed were simple, f l e x i b l e and amenable to economic, large scale, automated unit operations. 34 S i g n i f i c a n t advantages of this process include: a high recov- ery of nickel and cobalt of over 90 percent, the potential important source of chromium and iron due to chlorine leaching of the ore, rapid dissolution rates which minimize leaching c i r c u i t costs, the use of s a l t rather than fresh water and f i n a l l y , the i n t r i n s i c superiority of chloride systems in solvent extraction with i t s attractive process economics. The processes currently in commercial operation include: pressure acid leaching, matte smelting, production of ferro- nickel by e l e c t r i c or blast furnace and selective reduction roast/ammoniacal leach. 1.6.2 Pressure Acid Leaching In the pressure acid leach process, f i n e l y crushed and ground ore is s l u r r i e d and leached with dilute sulphuric acid at high temperature and pressure, usually in excess of 6 - 2 522°K and 4.1 x 10 Newtom metre" . At these temperatures iron and aluminium form basic sulphates of low residual solu- b i l i t y . The resulting metal containing solution is then separated from the s o l i d residues by means of dewatering thickners and the metal values precipitated by sulphide addi- tion (usually hydrogen sulphide). An alternative route to product recovery is solvent extraction of the pregnant leach liquor followed by electrowinning. 35 Recoveries of desired metal values by acid leach process- ing are usually very high. Furthermore, acid leaching is not very s e l e c t i v e . Magnesia and alumina consume excessive amounts of sulphuric acid, meaning that the feed ore must be low in these compounds or operating costs may become pro h i b i t i v e . Therefore, the application of acid leaching is limited to a narrow range of ore types. It should also be noted that the basic sulphate residue is a source of environmental p o l l u t i o n , because of dissolution of sulphate and metal ions into run-off waters. 1.6.3 Matte Smelting In this process, ground ore is charged to a blast furnace in the presence of coke, gypsum', and limestone. The coke is present as a reducing agent to convert nickel and iron oxides to the metals. The gypsum is reduced to calcium sulph- ide, which reacts with the nickel and iron to produce a metal sulphide matte. In addition to the molten matte, the high temperatures of the process result in the formation of a molten slag by the remainder of the ore and limestone. The i m i s c i - b i l i t y of the slag and matte f a c i l i t a t e s their physical separa- tion. The molten nickel-iron matte is then a i r blown in con- verters to oxidize the iron. Addition of s i l i c a flux produces a secondary iron-rich slag, which can be easily separated from what is now the nickel matte. Further refinement by oxidation roasting followed by reduction result in nickel metal. 36 Matte smelting is used on a wide variety of ores due to the inherent s e l e c t i v i t y of the process. However, the process is very energy intensive, and may be ruled out in certain geographical locations. Sulphur dioxide gas emission is an unwanted, but unavoidable, by-product of this process. 1.6.4 Ferronickel Production Ferronickel production is very similar to that of nickel matte production, the major difference being.that no s u l f i d i z i n g agent is added. As a r e s u l t , the i n i t i a l reduction step produces a molten crude ferronickel and s i l i c a t e slag. The crude ferronickel is then usually oxidized in a converter to remove impurities such as carbon, s i l i c o n , chromium, and phosphorus in a secondary slag. Here, a wide range of ores may be treated, but those high in s i l i c a t e s are desirable because of the a b i l i t y to from a good primary slag. This process is very energy inten- sive, p a r t i c u l a r l y i f e l e c t r i c smelting is used to melt a high- s i l i c a ore ( h i g h - s i l i c a ores have higher temperatures of fusion and require the accurate temperature control that can be provided by e l e c t r i c furnaces). 1.6.5 Selective Reduction Roast/Ammoniacal Leach The main objective in the selective roast/leach route is to reduce the desired metal oxides in l a t e r i t e ores 37 to the respective metals, using a gaseous reductant and a moderately high temperature. The metals are then oxidized in an ammoniacal-ammoniurn carbonate solution. The most important precaution is ensuring that the iron oxides in l a t e r i t e are not reduced to a form that is soluble in the leach. This affects the oxidation and subsequent s o l u b i l i z a t i o n of the nickel and cobalt. Furthermore, re- duction to metallic iron results in excessive consumption of costly reductants. The desired s o l u b i l i z e d metals are re- covered from solution by several established methods. Solvent extraction followed by electrowinning or hydrogen reduction is one such example. Other reductive roast/ammonia leach processes currently in advanced stages of development, but not in commercial 2 0 operation, includes the U.S. Bureau of Mines (USBM) process 3 and the UOP additive process. The USBM uses ammoniacal- ammonium sulphate solution for leaching. 38 Chapter 2 A REVIEW OF PREVIOUS INVESTIGATIONS ON DIRECT ACID LEACHING 2.1 Direct Acid Leaching of Nickeliferous Laterites 13 14 21 22 Many leaching methods ' ' ' have been proposed for obtaining nickel and cobalt from nickeliferous l a t e r i t e s . Although processes considered cover a wide variety of leaching procedures, they a l l have certain aspects in common. These are: 1) A pretreatment stage, usually reduction roasting and/or 2) high temperature-pressure leaching. In view of escalating energy costs, many workers have concentrated their e f f o r t s on bypassing the expensive pre- treatment reduction roast stage. However, most of these i n - vestigations have been on a laboratory scale, and in some cases up to the p i l o t plant l e v e l . Proposed processes for direct extraction of the nickel and cobalt values without roasting include sulphuric, hydro- 2 c h l o r i c , and n i t r i c acid leaching. Due to a lack of 39 s e l e c t i v i t y , and severe corrosive conditions, the use of mineral cac'id.s. has so far been unsatisfactory. Similar chemi- cal behaviour of the oxides results in mechanically complex unit operations, which may be prohibitive. 23 At the present writing, laboratory investigations i n d i - cate that a promising process may be devised for hydrochloric acid leaching of serpentine type ores (high magnesium s i l i c a t e s ) . This was based on the fact that the r e l a t i v e quantities of impurities, which dissolve and consume acid, are much lower in such ores. However, there was less optimism for direct leach- ing of nickeliferous l a t e r i t e (low magnesium). 2.1.1 Direct Leaching of Goethite Mineralogically, the predominant mineral species present in Nickeliferous l a t e r i t e s is goethite (a-FeO(OH), HFe02 or ^e^O^.H^O), with varying amounts of impurity ( S i 0 2 as quartz or s i l i c a t e ) . One of the more s t r i k i n g features of such l a t e r i t e s is the extremely fine p a r t i c l e size of goethite. Usually, 50 percent or more of the p a r t i c l e s is smaller than 10 ym. Due to the highly disseminated nature of such deposits, physical beneficiation of three quarters of the ore's nickel value from goethite is v i r t u a l l y impossible. As a r e s u l t , any attempt to. describe direct acid leaching of ni ckel i f erous l a t e r i t e s is incomplete without considering the leaching of goethi te. 40 Goethite has an orthorhombic s t r u c t u r e ' 1 ' * a n d is aniso- tropic in nature. Figure 2.1 shows i t s crystal structure, which is isomorphous with diaspore (a-Al0(OH),). The oxygen atoms are arranged in a hexagonally close-packed layer, with the iron atoms in octahedral i n t e r s t i c e s . 26 A drastic change in surface area due to p i t t i n g , is known to occur with hydrochloric acid attack of goethite. 27 28 29 Various workers ' ' have shown rates of attack, which i n - crease in the order - Perchloric acid <sulphuric acid h y d r o - chl o r i c acid, when considering solutions of equal normality 30 greater than IN...: Azuma and Kametani subsequently correlated these increasing absolute rates of leaching in di f f e r e n t acids with the increasing complexity constants of the respective anions for f e r r i c ion. Goethite has an activation energy for direct dissolution in various acids similar to that of hematite. Activation energies obtained for d i f f e r e n t acids are l i s t e d in Table 26 27 28 2.1. ' ' No s i g n i f i c a n t change in activation energy values with acid concentration was observed, although dis- solution rates in hydrochloric acid increased sharply up to 28 6N. Surana postulated that both goethite and hematite react by similar chemical mechanisms. V-c=z-aoA—; Figure 2.1 The structure of diaspore and goethite ( a f t e r Bragg). 42 Table 2.1 Activation Energies for Direct Dissolution (Kcal/mol) Material ACID HC1 H2S04 HC104 Massive red a F e2°3 23.4. 1.8.7 + 2.5 Botryoidal a F e2°3 23.2 20.2 + 1.8 21.1 + 5.2 Brazilian a F e2°3 22.7 Synthetic a F e2°3 22.6 18.4 + 2.0 22.1 + 4.5 Goethite 21.1 Synthetic a F e2°3 19.5 - 21.6 Single crystal a F e2°3 20 - .24 Goethite 22.5 19.8 Synthetic a F e2°3 21.9 18.2 19.2 - 21.4 43 2.1.2 Mechanisms of Leaching of Goethite Bath proposed a mechanism by which the surface of goethite is f i r s t assumed to undergo hydration. This is believed to be rapid. In order to describe the mechanisms, a hydrated surface w i l l be represented by: -0-Fe-OH The surface -may be protonated , 0-Fe - O H + H ! O , L 3 r •O-Fe + 2H20 The anion CI"may be adsorbed onthe s u r f a c e s i t e •O-Fe + C1 •O-Fe-Cl and this complex may then desorb, (rate determing step) •O-Fe-Cl ^ FeOCl aq The overall reaction is 0-Fe-OH + H.O + C l " — — > FeOCl n . + 2H90 3 aq 2 and the rate equation i s : { F e ) a a d a q * = K0 K,k dt 2 K l k 2 [ •0-Fe-OHl a ..a J H Cl 44 32 Ahmed and Maksimoir in their study of zero point of charge of hematite in various acids, postulated that the sur- face may undergo protonation to form an aquo complex or be attacked d i r e c t l y by the anion. A dual mechanism may thus be operative. The surface may f i r s t undergo a protonation as previously descri bed. -0-Fe-OH + H30 '1 •0-Fe(H20) + H20 This aquo complex may either desorb k •0-Fe(H20) U 0 - F e ( H 2 0 ) ; q or adsorb an anion X •0-Fe(H20) + X' •0-Fe-(H20)X The resulting complex then desorbs •0-Fe(H20)X - 0-Fe(H 20)X a q Assuming once again that desorption is the rate l i m i t i n g step the overall reaction may be written • O.Fe-OH + 2H3;0 + :x" > 0-Fe(H 20) a q + 0-Fe(H 20)X a q + 2H20 45 and the rate w i l l be given by: d [Fe] dt d.[0-Fe(H20) + ] + d.