Pre-Treatment of Tantalum and Niobium Ores From

PRE-TREATMENT OF TANTALUM AND NIOBIUM ORES

FROM DEMOCRATIC REPUBLIC OF CONGO (DRC) TO

REMOVE URANIUM AND THORIUM

Elie KABENDE BSc (Applied Physics) GradDip (Extractive Metallurgy)

2020

This thesis is presented for the degree of Masters of Philosophy at Murdoch University

DECLARATION

I declare that this thesis is my own account of my research contains as its main content work which has not previously been submitted for a degree at my tertiary education institution.

………………………………………….

Elie kabende

ABSTRACT

Tantalum (Ta) and niobium (Nb) have applications in high-technology electronic devices and steel manufacturing, respectively. Africa, South America and Australia collectively provide about 80% of the global supply of Ta-Nb concentrates. In Africa, Sociéte Minière de Bisunzu

(SMB) located in the Democratic Republic of Congo (DRC) remains the major supplier of Ta-

Nb concentrates containing about 33 wt % Ta2O5 and 5 wt % Nb2O5 as the two oxides of main economic values, associated with uranium (0.14 wt %) and thorium (0.02 wt %). The presence of U and Th complicates the transportation logistics of tantalite to international markets, due to stringent regulation on an allowable limit of U and Th at 0.1 wt %. High radiation levels also hinder further primary beneficiation of the Ta-Nb concentrates at the Bisunzu mine to an extent of at least 50 wt % Ta2O5 and Nb2O5 combined. Digestion of the concentrate using HF is the conventional method to remove U and Th followed by chemical treatment to recover Ta2O5 and

Nb2O5. The HF digestion process is hazardous and requires large investments. The main objective of this study is the mineral identification in the ore and concentrates of SMB mine sites, to investigate the possibilities of upgrading the Ta2O5 and the Nb2O5 by weight using physical separation and concentration, and removing U and Th using chemical treatment.

The mineral identification analysis conducted on eight concentrates sourced from different mining locations at the Bisunzu mine in DRC has shown that manganocolumbite, manganotantalite and ferrotantalite are the major phases with cassiterite, as well as pyrochlore and microlite group of minerals hosting uranium as minor components. The rare earth minerals, quartz and lepidolite were also identified in the ore and concentrates. The quantitative analysis revealed a chemical composition range of 20 – 47 wt % Ta2O5 and 5.4 – 22 wt % Nb2O5 as the principal oxides, associated with 0.005 – 0.802 wt % U and 0.001 – 0.072 wt % Th as radioactive elements. Due to high content of iron in concentrate, the magnetic separation of selected sample

has shown an increase of Ta2O5 and Nb2O5 to 50 wt % and 19 wt %, respectively, in the non- magnetic portion. Samples with low content of iron have shown weak magnetic properties.

Sieving of selected concentrates also increased 18 wt % Ta2O5 and 9 wt % Nb2O5 in coarse size fraction of –2800+1000 μm by about 100%. It was also observed that U and Th were more concentrated in the fines (–1000 μm) i.e., 4844 g/t U and 135 g/t Th compared to 1132 g/t U and

50.5 g/t Th in coarse particles.

The sulfuric acid bake of fine particles (–75 μm) at 300 oC for 2 h followed by leaching

o with H2SO4 and K2S2O8 at 90 C for 3 h lowered the concentration of U and Th in the feed from

0.8 wt % and 0.03 wt %, respectively, to ≤ 0.01 wt %. The measured concentrations of dissolved metal ions (U, Th, Ta and Nb) show reasonable agreement with published solubility diagrams.

An average of 7% Ta and 9% Nb dissolved in the bake-leach process with leaching efficiencies of 100% U and 60% Th. The reliability of the analytical methods and bake-leach results have been confirmed by the elemental mass balance calculation. The H2SO4 and CaF2 leaching at 90 oC dissolved 100% U and 90% Th along with 81% Ta and 43% Nb which warrants further investigations.

ACKNOWLEDGMENTS

First and foremost, words cannot express my gratitude towards my mentor and principal supervisor Associate Professor Gamini Senanayake through my MPhil study and successful completion of my thesis. Nevertheless, this is extended to Doctor Artur Deditius for accompanying me to collect samples from DR Congo mine sites. I would like to thank Professor

Bogdan Dlugogorski, Doctor Hans Oskierski and Associate Professor Mohammednoor

Altarawneh for their support and encouragement to complete my thesis.

To the laboratory staff especially Ken Seymour and Stewart Kelly your technical supports are a blessing in my quest to reach my goals. To Doctor Ibukun Oluwoye, James

Mulwanda and Sean Hunt from College of Science, Health, Engineering and Education your experience and knowledge sharing completes my studies and this thesis is a testimony for all your kind assistance.

To my beloved brothers, sisters and my fiancée (Joyeuse Nziza) thank you for all the love, prayers and support throughout my entire journey. A special word of gratitude goes to the one whom I consider as my father Edouard Hizi Mwangachuchu for his unmeasurable love and care to me and his invaluable contribution to my growth in general, and my studies in particular.

Last but not least I am truly grateful for the MPhil Scholarship from Sociéte Minière de Bisunzu (SMB Sarl) which provided the opportunity to pursue my study in Murdoch

University/Perth.

iii

TABLE OF CONTENTS

DECLARATION i

ABSTRACT i

ACKNOWLEDGMENTS iii

TABLE OF CONTENTS iv

LIST OF FIGURES viii

CHAPTER 1 INTRODUCTION 1

1. 1. Background 1

1.2. Mining challenges at SMB 8

CHAPTER 2 LITERATURE REVIEW 14

2.1. Geology, mineralogy, composition and properties of tantalum and niobium ores 14

2.2. Radioactivity and mineralogy of U and Th in tantalum and niobium ores 18

2.3. Mineralogy of tantalum-niobium concentrates from Africa 23

2.4. Beneficiation routes for upgrading tantalum-niobium concentrates 26

2.5 Physical separation and concentration at Niobec and Greenbushes 30

2.6. Chemical decomposition of tantalum and niobium concentrate 34

2.7. Dissolution of metals from columbite-tantalite 36

2.8. Potential-pH and solubility diagrams 40

CHAPTER 3 MATERIALS AND METHODS 47

3.1. Materials 47

3.1.1. Concentrates 47

3.1.2. Chemical reagents 48 iv

3.2. Methods 49

3.2.1. Sample preparation 49

3.2. Analytical techniques 52

3.2.1. X-ray Fluorescence 52

3.2.2. Geiger-Muller Counter 52

3.2.3. Scanning electron microscopy 53

3.2.4. X-ray diffraction 54

3.2.5. Inductive coupled plasma mass spectrometry (ICP-MS) 54

3.3. Batch baking and leaching experiments 55

3.3.1. Eh - pH measurements 58

3.3.2. Characterisation and elemental mass balance 58

CHAPTER 4 RESULTS AND DISCUSSIONS 61

4.1. Characterisation of ore and concentrate from SMB deposit 61

4.1.1. Geology and mineralogy of Bisunzu mines 61

4.1.2. Tantalite concentrates of different colours from SMB mines 65

4.1.3. Chemical composition of ore and concentrate from SMB deposit 68

4.1.4. XRD analysis of SMB ore and concentrates 79

4.1.5. SEM observations and EDS analysis of sample C 83

4.1.6. SEM observations and EDS analysis of sample G 87

4.1.7. Correlation of elemental assays 89

4.1.8. Comparison between different ores 91

4.1.9. Summary of characterisation 93

4.2. Radioactivity of tantalite samples collected at Bisunzu 94

4.3. Physical separation and concentration 96

4.3.1. Sieving to upgrade tantalum and niobium oxides and remove uranium and thorium 96

4.3.2. Magnetic separation and concentration of Ta-Nb oxides 99

4.3.3. Summary of physical separation and concentration 103

4.4. Leaching to remove uranium and thorium from tantalum and niobium concentrate 104

4.4.1. Summary of leaching results and elemental mass balance 104

4.4.2. Direct leaching with H2SO4 111

o o 4.4.3. Leaching with K2S2O8 at 40 C after roasting at 600 C 114

o 4.4.4. Leaching with K2S2O8 at 40 C after alkaline bake 115

o o 4.4.5. Leaching with K2S2O8 at 40 C after baking with H2SO4 at 400 C 118

o o 4.4.6. Leaching with K2S2O8 at 90 C after baking with H2SO4 at 400 C 121

4.4.7. Effect of baking conditions on uranium and thorium leaching with K2S2O8 at 90 oC 127

(a) Effect of bake time on uranium leaching 127

(b) Effect of particle size on uranium leaching 128

(d) Effect of bake conditions on thorium leaching 130

4.4.8. Effect of leaching conditions on uranium and thorium leaching with K2S2O8 after baking at 400 oC 133

(a) Effect of leaching temperature 133

(b) Effect of acid concentration 136

(d) Effect of oxidant mass to liquid volume ratio 138

(e) Effect of agitation 139

4.4.9. Effect of leaching conditions on minor elements after baking at 400 oC 140

4.4.10. Effect of baking at 100 – 300 oC and leaching at 90 oC 144

(a) Major elements 144

(b) Minor elements 147

4.4.11. Correlation between leaching efficiencies 148

4.4.12. Characterisation of bake product and leach residue of sample C 152

4.4.13. Acid decomposition and leaching of tantalite concentrate with CaF2 161

4.4.14. Summary 165

CHAPTER 5 SUMMARY, CONCLUSIONS AND FUTURE WORK 167

5.1. Summary and conclusions 167

5.2. Recommendations for future work 171

REFERENCES 173

APPENDIX 184

Appendix A 1 184

Appendix A 2 186

Appendix A 3 188

Appendix A 4 190

Appendix A 5 195

Appendix A 6 198

Appendix A 7 215

Appendix A 8 239

vii

LIST OF FIGURES

Figure 1. 1. Global distribution of Ta and Nb mineral deposits 3

Figure 1. 2. Global production of oxides of (a) tantalum and (b) niobium 6

Figure 1. 3. Price fluctuation of (a) concentrate containing Ta2O5 and (b) ferro-niobium 7