[0-Fe(H 20)X] dt dt = k ]K ] M -0-Fe-OH] a + + k ^ ! ^ [| -O^Fe-OH] a + a • H X K, f I - O - F e - O H T a , (k,+k0K0a ) 1 1 Is H 1 L L X~ In this case, the anion C l " is a good complexer with iron, and K2 w i l l be large. As the anion w i l l rapidly adsorb -0-Fe-H20 w i l l be low, and the rate equa-on the surface tion is simplified to, d [Fe] dt k„K 0K n J •0-Fe-OH a + a H Cl where K K a , a a H + 61 k.j K2 ((surface sites remaining: constant) It is interesting to note that this expression is identi cal to that obtained previously, which is known to f i t the experimental results for hydrochloric acid. At higher HC1 concentration, chloride ions may adsorb on the surface, and the equilibrium may be written: •O-Fe + C l " - O-Fe-Cl 46 At high enough hydrochloric acid concentration, the surface sites become saturated with anions. Dissolution may then proceed by a further protonation of the activated surface •0-Fe-Cl + H30 K3 ̂  OH-Fe-Cl + H20 This would be followed by the rate-determining desorption of the f e r r i c ion complex k, OH-Fe-Cl ^* FeOHCl + aq where is the reaction rate constant. The rate equation in strong acid solution is then d [ F e J ^ = k 2 K 3 d t •0-Fe-Cl a + J u The rate of leaching now depends on K3, which may depend on the "activation" of the surface by each anion, and this in turn could be related to the complexing power of anions for f e r r i c ions. 2.1.3 Direct leaching of Low Magnesium Limonites 23 Laboratory experiments involving the leaching of nickel l a t e r i t e ores in hydrochloric acid have shown that the extraction of nickel and iron is closely related to the 47 weight of goethite dissolved. Very l i t t l e extraction occurs at ambient temperatures, usually less than 10 percent. Dis- solution of the goethite increases d r a s t i c a l l y from about 348°K, and complete dissolution may be expected at 423°K for a leaching time of one hour, at hydrochloric acid concentra- tion greater than 2M. The general conclusions are that, at low temperatures, acid concentration is the dominating factor in the percentage metal extraction. However, leaching at higher temperatures, 353°K and over, results in almost the same percentage extrac- tion irrespective of acid concentration. However, maximum nickel extraction is only achieved with complete dissolution of the ore. 2.2 Hydrothermal Pr e c i p i t a t i o n in Solutions Containing Mainly Ferric Ions 2.2.1 Hydrolysis of Metal Species at Elevated Temperatures Very l i t t l e is known regarding the hydrolysis reactions of inorganic ions at higher temperatures. This is due to a dearth of thermodynamic data for aqueous solutions in the temperature range 373 - 573°K. Many of the older data on the temperature c o e f f i c i e n t s of simple cations have been obtained before i t was recognized that dimerization and 48 polymerization had to be considered. 33 34 The correspondence p r i n c i p l e of Criss and Cobble ' has therefore been used to more accurately predict thermodynamic e q u i l i b r i a at elevated temperatures. In e f f e c t , entropies and heat capacities of ionic species partic i p a t i n g in a given reaction are u t i l i z e d in computing the standard free energy changes at temperatures other than 298°K. A useful method of presenting thermodynamic data related to the hydrolysis of metal species at elevated temperatures, 3+ " is the use of temperature/pH diagrams, such as the Fe —H^O 35 system shown in Figure 2.2. The pr e c i p i t a t i o n temperature and the composition of the precipitate formed were shown to be dependent on the pH of the i n i t i a l nitrate solution. A l - though pre c i p i t a t i o n of hematite from f e r r i c chloride solu- tion is not expected to show identical behaviour, the e q u i l i - brium datagram;may serve as a guide. 2.2.2 Precipitation of Iron as Ferric Oxide An examination of the temperature versus pH plot suggests that iron can be removed from leach solutions by hydrolysis at high temperatures of about 473°K. In practice however, i t has been found that in the absence of j a r o s i t e precipitants , the extent of hydrolysis and pr e c i p i t a t i o n of 49 50 f e r r i c oxide at 473°K is not always s u f f i c i e n t for the iron content of the f i n a l solution to be lowered to a l e v e l , which is of practical importance. There are two methods of increasing the extent of hydro- l y s i s and p r e c i p i t a t i o n of f e r r i c oxide. One method is to heat the leach solution to well over 473°K. This however results in an increase in operating pressure, and increased corrosion problems. A l t e r n a t i v e l y , an increase in the extent of pr e c i p i t a t i o n is accomplished by a decrease in the 'free acid' level of the hydrolyzed liquor. In considering s t a b i l i t y relations of the compounds 3+ a~F e2^3' and a-FeO(QH) in aqueous solutions of Fe at various temperatures, the general form of the pr e c i p i t a t i o n reaction can be written: 2M X + + yH2.0 = HO- (y-x)H 20 + 2xH + x+ Considering the systems in the form M —M„0 —H„0 leads L. X (— to the p o s s i b i l i t y of representing reactions on temperature- pH plots of the type shown in Figure 2.2. The standard iso- therm may be written as: 51 AG o 6 -RT In aM 2Cy (y-x)H 20 a 2x T H + y and s impli f i ed to : b 2xpH = AG T - 2 .Log m x+ 2 Log Y x+ M M 4.575T The pH at which various compounds are in equilibrium with aqueous solutions of the metal ions can thus be calculated as a function of temperature. However, at temperatures above 373°K, very few values of a c t i v i t y c o e f f i c i e n t (y±) are known in order to calculate the a c t i v i t y of the metal ion. 37 Lietzke and Stoughton have shown rather conclusively that a c t i v i t y c o e f f i c i e n t s can be correlated by various Debye-Huckel expressions up to 523°K as easily as they are at room temperature. This method was further developed by 38 Cobble to predict s o l u b i l i t i e s of salts in water up to 523°K. These Debye-Huckel expressions show a decrease in a c t i v i t y c o e f f i c i e n t with temperature, which can become very s i g n i f i - cant at higher concentrations. 39 Meissner, Kusik and Tester have described a method, which u t i l i z e s vapour pressure measurement and the Gibbs- Duhem equation. However, this did not serve the purpose of 52 calculating the value of the a c t i v i t y c o e f f i c i e n t of f e r r i c chloride, due to the u n a v a i l a b i l i t y of data on i t s vapour pressure. An attempt was therefore made to establish a r e l a t i o n - ship between the a c t i v i t y of f e r r i c ion in equilibrium with hematite, and goethite. The s o l u b i i i t i e s of iron compounds at various elevated temperatures were estimated from a knowledge of the e q u i l i b r i a constants, which in turn were calculated from the standard free energy changes for each reaction according to equations: AG. RT In K, where AG. = A G298 + A C p j o AG^ AC. 298 (T-298) A S 2 9 8 (T-298) - T A>Cp In T 298 298 AG2g --= standard free energy of reaction at 298°K = standard free energy of reaction at T°K AS 2gg= standard entropy change of reaction at 298°K. C M T 298:: the average heat capacity change for the reaction between 298° and T°K. 53 2.2.3 Hydrothermal Precipitation of Iron III Compounds The degree to which leach solutions can be freed from f e r r i c ion is dependent on the s o l u b i l i t y of the pre- 40 c i p i t a t e . Feitknecht and Schindler have c r i t i c a l l y reviewed the s o l u b i l i t y relationships of iron III oxides and hydroxides at 298°K, and their findings are summarized on the basis of the following aging scheme: > FeQ(OH) AMORPH.. Fe(0H)3. (ACTIVE)-^- -> AMORPH. Fe(0H)3 (INACTIVE) * F e2°3 They proposed that freshly! preci pi tated active amorphous hydroxide slowly converts to goethite. In addition, a s o l i d transformation occurs resulting in a more stable amorphous 41 hydroxide. Biedermann and Schindler have shown that the complete transformation requires about 6-8 days, but less 42 than one day at 373°K. Goethite is stable at room termpera- 43 ture but dehydrates to hematite above about 403°K. It is anticipated, that in leaching experiments at about 423°K, hematite is nucleated homogeneously by the hydrolysis reaction F e 3 + + 3 H20 = 1 Fe 20 3 + 3 H + Ferric hydroxide is.unstable under such conditions of ac i d i t y and temperature, and w i l l not be further considered. 54 2.2.4 Hematit.e-GoethT.te Relations in Acid Solutions 44 Tunell and Posnjak cited an experiment in which goethite in 0.1 MCI solution was converted to hematite at 373°K. 45 This decomposition required a few weeks. Gruner, usingigoethite in d i s t i l l e d water, took about ninety days to effect this decom- position to hematite at temperatures of over 473°K. The qua l i - 46 tative e f f e c t of pH on the decomposition of goethite, is that the decomposition temperature is near 373°K in acid solutions and above 423°K in alkaline solutions. In l i g h t of the fact that the decomposition of goethite to hematite is k i n e t i c a l l y a slow process, i t is imperative that complete dissolution occurs to produce the f e r r i c chloride intermediate for subsequent hydrolysis to hematite. Hydrolysis 47 of f e r r i c chloride is known to occur at as low as 393°K. 2.2.5 The Effect of pH on the S o l u b i l i t y of Iron III Compounds Without Complexing With knowledge of the working pH range during leaching experiments, and careful scrutiny of the iron E - pH 51 3+ diagrams, only the F e a q species was considered. For a r e l a t i v e l y complete treatment of the iron-water r e l a t i o n s , over 50 equilibria reactions are usually required to describe the s o l u b i l i t y of a single compound. Furthermore, i t should be pointed out that not only dissolved species for which free 55 energy values are available should be considered but, for example, Fe(0H)2 or undissociated dissolved f e r r i c hydroxide, because i t may be an important contributor to iron s o l u b i l i t y (J. Winchester, personal communication). Neverthe- less, an attempt was made to show the relations of f e r r i c ions, 3+ Fe , to the oxides and hydroxides of iron, aq J Table 2.2 shows values of the equilibrium constant obtained using equation 2:1, together with the associated pH, which were calculated using equation 2.4 and assuming that the a c t i v i t y of f e r r i c ions in solution was unity. It should be noted that lower pH values would be obtained for the hydrolysis reaction at higher concentrations, because Debye-Huckel expressions indicate a decrease in 35 a c t i v i t y c o e f f i c i e n t with temperature. 2.2.6 Considering the Ferric Ion--Hematite Equilibrium The standard free energies for the hydrolysis of hematite at various temperatures (see Appendix A.3) -.were.- used to calculate the equilibrium constant via the relationship: Fe 3+ aq + 3 H90 2 L aq o Log K -AG T . . . 2 .1 2.3 RT 56 Eliminating Fe^O and H90 from the constant because the i r a c t i v i t y was assumed to be close to unity. 3 - a + K = — ...2.2 a r 3+ Fe and Log K = 3 Log a - Log a ^ H+ Fe therefore .2'.':-3 Log K = -3 pH - Log a 3 + ...2.4 Fe The log of the a c t i v i t y of Ferric ion in equi1ibriurn with Fe^Og is seen to be a linear function of pH, with a slope of c minus 3. As a r e s u l t , i f a temperature and f e r r i c ion a c t i v i t y are stipulated, then the pH is fixed. At a leaching temperature of 423°K Log [ F e 3 + ] = -4.8 - 3 pH . . .2.5 The pH associated with various a c t i v i t i e s of f e r r i c ion in equilibrium with hematite were calculated using equation 2.;5,. A c t i v i t y values higher than 10 _ 1 f a l l in such a high concentra- tion range that a marked departure from molalities can be expected. 57 Table 2.2 Equilibrium Data , for the Fe — F e ? 0 —H ?0 System Temperature °K Calories Log K PH 298 -2550 1 .87 ' -0.62 333 -4774 3.13 -1.04 373 -6621 3.88 -1.29 423 -9295 4.80 -1.60 473 -11880 5.49 -1.83 58 2.2.7 Considering the Ferri c Ion >T-Goethi te Equilibrium F e 3 + + 2H20 = FeOOH + 3H+ ;i The free energy and pH values shown in Tables 2.4 and 2.5 were obtained in the similar manner described in section 2.2.6, and making the same assumption. At a leaching temperature of 423°K. Log [ F e 3 + ] = 4.44 - 3pH 2.2.8 Considering the Ferri c Ion r-Ferri c Hydroxide Equi1i bri urn 3 + Fe' + 3H20 = Fe(0H) 3 + 3H At a leaching temperature of 423°K Log [ F e 3 + ] = -0.13 - 3pH In constructing a pseudo phase diagram for the 3+ Fe — ^ e 2 ^ 3 — ^ 2 ^ s y s t e m ' t n e following expression is u t i l i - zed to calculate the equilibrium pH at various temperatures and a c t i v i t i e s : pH = 1 3 log K - log a .. - Fe"5 The same expression was u t i l i z e d to calculate the equi librium pH at various temperatures and a c t i v i t i e s for the 3+ Fe —FeOOH — H20 metastable system. 59 Table 2.3 The pH Corresponding to Various A c t i v i t i e s of Ferric Ion in Equilibrium with Hematite Log [ F e 3 + ] PH 0 -1 .60 -1 -1.27 -2 -0.93 -3 -0.60 -4 -0.27 -5 -0.07 60 Table 2.4 Equilibrium Data for the Fe — FeOOH —H 20 Metastable 1 System. Temperature ° K A G ° T Calories Log K P H 298 -2500 1 .83 -0.61 333 -4250 2.79 -0.93 373 -6185 3.62 -1.21 423 -8598 4.44 -1.48 473 -10859 5.02 -1.67 61 Table 2.5 The pH Corresponding to Various A c t i v i t i e s of Ferric Ion in Equilibrium with Goethite Log [ F e 3 + ] PH 0 -1 .48 -1 -1.15 -2 -0.81 -3 -0.48 -4 -0.15 -5 -0.19 62 fable 2.6 Equilibrium Data for the Fe — Fe(OH), — FLO Metastable System. Temperature °K AG0 j Calori es Log K pH 298 4700 -3.45 1.15 333 3176 -2.08 0.69 373 1585 -0.93 0.31 423 -250 0.13 -0.04 473 -1826 0.84 -0.28 63 Table 2.7 The pH Corresponding to Various A c t i v i t i e s of Ferric Ions in Equilibrium with Ferric Hydroxide Log [ F e 3 + ] pH 0 -0.04 -1 0.29 -2 0.62 -3 0.96 -4 1 .29 -5 1 .62 64 Table 2.8 Theoretical Equilibrium pH at Various Temperatures 3+ and A c t i v i t i e s for the Fe — F e 9 0 „ — H 9 0 System Temperature °K Log K Activity of Ferric Ions 1M 0.5M 0.2M 0.1M PH PH PH PH 298 1.87 -0.62 0.52 -0.39 -0.26 333 3.13 -1.04 -0.94 -0.81 -0.71 373 3.88 -1.29 -1.19 -1.06 -0.96 423 4.80 -1.60 -1.49 -1.37 -1.27 473 5.49 -1.83 -1.73 -1.59 -1.49 65 Table 2.9 Theoretical Equilibrium pH at Various Temperatures 3+ and A c t i v i t i e s for the Fe —FeOOH—H 20 Metastable System Temperature °K Log K Activity of Ferric Ions 1M 0.5M 0.2M 0.1M PH PH PH PH 298 1.83 -0.61 -0.51 -0.38 -0.28 333 2.79 -0.93 -0.83 -0.70 -0.60 373 3.62 -1.21 -1.11 -0.97 -0.87 423 4.44 -1.48 -1.38 -1.25 -1.15 473 5.02 -1.67 -1.57 -1.44 -1.34 323 373 423 473 Temperature (K) c n F i g u r e 2 . 3 Free energy versus temperature f o r i r on ( I I I ) h y d r o ! i s i s r e a c t i o n s 0 1 Figure 2.4 Influcence of pH on the solubi l i ty of iron (III) hydroxides and oxide at 423°K.  69 2.3 Thermodynamic Considerations Underlying the Hydrothermal Treatment.ofiNickel Laterite With Ferric Chloride 43 It has been shown that the dir e c t conversion of goethite to hematite is a r e l a t i v e l y slow reaction. This can be thermo- dynamically explained by considering the free energy versus temperature relationship for hydrolysis reactions (see Figure 2.3). F e 2 0 3 . H20 (+2500 calories) + HC1 -50 calories si ow » Fe 20 3 + H20 •HC1 (-2550 calories) FeCl 3 fast At 298°K, the conversion of goethite to hematite proceeds with a reduction of free energy of only approximately -50 c a l o r i e s . Figure 2.3 also i l l u s t r a t e s that the hydrolysis of hema- t i t e and goethite have similar thermodynamic driving forces, which increase with temperature. The hydrolysis of f e r r i c hydroxide appears to be the.rmodynamical ly favored under standard conditions. Its presence however, is not expected under experimental conditions because of slow ki n e t i c s . 3 6 Generally, the extent of hydrolysis is enhanced by an i n - crease in temperature and a decrease in the free acid level of the hydrolyzed liquor. 70 The overall conclusion therefore, is that the hydrolysis of hematite has the largest thermodynamic driving force, and this provides a p o s s i b i l i t y for a fast conversion of goethite to hematite under conditions where goethite can dissolve rapidly. 2.3.1 The Effect of Chloride Complexing Due to the complexing power of chloride ion for f e r r i c ions, the effect on f e r r i c ion s o l u b i l i t y should also be considered. Each complex predomn^nste's under dif f e r e n t conditions of pH and f e r r i c ion a c t i v i t y , and i t s e f f e c t on s o l u b i l i t y can be estimated from equilibrium considerations. Figure 2.4 i l l u s t r a t e s the influence of pH on the solu- b i l i t y of hematite, goethite and f e r r i c hydroxide. It should however be pointed out that these data are in terms of the f e r r i c ion a c t i v i t y ; the f e r r i c ion a c t i v i t y is dependent on the degree to which i t is complexed by chloride ions as well as by departures from i d e a l i t y in strong solutions. 2.3.2 Temperature/pH Consideration The pseudo-JDhase diagram of Figure 2.5 cannot serve to indicate the temperature of transformation of goethite to hematite. This must be done experimentally, by i d e n t i f i c a t i o n of the compounds precipitated a t various 71 temperatures and pH. Furthermore, the curves were obtained by assuming con- stant values of f e r r i c ion a c t i v i t y with increased tempera- ture. In fact, one can only assume the concentration remains constant, because., the a c t i v i t y c o e f f i c i e n t decreases with temperature. As a r e s u l t , lower values of pH were calculated using the equation: o 2xpH = AG - 2 Log [ M x + ] * 4:575T instead of: o 2xpH = AG - 2 Log m Y, - 2 Log 4.575T + - c L u y T While the pseudo-phase diagram may serve as a guide in describing the temperature-pH relationship, i t is inaccurate, due to the u n a v a i l a b i l i t y of data on the a c t i v i t y c o e f f i c i e n t as a function of temperature. * Concentration (M) 72 Chapter 3 EXPERIMENTAL 3.1 Mineralogical Investigation 3.1.1 ENERGY DISPERSIVE X-ray ANALYSIS VIA SEM The scanning electron microscope was u t i l i z e d in determining the nature of a wide variety of pa r t i c l e s found in the mineral sample supplied by Sh e r r i t t Gordon Mines Limited/Sherritt Research Centre, Fort Saskatchewan. Three discrete types of particles were i d e n t i f i e d . 1) Quartz or Si 1icate (Pale-yellow p a r t i c l e s , which were r e l a t i v e l y the 1argest observed.) 2) Chromite (Black or dark-brown p a r t i c l e s , which were magnetically separated and smaller in size than those of quartz.) 3) Goethite (Redish-brown p a r t i c l e s of very fine s i z e , which possessed the highest nickel content.) 73 3.1.2 X-ray D i f f r a c t i o n Analysis X-ray d i f f r a c t i o n patterns of the ore sample, precipitated hematite and magnetic pa r t i c l e s were obtained. The 2Q: and re l a t i v e intensity values were compared with those of the respective minerals, as given in the ASTM card. These values are reported i n Ta'bl.es B . 1 , B.2 and B.3 of Appendix B. The ore sample was found to be composed mainly of goethite, which was .poorly c r y s t a l l i n e . . There was no indica- tion of a discreet nickel mineral such as gar n i e r i t e . The presence of an appreciable amount of hematite was detected. In each case, the re l a t i v e i n t e n s i t i e s showed some variation. There was excellent agreement on the 2;e values of the peaks for hematite and chromite but not for goethite. 3.2 Quantitative Chemical Analysis Several samples of nickel l a t e r i t e were digested u t i l i - zing two standard methods of dis s o l u t i o n : 1) Hydrochloric/Nitric acid solution in the ratio 3:1 2) Perchloric acid dissolution after removal of organic matter with n i t r i c acid. Determinations were conducted by atomic absorption 74 spectrometry. The results obtained by the above methods were compared with those obtained from general testing laboratories. Total Analysis of the ore is shown in Table 3.1. 3.3 Apparatus Design An a l l titanium autoclave of 100 mis capacity was u t i l i - zed. It was placed in a horizontally shaken heating jacket, which was connected to a thermistor temperature co n t r o l l e r and voltage regulator. A potentiometer with a thermocouple was also employed to occasionally monitor the temperature. The main features of the apparatus are shown in Figure 3.2. 3.4 Pressure Leaching Experiments 3.4.1 Experimental Procedure The experimental procedure consisted of the follow- ing steps: 1) The powdered sample (usually 16 gms) was added to 40 mis of solution. 6 2 2) A pressure of 4.1 x 10 Newton metre" nitrogen was em- ployed to minimize vapour transport into tubing con- nect ions and valves. 3) The thermistor was set at the required temperature. 75 e 3.1 Chemical Analysis of the Nickeliferous Laterite Element Percentage wt. Obtained Percentage wt. Obtained (General Testing Lab) Fe 44.5 2 44.60 Ni 1 .253 1 .24 Co 0.15 0.14 Cr 2.24 2.28 Mn 1 .08 1 .08 Al 4.70 4.66 Mg 2.30 2.32 Ca - 0.12 Si - 2.72 LOI 9.68 9.68 Marinduque Nickle Mines, Philippines suggested 43.9 suggested 1.27 Source : Supplier Supplier 76 M E T 1 0 P O T Figure 3.1 The leaching apparatus, showing from (L to R): the potentiometer, the titanium autoclave mounted on a shaking device, voltage regulator and the thermister control 1er. F i g u r e 3.2 The p r e s s u r e f i l t e r d e v i c e . 78 4) Shaking was started with a constant heating rate cor- responding to the maximum variac setting. 5) Ten degrees below the required temperature, the variac setting was adjusted to that corresponding to the re- quired calibrated temperature. 6) The autoclave was quenched in a pail of water exactly one hour after having reached the desired leaching temperature. 3.4.2 Liquid/Solid Separation Figure 3.1 shows the pressure f i l t r a t i o n device, which was employed to speed up f i l t r a t i o n , and at the same time prevent any evaporation of the f i l t r a t e . Although f i l t r a t i o n was much faster with a buchner funnel and f i l t e r pulp, there were two disadvantages. F i r s t l y , recovery of the iron residue was d i f f i c u l t when mixed with the pulp. Secondly, the hot f i l t r a t e evaporated to a certain extent due to the low pressure on the vacuum side of the f i l t e r i n g medium. In order to reduce the contact time between the iron residue and the acidic f i l t r a t e , a laboratory centrifuge was v employed. After about 5 minutes of centrifuging , the solu- tion was decanted and the solids immediately swamped with water. 79 3.5 Analytical Methods 3.5.1 Atomic Absorption Analysis Samples of f i l t r a t e were pipetted and diluted to the appropriate concentration for accurate measurement of their contents by atomic absorption spectrometry. Samples of the corresponding residues were digested and analyzed after thorough washing. A mass-balance technique was adopted as a means of checking the accuracy of the determinations. Nickel, cobalt, iron, manganese, magnesium, aluminium and chromium were determined on a Perkin-Elmer 306 atomic absorp- tion instrument. For a l l analyses, standards were f i r s t made up to approximate compositions of the unknown f i l t r a t e solu- tions. It was hoped that this procedure would compensate for the flame matrix effects of the f i l t r a t e solutions. F l e x i b i l i t y in sample d i l u t i o n was made possible by rotating the atomic. absorption burner head, thereby changing the flame path length through which the beam was passed. ;Using this procedure, the linear portion of each element's absorp- tion versus concentration curve was increased by approxi- mately 20 times. (That i s , by having the burner normal to the lamp beam line rather than co-linear with i t . ) Fewer sample dilut i o n s were required by employing this technique. 