Figure 1. 4. SMB deposit location in the eastern side of DR Congo 9

Figure 1. 5. (a) Topographic map of the Bisunzu area with geological observations, (b) topographic map showing the discrepancy between the actual position of the Bisunzu mine/s and the position indicated by Google maps 9

Figure 1. 6. Ground sluicing at Nyabitare beryl–Sn–Ta–Nb-bearing weathered pegmatite, producing high grade tantalum-tin concentrate in western Rwanda 10

Figure 1. 7. Tantalite pegmatite ore treatment roadmap to produce export product at Bisunzu mine 11

Figure 2. 1. Generalised physical beneficiation flow diagram for Ta-Nb minerals, based upon the Niobec and Greenbushes operations 31

Figure 2. 2. Pilot test flow diagram of Ta-Nb concentration with falcon concentrator 33

Figure 2. 3. Flowsheet for recovery of tantalum and niobium oxides using KOH media 34

Figure 2. 4. Dissolution and recovery of Ta, Nb, U and RE after sulfuric acid baking/sulfation and leaching 40

o -3 -3 Figure 2. 5. Solubility data at 25 C for UO2 in 8.10 mol dm sodium perchlorate, ThO2 at I = 0.1 M and 0.5 M (HCl/NaCl) 41

Figure 2. 6. Eh–pH diagram for U–S–H2O system at 25 °C 42

−3 Figure 2. 7. Eh–pH diagrams for Th–SO4–H2O system at 25 °C, {Th}=10 M, {SO4} = 1.0 M 42

o - Figure 2. 8. Solubility data for Nb2O5.nH2O(s) and Ta2O5.nH2O(s), temp = 25 C, I = 0.12 mol L 1 o (NaCl/NaCO3/NaOH); Solubility data of Nb2O5.nH2O(s) and Ta2O5.nH2O(s), temp = 19 C, I = -1 1 mol L (KNO3) 43

viii

Figure 2. 9. Distribution of hydrolysed Nb(V) and Ta(V) species in acidic solution 44

Figure 2. 10. Pourbaix diagram for (a) Nb-O-H system at 25, 75 and 95 oC, (b) Eh-pH diagram for Ta-O-H system 46

Figure 3. 1. Scheme for tantalite concentrate beneficiation using magnetic separation 50

Figure 3. 2. Elemental mass balance diagram 59

Figure 4. 1. (a) Ta-Nb fine-grained lepidolite (pink/purple) and kaolinite (white) in pegmatite, (b) mixture of fine-grained lepidolite and spodumene, (c) coarse-grained rock hosting cassiterite, (d) a sharp contact between kaolinite and Li-mineralization (lepidolite and minor spodumene) (e) gem-quality crystal of tourmaline and (f) contact between pegmatite and fyllite/schist with quartz veins penetrating discordantly 62

Figure 4. 2. Grades of Ta and Nb of different samples 71

Figure 4. 3. Grades of U and Th of different samples 71

Figure 4. 4. Grades of Si of different samples 73

Figure 4. 5. Grades of Fe of different samples 73

Figure 4. 6. Grades of Mn of different samples 74

Figure 4. 7. Grades of Al and Ca of different samples 74

Figure 4. 8. Grades of Li and Cs of different samples 76

Figure 4. 9. Grades of Sn of different samples 76

Figure 4. 10. Correlation between chemical assays based on ICP-MS 1 and XRF (a) Ta, (b) Nb, (c) U and (d) Th 78

Figure 4. 11. XRD pattern of pegmatite ore at Bisunzu mine 79

Figure 4. 12. XRD pattern of tantalite concentrate of sample C 82

Figure 4. 13. XRD pattern of tantalite concentrate of sample G 82

Figure 4. 14. Backscattered SEM image of tantalite sample C from Bisunzu mine 84

Figure 4. 15. (a) and (b) Backscattered SEM images and their correspondent EDS elemental analysis of tantalite concentrate from Bisunzu mine (sample C) 85

Figure 4. 16. (c) Backscattered SEM image and their correspondent EDS elemental analysis of tantalite concentrate from Bisunzu mine (sample C) 86

Figure 4. 17. Backscattered SEM image of tantalite sample G from Bisunzu mine 88

Figure 4. 18. Correlation between of chemical assays (a) Al and K, (b) Al and Li, (c) Li and Cs, (d) Na and Th 90

Figure 4. 19. Sieving beneficiation of samples D and G (a) Ta2O5, (b) Nb2O5, (c) Fe2O3, (d) TiO2, (e) SiO2 and (f) MnO 97

Figure 4. 20. Sieving removal of (a) U and (b) Th from samples D and G 98

Figure 4. 21. Composition of samples FH, FM and FL before and after magnetic separation (a) Ta2O5, (b) Nb2O5, (c) Fe2O3 and (d) TiO2 100

Figure 4. 22. Composition of samples FH, FM and FL before and after magnetic separation (a) uranium (b) thorium 102

-3 Figure 4. 23. Eh–pH diagram for Th–SO4–H2O system at 25 °C, [Th] = 10 M, {SO4} = 1.0 M 113

-3 -1 Figure 4. 24a-b. Total uranium in solution for the dissolution of UO2 in 8.10 mol L sodium o o perchlorate at 25 C and Th in acid solution at I = 0.5 M (NaCl/NaOH) and 22 C. UO2 and ThO2 o literature data, *. 400 and 40 C bake and leach temperatures, respectively, in 4 M H2SO4 114

o Figure 4. 25(a)-(b). Solubility data for Nb2O5.nH2O(s) and Ta2O5.nH2O(s) at 25 C and I = 0.12 -1 mol L (NaCl/NaCO3/NaOH); Solubility data of Nb2O5.nH2O(s) and Ta2O5.nH2O(s), temp = 19 o -1 o C, I = 1 mol L (KNO3); Ta(V) and Nb(V) at 40 C in 4 M H2SO4 117

Figure 4. 26. Eh-pH diagram for uranium in sulfate solution at 25 °C, [U] = 4.0 × 10-3 M, [S] = 1.0 M) 122

o -3 -3 Figure 4. 27a-b. Solubility data at 25 C for UO2 in 8.10 mol dm sodium perchlorate; ThO2 at I = 0.1 M and 0.5 M (HCl/NaCl); * bake at 300 oC, ** bake at 100 oC and leach at 90 oC in 4 M H2SO4 124

Figure 4. 28. TG-DSC of 96 wt % sulfuric acid in air (heating rate = 5 oC/min) 126

Figure 4. 29. Effect of bake conditions on U and Th leaching efficiencies of sample C 129

Figure 4. 30. Effect of bake time on leaching efficiency of U at short and long leaching times (a) 15 minutes, (b) 3 h 131

Figure 4. 31. Effect of bake time on leaching efficiency of Th at short leaching time (15 minutes) 132

Figure 4. 32. Effect of bake time on leaching efficiency of Th at long leaching time (3 h) 132

Figure 4. 33. Effect of various leaching conditions on U dissolution of sample C 134

Figure 4. 34. Effect of leaching temperatures on U leaching efficiency at short and long leaching times (a) 15 minutes, (b) 3 h 135 x

Figure 4. 35. Effect of various leaching conditions on Th dissolution of sample C 136

Figure 4. 36. Effect of agitation speed on leaching efficiency of (a) U and (b) Th of sample C 139

Figure 4. 37. Dissolution of minor elements at suitable conditions of sample C 141

Figure 4. 38. Effect of baking temperature of 100 – 400 oC on leaching efficiency of U, Th, Ta and Nb of sample C 145

Figure 4. 39. Experimental scheme for leaching uranium and thorium from tantalite concentrate 149

Figure 4. 40. Correlation between Ta and Nb leaching efficiencies at different conditions 150

Figure 4. 41. Correlation between Li and Cs leaching efficiencies at different conditions 150

Figure 4. 42. Correlation between (a) U and Fe, (b) U and Mn leaching efficiencies at different conditions 151

Figure 4. 43. XRD patterns (a1) feed, (a2) bake product and (a3) leach residue of sample C 153

Figure 4. 44. (a) Secondary electron (SE), (b) analysis of bake product of sample C 156

Figure 4. 45. (a) Secondary electron (SE), (b) analysis of leach residue of sample C 157

Figure 4. 46. (a) Secondary electron (SE), (b) EDS analysis of bake product of sample G 158

Figure 4. 47. (a) Secondary electron (SE), (b) EDS analysis of leach residue of sample G 159

Figure 4. 48. Leaching efficiency of U, Th, Ta and Nb in CaF2 and H2SO4 of sample C 162

Figure 4. 49. XRD patterns of sample C (a) tantalite concentrate and (b) leach residue 163

LIST OF TABLES

Table 1. 1. Global estimated reserves of tantalum oxides 4

Table 1. 2. Important deposits of tantalum and niobium minerals across the world and their oxides grades 5

Table 1. 3. Chemical analysis of pegmatite ore and exported sample from SMB after milling and mixing of concentrate 12

Table 2. 1. Chemical composition and crystal structure of most important tantalum-niobium minerals 15

Table 2. 2. Ionic radius of uranium, thorium, tantalum and niobium species 17

Table 2. 3. Half-lives and abundances of uranium and thorium isotopes 19

Table 2. 4. Elements associated with rocks hosting uranium and thorium 20

Table 2. 5. Uranium concentrations in ore deposits, earth crust and sea water 21

Table 2. 6. Niobate, tantalate and titanate ores bearing uranium and thorium 22

Table 2. 7. Mineral identification of tantalum-niobium concentrates from different locations 24

Table 2. 8. Constituents and grades of tantalite ores from different locations 25

Table 2. 9. Choice of technology based on tantalum grade and raw material input 27

Table 2. 10. Suitability of concentration criterion for particles of different sizes 27

Table 2. 11. Classification of minerals based on their magnetic properties 29

Table 2. 12. Chemical reactions for decomposition of tantalum-niobium ores 36

Table 2. 13. Leaching efficiencies of metals from coltan in different media 37

Table 2. 14. Chemical reactions for acid baking of tantalum and niobium ores 39

Table 3. 1. Columbite tantalite concentrates collected at different SMB mining sites 47