80 3.5.2 Determination of "Free Acid" The pH .at which the precipitation of cations in- ch! oride solutions.occur were f i r s t determined, by adding metal chloride solutions to the same aliquot portion of standard hydrochloric acid, and t i t r a t i n g with sodium hydroxide. An a u t o t i t r a t o r in the i n f l e c t i o n mode of operation pro- duced the curves shown in Figure 3.3. These curves served to indicate the relationship between the pH value at which hydro- l y s i s occurs and volume of t i t r a n t . A symmetric t i t r a t i o n curve is produced when the indica- tor electrode is reversible, and when there are an equal number of t i t r a n t reagent and reactant species in the equiva- lence equation. In a simple acid-base pH t i t r a t i o n , the pH t i t r a t i o n curve for equation (3.1 ) i s symmetri c. A t i t r a t i o n error can however, be generated i f the solu- tion contains a species, which chemically interferes with the t i t r a n t or reactant and in so doing, distorts the symmetry of the t i t r a t i o n curve (equation (3.2)). This error is greatest when the equilibrium constant for the reaction between the reactant (or the t i t r a n t ) and an in t e r f e r i n g species is H f + OH" = H20 F e 3 + + 30H" = Fe(0H) 3 ...(3.1) ...(3.2) 81 Table 3.2 pH Values of Hydroxides in Equilibrium with Their Metal Ions Hydroxides PH 0.5M[Mn+] Published Data lM[M n +] .001 M[M n +] Fe(0H) 3 2.1 1 .61 2.61 Al(0H) 3 3.5 3.22 4.22 Cr(0H) 3 4.5 3.93 4.93 Ni(OH) 2 7.2 6.09 7.49 Mn(0H)2 8.4 7.65 9.15 Mg(0H)3 9.5 8.48 9.98 Fe(0H) 2 6.5 6.65 8.15 82 co CM X CL Na OH solution, cm F i g u r e 3.3 pH t i t r a t i o n c u r v e s o f metal c h l o r i d e s i n hydro- c h l o r i c a c i d v e r s u s sodium h y d r o x i d e . 83 close to that of the reaction between the t i t r a n t and the reactant. If the interference reaction is with the t i t r a n t a second end point may be seen. In acid-base ph t i t r a t i o n s , an acid strength difference of six orders of magnitude (ApK = 6) is required for the clean separation of two end points (Basset et a l . , 1978). However, the magnitude of this type of t i t r a t i o n error depends heavily on the r e l a t i v e quantities of the species involved. Table 3.2 shows the pH at which the hydrolysis of several chlorides occur. It also shows the variation in pH values with the r e l a t i v e quantities of species involved in the Mn +—M:(:0H)n equilibrium. Figure 3.3 i l l u s t r a t e s how close the pH value corresponding to neutralization of the "Free Acid" is to that for the hydrolysis of f e r r i c ions in solution. For this reason, direct t i t r a t i o n of the f i l t r a t e with 2N NaOH solution, with the auto- t i t r a t o r in the derivative mode, re- sulted in very broad peaks. In most cases, the points of i n f l e c t i o n on pH t i t r a t i o n curves were also unsatisfactory. An alternative method was therefore devised in determin- ing the "Free Acid" concentration of the f i l t r a t e . In this procedure, an accurately measured volume of standard 2N hydro- ch l o r i c acid was added to an aliquot of the f i l t r a t e . Any extra volume of 2N sodium hydroxide, beyond that which was required to neutralize the acid, was obtained by finding the di fference. 84 Chapter 4 RESULTS 4.1 The leaching of Nickel iferous Laterite in Aqueous Ferric Chloride Al l leaching tests were performed in the 100 ml capacity titanium autoclave, the temperature of which was controlled to + 2°K. The extractions of n i c k e l , cobalt, manganese and chromium were determined as a function of: 1) Temperature ranging from 373°K to 473°K. 2) Concentration of f e r r i c chloride ranging from 0 . 5 M " to 4 . ' 0 M . , 3) Pulp density ranging from 100 g/1 to 400 g/1 . 4.1.1 Effect of Temperature Using 1 and 2 molar f e r r i c chloride solutions with a constant 1 hour reaction time, the effect of reaction tem- perature is shown in Figure 4.1. In general, metal extraction increased with increasing temperature (see leaching results of Table 4.1). At 423°K, nickel extraction was over 90 percent, provided the concentration of f e r r i c chloride was greater than  Tab!e 4.1 Metal E x t r a c t i o n From N i c k e l i f e r o u s L a t e r i t e with Aqueous F e r r i c C h l o r i d e So lu t ion Type FeCl 3 (M) Temp. °K Pulp Density 9/1 Metal Cone. 1r Fi 1trate gpi % Ni E x t r a c t i o n Ni Co Cr Mn 0.5 448 100 1.2 0.15 0.16 1 .0 96 ; 0.5 448 200 2.1 0.30 0.19 1 .9 84 1.0 373 200 0.6 1.10 0.14 0.7 24 1.0 398 200 1.2 0.20 0.21 1 .3 48 1.0 423 200 2.4 . 0.30 0.38 2.0 96 1.0 448 200 2.5 0.30 0.21 2.2 98 1.0 448 300 3.4 0.45 0. 20 3.0 91 2.0 373 200 0.7 0.10 0.19 0.8 28 2.0 398 200 1.2 0.23 0.28 1 .5 50 2.0 423 200 2.4 0.30 0.50 2.1 96 2.0 448 200 2.5 0.30 0.40 2.2 98 2.0 448 300 3.6 0.45 0,. 28 3.1 96 1.0 373 100 0.31 25 1.0 423 100 1 .24 99 1.0 473 100 1 .24 99 1.0 523 100 1.24 99 1.0 423 400 4.3 88 2.0 423 4C0 4.9 0.60 0.45 4.0 96 87 1M. The concentration of chromium in the f i l t r a t e increased with increasing temperature up to around 423°K but decreased at higher temperatures. This behaviour was suspected to be due to the pr e c i p i t a t i o n of a ferric-chromium complex. 4.1.2 Effect of Concentration Figures 4.1 and 4.2 show the effect of f e r r i c chloride concentration on nickel extraction. At low tempera- tures, 348°K, the percentage metal extraction was more depen- dent on f e r r i c chloride concentration whereas leaching at higher temperatures resulted in almost the same percentage extraction irrespective of concentration. 4.1.3 Effect of Pulp Density The pulp density of the leach did not appear to have any s i g n i f i c a n t effect on metal extraction, provided enough reaction time was allowed and s u f f i c i e n t chloride ions were available. With pulp densitites of 100 g/1, 200 g/1 and 400 g/1, nickel extraction at 423°K, with 1M f e r r i c chloride solution was 98 percent, 96 percent and 90 percent respectively. Figure 4.3 i l l u s t r a t e s this effect in the form of a curve, which is nearly horizontal. Figure 4.2 shows low values of percentage nickel extraction, which were obtained at pulp densities of 300 g/1 and 400 g/1. Such low percentages were FeCI 3,(M) Figure 4.2 Effect of FeCL-, concentration on nickel extraction at 448°K. 100 200 300 400 Pulp density, g/1 Figure 4.3 Effect of pulp density on" nickel extraction with IMMFeClg at 423°K. 90 to be expected because of the low concentration of f e r r i c chloride solution employed and the subsequent u n a v a i l a b i l i t y of s u f f i c i e n t chloride ions to react with metal values con- tained in the ore. 4.2 The Leaching of Nickeliferous Laterite in Ferric Chloride/Hydrochloric Acid Solutions It was postulated that the presence of higher amounts of "free acid" would not only increase metal recovery at lower temperatures, but also reduce the consumption of f e r r i c chloride - a more expensive reagent. Various a c i d i f i e d f e r r i c chloride solutions were prepared from a stock solution of hydrochloric acid, and used as the l i i x i v i a n t . As shown in Figure 4.4. nickel extraction was more sensi- t i v e to f e r r i c chloride concentration than to that of hydro- ch l o r i c acid. In fact, i t was apparent that the presence of acid s l i g h t l y decreased nickel extraction and greatly increased the amount of iron leached into solution. The l a t t e r , however, occurred only when there was an excess of' "free acid" over that which was required to react with a l l the metal consitituents of the l a t e r i t e . It should be noted that maximum nickel extrac- tion in the presence of "free acid" occurred only with a f e r r i c chloride concentration greater than 1.5 M. 100 25 0 Cone of FeCI3 O 0.5 M • 1.0 M O |.5 M A 2.0 M i 0 2 HCI (M ) Figure 4.4 Effect of "Free Acid" concentration at several FeCl, concentrations on nickel extraction at 423°K and a pulp density of 400 g/1 CO Tab!e 4.2 Leaching Results In the Presence of "Free Ac id" at 423°K S o l u t i o n T /pe Pulp Dens 1ty .9/1 N1 Cone. 1n, F11 t rate gpl % Nickel Ex t rac t ion Fe Cone. In L i x l v l a n t gpl Fe Cone. 1n F i l t r a t e gpl "Free A c i d " In FI 1 t rate "Free A c i d " Consumed HCl(M) FeCl 3 (MJ HCI(M) 0.24 2 .00 200 2 1 82 13 23 0. 17 1.83 0.45 1 .93 200 2 2 88 25 33 0. 17 1.76 0.86 1 . 79 200 2 3 90 48 53 0. 1 1. 1.68 1 .57 1 . 50 200 2 4 96 88 88 0. 05: 1.45 0. 24 2 .00 400 3 0 60 1 3 fi.3 0. 09 1.91 0.45 0 .55 400 2 7 54 26 0.2 0. 08- 0.47 0.45 0 .98 400 2 9 58 25 1.2 0. 09 0.89 0.45 1 .93 400 3 8 76 25 14 0. 09 1 .84 0. 58 3 .28 400 4 3 86 33 44 0. 09 3.19 1.57 1 . 50 400 4 7 94 88 62 0. 05 1.45 0.47 3 .39 400 4 4 88 26 38 0. 09 3.30 0 .86 1 3 .16 400 2 5 50 48 75 0. 04 3.12 0.86 3 .16 400 4 5 90 48 55 0. 02 3.14 1 .52 2 .84 400 4 6 92 85 85 0. 02 2.82 1.92 3 .00 400 4 2 84 107 110 0. 09 2.91 Leaching time 15 minutes. Leaching In a la rge autoclave (2 l i t r e s capac i ty ) with very l i t t l e a g i t a t i o n . to ro 93 Table 4.2 shows the leaching results obtained using f e r r i c chloride/hydrochloric acid solutions. Higher concentra- tions of iron in the f i l t r a t e as compared to the corresponding l i x i v i a n t were due to dissolution of precipitated hematite. The concentration of iron remained the same, whenever the acid concentration employed was equal to the theoretical amount, which is required to react with the metal constituents of the ore (as described in Section 4.4). The use of acid concentra- tions less than this theoretical value resulted in the hydrolysis of f e r r i c chloride. The extent of hydrolysis depended on the concentration of acid used and was reflected in the decrease of iron concentration in the f i l t r a t e as compared to that in the 1i xi vi ant. 4.3 Leaching of Nickeliferous Laterite in the Presence of Ferrous Chloride Ferrous chloride reagent was u t i l i z e d in an e f f o r t to retard the rate of p r e c i p i t a t i o n of hematite. It was envisaged that the p a r t i c l e size of the precipitate would increase, thereby speeding up f i l t r a t i o n . It was observed however, that the presence of ferrous chloride, to a greater extent than hydrochloric acid, inhibited the extraction of n i c k e l . Table 4.3 shows the effect of ferrous chloride on nickel extraction. 