Table 3. 2. List of chemical reagents 48

Table 3. 3. Conditions for magnetic separation and concentration 49

Table 3. 4. Calculating of radioactivity of uranium and thorium from an assay 53

Table 3. 5. Direct leaching of U, Th, Ta and Nb under various conditions 56

xii

Table 3. 6. Direct leaching of U, Ta, Nb and Th using CaF2 in H2SO4 56

Table 3. 7. Sulfuric acid baking and water or acid leaching conditions 57

Table 4. 1. Names and chemical formulae of minerals associated with columbite-tantalite deposits 64

Table 4. 2. Feature of columbite-tantalite concentrate in Namibia 66

Table 4. 3. Feature of columbite-tantalite concentrate at the Bisunzu mines 67

Table 4. 4. Chemical composition of pegmatite ore and exported concentrate sample at Bisunzu mine 68

Table 4. 5. Chemical composition of concentrates collected at Bisunzu mine based on ICP-MS 69

Table 4. 6. Chemical composition of concentrates collected at Bisunzu mine based on XRF 69

Table 4. 7. Correlation parameters of Ta, Nb, U and Th analyses from ICP-MS 1 and XRF 78

Table 4. 8. Mineral identification from eight concentrates collected at the Bisunzu mine 81

Table 4. 9. Correlation of elemental assays of different samples 91

Table 4. 10. Mineralogical of columbite-tantalite deposits from Nigeria, Mozambique and DR Congo 92

Table 4. 11. Chemical composition of columbite-tantalite from various locations 92

Table 4. 12. Radioactivity due to uranium and thorium oxides in coltan concentrates 95

Table 4. 13. Comparison of elemental leaching efficiencies 105

Table 4. 14. Repeat test results of elemental leaching efficiencies 105

Table 4. 15. Elemental mass balance for bake – leach results of sample C, T 25 107

Table 4. 16. Leaching efficiencies under direct leaching with base, acid and oxidants 108

Table 4. 17. Leaching efficiencies under direct leaching with acid and fluoride 108

Table 4. 18. Leaching efficiencies after acid or base baking and leaching with acid and oxidants (part I) 109

Table 4. 19. Leaching efficiencies after acid baking and leaching with acids and oxidants (part II) 110

Table 4. 20. Chemical reactions of UO2 and ThO2 with H2SO4 112

xiii

Table 4. 21. Solution concentrations and leaching efficiencies 112

Table 4. 22. Leaching efficiency after roasting and leaching with acid 115

Table 4. 23. Leaching efficiency of minor elements after roasting and leaching with acid 115

Table 4. 24. Leaching efficiency after baking with KOH and leaching with water 116

Table 4. 25. Chemical reactions of UO2, ThO2, Ta2O5 and Nb2O5 with H2SO4 118

Table 4. 26. Leaching efficiency after acid baking and leaching with water 119

Table 4. 27. Leaching efficiency of minor elements after acid baking and leaching with water 119

Table 4. 28. Leaching efficiency after acid baking and leaching with acid 120

Table 4. 29. Leaching efficiencies of minor elements after acid baking and leaching with acid 121

Table 4. 30. Leaching efficiencies obtained from eight different concentrates after acid baking and leaching with acid 123

Table 4. 31. Chemical reactions of minor oxides with H2SO4 140

Table 4. 32. Elemental leaching efficiencies of minor elements 142

Table 4. 33. Effect of particle size on leaching of minor elements 142

Table 4. 34. Effect of leaching temperature on dissolution of minor elements 144

Table 4. 35. Effect of bake temperature on leaching efficiency of minor elements 147

Table 4. 36. Mineral identification of feed, bake product and leach residue of sample C 153

Table 4. 37. EDS quantitative analysis of feed, bake product and leach residue of samples C and G 160

Table 4. 38. Elemental analysis of feed, bake product and leach residue of samples C and G 160

Table 4. 39. Decomposition reactions of tantalite concentrate in CaF2 and H2SO4 161

xiv

Chapter 1: Introduction

CHAPTER 1

INTRODUCTION

1. 1. Background

The history and discovery of tantalum started in 1802 by the Swedish analytical chemist and mineralogist Anders Ekeberg in mineral samples obtained from Ytterby in Sweden. The minerals from Ytterby are rich in several rare earth elements, including yttrium (Y) (Linnen et al., 2014). According to British Geological Survey (BGS) (2011), the name ‘tantalum’ took roots from the ancient Greek myth of Tantalus, a powerful hero who challenged the gods, lost and was punished by eternal thirst and hunger. When Ekeberg discovered tantalum, he needed to dissolve it to test its properties. However, the common mineral acids could not dissolve it, and when he eventually succeeded, he named the new element after Tantalus, remembering his struggles.

According to the same ancient Greek myth, Niobe was a daughter of Tantalus, and this was a fitting connection when niobium capacitors first started to be developed in Russia. A shortage of tantalum forced the development of niobium capacitors in Russia. Eventually, niobium capacitors were designed as an alternative to tantalum capacitors, and they are still manufactured in Russia today with less efficiency compared to tantalum capacitors (TIC, 2018; Nikishina et al., 2013).

Tantalum (Ta) is a driver of the high-technology microelectronic devices, with a new trend of being used as an implant in human tissue. Niobium (Nb), on the other hand, is a valuable additive to high - strength low - alloy (HSLA) steel manufacturing (Nikishina et al., 2013; Nete et al., 2012) which made them critical and strategic metals of the last few decades (Nikishina et al., 2014; EL-Hussaini, 2001; Mackay et al., 2015; Simandl, 2018). The detailed specification of both tantalum and niobium (and their derivatives) employed in various applications based on their purity is shown in Appendix A 1 and Appendix A 2, respectively. Considering the major

Chapter 1: Introduction

elements in the ore, the deposits bearing tantalum are wider than those of niobium (BGS, 2011;

United States Geological Survey, 2018; Nikishina et al., 2014). The reserves of these metals are abundant in Brazil, Australia, China and Africa, in countries such as the Democratic Republic of

Congo (DR Congo), Zambia, Ethiopia and Nigeria, among others. Figure 1.1 shows the global distribution of Ta/Nb mineral deposits.

In general, projects involving critical and strategic metals/materials have high popularity with potential investments (Simandl et al., 2018). Brazil has the largest share of the global tantalum resources with approximately 41%; the remaining 59% of identified tantalum oxides reserves occur in decreasing order of tonnage: Australia with 21%, while only 1% found in

Europe (BGS, 2011; USGS, 2019) as listed in Table 1.1. Tantalum and niobium ores of high- grade of a commercial interest concerning oxides and reserves are listed in Table 1.2. The currently identified resources of niobium are predominantly located in Brazil 95% and Canada

3.5% (Schulz et al., 2017). Small quantities of resources reside in Angola, Australia, China, DR

Congo, Greenland, Malawi, Russia, South Africa, and the United States. Unlike niobium, the production of tantalum appears extensive in countries in Central Africa. According to US

Geological Survey reports (USGS) published in 2016, 2017 and 2019, DR Congo and Rwanda remains the current leading chain suppliers of tantalum, with production of about 66.7% and

68.2% in 2014 and 2018, respectively, as illustrated in Figure 1.2a. In 2002, Australia accounted for 60% of the global Ta mineral production, but this figure dropped to less than 10% due to the temporary closure of Wodgina and Greenbushes mines (Viruet et al., 2013). On the other hand,

Figure 1.2b depicts Brazil as the top producer of niobium over the last 10 years, with approximately 90.6% of the total world production followed by Canada (USGS, 2016).

Chapter 1: Introduction

Figure 1. 1. Global distribution of Ta and Nb mineral deposits (Linnen et al., 2014)

Chapter 1: Introduction

The price of material containing tantalum and niobium oxides on the international market depends heavily on the purity of both Ta2O5 and Nb2O5. According to the report published by

Tantalum-Niobium International Study Centre (TIC) in 2017, the columbite-tantalite ore containing 30% tantalum oxide (Ta2O5) or 50% niobium oxide (Nb2O5) are saleable in the international market. However, niobium ore is converted into ferroniobium (Fe-Nb), which is the marketable product (Simandl, 2018; USGS, 2018; Schulz et al., 2017). The Ta2O5 prices have been characterised by a habit of volatility over the last two decades, while the ferroniobium (Fe-

Nb) prices were relatively unchanged. Figures 1.3a and 1.3b show the change of price of Ta2O5 and Fe-Nb over a period of 20 years. Despite the challenges of production, the demand for Ta commodity worldwide is expected to increase by about 5.81% for the next five years, while the niobium as an alloy of Fe-Nb is expected to increase by approximately 7% from 2019 to 2024

(Roskill information services, 2017 & 2018). In addition, the trend shows that new projects are ongoing to produce tantalite either as the main product or as a by-product, specifically from the lithium bearing ores (Simandl et al., 2018).