94 Table 4.3 Leaching Results in the Presence of Ferrous Chloride at a Pulp Density of 400 g/1. Solution Type % Nickel Extraction Fe Cone. in Lixiviant" gpl Fe Cone, in Filtrate gpl Acid Consumed HCl(M), IM. FeCl 2 36 51 43 - IM FeCl 2, 3M HCl 71 55 59 2.80 0.5M FeCl 9, 0.5M FeCl,, 3M HCl L 6 88 55 55 2.76 IM FeCl 3, 3M HCl 94 55 55 2.82 IM FeCl 3 90 55 12 - 4.4 Acid Consumption During a Batch . Leach (423°K) Table 4.4. shows the amount of acid consumed in reacting with each metal species present in the l a t e r i t e , with the exception of iron. These values were calculated by using the concentration of each metal in the f i l t r a t e . The total value represented the quantity of acid to be replenished after each leaching cycle, providing there was no additional hydrolysis reaction to supply acid. 95 Table 4.4 Acid Consumption During a Batch Leach at 423°K with a Pulp Density of 400 g/1. Metal Metal Cone, in Filtrate gpl Acid Consumption by Individual Metal qpl Equivalent Acid Concentration HC1(M) Co 0.6 0.74 0.02 Cr 0.5 1.05 0.03 Mn 4.1 5.58 0.15 Ni 5.0 6.21 0.17 Mg 8.0 24.00 0.66 1 Al 14.6 59.19 1.62 Total 96.77 2.65 The calculated value of 2.65M hydrochloric acid com- pare favorably with 2.82M, which was obtained using the auto- t i t r a t i o n method. The s l i g h t l y higher value of acid consumed was to be expected, taking into consideration further dis- solution of f i n e l y precipitated hematite, during the period of time taken to dismantle the apparatus and f i l t e r the s o l i d s . An attempt was therefore made to determine i f in fact a portion of the "free acid" did become consumed in the dis- solution of f i n e l y precipitated hematite after leaching. 96 A laboratory centrifuge was employed to reduce the period of time during which the solution was in contact with the iron residue. By reducing this period from 1 hour to 15 minutes, there was an increase in "free acid" concentration from about 0.05M to approximately 0.2M. This was however, s t i l l lower than the theoretical value of about 0.5M. This was confirmed by determining the concentration of "free acid" produced by heating f e r r i c chloride solutions alone, at temperatures of 423°K and 453°K. The acid results obtained due to hydrolysis were compared with those obtained after leaching with f e r r i c chloride and acid / f e r r i c chloride mixtures. In one experiment, the slurry obtained after leach- ing was allowed to stand for 13 hours, and the."free acid" con- centration in the f i l t r a t e f e l l to a lower value,, The concentra- tion of iron increased s l i g h t l y . The results obtained are shown in Table 4.5. ". . 4.4.1 The Effect of Hydrolysis on Acid Consumption 47 Hydrolysis of f e r r i c chloride is known to begin at around 393°K, producing hematite and hydrochloric acid. Table 4.5 shows that at room temperature, the "free acid" concentration in f e r r i c chloride solution was 0.06M. However, hydrolysis of this solution at temperatures, 423°K and 453°K resulted in "free acid" concentration of 0.38M and 0.50M Table 4.5 Comparison of "Free A d d " Concent ra t ion In F i l t r a t e s wi th Ac id Coneentrat lon Due to the Hyd ro l ys i s of F e r r i c Ch lo r i de S o l u t i o n FeCl 3 Type HCT Temperature Pul p Dens 1ty 9/1 Fe Cone, in L i x l v l a n t Fe Cone. 1n F i l t r a t e gpl "Free A d d " 1 n L i x l v l a n t HCl(M) "Free A d d " 1n F i l t r a t e HCl(M) 2M - 42 3 - 112 109 0.06 0.38 2M - 453 - 112 106 0.06 0.50 2M - 423 400 112 60 0.06 0.07 2M - 453 40 0 112 58 0.06 0.03 2M, 3H HCl 423 400 110 i n - 3.00 0.02 2M, 1 3M HCl 453 400 no ns 3.00 0.003 ^Slurry from this run was al lowed to stand fo r 13 hours. 98 respectively. This increase in acid concentration coincided with a decrease in iron concentration of the f i l t r a t e . Leaching experiments were conducted at the same tempera- tures, employing the same concentration of f e r r i c chloride. Lower values of iron concentration in the f i l t r a t e indicated that further hydrolysis occurred, in order to supply more acid necessary to react with metal values present in the ore. The "free acid" level in the f i l t r a t e approximated to that in the 1 i x i v i a n t. In anotherleaching experiment, an excess of hydrochloric acid (over that which was required for complete metal extrac- t i o n ) , was added to a l i x i v i a n t solution containing the same amount of f e r r i c chloride used before. A s l i g h t l y higher concentration of iron in the f i l t r a t e indicated some dissolu- tion of the iron residue. Further dissolution was observed on leaving the leach slurry to stand for 13 hours. It follows therefore, that dissolution of the ore gradu- a l l y occurred, producing a f e r r i c chloride intermediate, which was promptly hydrolysed to hematite. Once enough acid was present in the l i x i v i a n t , hydrolysis of f e r r i c chloride con- tained in the l a t t e r was probably i n s i g n i f i c a n t . It is not known at this stage, what proportion of the hematite p r e c i p i - tated from f e r r i c chloride of the l i x i v i a n t or from dissolved 1ateri te. 99 4.4.2 "Free Acid" Concentrations in Ferric Chloride Soluti ons Several concentrations of f e r r i c chloride solution were heated under the same conditions used for leaching experi- ments. The hematite collected was weighed and the f i l t r a t e analysed for iron and "free acid" concentration. Table 4.6 shows the percentage of iron precipitated from these solutions along with their acid concentrations. The extent to which hematite is precipitated from f e r r i c chloride solutions, and the amount of "free acid" thereby being released are i l l u s t r a - ted in Figure 4.5. At ferric'chl oride concentrations greater than IM, the quantity of acid liberated due to hydrolysis was more dependent on temperature than on concentration of the former. In one experiment, using a 1 molar f e r r i c chloride solution, the concentration of hydrochloric acid determined was 0.39M at 423°K, as compared to 0.53M at 453°K. 4.5 The Determination of Leaching Equilibrium Calculated amounts of nickel chloride hexahydrate were added to each batch of leach solution to give 20 gpl, 40 gpl and 60 gpl nickel respectively. Leaching experiments were conducted at 423°K and 448°K, and the iron residue obtained was c a r e f u l l y washed and analysed. Table 4.6 Iron Precipitation From Ferric Chloride Solutions at 423°K. Fe.Cone. in Solution gpl Fe Cone, in Fi1trate gpl wtiJ of Fe Preci pi tated gms Percentage Preci pi t a t i on of Fe "Free Acid" Cone, in Fi1trate HCT(M) 27.2 19.8 7.4 27 0. 51 54.4 47.5 7 13 0. 50 109 105 4 3.7 0.39 110 104 6 5.4 0.531 21 7 215 2 0.01 0.41 2 Determined at 453°K. Determined at 433°K. o o 0.5 0.4 h £ 0.3 4) U- 0.2 0. 0 0 \ A \ \ Precipitation Free acid \ \ \ \ \ \ A — o — H 20 16 ° "0 CO o 12 ° . o 3 H 8 « 04 + . 40 80 120 L +](gpi) 60 200 240 Figure 4.5 "Free Acid" concentration and percentage iron precipitation versus iron (III) concentration at 423°K. 102 Results shown in Table 4.7 indicate that at 423°K over 90 percent nickel can be extracted with 1M f e r r i c chloride reagent, in the presence of up to 60 gpl n i c k e l . Table 4.7 Percentage Nickel Extraction from Nickeliferous Laterite in the Presence of High Concentrations of Nickel Solution Type Pulp Dens i ty , .g-.p-.l-. Nickel Cone, in F i l t r a t e gpl Percentage Nickel Extracti on 20 gpl Ni, 2M F e C l 3 200 22.5 98.4 40 gpl Ni , 2M F e C l 3 200 42.2 97.6 60 gpl Ni , 2M F e C l 3 200 62.5 95.2 60 gpl Ni , 1M F e C l 3 200 62.0 91 .2 4.6 Simulation of a Continuous Leaching C i r c u i t A constant chloride ion a c t i v i t y was maintained by adding calculated amounts of 6N magnesium chloride, aluminium chloride, and f e r r i c chloride to a constant volume of 9.6N hydrochloric acid. These chlorides wen.e known to be the main constituents of the f i l t r a t e . The tofral volume of the l i x i v i a n t in each leaching cycle was the same, and the f i n a l acid concentration 103 Was 2.4N hydrochloric acid. A Similar run with only 6N f e r r i c chloride solution was conducted and the results obtained are shown for comparison (see Table 4.8). •Figure 4.5 i l l u s t r a t e s how a decrease in f e r r i c chloride concentration, despite a constant acid concentration, resulted in a decrease in the percentage of nickel extracted. Cobalt and manganese showed similar extraction behaviours, whilst very l i t t l e chromium was extracted. The extraction of alumi- nium and magnesium decreased gradually as a result of the de- crease in f e r r i c chloride concentration, and/or because of the common ion e f f e c t . 4.7 Morphology of the Iron Residue 4.7.1 Hematite from the Hydrolysis of Ferric Chloride The Scanning Electron Microscope was employed to scrutinize p a r t i c l e s obtained from each of the experiments described in Section 4.4.2. There was no detectable difference, in the nature, shape and size of the p a r t i c l e s obtained at dif f e r e n t temperatures - only in the quantity of precipitate (see Table 4.6). There was however, an increase in p a r t i c l e size of the hematite precipitated from f e r r i c chloride solutions ranging Table 4.8 Percentage Metal Ex t rac t ion in a Simulated Continuous Leaching C i r c u i t 1 i Y i v i a n t ComDOsition Metal Cone, in Li xi vi ant gpl Fe Cone. Percen tage Met al Extra : t i on S o l u t i o n Type Volume (mis) in F i l t r a t e gpi Ni Co . .Mn Cr Al Mg 6N F e C l 3 40 110 Staqe 1 60 98 91 91 20 78 76 9.6N HCl 6M F e C l 3 6N A l C I 3 6N MgCl 2 10 30 82.5 Stage 2 75 91 90 91 20 76 75 9.6N HCl 6N F e C l 3 6N A l C I 3 6N MgCU 1 0 22.5 5 2.5 52 6 5 Stage 3 58 85 89 90 20 74 74 9.6M HCl 6N F e C l 3 6N A l C I 3 6N MgCl 2 10 15 10 5 41 12 10 Stage 4 60 65 85 88 20 59 49 i 9.6N HCl 6N F e C l 3 6N A l C I 3 6N MgCl 2 ' 10 7.5 1 5 7.5 21 18 15 Stage 5 42 56 75 77 20 48 25 9.6N HCl 6N F e C l 3 6N A l C I 3 6N MgCl 2 10 20 10 24 19 Stage 6 26 50 48 62 20 16 15 1 Lixiviant composition (gpl) Figure 4.6 Metal extraction versus lixiviant composition at 423°K and a pulp density of 400 Table 4.9 Chemical Analysis of the Iron Residue From Each Stage Element Stages • Laterite . Sample 1 ... -2 3 4 5 6 Weight Percentage Fe 57.5 57.5 56.2 48.2 45.4 44; 7 44.55 Ni 0.02 0.17 0.38 0.54 . 0.64 0.7 1 .25 Co 0.014 0.01.5 0.017 0.022 0.038 0.078 0.15 Cr 1.81 1 .811 - - - - 2.26 Mn 0.10 0.10 0.11 0.13 0.25 0.41 1 .08 Mg 0.34 0.44 0.49 0.95 1 .72 2.07 2.31 Al 9.90 0.88 1 .08 2.72 3.09 3.58 4.68 Si 2.46 - - - - - 2.65 Estimated. o cn 107 in concentration from 0.5M to 2M. Figures 4.7 (a), ((b), (c) and (d) show SEM photomicrographs of hematite pa r t i c l e s at 5000 times magnification, with p a r t i c l e size increasing from 0.8ym to 2vm- The par t i c l e s were spherical and uniformly shaped, with a tendency to agglomerate. u :, 4.7.2 Iron Residue from the Hydrothermal Treatment of Nickeliferous Laterite with Ferric Chloride Solutions Hematite pa r t i c l e s precipitated during leaching experiments, in which the l i x i v i a n t employed was f e r r i c chloride, possessed highly etched surfaces. Particles were larger than those obtained from the hydrolysis of f e r r i c chloride solution alone, and were irregular in shape. The surfaces of pa r t i c l e s immediately after p r e c i p i t a t i o n may be somewhat different from those shown in the photomicro- graphs (Figure 4.8). There were indications of surface dis- solution in particular directions corresponding to the direc- tion of flow of the f i l t r a t e during f i l t r a t i o n . This dis- solution was p a r t i c u l a r l y v i s i b l e in photomicrographs of par t i c l e s obtained through pressure f i l t r a t i o n , which required an average of 30 minutes for completion. Such "laminar dis- solution" was not apparent on par t i c l e s recovered using the centrifugal technique. (a) (b) ( c ) (d) Figure 4.7 Hematite particles precipitated from f e r r i c solutions at 423°K. a) from 0.5M FeCl 3 (4000 x) b) from 1M FeCl 3 (5000 x) c) from 2M FeCl 3 (5000 x) d) from 4M FeCl, (5000 x) Figure 4.8 Iron residue obtained from the hydrothermal treatment of n i ckeliferous l a t e r i t e with f e r r i c chloride solutions. a) from 2M FeClg, 433°K ( 380 x) b) from 1M F e C l 3 > 423°K (2000 x ). o n o The cause of such redissolution was the reversal of the hydrolysis reaction on cooling. This behaviour was predicted by the thermodynamic information developed in Section 2.2.6, and evidenced by the loss of "free acid" during extended con- tact between residues and solution, as shown in Section 4.4. Temperature and concentration of the f e r r i c chloride solution employed did not have any s i g n i f i c a n t effect on the morphology of the p a r t i c l e s . 4.7.3 Iron Residue from the Hydrothermal Treatment of Nickeliferous Laterite with Ferric Chloride/ Hydrochloric Acid Solutions The photomicrographs of Figure 4.9 show the sur- faces of pa r t i c l e s obtained from the above leaching experi- ments to be porous and highly etched. The pa r t i c l e s were fine r than those obtained from the preceding experiment. This reduction was attributed to a greater extent of redis- solution of the precipitate on cooling. 4.8 Comparison of Nickeliferous Laterite and the Iron Residue by Electron Microanalysis Figure 4.11 shows SEM X-ray analyser spectra taken for nickel-bearing l a t e r i t e p a r t i c l e s (see Figure 4.10),and iron residues precipitated under dif f e r e n t leaching conditions. Figure 4.9 Iron residue obtained from the hydrothermal treatment of n i c k e l i f e r o u s l a t e r i t e with f e r r i c chloride / hydrochloric acid solutions. a) from IM FeClg, 3M HCl, at 423°K (2000 x) b) from IM FeCl,, 2.5M HCl, at 423°K (800 x ) . (a) (b) Figure 4.10 a) Main nickel-bearing p a r t i c l e s (of redish brown colour, 2000 x) white areas depict the mineral goethite; or quartz darker areas represent magnetic regions high in chromium, b) Predominantly goethite (4000 x). (a) to +J c 3 o c_> CD o N i \*V — (b) Mg Al CI Ca T t " Cr Cr Fe Fe Ni Ni -> X-ray Energy (keV) Figure 4.11 SEM x-ray analyser spectra for nicke (a) ...Iron residue from leach with 2M (b) Nickeliferous laterite (c) Iron residue from leach with 2M iferous laterite and iron residue FeCl 3 at 453°K FeCl 3, 3M HCl, at 453°K 114 The spectra, containing 500,000 counts each, are shown together in order to accentuate their differences. The most s t r i k i n g feature i l l u s t r a t e d is the disappearance of aluminium, magnes- ium and nickel from the l a t e r i t e . 4.9 Comparison of Particles Found in Nickeliferous Laterite and the Iron Residue by Electron Microanalysis. The photomicrographs shown in Figures 4.12 and 4*13 i l l u s - trate the various shapes and r e l a t i v e sizes of par t i c l e s found in the l a t e r i t e and iron residue. Figure 4.14 shows the cor- responding SEM X-ray analyser spectra for some of those p a r t i - cles before leaching. Each spectra served to iindicate the pre- sence and r e l a t i v e abundance of elements without being t r u l y - quanti t a t i ve. Quartz and chromite pa r t i c l e s were insoluble under leach- ing conditions. However, there were indications of the removal of magnesium, aluminium and minor amounts of nickel from these p a r t i c l e s . Some par t i c l e s referred to as magnetic, contained a s i g n i f i c a n t amount of magnesium, as d i s t i n c t from chromite, which showed stronger peaks of chromium with less magnesium. The spectra shown in Figure 4.15 i l l u s t r a t e s the dis- appearance of magnesium, aluminium and nickel from pa r t i c l e s after leaching. ( a ) ( b ) F i g u r e 4.12 a ) S i l i c a t e p a r t i c l e s a f t e r l e a c h i n g i n HC1/HN03 (200 x) b) P a l e - y e l l o w s i l i c a t e (400 x). 116 (a) (b) (c) (d) Figure 4.13 Magnetic particles found in nickeliferous laterite and the iron residue a) magnetic particle before leaching (2000 x) b) from 2M FeCl 3, at 453°K (340 x) c) from IM FeCl 3, 3.6HC1, at 423°K (1500 x) d) from 2M FeCl_v at 423°K (4200 x) (a) CO o o O (b) (c) Mg Al Si Cl Cr Cr Fe Fe :  !.7"Ni Ni* " '• X-ray Energy (keV) Figure 4.14 SEM x-ray analyser spectra for s i l i c a t e and magnetic pa r t i c l e s (a) Insoluble s i l i c a t e (b) . Magnetic ' p a r t i c l e before leaching (c) Chromite pa r t i c l e before leaching Cr Cr Fe Fe • > X-ray Energy (keV) SEM x-ray analyser spectra for magnetic pa r t i c l e s in the iron residue. (a) Magnetic part i c l e from leach with TM FeC13, 3.6M HCl, at 423°K. (b) Chromite pa r t i c l e 119 Chapter 5 DISCUSSION 5.1 Review of Leaching Results It was established in the l i t e r a t u r e that the rate of iron extraction and therefore the rate of nickel extraction from goethite were proportional to the product of the hydrogen and chloride ion a c t i v i t i e s . However, leaching experiments done by the author indicate that extraction was more heavily dependent on chloride ion a c t i v i t y than hydrogen ion a c t i v i t y . This was confirmed by leaching experiments in which a constant chloride ion concentration of 5.5N was employed. In one case, nickel extraction was 91 percent using a 5.5N hydrochloric acid solution, as compared to 94 percent using a f e r r i c chloride/hydrochloric acid mixture of a total chloride concentration of 5.5N. The use of f e r r i c chloride alone resulted in a nickel extraction of 96 percent. This concentration of chloride ion was employed, as i t approxi- mated the total amount of acid consumed in producing metal chlorides (see Section 4.4) at a pulp density of 400 g/1. 120 Magnesium and aluminium showed similar extraction be- haviour, with approximately 75 percent of the metal extracted under most conditions of acid and f e r r i c chloride concentra- tions. Maximum extraction of cobalt and manganese was achieved under conditions previously described. This was however, to be expected, because of the close association of the two metal oxides. 5.2 The Significance of Acidity During Hydrolysis It was shown in. Chapter 2, that the log of the a c t i v i t y of f e r r i c ion in equilibrium with hematite was a linear function of pH with a slope of minus 3. At a temperature of 423°K with a f e r r i c ion a c t i v i t y of about 1M (approx. 56 g/1 Fe) in solution, the pH due to hydrolysis of f e r r i c chloride would be around -1.6. With reference to the Potential/pH diagram 5 1 for the Fe — H20 system at 423°K, f e r r i c ions were observed to be in equilibrium with hematite at about the same pH. However, these s l u r r i e s were cooled below 353°K before f i l t r a t i o n or centrifuge separation of solids was attempted, and at this lower temperature the equilibrium pH rises from -1.6 at 423°K to -1.04 at 333°K (see Table 2.2), a decrease 121 [ in free acid of a factor of 3.6. Redissolution of hematite is therefore anticipated, but because i t is slow, i t is more serious under conditions of long f i l t r a t i o n times than under conditions of short centrifugal separation times. 5.3 Nature of the Iron Residue Table 5.1 compares the chemical analysis of the iron residue, with that of currently marketed regular iron ore. Although the analyses compared favorably, the iron residue contained too much chromium for large-tonnage use by the Steel Industry. Other deleterious impurities such as n i c k e l , magnesium and aluminium would also require separate considera- t i o n , reserved for the steelmaker. Judging from the intensity of the peaks obtained in x-ray d i f f r a c t i o n patterns, iron residues from leaches with f e r r i c chloride solutions only may have been more c r y s t a l l i n e than residues obtained from leaches in the presence of hydro- ch l o r i c acid. However, the p a r t i c l e size of residues ob- tained in the presence of acid was smaller than that in the former. Therefore, i t is not known at this point, whether the decrease in peak intensity was due to a decrease in p a r t i c l e size or reduced crystal 1 i n i t y . The SEM X-ray analyser spectra of magnetic p a r t i c l e s 122 Tab!e 5.1 Comparison of Raw Materials Used in the Iron and Steelmaking Industry Analysi s • Dry % Leached Nickeliferous Laterite Iron Ore Pel 1ets Normal Iron Ore Fe 57.5 64 59 Si 2.46 2.8 4.2 Al 0.90 0.27 0.8 Ca - - 0.2 Mg 0.34 0.3 0.1 2 Mn 0.10 0.2 0.5 Cr 1 .81 0.01 0.01 Co 0.01 0.01 00.01 Ni 0.1 0.01 0.01 123 before and after leaching experiments showed very l i t t l e extraction of chromium. It was apparent that nickel was more ea s i l y released from i t s l a t t i c e positions in goethite, and that the residual nickel occurred mainly as replacement atoms in s i l i c a t e and chromite p a r t i c l e s . 5.4 Hydrothermal Precipitation of Hematite in Supersaturated Ferric Chloride Solutions 5.4.1 Homogeneous Nucleation Thermodynamical1y, a homogeneous supersaturated solution is in a metastable state and may remain so i n d e f i - n i t e l y . Before nuclei can grow, the system must overshoot a c r i t i c a l equilibrium point and pass into the region where AGy (volume free energy) is negative. This equilibrium corres- ponds to a f i n i t e c r i t i c a l s i z e , at which a nucleus i f so formed can grow. Such nuclei can be formed spontaneously within the system due to s t a t i s t i c a l fluctuations in the free energy, or may be provided a r t i f i c i a l l y by seeding. The Gibbs free energy of nucleation is given by the 52 expresslon: AG = d3AG + a d 2 Y ...5.1 11 • 124 where d = The approximate d iameter of the nucleus 3 d = volume of the nuc leus . a d 2 = su r f ace a rea , a = A f a c t o r r e l a t e d to the shape of the nucleus (approx imate ly 5 f o r homogeneous n u c l e a t i o n ) . AG y = Volume f r e e energy ( f r e e e n e r g y / u n i t volume r e q u i r e d to form the s o l i d phase assumed to be in 1 p i e c e ; negat i ve f o r supe r sa tu ra ted s o l u t i o n s ) . Y = Sur face f r e e energy per un i t area of the s o l i d / l i q u i d i n t e r f a c e (always p o s i t i v e ) . Assume y . ' andAG v to be independent of d and-a . Equat ion 5.1 shows su r f a ce energy to be dependent on the square of the nuc l ea r dimensions whereas the bulk f r e e energy depends on i t s cube. T h e r e f o r e , AG v a r i e s acco rd ing to the abso lu te volume of the nuc l eu s . When the temperature i s such that hemati te i s s t a b l e r e - l a t i v e to f e r r i c ions in s o l u t i o n , AG y i s n e g a t i v e . At small va lues of d, the su r f a ce term dominates and AG i s p o s i t i v e ; at l a r g e d the volume f r e e energy term dominates because t h i s 3 i s p r o p o r t i o n a l to d . At a c r i t i c a l p a r t i c l e s i z e d^, AG passes through a maximum denoted by W. The va lue of W and d^ depend on AG y and thus upon the temperature. Growth of hemati te p a r t i c l e s sma l l e r than d leads to an i n c r e a s e in 125 the free energy and thus,there is a greater tendency for such pa r t i c l e s to dissolve rather than grow; par t i c l e s larger than r c are stable because growth is accompanied by a decrease in free energy. Particles of diameter d c are unstable having an equal chance of dissolving or growing. Figure 5.1 graphically i l l u s t r a t e s the free energy of formation of spherical embryos as a function of the diameter for a series of temperatures. At temperatures T<Tr;, although there is a greater tendency for par t i c l e s of diameters d(, and &2 t° shrink, some may overcome the free energy barrier W due to thermal fluctuations and grow. 5.4.2 Growth of N u c l e i 5 3 The growth of precipitates from supersaturated solutions involves the formation of the precipitate phase, due to the transfer of atoms across an interface with a re- di s t r i b u t i o n of solute species. In leaching experiments con- ducted, i t was expected that the growth rate may have depended upon some or a l l of the following factors: 1) The detailed mechanism by which the interface propagated. 