Table 1. 1. Global estimated reserves of tantalum oxides

Region Reserves of Ta2O5 References Tonnage % USGS, 2011, 2015 South America 129274 41 & 2019 Australia 65771 21 China and South Asia 33112 10 Russia and East Europe 31298 10 Central Africa 28576 9 Other African countries 21318 7 North America 5443 9 Other Europe countries 2268 1

Chapter 1: Introduction

Table 1. 2. Important deposits of tantalum and niobium minerals across the world and their oxides grades

Country Deposit name Company Status Type of ore Mining method Commodity Reserves % % (Mt) Nb2O5 Ta2O5 Australia Greenbushes Global advanced metals and Talison Operation Pegmatite Open/underground Ta, Nb, Sn, Cs & Li 6 0.023 0.026 Wodgina Global Advanced metals Discontinued Pegmatite Open pit Ta, Nb, Be & Sn 28 NR 0.042 Mt Cattlin Galaxy Resources Ltd. Operation Pegmatite Open pit Li & Ta 10 NR 0.015 Tabba Tabba Pilbara Minerals Ltd. Operation Pegmatite Open pit Ta 0.32 NR 0.095 Toongi-Dubbo Alkane Resources Ltd. Development Trachyte Open pit Zr, Hf, Nb, Y, Ta & REE 73 0.46 0.03 Brazil Araxá CBMM Operation Weathered Open pit Nb 462 2.48 NR Carbonatite Catalao-Boa vista China Molybdenum Co. Operation Weathered Open pit Nb 42 1.2 NR Carbonatite Mibra/Volta Grande Advanced Metallurgical Group Operation Pegmatite Open pit Ta, Nb, Sn & Li 6 0.009 0.038 mine Pitinga Mine Minsur/Mineração Operation Peralkaline granite Open pit Sn, Nb & Ta 267 0.22 0.027 Canada Niobec Mine Magris Resources Inc. Operation Mineralised Underground Nb 630 0.42 NR carbonatite Aley Taseko Mines Ltd.Corp. EIAa process Mineralised Open pit Nb 84 0.50 NR carbonatite Tanco Cabot corporation Operation Pegmatite Underground Ta, Cs & Li 2 NR 0.22 DR Congo Bisunzu Societe Miniere de Bisunzu (SMB) Operation Pegmatite Open pit Ta, Nb, Sn & Li 2700 NR 0.045 Ethiopia Kenticha Mine Elenilto Mining Operation Pegmatite Open pit Ta & Nb 116 NR 0.02 Malawi Kanyika Globe Advanced Metals & Minerals Ltd Bankable Nepheline Syenite Open pit Nb, Ta, U & Zr 21 0.330 0.015 Egypt Abu Dabbab Gippsland Ltd. Bankable Granite Open pit Ta & Sn 32 NR 0.027 China Yichun Jiangxi Tungsten Industry Group Co. Ltd Operation Granite Open pit Ta, Li, Nb & Sn 109 0.0213 0.0213 Spain Penouta mine Strategic Minerals Spain Exploration Alkaline Granite NR Ta, Nb & Snc 95.6 0.009 0.0094 Spain Alberta II Strategic Minerals Spain Exploration Pegmatites NR Ta, Nb, Snd & Lie 12.3 NR 0.0121 f Denmark Motzfeldt Regency Mines/Greenland Exploration Syenite NR Ta, Nb & Zr 340 0.19 0.012 Finland Sokli Yara international ASA NR Carbonatite NR Ta, Nb & Pg 110 0.1 NR France Echassières Imerys NR Granite NR Ta 5h NR 0.012

Pappy, 2012; Melcher et al., 2015. NR: Not reported, a. EIA: environment impact assessment, b. bankable (project to be funded), c. Sn = 0.044%, d. Sn = 0.044, e. Li = 0.204%, f. Zr = 0.34%, g. P = 7.2% & h. 5 Mt Ta2O5 / year

Chapter 1: Introduction

(a)

800 2014 2015 2016 700 2017 2018

600 (metric tonnes) (metric

5 500

O 2

400

300

Production of Ta of Production 200

100

0 Australia Brazil China DR Rwanda Ethiopia Nigeria Others Congo Countries

(b)

70000 2014 2015 2016 2017 2018

60000

50000

(metric tonnes) (metric

5 O 2 40000

30000

Production of Nb of Production 20000

10000

0 Brazil Canada Others

Countries

Figure 1. 2. Global production of oxides of (a) tantalum and (b) niobium (USGS, 2016; USGS, 2019)

Chapter 1: Introduction

(a)

(b)

Figure 1. 3. Price fluctuation of (a) concentrate containing Ta2O5 and (b) ferro-niobium (Linnen et ala, b., 2014; Mackay and Simandla, b, 2015; USGSa, b, 2018)

Chapter 1: Introduction

1.2. Mining challenges at SMB

Societé Minière de Bisunzu (SMB), located 62 km east from the City of Goma in the

Democratic Republic of Congo (DR Congo) as indicated in Figure 1.4, is one of the major suppliers of tantalum to the global market. The production at SMB is dominated by artisanal and small-scale mining (ASM). Figure 1.5 shows six large and potential mining sites, while other small sites are scattered across the mine. Note that all mining sites within the SMB perimeter are in full operation as an open pit. Apart from being the important deposit of columbite-tantalite

(Coltan), it could be the second largest to host a deposit with a higher concentration of 0.045 wt

% Ta2O5 (Table 1.2) in the SMB ore (SMB report, 2017), after Tabba Tabba in Western Australia with an average of 0.095 wt % Ta2O5 (Pilbara Minerals Ltd Open Report, 2015).

The coltan mining activities at SMB involves traditional digging and washing in order to extract the concentrate-bearing mineral. The pegmatite-rich ore containing Ta and Nb is identified by its colours, instead of the mineralisation based geochemical studies. The ore colour is white, pink, yellow or black. Figure 1.6 shows the artisanal production of tantalite concentrate from a coloured weathered pegmatite ore using ground sluice considered as a primary concentrator, favoured by artisanal miners, as no electrical power is required. The rich pegmatite ore is mixed with water to form a slurry that is then sloshed in a wash tube or ground sluice to separate the concentrate from impurities. Based on its high density, tantalum and niobium bearing minerals sink to bottom along with other heavy minerals commonly cassiterite, ilmenite and others, whereas the lights are washed away. Eventually, this washing system can be summarised as primary concentration of tantalite concentrate using gravity technique to obtain a pre-concentrate (stage

1).

Chapter 1: Introduction

Figure 1. 4. SMB deposit location in the eastern side of DR Congo

Chapter 1: Introduction

Figure 1. 6. Ground sluicing at Nyabitare beryl–Sn–Ta–Nb-bearing weathered pegmatite, producing high grade tantalum-tin concentrate in western Rwanda (Sindandl et al., 2018)

Figure 1.7 describes the flow of various stages for the production of tantalite concentrate at the SMB. The pre-concentrate is collected for the second wash to obtain a concentrate (stage

2). The concentrate is exposed to open air for drying (stage 3). The dry concentrate is subjected to magnetic separation to reduce the content of some magnetic materials such as iron, manganese, magnetite and titanium phases using handheld magnetic tweezers considered as the second treatment. In stage 4, the coarse grains are hand sorted and crushed using a cone to reduce the particle size to a moderate size of about ≤ 1000 μm. At SMB, the X-Ray Fluorescence (XRF) analysis is routinely performed on a representative sample collected from a bulk concentrate after being mixed, homogenised and ready for exportation. The results summarised in Table 1.3 indicate that the concentrate consists of Ta2O5 and Nb2O5 along with SiO2, Al2O3, Fe2O3 and SnO2 as the major components, and uranium and thorium as minor radioactive components. The presence of these naturally occurring radioactive materials (NORM) poses problems with transportation within the international borders. According to the reports published by Tantalum-Niobium

International Center (TIC) in 2016 and the European commission in 2002, the overall content of 10

Chapter 1: Introduction

0.1 wt % each of U3O8 and ThO2 which is equivalent to 10 Bq/g is the limit of allowable content of naturally occurring radioactive material (NORM) in the columbite-tantalite minerals and concentrates for transportation.

Figure 1. 7. Tantalite pegmatite ore treatment roadmap to produce export product at Bisunzu mine

(Unpublished, SMB company report)

Chapter 1: Introduction

Table 1. 3. Chemical analysis of pegmatite ore and exported sample from SMB after milling and mixing of concentrate

Pegmatite ore Exported sample Elements wt % Oxides wt % Elements wt % Tantalum (Ta) 0.300 Tantalum pentoxide (Ta2O5) 33.0 Tantalum (Ta) 27.0 Niobium (Nb) 0.080 Niobium pentoxide (Nb2O5) 5.40 Niobium (Nb) 4.00 Tin (Sn) 0.040 Tin oxide (SnO2) 2.80 Tin (Sn) 2.00 Uranium (U) 0.001 Uranium (U3O8) 0.17 Uranium (U) 0.14 Thorium (Th) ND Thorianite (ThO2) 0.02 Thorium (Th) 0.012 Titanium (Ti) 0.005 Rutile (TiO2) 0.96 Titanium (Ti) 0.58 Silicon (Si) - Silica (SiO2) 25.3 Silicon (Si) 12.0 Iron (Fe) 0.060 Hematite (Fe2O3) 3.60 Iron (Fe) 3.00 Tungsten (W) - Tungsten trioxide (WO3) 0.02 Tungsten (W) 0.02 Aluminium (Al) 42 Aluminium oxide (Al2O3) 5.17 Aluminium (Al) 3.00 Zirconium (Zr) - Zirconium oxide (ZrO2) 0.37 Zirconium (Zr) 0.27 Manganese (Mn) 0.050 Manganese oxide (MnO) 0.23 Manganese (Mn) 0.18 Potassium (K) 0.600 Potassium oxide (K2O) - Potassium (K) - Caesium (Cs) 0.010 Caesium oxide (Cs2O) - Caesium (Cs) - Moisture - Moisture 0.10 Moisture -

(Unpublished, Alex Stewart International Rwanda Ltd. report, 2006); ND: Not detected

The concentrates with 1 Bq/g of NORM content are considered as of concern for mining and handling activities (Nete et al., 2014a). In fact, if the radioactivity of the mineral is below 10

Bq/g, the material is exempted from class 7 transport regulations. However, the material should be accompanied by an analysis certificate to demonstrate radioactivity levels. On the other hand, if it is above 10 Bq/g, the material is subjected to class 7 transport regulations, and a certificate has to be obtained from the National Nuclear Regulator to handle such material in a large quantity

(TIC, 2016). Referring to the UN 2910 code, material exhibiting a radiation level below 5 μSv/h is considered as “excepted package” which reflect low risk of the material for transportation (TIC,

2016). Otherwise, regulatory requirements for transportation and handling have to be met.

Furthermore, due to the traditional way of extraction, companies are facing a loss of fine and ultrafine grains bearing valuable minerals due to the uncontrolled stream of water used during the washing stage to evacuate the burden soil. Finally, low income from export is a result of low grade of Ta2O5 and Nb2O5 currently produced.