2) The rate of di f f u s i o n of the atoms in both phases. 3) Size e f f e c t s : Small particles of a phase have a higher s o l u b i l i t y than larger ones, due to differences in the ra d i i of curvature of the interfaces. Figure 5.1 The free energy of formation of spherical embryos as a function of the diameter for a series of temperatures. 127 The heating of f e r r i c chloride solution above 393°K re- sults in a system containing mixed sizes of hematite nuclei (see Figure 5.1). A reduction in the overall free energy of the system could only be accomplished by a reduction in the total area of internal interfaces. This process requires d i f f u s i o n of solute from regions close to small p a r t i c l e s to regions around large ones (because the concentration of solute in the solution in equilibrium with a precipitate is larger for a small p a r t i c l e of precipitate than for a large one). Therefore, removal of solute from the solution near to small pa r t i c l e s causes the l a t t e r to dissolve. The coarsening of precipitates with increasing f e r r i c chloride concentration as shown in Figure 4.7, may have been due to the gradual nature of the increase in the quantity of part i c l e s of varying sizes within each system. Furthermore, because small precipitate p a r t i c l e s are more soluble, they dissolved to provide the solute for growth of the larger ones. 5.5 The Degree of Supersaturation for Hematite in a Goethite Saturated Solution. The free energy per unit volume for nucleation can be 54 expressed by: AG = RT In a ao .5.2 128 where a a c t i v i t y of the dissolved matter s o l u b i l i t y at the actual temperature. using S = a.• (supersaturation ratio) A G RT In S v Equation 5.3 shows that an increase in the supersatura- tion ratio results in an increase of the free energy of the solution. This added free energy is then available to over- come the nucleation free energy barrier. If the free energy available from this source exceeds the c r i t i c a l maximum free energy for the most unstable nuclei in Figure 5.1, a l l sizes of nuclei become stable and the barrier for nucleation van- ishes. Under these conditions nucleation is so fast that growth rates are almost i r r e l e v a n t ; this solution produces a large number of very small p a r t i c l e s . In leaching experiments conducted, -* i t appeared that the goethite dissolved rapidly to produce s u f f i c i e n t l y super- saturated solutions with the inherent fast rate of nucleation. A large amount of very fine hematite p a r t i c l e s precipitated rapidly under conditions where growth was limited because of the large i n t e r f a c i a l area generated by nucleation. Such part i c l e s receive a limited amount of prec i p i t a t i n g solute which resulted in fine precipitates and poor f i 1 t e r a b i 1 i t y . 129 However, equation 5.3 indicates that i f a lower degree of supersaturation can be established during goethite dissolution (and therefore, a decrease in the free energy of formation AG), hematite pa r t i c l e s w i l l not nucleate spontaneously. Those that do surmount the free energy barrier and grow w i l l become much larger than under spontaneous nucleation conditions. Improved f i 1 t e r a b i 1 i t y can therefore be expected. By considering the theoretical equilibrium pH at several a c t i v i t i e s of f e r r i c ion as described in sections 2.2.5 to 2.2.7, the free energy for the pr e c i p i t a t i o n of hematite from a goethite saturated solution can be calculated. Values of f e r r i c ion a c t i v i t y in equilibrium with hematite were calcu- lated at room temperature using the relationship: pH = 1 3 Assuming total dissolution of goethite, the associated pH, were f i r s t calculated, employing the equilibrium constant 3+ for the Fe - FeOOH equilibrium at 423°K. These pH values were then used to calculate the a c t i v i t y of f e r r i c ion in equilibrium with hematite, this time, employing the e q u i l i - 3+ brium constant for the Fe - F^O^ equilibrium at the same temperature. The results obtained are shown below: Log [Fe 3 +] pH Log [Fe 3 +] [Fe 3 +] (Hematite) 3+ (from dissolved FeOOH) (in'equilibrium with Fe^Og) , [Fe ] (Goethite) 0 -1.48 0.36 2.29 x -1 -1.15 -1.35 2.24 x -2 -0.81 -2.37 2.34 x •log K log 3 + Fe 130 Using equation 5.3, (Max.) RT In 2.3 700 calories MOL Fe 3+ at 423°K. 3+ These data make i t clear that 700 calories per mole Fe are available from a saturated solution of goethite to overcome the nucleation free energy barrier of hematite. This value is probably high enough to cause spontaneous nucleation of hema- t i t e y i e l d i n g very f i n e , poorly f i l t e r a b l e precipitates. 131 Chapter 6 CONCLUSION T!he results of the leaching experiments indicate that a promising process might be devised for the hydrothermal treatment of nickeliferous l a t e r i t e with f e r r i c chloride solutions. Direct ore attack obviates the need for prior pyro-metal1urgical reduction, but at the expense of unit operations to recover magnesium, aluminium, manganese, and chromium. Of the chlorides present in solution, iron, aluminium and magnesium hydrolyze readily at moderate temperatures. Hydrochloric acid can therefore be recovered from the mother liquor by hydrolysis in an Aman-type reactor (AMAN, J. 1956, 1958), and recycled in the system. Close scrutiny of the leaching results obtained in the simulation of a continuous leaching c i r c u i t , indicate that two stages of leaching would have to be u t i l i z e d , in order to obtain at least 90 percent nickel extraction. 132 Owing to the amorphous and f r i a b l e nature of the pre- c i p i t a t e d residue, f i l t r a t i o n was extremely d i f f i c u l t . In order to thoroughly wash the residue, i t was necessary to repulp at least twice. This v i r t u a l l y removed a l l of the chlorides. In industrial practice, an economic balance would have to be struck between the cost of evaporating excess wash water and the loss of chlorides in the residue. It appears that the iron ore residue would not meet acceptable specifications with regard to chromium, and probably nickel without further beneficiation. Further- more, recovery of, at most, half of the ore's chromium con- tent by physical means is a p o s s i b i l i t y but of doubtful ind u s t r i a l merit. The marked difference in the morphology of hematite precipitated from f e r r i c chloride reagent, and that of the iron residue was due to the following: 1) Dissolution of the precipitate on cooling. 2) Rapid p r e c i p i t a t i o n under super-saturated conditions. These conditions resulted in a precipitate of poor f i l t e r - a b i l i t y . It appears however, that once conditions of preci- pitation can be controlled, some improvement in p a r t i c l e size and/or shape may be expected. This should result in better f i 1 t e r a b i 1 i t y of the iron residue. 133 Chapter 7 SUGGESTIONS FOR FUTURE WORK 7.1 Improvement in Experimental Procedure In this investigation, leaching was r e s t r i c t e d to batch runs using a small titaniurn autoclave. , Such runs had the drawback that a s i g n i f i c a n t amount of hydrolysis occurred in the heating up period, which took about 20 minutes. It was therefore d i f f i c u l t to assess the effect of variables such as retention time and temperature from such experiments. A continuous experimental arrangement should therefore be devised, in which the leach solution to be hydrolysed is pumped continuously into a heated autoclave. The reacted slurry is allowed to overflow continuously into a receiving vessel, the contents of which may be p e r i o d i c a l l y adjusted and recycled. With such back mixing, hematite nuclei would be in a regime suitable for continued growth, because condi- tions of severe supersaturation can be avoided. Lower levels of supersaturation w i l l lead to slow nucleation and coarse precipitates. 134 REFERENCES 1 Queneau, P.E. and Roorda, H.J. Cobalt and the N i c k e l i - ferous Limonites. De Ingeneur, Vol. 83, No. 28, (1971), pp. 1-9. 2 Bolt, J.R., Queneau, P.E. The Winning of Nickel. D. Van Nostrand Co., Princeton, N.J., (1967). 3 Stevens, L.G., Goeller, L.A., M i l l e r , M. "The UOP Nickel Extraction Process" CIMM, 14th Annual Conference Edmonton, Alberta, Canada, (1975). 4. Weege, R.J., Beres, W.J., "The Exploration and Develop- ment of Nickel Laterites" AIME Annual Meeting, Denver, Colorado, Feb. 27, 1978. 5 Posnjack, E. and Merwin, H.E. Am. J. Sc. Vol. 47, (1919), pp.311-348. 6 Kulp, J.L., and T i t e s , A.F., Am. Miner. Vol. 36, (1951), pp. 23-44. 7 White-Howard, F.B. Nickel - A H i s t o r i c a l Review. D. Van Nostrand Co., Inc., New York, (1963), p.352. 8 Mineral Commodity P r o f i l e s , May 1979. United States Department of the Interior. 9 Engineering and Mining Journal, March 1981, p. 137. 10 Cornwall, H.R. Nickel Changes in United States Mineral Resources, U.S. Geol. Survey Prof. Paper 820, (1973), pp. 437-442. 11 Roorda, H.J., Queneau, P.E. M.I.M Manuscript, June 1973. 12 Agarwal, J . C , Reddy, B.J. Role of Cobalt in Industrial Society. CIM Report 10th Annual Hydromet Meeting, Edmonton, Alberta, Oct. 1980. 13 Cuban American Nickel, Engineering and Mining Journal, 158, Vol. 79, (1957). 14 Mancke, E.B., U.S. Patent, 2, 762, 703 (1956). 15 Wescott, E.H., i b i d . 2, 036, 664 (1936). 16 Graham, M.E., Reed, R. , and Cameron, J.R. i b i d . 2, 766, 115 (1956). 135 1.7 Van Nes, C.L. Dutch Patent Application 1 93 , 61 2 ( 1 955 ) N.O. 89510 (1958). 18 Caron, M.H., i b i d . 1, 487, 145 (1924). 19 Heertjes, P.M. and Van Nes, C.L. Recovery of Nickel and Cobalt from L a t e r i t i c Iron Ores. Reel. Trav. Chim. Pays-Bas Belg. Vol. 79, (1960), pp. 593-604. 20 Seimens, R.E., Corrick, J.D. "Process for Recovery of Nickel, Cobalt and Copper from Domestic Lat e r i t e s , " Mining Congress Journal, January, 1977, pp. 29-34. 21 Wilson, F. Min. Eng. Vol. 10, (1958), pp. 563-565. 22 Anon. Report on Moa Bay. New York Times, 5 Feb. 1968. 23 Rice, N.M., Strong, L.W. The Leaching of Laterite Nickel Ores in Hydrochloric Acid. CIM Quarterly. Vol. 13, No. 3, 1 974. 24 Goldztaub, S. Compt. Rend. Acad. S c l . URSS Vol. 195, (1932), p. 264. 25 Jurinak, J.J. Journal of Colloid Science Vol. 19, (1964), pp. 477-487. 26 Bath, M.D., M.A.Sc. Thesis, Dept. of Metallurgy, U.B.C. (1968). 27 Roach, G.I.D. M.A.Sc. Thesis, Dept. of Metallurgy, U.B.C. (1968). 28 Surana, V.S. M.A.Sc. Thesis, Dept. of Metallurgy, U.B.C. (1969). 29 Devuyst, E. Ph.D. Thesis, Dept. of Metallurgy, U.B.C. (1973). , 30 Azuma, K. and Kametani, H. Trans. AIME., Vol. 230, (1 964), p.853. 