Chapter 1: Introduction

Addressing the above challenges could potentially boost the production supply chain of Ta and Nb minerals, resulting in low market prices of both minerals and end-user products. Thus, it is important to extend the mineral identification studies on the samples of SMB pegmatite ore and concentrates from different sites to identify the important Ta-Nb mineral rich end-members and minerals carrying uranium and thorium. It is also essential to study the possibilities of upgrading the currently produced concentrate by physical or chemical methods in order to maximise the export income. Thus, it is important to study different leaching process to obtain a concentrate free from U and Th for their transportation and safety for further processing by SMB or buyers.

The main objectives of the present study are:

1. Review of literature on columbite-tantalite ores.

2. Collection of samples from SMB sites in DR Congo.

3. Measurement of radiation levels of different samples.

4. Characterisation of ore and concentrate samples.

5. Physical pre-treatment to upgrade the concentrate.

6. Chemical pre-treatment to remove radioactive materials from concentrate through direct

leaching and or acid baking followed by leaching.

Chapter 2: Literature Review

CHAPTER 2 LITERATURE REVIEW

2.1. Geology, mineralogy, composition and properties of tantalum and niobium ores

Tantalum bearing minerals occur mainly in the pegmatite and other related mineral rock deposits as tantalite, columbite, pyrochlore, microlite, wodginite and struvenite among others.

Table 2.1 presents some of the important mineral groups hosting tantalum and niobium. The abundances of tantalum and niobium in the upper continental crust are 0.7 g/t and 8 g/t, respectively (Linnen et al., 2014). Naturally, tantalum and niobium bearing minerals are found in pegmatite in conjunction with other economic minerals such as beryl, spodumene, lepidolite, feldspar, mica, scheelite, cassiterite, uraninite and monazite (Amuda et al., 2007). Tantalum and niobium bearing minerals are discovered as primary or secondary deposits. In fact, all primary niobium and tantalum mineral deposits are associated with igneous rocks and they are classified into three categories (Linnen et al., 2014; BGS, 2011).

1. Carbonatite hosted deposits (niobium and industrial minerals);

2. Alkaline to peralkaline granites and syenites (niobium, yttrium, rare earths, zirconium, tin and tantalum);

3. Peraluminous pegmatites and granites (tantalum, niobium, tin, tungsten, caesium, lithium).

Moreover, tantalum and niobium elements are mechanically and chemically resistant to weathering and they possess high specific gravities which give them a favour to be accumulated as secondary placer and alluvial deposits (Pohl, 2016).

Chapter 2: Literature Review

The tantalum, tin and lithium minerals are derived from rare metal pegmatites dated radiometrically at 2.6 billion years (Linnen et al., 2014). The albite (feldspar) containing tantalum/tin minerals and the spodumene enriched in lithium are the two main ore zones within the pegmatites. Thus, the ore bodies are complex in their chemical composition. Their continuity may occupy along a strike length of 3-5 kilometers (Figure 1.5). In pegmatite ore, a large amount of rock must be mined for a relatively small recovery of mineral as the distribution of economic minerals in them is very irregular. Quartz, potash, soda feldspar and muscovite are mostly consistent components in the pegmatite ores. Lenticular quartz bodies surrounded by alteration of zones containing minerals such as kaolin, sericite, tourmaline, flurvite, beryl among others are the main mineralisation in the pegmatite ores (Linnen et al., 2014).

Table 2. 1. Chemical composition and crystal structure of most important tantalum-niobium minerals

Mineral group Mineral name Formula Nb2O5, Ta2O5, Crystal structure wt % wt % Columbite- Columbite (Fe,Mn)(Nb,Ta)2O6 40-75 1-40 Orthorhombic Tantalite Tantalite (Mn,Fe)(Ta,Nb)2O6 2-40 42-84 Orthorhombic Pyrochlore Pyrochlore (Na,Ca)2Nb2O6(O,OH,F) 37-66 0-6 Octahedral Microlite (Ca,Na,U)2Ta2O6(O,OH,F) 0-7 66-77 Isometric 3+ Yttrotantalite Yttrotantalite (Y,U,Ca)(Ta,Nb,Fe )2O6 41-56 14-27 Monoclinic 3+ Rutile Strüverite (Ti,Ta,Nb,Fe )2O6 9-14 6-13 Tetragonal Ilmenorutile Fex(Nb,Ta)2x4Ti1-xO2 27.9 NR Tetragonal 3+ Fergusonite Fergusonite (Re )(Nb,Ta)O4 14-46 4-43 Tetragonal Samarsakite Samarsakite (Fe,Ca,U,Y,Ce)2(Ta,Nb)2O6 40-55 15-30 Monoclinic Simpsonite Simpsonite Al4(Ta,Nb)3O13(OH) 0.3-6 60-80 Trigonal Thoreaulite Thoreaulite SnTa2O6 NR 73-77 Monoclinic Tapiolite Tapiolite (Fe,Mn)(Ta,Nb)2O6 8-15 40-85 Tetragonal Ixiolite Ixiolite (Ta,Nb,Sn,Mn,Fe)4O8 8.30 68.96 Orthorhombic Wodginite Wodginite (Ta,Nb,Sn,Mn,Fe)O2 8.37 69.58 Monoclinic Perovskite Loparite (Ce,La,Na,Ca,Sr)(Ti,Ta,Nb)2O6 4-20 0.5-3 Orthorhombic Lueshite NaNbO3 81.09 NR Orthorhombic Euxenite Euxenite (Y, Ca, Ce, U, Th)(Nb, Ti, Ta)2O6 4-47 0-47 Orthorhombic

(Linnen et al., 2014; BGS, 2011; Othmer, 1997), NR: Not reported

Chapter 2: Literature Review

The tantalite mineralisation is usually concentrated in the quartz-rich part of the pegmatite. In addition, the highest grade of tantalum deposits occur in pegmatite that contain relatively high concentration of the tin, lithium and anomalous concentration of beryl. The grain size of mineral tantalite is very fine and not visible in hand specimens (Amuda et al., 2005).

Tantalite is a compound of tantalum, iron and manganese and is often associated with columbite.

Eventually, the columbite-tantalite mineral group is the most common group of tantalum and niobium bearing minerals. On the other hand, the pyrochlore is of great economic importance particularly for niobium (BGS, 2011; Nete et al., 2012). Both elements exist in more than 150 minerals as main or minor species (Nete, 2013). The common oxide mineral group is (Ta/Nb)2O5 and hydroxides with the exception of behierite (Ta,Nb)(BO4) as borate mineral. The silicates of

Ta and Nb do exist, but are relatively rare (BGS, 2011; Cẽrný and Ercit, 1989). Currently, one non-oxide mineral has been identified to host tantalum carbide (TaC) (Nete et al., 2014b).

Referring to Table 1.1, the SMB deposit contains 33 wt % Ta2O5 and 5 wt % Nb2O5. These values reflect the content of Ta2O5 and Nb2O5 in tantalite or columbite minerals (Table 2.1).

Thus, it is likely that the minerals bearing Ta and Nb at SMB belongs to columbite-tantalite group.

The variability of the chemical composition of Ta-Nb oxides on different scales in pegmatite ores is due to the existence of extremely variable magmatic rocks in terms of textures and geochemical aspects. The variability is reflected by the complex zoning patterns within a single pegmatite, a wide variety of rare elements and, among them, different compositions of individual minerals. These variations develop through fractional crystallisation, contamination, postmagmatic overprint, penetration of different fluids into the rock and other processes

(Melcher et al., 2015). Tantalite mineral is mainly divided into two main groups either with iron- rich, ferrotantalite or manganese rich-end-member, manganotantalite. The properties including

Chapter 2: Literature Review

colour, transparency and streak are of relevance to distinguish them. For example, ferrotantalite is black and opaque, while manganotantalite varies between blackish to reddish-brownish. An even rare member of this series is the magnesium rich end-member, magnesiotantalite. The tantalite fractures are uneven and the luminescence varies from opaque to translucent or transparent (Knorring, 1958). In general, despite being discredited, the name tantalite is still used without designation, and often, it is labelled as columbite-tantalite (Coltan), specifically, in countries in Central Africa, since the exact species can be difficult to determine. In addition, tantalite and tapiolite ((Fe,Mn)(Ta,Nb)2O6 are dimorph minerals with an orthorhombic and tetragonal crystal structure, respectively (Linnen et al, 2014). Tantalite mineral has a hardness of

6 on the Moh’s scale with a specific gravity of 5.2-8.0. The ionic charge to radius ratio makes tantalum and niobium to possess a strong affinity towards oxygen. Thus, they are naturally occurring as oxides. In addition, they are characterised by similar chemical and physical properties. Table 2.2 presents the ionic radius of U, Th, Ta and Nb species indicating that they have same ionic radius which allows them to occur naturally together (Linnen et al., 2014;

Bulatovic, 2010; Küster, 2009; De Paolo, 1988; Siderenko and Doynikova, 2010; Nete et al.,

2016)

Table 2. 2. Ionic radius of uranium, thorium, tantalum and niobium species

Ion Radius (pm) U4+ 89 U5+ 90 U6+ 87 Th4+ 94 Ta4+ 68 Ta5+ 70 Nb4+ 69 Nb5+ 70

(Dickin, 1995)

Chapter 2: Literature Review

2.2. Radioactivity and mineralogy of U and Th in tantalum and niobium ores

Table 2.3 gives the naturally occurring three major U and Th isotopes, half-lives and their abundances. Uranium is weakly radioactive because all naturally occurring isotopes of uranium

(238U, 235U and 234U) are unstable, with half-lives varying from 159,200 years to 4.5 billion years.

There are 27 known isotopes of uranium with atomic weights ranging from 217-219, 222-240 and 242, with half-lives varying from billions of years to a few nanoseconds. Uranium is mobile and enter into the structures of major rock forming minerals. Thus, it is continuously enriched in different rocks, whereas thorium is immobile under most circumstances. Uranium is widespread and geochemically persistent due to the following three factors:

(i) During mineral formation, U4+ ion behaves as a typical cation demonstrating isomorphism with Th4+, Zr, REE and rarely with Ca and Fe2+. The U4+ ion appears, even in little quantities, in various complex oxides, silicates, and in phosphates of high temperature (Table 2.4) (Sidorenko and Doynikova, 2010).