31 Parks, G.A. and De Bruyn, P.L. J. Phys. Chem., Vol. 66, (1962), p. 967. 32 Ahmed, S.M. and Marksimov, D. Studies of the Double Layer at the Oxide-solution Interface, Department of Energy, Mines and Resources, Mines Branch, Ottawa, 1968. 33 C r i s s , CM. and Cobble, J.M. The Thermodynamic Properties of High Temperature Aqueous Solutions. IV : Entropies of the Ions up to 200° and the Corresponcence P r i n c i p l e , J. Amer. Chem. Soc. Vol. 86, (1964), pp. 5385-5390. 136 34 Criss , CM. and Cobble, J.M. The Thermodynamic Properties of High Temperature Aqueous Solutions V : The Calculation of Ionic Heat Capacities up to 200°, Entropies and Heat Capacities Above 200°, J. Amer. Chem. S o c , Vol. 86, (1964), pp. 5390-5393. 35 Robins., R.G. J. Inorg. Nucl . Chem. Vol. 29 , (1 967) pp. 431-435. 36 Haigh, C.J. Hydrolysis of Iron in Acid Solutions Proc. Aust. Inst. Min. Met., Sept., 1967. 37 Lietzke, M.H. and Stoughton, R.W. J. Phys. Chem. Vol. 63, (1959), pp. 1183, 1186, 1188, 1190, 1984. 38 Cobble, J.W. J. Amer. Chem. Soc. Vol. 86, (1964), p. 5394. 39 Meissner, Kusik and Tester. AICHE Journal Vol. 18, No. 3, May 1972 , p. 661 . 40 Feitknecht, W. and Schindler, P. S o l u b i l i t y Product of Metal Oxides. Pure Appl. Chem. Vol. 6, (1963), pp. 130- 199. 41. Biedermann, G. and Schindler, P. ACTA Chem. Scand. Vol. 11, pp. 731-740. 42 Weiser, H.B. Inorganic Colloid Chemistry Vol. 11, (1935), p. 40. 43 Posnjak, E. and Merwin, H.E. The System Fe?,03 — S0_ — H„0, J. Amer. Chem. Soc. Vol. 44, (1922), pp. 1965-1993. d 44. Tunell, G. and Posnjack, E. The S t a b i l i t y Relations of Goethite and Hematite (Discussions) - Econ. Geo!., Vol. 26, (1931), pp. 337-343. 45 Gruner, J.W. The S t a b i l i t y Relations of Goethite and Hematite (Discussion) - Econ. Geol ., Vol. 26 , (1 931 ), pp. 442-445. 46 Smith, F.G. and Kidd, D.J. Am. Mineralogist, Vol. 34, (1949), p. 403. 47 Katsurai , T. and Watanabe, T. The Structure of Iron Oxide Prepared by the Autoclave Treatment. S c i . Pap. Inst. Phys. Chem. Res., Tokyo, Vol. 13, pp. 89-92. 48 Queneau, P.E. Cobalt and the Nickeliferous Limonites Doctoral Thesis, University of Delft, 1971, p. 254. 137 49 Kelley, K.K. U.S. Bureau of Mines Bulle t i n 476 U.S. Government Printing O f f i c e , Washington, D.C., 1949. 50 Barner, H.E. and Scheurman, R.V. Handbook of Thermo- chemical Data for Compounds and Aqueous Species 1978. 51 Biernat, R.J. and Robins, R.G. High Temperature Potential/pH Diagrams for the Iron-Water System Electrochimica Acta, Vol. 17, (1972), pp. 1261-1283. 52 C o t t r e l l , A. An Introduction to Metallurgy, Second Edition, SI Units. 53 Burke, J. The Kinetics of Phase Transformations in Metals. 54 Nielsen, A.E. Kinetics of Precipitation Pergamon. Appendix A CRISS AND COBBLE CALCULATIONS 139 , Appendix A The Effect of Temperature on Thermodynamic Parameters 1) Consider the Effect of Temperature on Enthalpy of a Single Species Involved in a Reaction S and AH = / 2 C p ( T 2 - T , ) C p average being used due to the fact that C p is not normally constant between the temperatures. 2) The Effect of Temperature on Entropy AS„ = C In T 0 P P _2 140 3) The Effect of Temperature on Free Energy At temperature, T ^ = - , T-jS^ At temperature, T ^ 2̂ = 2̂ ~ ^2^2 therefore. A G = A H - ( T 2 " T J S l " T 2 A S where A H and AS have already been defined. Considering now the change in free energy of a reaction in going from 298°K to temperature T . AGj - A G ° g 8 = AH° -AH° g 8. -. ( T A S ° - 298AS° gg) ...(a) however, AS° - A S ° 9 8 =f AC° d (In T) 298 therefore, TAS° = T A S ° g 8 + T f AC° d (In T) ...(b) '2 98 now A H T * A H298 =f A C p d T • • ' ( c ) J298 by substituting (c) and (b) into (a) we get: AG° = A G ° 9 8 t f AC° dT - T A S ° g 8 - T » o r\ t 298 f AC° d(ln T) + 298 A S ° g 8 '298 141 by rearranging AGT = AG 2 9 8 - ATAS 2 9 8 + /"T AC° dT - T fJ AC° d(ln T) ^298 ^298 (d) o The detailed variation of C p with temperature is not usually well known, but satisfactory estimates may be obtained from average values of Cp over the temperature range. Therefore, using average values of Cp, equation (d) becomes: o o o AGT = AG 2 g 8 - ATAS 2 g 8 J 0,T + AC. 1 AT - TAC°] In T/298 ...(e) p J298 p 298 where AT = T - 298 o o AG and AS values are well established by experiment at 298° K. Therefore, in order to solve equation (e), i t is necessary to estimate AC P '298 142 A.2 Criss and Cobble Entropy Correspondence Principle 33 34 Criss and Cobble ' using data at di f f e r e n t temperatures established a correspondence p r i n c i p l e , whereby the entropy at o a given temperature, S T can be expanded about a reference tem- i 2 perature, such that: o o o p o o S T = a T + b S, + C T ( S T - r + d T (S T ) 2 XZ *2 i l '2 ' l '2 ' l It was shown that a standard state can be chosen at every temperature, such that the par t i a l molal entropies of one class of ions in solution at that temperature are l i n e a r l y related to the corresponding entropies at some reference temperature. o o S T = a T + b T S 2 Q 8 o S 2gg in the above expression is expressed in absolute units, which is defined as: o o S298 flh = S298 r +. " 5'°(Z> C a l s M o 1 _ 1 D e g _ 1 Abs Conventional where Z = Ionic Charge In a subsequent publication, the following procedure was adopted: When the entropy of an ion is known (or can be accurately pre- dicted at two temperatures, then the average value of the heat 143 - 1T2 capacity C can be calculated therefore, and where and 0 0 _ ST " S298 = Cp 298 In T/298 a T + ( b T - l ) S 2 9 8 = C p T 298 o P J298 a I 298 In T/298 aT ' S298 ( 1 - b T } In T/298 aT + 6T S298 In T/298 - D - b T ] In T/298 The heat capacity constants a-p, gy of simple cations, anions, and oxyanions etc. at various temperatures are shown o,T in Table A . l . Therefore, from calculations of C | , i t is o possible to determi ne ^C^ equation (e;. 298 , which can be inserted in 298 Table A.I Best Values of a t,b ,a ,B and S t (H +) at Several Temperatures Temperature s t <H+> C B ] ^ 9 8 (H + ) 298 °K -5 0 333°K -2.5 23 373°K + 2.0 31 423°K + 6.5 33 473°K (11.1) (35) Cations: a ̂ a t 0 1 .000 3.9 0.955 35 .. -0.41 10.3 0.876 46 -0.55 16.2 0. 792 46 -0.59 (23.3) (0.711 ) (50) (-0.63) Anions a^ b t a t h 0 1 .000 -5.1 0.969 -46 -0.28 -13.0 1 .000 -58 0.000 -21.3 0.989 -61 -0.03 (-30.2) (0.981 ) (-65) (-0.04) Oxy anions â . i.e. CO^ b^ S 0 4 . a t c 2 o 4 e t 0 1 .000 -14.0 1 .217 -127 1 .96 -31 .0 1.476 -1 38 2.24 -46.4 1 .687 -133 2.27 (-)67.0) (2.020) (-145) (2.53) Acid Oxyanions a^ i.e. HCQ~ b t H 2 P 0 4 a t HP0= e t HSO" 0 1 .000 -13.5 1 . 380 -1 22 3.44 -30.3 1 .894 -135 3.97 (-50) (2.381 ) (-143) (3.95) (-70) (2.960) (41-52) (4.24) Table A.2 Standard Free Energies, Entropies and Partial Molal Ionic Heat Capacities Adopted for Species Participating in Reactions Considered in the 3 + Fe —H^O System 333°K 373°K 423°K 473°K Speci es ,o T p 298 o T C p 298 o T- C P 298 o T c n p 298 o S298 o G298 K.Cals H+ 23 31 33 35 -5 0 Fe 69.9 92.8 96.2 103.6 -85.1 -1.1 H20 18.04 18.04 18.15 18.15 16.7 -56.7 F e2°3 26.5 27.9 29.4 30. 7 21.5 -177.4 FeOOH 19.0 20 21 .5 22 22 -117 Fe(0H) 3 20 20 20 20 25.5 -166.5 on 146 A.3 Sample Calculation for the Estimation of Thermodynamic Data at Elevated Temperatures Hydrolysis of Hematite . . 3 + + 3 H„0 = 1 Fe o 0_ + 3H + Fe 2 2 At 2 9 8 ° K , the standard free energy change for the reaction is calculated using the G^gg value for each species '298 l [ ( - 1 7 7 . 4 ) + 3(0) ] - [ - 1 . 1 + 3 ( - 5 6 . 7 J •2550 cals S i m i l a r l y , the standard entropy change for the reaction o is calculated from the S^gg value for each species. AS AC 4C P] 298 .333 298 373 298 423 = j~l_v(21.5) + 3(-5)J = r l (26.5) + 3.(23)] = [ l (27.9) + 3(31)] A c ] = | l (29.4) + 3(33)1 - p J 2 9 8 L 2 J 473 = [ l (30.7) + 3(35) - 298 L 2 •85.1 + 3 2 (16.7)] 55.8 e. 69.9 + 3 (18.04)1 = 15 Cal Mol "°K 2 92 .8 +.3 (18.04)] = -13 9 J 96.2 + 3 (18.15)] = -10 2 J 103.6+ 3 2 (18.15)] = •10 147 The free energy changes at temperatures 333°, 373°, 423° and 473°K are then calculated using equation (e). o A G T = G298 + Cp o T (T-298) - S?pR(T-298) - T C 298 ™ P T In T 298 298 AG 3 3 3 = -2550 + [-15x35] - [55.8x35] - [222x-15x.Hl] = -4774 Cals AG° 7 3 = -2550 + [-13x75] - [55.8x75] - [373x73x.2245] = -6621 Cals AG° 2 3 = -2550 + [-10x125] - [55.85x125] - [423x-10x.35] = -9295 Cals AG° 7 3 = -2550 + [-10x175] - [55.85x1.75] - [473x-10x.462] = -11880 Cals The summary of available measured data on free energy and entropy of species partic i p a t i n g in the hydrolysis re-., actions considered, is shown in Table A.2. Partial molal ionic heat capacities were calculated using; p J 2 9 8 T T 298 and u t i l i z i n g values of a T and e T , which are given in Table A.l . Table A.3 shows the average heat capacities over the temperature ranges considered, for each hydrolysis reaction. These values were u t i l i z e d in equation (e), for the computa- o tion of the free energy, AĜ . of each reaction at the speci- fi e d temperature. The summary of the computed free energies of each reaction is shown in Table A.4 Thermodynamic data were considered for the following e q u i l i b r i a reactions. F e + + + + 3 H20 ;==* 1 f e 2 0 3 + 3H+ (1) 4- 4-4- 4- Fe + 2H20 i = t FeOOH + 3H (2) F e + + + + 3H2Q Fe(0H) 3 + 3H+ (3) Table A.3 Summary of Average Heat Capacities Over the Ranges of 298°K to Upper Limits of 333°, 373°, 423° and 473°K (Estimated by the Correspondence Principle) AC . Cal p Ave Mol - 1 Deg" 1 Reaction A S298 e. u. o. 333 AC p J 298 0 1373 AC . •PJ298 o .423 AG P J298 o,473 AC n P J 298 (1) 55.8 -15 -13 -10 -10 (2) 51 -18 -16 -12 -13 (3) 45.5 -35 -34 -32 -33 Table A.4 Summary of Computations of the Free Energy of Hydrolysis Reactions 3 + Considered in the Fe -H„0 System Reacti on AG°(T), Cal at Indicated Temperatures - 298° K 333°K 373° K 423° K 473°K (1) -2550 -4774 -6621 -9295 -11880 (2) -2500 -4250 -6185 -8598 -10859 (3) 4700 3176 1585 -250 -1826 O Appendix B X-RAY DIFFRACTION RESULTS Appendix B Table B.I X-ray Diffraction Patterns of Goethite Using the Fe Radi ati on Reported Sample d° A I / l ! d° A I / ! l 5.0 20 4.92 30 4.21 100 4.29 100 3.37 20 - - 2 .69 80 2.87 75 2.57 20 - - 2.48 20 2.71 100 2 .44 70 2.65 65 2.25 20 - - 2.18 40 2.44 40 1.719 50 - - en ro Table B.2 X-ray Diffr a c t i o n Patterns of Hematite Using the Fe k Radi ati on Reported Sample (Fe i C l 3 Leach) Sample (FeCl 3/HCl + * d°A * d°A I/I- d°A I / r i 3.66 25 3.92 50 3.81 50 2.69 100 2.89 1 00 2.88 90 2.51 50 2.72 100 2.72 100 2.201 30 , 2.45 35 2.45 40 1 .838 40 2.16 40 2.16 40 1 .690 60 2.06 60 2.06 60 1.596 16 2.01 16 2.01 16 1 .484 35 1 .96 35 1 .96 35 1 .452 35 1 .95 35 1 .95 35 + Relative visual i n t e n s i t i e s , powder camera, CoK radiation Relative diffractometer peak heights cn Co Table B.3 X-ray Diffraction Patterns of Chromite Using Fe K Radi ati on Reported Magnetic Part i c l e s d°A i / i / d°A * I/I, 4.82 50 4.92 10 2.95 60 3.12 40 2 .52 100 2.73 100 2.40 10 - - 2.07 70 2.34 70 1 .69 • 40 2.06 40 1 .60 90 2.0 90 + Relative visual i n t e n s i t i e s , powder camera, MoK^ radiation * Relative diffractometer peak heights.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 9 15
Japan 8 0
United States 8 1
Philippines 2 0
Germany 2 0
City Views Downloads
Beijing 9 0
Tokyo 8 0
Mountain View 6 1
Unknown 4 1
Ashburn 2 0

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

Share

Share to:

Comment

Related Items