(ii) Unlike the tetravalent U4+ ion which forms insoluble species, the hexavalent U6+ ion is stable under a wide range of conditions (high to atmospheric pressure and temperature) at the surface

2− 6+ of the earth (Table 2.4) and forms the soluble oxyanion complex, UO4 . As a result, U can be quite mobile (De Moura, 2009; Fiedor et al., 1998).

4+ 6+ 2+ (iii) The easy oxidation of U to U , with the formation of uranyl ion UO2 generate complexes

2− with carbonate CO3 and leads to precipitate various uranyl mineral species at low temperature as a results of decomposition and re-precipitation (Frondel and Fleischer, 1955; De Moura,

2009).

Chapter 2: Literature Review

Table 2. 3. Half-lives and abundances of uranium and thorium isotopes

Isotope Half-life (years) % in nature Reference 238U 4.468×109 99.28 Ovaskainen, 1999 235U 7.13×108 0.72 234U 2.48×105 0.0054 232Th 14.05×109 -

Table 2.4, De Moura (2009) summarises the main categories of deposits and chemical elements accompanying U and Th commonly in the form of oxides. These oxides are responsible for the different percentages of U in different ores shown in Table 2.5. Referring to Table 1.1, the U content in the concentrate at SMB is approximately 1700 g/t. Therefore, the tantalite deposit at SMB can be classified as an ore in between high- and low-grade range with respect to uranium.

The tetravalent uranium as oxides are mostly associated with rare earth elements and the tin, niobium, tantalum, titanium family and, often, all three groups together. Due to their complex nature with extensive crystal lattice substitution, they are highly refractory minerals. The increase of tantalum and niobium contents increases the refractoriness. In contrast, the refractoriness decreases with the decrease of titanium and iron contents (Yan and Connely, 2008;

Lunt et al., 2007). Referring to Table 2.2, the ionic radii of U and Th are not particularly large, but the combination of somewhat large radius and high-charge is not readily accommodated in crystal lattices of most common rock-forming minerals; so both U and Th are highly incompatible elements (Sidorenko and Doynikova, 2010). Consequently, the content of U and

Th in various minerals differ from each other. For instance, U is more concentrated than Th in zircon (Zr,U)SiO4, whereas Th is more abundant than U in monazite (Ce,La,Nd,Th,U)PO4

(Venter and Boylett, 2009).

Chapter 2: Literature Review

Table 2. 4. Elements associated with rocks hosting uranium and thorium

Charac- Type of deposit teristic Granitic Carbo- Elevated Interme- Low Clastic Urani- Urani- element pegmatite natite temp. diate temp. rocks ferrous ferrous temp. Veins phosopho- black rites shales U X X X X X X X X Th X X X Ce X X Y X m Zr X X Ta-Nb X X Ti X X X X X Fe2+ X X X X m X m X Mg X X X m Ca X X X X X X X m Ba X X m Sr X Na X X m K X X m X Al X X X m m Si X m X X X m C X X CO3 X m X m X X m P X m X m S X X X X X m X F X m X m X m Cl X Cu m X m X Pb m m m Zn m m Cd m Mo m m m m Au X Ag X m V X Bi X Co X Ni X Cr m As m Mn m Se m m

(De Moura, 2009), X: always present; m: minor quantity

Chapter 2: Literature Review

Uranium and thorium are also present in xenotime (Y,Th,Dy,Yb,Er,Gd,U)PO4, bastnasite

(La,Ce,Nd,Th)CO3F, among others in association with columbite-tantalite (Venter and Boylett,

2009). Referring to Table 1.1, U is more concentrated than Th in the Bisunzu deposit which reflect the high content of U compared to Th. According to International Atomic Energy Agency

(IAEA) (1993, 2003 & 2004), thorium is not widespread as uranium, it is distributed within a significant pressure and temperature, mainly in minerals including monazite, thorianite and thorite. Table 2.6 shows various associations of niobate - tantalate - titanate ores bearing uranium and thorium. The monazite mineral is often found in the same deposit as columbite-tantalite in a small amount, most often, in pegmatite ore (El-Husaini et al., 2001; IAEA, 1993).

Table 2. 5. Uranium concentrations in ore deposits, earth crust and sea water

Category Location Concentration Concentration References of U (wt %) of U (g/t) Very high-grade ore Canada 20 200,000 The National High-grade ore Mozambique 2 20,000 Academics Press, Low-grade ore Ethiopia 0.1 1,000 2012; Very low-grade ore Namibia 0.001 10 Ovaskainen, 1999 Granite - 4-5 Sedimentary rock - 2 Earth’s continental - 2.8 crust Seawater - 0.003

Chapter 2: Literature Review

Table 2. 6. Niobate, tantalate and titanate ores bearing uranium and thorium

Mineral name Chemical formula References Ampangabeite (Y,Er,U,Ca,Th)2(Nb,Ta,Fe,Ti)7O8 Černý and Ercit, 2005 Betafite (U,Ca)(Nb,Ta,Ti)2O7 McMaster et al., 2015 Brannerite A3B5O16 with A = U, Ca, Fe, Th, Y, Bi and Costine et al., 2013 B = Ti Calciosamarskite (Ca,X,U,Th)3(Nb,Ta,Fe,Ti,Sn)8O15 Frondel and Fleischer, with X = yttrium, other rare earths 1955 2+ Davidite AB3(O,OH)7 with A = Fe , rare earths, El-Hussaini and Mahdy, U6+, Ca, Zr, Th and B = Ti, Fe3+, V, Cr 2002 2+ Dolorenzite (Y,U,Fe )(Ti,Sn)3O3 Frondel and Fleischer, 1955; Weil, 2012 2+ Aeschynite (Ce,Ca,Fe ,Th)(Ti,Nb)2O6 Černý and Ercit, 2005 Euxenite (Y,Cu,Ce,U,Th)(Nb,Ta,Ti)2O6 Černý and Ercit, 2005 Fergusonite (Y,U,Th,Ca)(Ta,Nb,Ti)O4 Černý and Ercit, 2005 Irinite (Na,Ce,Th)(Ti,Nb)[O3-x(OH)]X Frondel and Fleischer, 1955 Ishikawaite (U,Fe,X)(Nb,Ta)O4 (X = rare earths) Frondel and Fleischer, 1955 4+ Khlopinite (Y,U ,Th)3(Nb,Ta,Ti,Fe)7O20 Frondel and Fleischer, 1955 2+ 3+ Nohlite (Ca,Mg,Fe ,X,U)2(Nb,Zr,Fe )3O10 Frondel and Fleischer, with X = yttrium, other rare earths 1955 Polycrase (Y,Ca,Ce,U,Th)(Ti,Nb,Ta)2O6 Ditz et al., 1987 2+ Priorite (Y,Er,Ca,Fe ,Th)(Ti,Nb)2O6 Murata et al., 1957 Samarsakite (Y,Ce,U,Fe,Pb,Th)(Nb,Ta,Ti,Sn)2O6 Hanson et al., 1999

Commonly, the treatment of columbite-tantalite ores begins with the physical treatments followed by chemical treatment depending upon the market needs. The physical beneficiation often involves the pre-concentration, primary concentration and clean-up of the concentrate.

However, the presence of radioactive elements such as uranium and thorium in tantalite ore complicates the beneficiation process. The removal of the radioactive species at an early stage of beneficiation stage will lower the cost and increase the confidence for transportation and ensure the safety of processes of further treatement. The chemical treatment of tantalite ores bearing uranium and thorium generally involves the direct leaching or leaching after roasting or

Chapter 2: Literature Review

acid baking. These methods will be discussed in detail in the next few sections. Further treatments on the tantalite ores may take place depending on the market demands. However, the mineralogical identification of the concentrate is of prime importance which will lead to an appropriate method to be employed for further treatement.

2.3. Mineralogy of tantalum-niobium concentrates from Africa

The first step in the mineral processing investigation is the characterisation of the ores and concentrates which will lead to the effective environmentally friendly and low cost recovery of the target metals. The characterisation of tantalum and niobium ores has been performed on samples from various locations including Egypt, Nigeria, Ethiopia and Mozambique to name few as presented in Table 2.7. The mineral and chemical composition characterisation of Ta-Nb concentrates is complicated due to the fact that these two metals are closely associated in nature as the two most characteristic rare-metal accessories in granite pegmatites, forming a large exotic number of minerals (Knorring, 1958).

Chapter 2: Literature Review

Table 2. 7. Mineral identification of tantalum-niobium concentrates from different locations

Site Mineral name Formula

a Kenticha, Ethiopia Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Ferrocolumbite (Fe,Mn)(Nb,Ta)2O6 Ilmenite FeTiO3 Hematite Fe2O3 b Kenticha, Ethiopia Tantalite (associated minerals were not reported) (Mn,Fe)(Ta,Nb)2O6 c Naquissupa, Mozambique Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Ferrocolumbite (Fe,Mn)(Nb,Ta)2O6 Euxenite (Y,Er,La,Ce,U,Th)(Nb,Ta,Ti)2O6 Microlite (Ca,Na)2(Ta,Nb)2O6(OH,F) Quartz SiO2 Mica and hornblende (amphiboles) NaCa2(Mg,Fe,Al)5(Al,Si)8O22(OH)2 d Eastern-Desert, Egypt Euxenite (Y,Er,La,Ce,U,Th)(Nb,Ta,Ti)2O6 Betafite (U,Ca)2(Nb,Ta,Ti)2O7 Thorite ThSiO4 Zircon (Zr,U,Th,Hf)SiO4 e Kenticha, Ethiopia Manganocolumbite and ferrocolumbite as major (Mn,Fe)(Ta,Nb)2O6 (spodumene ore) phases (associated phases were not reported) (Fe,Mn)(Nb,Ta)2O6 Northern to central Eastern Zircon (Zr,U,Th,Hf)SiO4 f desert, Egypt Euxenite (Y,Er,La,Ce,U,Th)(Nb,Ta,Ti)2O6 Davidite (Fe,La,U,Ca)(Ti,Fe,V,Cr)(O,OH) Quartz SiO2 Muscovite KAl2(Si3Al)O10(OH,F)2 Phlogopite KMg3AlSi3O10(F,OH)2 Fluorite (Ce,Ca,Y)F2 Hematite Fe2O3 g Nigeria Tantalite (Mn,Fe)(Ta,Nb)2O6 Columbite (Fe,Mn)(Nb,Ta)2O6 Ilmenite FeTiO3 Hematite Fe2O3 Wolframite (Mn,Fe)WO4 Cassiterite SnO2 zircon (Zr,U,Th,Hf)SiO4 a and b. Cheru et al., 2018, c. Nete et al., 2014a, d. El-Husaini and EL-Hazek, 2004, e. Gebreyohannes et al., 2017, f. El-Hussaini and Mahdy, 2002, and g. Adetunji et al., 2005

Table 2.8 gives an overview of chemical composition of tantalum and niobium concentrates from various locations described in Table 2.7 indicating substantial differences in the type of minerals that contain Ta and Nb and their associated major and minor impurities.

Except the tantalite of low Ta2O5 content from Egypt, the chemical composition from other locations closely resembles the chemical composition of concentrates from SMB deposit.

Despite high content of SiO2 in SMB deposit compared to other deposits which may render the beneficiation difficult, the presence of ZrO2 and SnO2 may bring an additional income to the

Chapter 2: Literature Review

company. The characterisation results obtained from various deposits in the present study will be compared to mineralogical findings, percentages of Ta2O5 and Nb2O5 as well as associated elements including U and Th. Such a comparison would be useful in planning the beneficiation processes.

As shown in Tables 2. 6 and 2.7, majority of valuables minerals do exist in the form of oxide and silicate minerals. The columbite-tantalite ores belong to oxide group. Their separation/concentration is achieved either by conventional separation and/or enrichment techniques followed by chemical treatment.

Table 2. 8. Constituents and grades of tantalite ores from different locations

Oxides Chemical composition analysis of elements in each concentrate (%) a b c d e f g h* Ta2O5 46.2-60.1 46.2 - 60.1 27 2.11 26.3 – 69.7 16.2 8-60 33.0 Nb2O5 16 11.5 - 20.0 27 1.76 10.6 - 31.0 12.08 20-38 5.40 TiO2 0.88 0.43 - 0.73 2.8 NR 0.31 - 5.17 4.48 Up to 33.38 0.96 FeO NR NR NR NR NR 23.40 NR - REE NR NR NR NR NR 0.09 NR - ThO2 0.65 0.07 - 0.60 0.5 2.02 0.04 0.04 0.01-0.34 0.02 U3O8 1.23 0.99 - 1.32 3 0.093 0.26 - 1.19 0.23 0.17-1.57 0.17 ZrO2 NR NR NR NR NR 0.19 NR 0.37 SnO2 0.10 NR NR NR 0.04 – 0.69 NR NR 2.80 MnO NR NR NR NR NR NR 0.74-10.1 0.23 MnO2 7.60 NR NR NR 0.015 – 0.200 NR NR - Fe2O3 17.95 NR 8 NR ND NR 2.85-10.69 3.60 Mn3O4 NR NR 8 NR NR NR NR - RE2O3 NR NR NR 3.7 NR NR NR - ZrO2 NR NR NR 43.5 NR NR NR - K2O NR NR NR NR 0.67 NR NR - SiO2 NR NR NR NR NR NR NR 25.3 LOI NR NR NR NR NR 0.31 NR -

References: Same as in Table 2.7 a and b. Cheru et al. 2018, c. Nete et al. 2014a, d. El-Husaini and EL-Hazek, 2004, e. Gebreyohannes et al. 2017, f. El-Hussaini and Mahdy, 2002, g. Adetunji et al. 2005, h*. SMB exported sample, NR: Not reported and ND: Not detected

Chapter 2: Literature Review

2.4. Beneficiation routes for upgrading tantalum-niobium concentrates

The mined ores contain approximately 0.1 wt % Ta2O5 (Adetunji et al., 2005), which would essentially require enrichment through pre-concentration, primary concentration and concentrate clean-up technique (Amuda et al., 2007). The choice of any or all of these processes would depend on the nature of the ore, particularly the content of Ta2O5 in the ore relative to its associated minerals and impurities (Adetunji et al., 2005). Table 2.9 presents different beneficiation routes based on the wt % Ta2O5 content. The concentration processes may be carried out by wet gravity, magnetic or electrostatic separation and classification or by flotation to produce concentrates containing approximately 70 wt % combined Ta2O5 and Nb2O5 (Linnen et al., 2014 and Adetunji et al., 2005). In fact, the gravity separation is the common beneficiation route employed to upgrade the tantalite ores (Wills, 1992).

The gravity separation is a technique for separating two or more minerals as a result of differences in specific gravity and their movement in response to the force of gravity. A concentrate suitable for further processing to recover Ta are generally required to exceed 25 wt

% Ta2O5, with 50 wt % combined Ta2O5 and Nb2O5 (Burt, 1984 & 1996). Primary ores with these Ta2O5 and Nb2O5 contents would only need clean-up operations before extraction of Ta2O5

(Adetunji et al., 2005). Thus, the Ta-Nb concentrate at SMB of 33 wt % Ta2O5 and 5 wt % Nb2O5 can be subjected to a clean-up beneficiation route.

Chapter 2: Literature Review

Table 2. 9. Choice of technology based on tantalum grade and raw material input

Ta2O5 (wt %) Raw material input Beneficiation routes References <10 Low- medium Pre-concentration (sizing, gravity Amuda et al., 2007 grade tin slags separation) > 15 High grade tin slags Primary concentration and classification Singh et al., 1997; treatment stages followed by flotation, Nete et al., 2012 magnetic separation or electro-static separation as secondary concentration <10 Natural ores Pre-concentration (sizing, gravity Adetunji et al., 2005 separation) 10 - 25 Natural ores Primary concentration (jigs and sluices) Amuda et al., 2007 Hydraulic classification, series of spirals and shaking tables) 20 – 40 Natural ores Concentrate clean-up (flotation, magnetic Adetunji et al., 2005 separation or electrostatic separation)

It is important to determine the suitability of a gravity concentration process before it is used for concentration of an ore based on the mineral particle size as shown in Table 2.10. Equation 1 deals with the concentration criterion of gravity methods, where CC is the concentration criterion and SG represents the specific gravity.

푆퐺 (ℎ푒푎푣푦 푚𝑖푛푒푟푎푙)−푆퐺 (푓푙푢𝑖푑) 퐶퐶 = (1) 푆퐺 (푙𝑖푔ℎ푡 푚푎푡푒푟𝑖푎푙)−푆퐺 (푓푙푢𝑖푑)

Table 2. 10. Suitability of concentration criterion for particles of different sizes

Concentration Criterion (CC) Particle Size (μ) Reference > 2.5 > 75 Balasubramaniam, 2017 1.75 < CC < 2.5 > 150 1.50 < CC < 1.75 > 1700 < 1.25 Suitable for any size

Chapter 2: Literature Review

According to Falconner (2003) and Balasubramaniam (2017), some advantages of using gravity separation are listed below:

 To reject the barren waste as an initial separation

 To recover malleable and/or friable coarse heavy minerals from grinding circuit

circulating loads

 To pre-concentrate heavy minerals to minimize downstream processing costs

 To concentrate heavy minerals

 To clean low weight yield bulk concentrates

 To scavenge plant tailings

 To generate a concentrate that can go direct to a refinery rather than a smelter.

After the gravity separation, the tantalite ores undergo the second stage of concentrate clean-up using magnetic separation to remove light materials with the aim to achieve at least a total of 70 wt % Ta2O5 and Nb2O5. The magnetic separation technique exploits the magnetic susceptibility of components which allows the separation of minerals. Table 2.11 lists the magnetic properties of materials associated with tantalite. In fact, the portion of minerals with elements of high spin properties of electrons is more attracted by the magnetic separator. This theory predicts magnetic properties for high spin d4, d5, d6 and d7 electron configurations and less or diamagnetic properties for low spin electron configurations. For instance, minerals containing Fe, Ti and Mn are more attracted to a magnetic field, due to the high oxidation state of these elements (Nete et al., 2014a). Based on the magnetic properties, all substances can be classified into one of the three groups as discussed below:

Chapter 2: Literature Review

 Ferromagnetic materials are strongly attracted to magnetic fields and retain their

magnetic properties even after the externally applied field has been removed.

 Paramagnetic materials refer to minerals that are weakly attracted to magnetic field but

possess the magnetic properties only in the presence of the applied magnetic field. The

paramagnetic properties in these materials arise from the presence of unpaired electrons

and the realignment of the electron paths in the direction of the externally applied field.

Most metals which include Mg, Mo, Li and Ta have the paramagnetic properties.

 Diamagnetic materials are nonmagnetic at 1.70 amps and would be rejected by

electromagnets, even with strong currents (Rosenblum and Brownfield, open-file report

99-529).

Table 2. 11. Classification of minerals based on their magnetic properties

Strongly Medium magnetic materials Weakly magnetic Non-magnetic Diamagnetic References magnetic materials materials materials materials

Titanium Chromite (FeCr2O4) Pyrochlore Fluorite (CaF2) Zircon Adetunji et (Ti) Columbite (Na,Ca)2Nb2O6(OH,F) Quartz (SiO2) (ZrSiO4) al., 2015; Magnetite (Fe,Mn)(Nb,Ta)2O6 Uranium (U) Feldspar Tin (Sn) Brownfield (Fe3O4) Wolframite (Fe,Mn)WO4 (Na,Ca)(Si,Al)4O8 open–File Manganese Tourmaline Rutile (TiO2) Report 99- (Mn) 3+ NaFe Al6(BO3)3SiO18(OH)4 Silicon (Si) 529

Ilmenite (FeTiO3) Aluminium (Al) Tantalite (Mn,Fe)(Ta,Nb)2O6 Niobium (Nb)

The Magnetic Susceptibility Balance (MSB) displays results for the sample, R and the blank tube, Ro, which are then used to calculate the mass susceptibility using formula 2 (Cano et al., 2008; Bock, 1979b). Some advantages of magnetic separation are high efficiency of purification at low cost without using chemicals. If magnetic force is stronger than the friction forces, weak particles can also be attracted. Common examples of separation by physical

Chapter 2: Literature Review

beneficiation and concentration techniques of tantalum and niobium minerals from Niobec in

Canada and Greenbushes in Western Australia are discussed in the next section.

C * I * (R  R )   Bal o (2) g (109 *m)

where  g  mass magnetic susceptibility

I  sample length (cm) m  sample mass (g)

C Bal  balance calibration constant

2.5 Physical separation and concentration at Niobec and Greenbushes

At the Niobec plant in Canada, niobium is separated and concentrated from pyrochlore ores. Pyrochlore is associated with dolomite minerals in the carbonatite rock (BGS, 2011). The dolomite subordinate with calcite (CaCO3), ankerite Ca(Fe,Mg,Mn)(CO3)2, barite BaSO4, apatite Ca5F(PO4)3, magnetite Fe3O4, perovskite (CaTiO3), quartz (SiO2), pyrite (FeS2), phlogopite (KMg3AlSi3O10(F,OH)2) and sodic amphibole (Mg7Si8O22(OH)2) (Linnen et al.,

2014). As shown in Figure 2.1, the pyrochlore ore is first crushed in jaw, cone or impact crushers and milled in rod or ball mill operating in closed circuits with vibrating and screw classifiers to liberate niobium particles. After the ore is screened and classified, the resultant slurry is sent for desliming. Carbonate material is removed by two stages of froth flotation, followed by an additional desliming stage. Magnetite is removed from the slurry by low froth flotation and sent to waste. The sought–after pyrochlore is collected from the slurry by froth flotation using

Chapter 2: Literature Review

diamine collectors. A final stage of froth flotation is used to remove sulfides, such as pyrite. The leaching process concentrate the Nb2O5 up to 54 wt %.

Figure 2. 1. Generalised physical beneficiation flow diagram for Ta-Nb minerals, based upon the Niobec and Greenbushes operations (BGS, 2011)

Chapter 2: Literature Review

The studies of the pegmatite archean deposits at the Greenbushes indicated that the ore hosts about ten various significant Ta-bearing minerals (BGS, 2011; Sweetapple et al., 2017).

The main associated minerals are cassiterite (Sn) and spodumene (Li) where the latter is the main lithium bearing phase (Partington et al., 1995). The Wodgina pegmatite mine located approximately 100 km southeast of Port Hedland in the northwestern part of Western Australia is an important tantalite ore. The Wodgina processing plant produces primary tantalum concentrate of grade between 8-19 wt % Ta2O5, which is transported to Greenbushes for further processing (BGS, 2011; Pilbara Minerals Ltd open report, 2013). The Galaxy Resources Ltd. of

Australia owns the Mt. Cattlin Mine, which is located about 2 km north of Ravensthorpe in

Western Australia. It is an important deposit of tantalum ore and has annual production capacity of about 25 t of Ta2O5.

The mined tantalite ore from the three locations are processed at the Greenbushes primary and secondary physical separation facilities. Tin-tantalum ores are processed using a circuit of shaking tables, spirals and jigs. The rougher concentrate is de-watered and dried to produce a tantalum concentre containing about 4-6 wt % Ta2O5 as a primary treatment technique.

The secondary physical processing plant facilities at Greenbushes (Figure 2.1) can be described as follows: (i) high intensity magnetic separation used to separate the paramagnetic tantalum grains from the non-magnetic tin-tantalum grains, (ii) further concentration of the paramagnetic fraction to a high grade of about 30 wt % Ta2O5. Figure 2.2 shows a gravity falcon separator used for a higher recovery of Ta-fines. Natural Ta-Nb bearing ores possess specific properties that promote the formation of fines and ultrafines during processing which is a source of metal loss. Such properties are brittleness, the tendency to produce fines under mechanical stress, particularly during the comminution of the ore, high specific gravity and high intergrowth with sulfides and other oxide minerals (Deveau and Young, 2004). Thus, majority of fines are

Chapter 2: Literature Review

eventually lost during the concentration. To tackle this issue, a gravity falcon separator has been proposed to recover the Ta-fines resulting a high recovery of about 73% (Deveau and Young,

2004).

Figure 2. 2. Pilot test flow diagram of Ta-Nb concentration with falcon concentrator (Deveau and Young, 2004; Deveau, 2006) Abbreviations. Rgh: Rougher, Scav: Scavenger, Clnr: Cleaner, Rec: Recovery

Chapter 2: Literature Review

2.6. Chemical decomposition of tantalum and niobium concentrate

The alkali fusion was the first method at industrial level to be adopted for a complete break-down of the low-grade columbite-tantalite raw material as described in Figure 2.3. A large number of alkaline fluxes including caustic soda, soda ash, caustic potash, potassium carbonate, and a combination of these, with or without the addition of oxidizing agents such as sodium nitrate and sodium peroxide were tested for this purpose (Ayanda and Adekola, 2011). However, this treatment method was challenged due to the use of high temperature, typically 800 oC (Nete et al., 2014b; Bose and Gupta, 2002). To tackle the implication, a hydrothermal method has been developed using high pressure (Cardon, 1962), although the autoclave equipment was significantly expensive (Zelikman and Orekhov, 1965; Gaballah et al., 1997).

Figure 2. 3. Flowsheet for recovery of tantalum and niobium oxides using KOH media (Wang et al., 2009)

Chapter 2: Literature Review

The alternative method of Ta-Nb decomposition was by chlorination route (Othmer,

1997), followed by acid leaching (Gupta and Suri, 1994; Gupta and Yan, 2016). The main chemical equations on decomposition and dissolution of columbite-tantalite concentrate in the presence of a source of fluoride, chloride or alkali are described in equations (1-11), Table 2.12.

In chlorination breakdown of the Ta-Nb ore, the constituents of the concentrates are released due to high volatility, and high water solubility of most chlorides (Damodaran et al., 1969). The gaseous chlorides generated are readily separated due to differences in their pressure, differences in reactivity with oxygen and water vapour (Ayanda and Adekola, 2011). However, this digestion method is only appropriate to high–grade tantalum-niobium ores (Miller, 1995). Thus,

Wang et al. (2009) proposed a method to decompose the low grade Ta-Nb ores by molten potassium hydroxide to produce K3(Ta,Nb)O4, followed by hydrolysis to K8(Ta,Nb)6O19.nH2O and then to be leached with water. The process is improved by adding a certain amount of Nb2O5 to the ore to be decomposed, then decomposing the enriched ore with solid KOH of mass ratio

2:1 at about 400 °C and finally leaching the calcined product with water. No Gibbs free energy

(ΔGo) data were provided for the chemical equations in Table 2.12 from the literature.

Currently, most of tantalum-niobium ores are decomposed by concentrated hydrofluoric acid (El-Hussaini, 2001; Gupta and Suri, 1994). Two possible complex fluoride species are

2- - 2- - formed, TaF7 / TaF6 and NbOF5 / NbF6 as shown in Table 2.12. According to Zhu and Cheng

(2011) as well as Damodaran et al. (1969), the generation of these two species are determined by the concentration ratio of HF/Nb as well as HF/Ta. High concentration of HF will shift the

- - equilibrium from left to right resulting in the formation of TaF6 and NbF6 according to chemical equations 4-5.

Chapter 2: Literature Review

Table 2. 12. Chemical reactions for decomposition of tantalum-niobium ores

Equations No Decomposition of Ta-Nb ore in a source of fluoride or chloride Reference 1* (M) (Ta, Nb)2O6 + 12NH4HF2 a →NH4MF3 + (NH4)2TaF7 + (NH4)3NbOF6 + 5H2O + 6NH3 + 8HF 2 (Mn0.46 Fe0.54) (Nb0.65 Ta0.35)2O6(s) + 16HF(aq) b →1.3H2NbF7(aq)+0.7H2TaF7(aq) + 0.46MnF2(aq) + 0.54FeF2(aq) + 6H2O 2- 2- 3 (Ta/Nb)2O5 + (HF/H2SO4) → TaF7 / NbOF5 b 2- - - 4 TaF7 + HF → TaF6 + HF2 a 2- - - 5 NbOF5 + 3HF → NbF6 + HF2 + H2O a 6 (Ta/Nb)2O5 + chloride (Cl2, NaCl and FeCl3) → (Ta/Nb) Cl5 b Decomposition of Ta-Nb ore in molten KOH 7 3(Fe,Mn)O.(Ta,Nb)2O5 + 8KOH + (n – 4) → K8[(Ta,Nb)6O19.nH2O] + 3(Fe,Mn)O c 8 K8[(Ta,Nb)6O19.nH2O] → 6K(Ta,Nb)O3 + 2KOH + (n + 1)H2O c 9 Fe2O3 + 2KOH → 2KFeO2 + H2O c 10 MnO2 + 2KOH → K2MnO3 + H2O c 11 SiO2 + 2KOH → K2SiO3 + H2O c a. Kabangu and Crouse, 2011; Rodriguez et al. 2015; Olushola and Ayanda, 2011; Nete et al. 2014b; b. Agulyansky et al. 2004; Gupta and Suri, 1993; Zhu and Cheng, 2011; c. Wang et al. 2009; El-Hussaini and Rice, 2004; *. M = Mn or Fe

These fluoride complex species are of lower charges and are most likely to be extracted into the organic phase during solvent extraction (Agulyansky et al., 2004). Generalised flow chart and previous work on recovery of Ta and Nb metals/oxides are presented in Figure A 3 and

Table A 3, respectively in Appendix A 3.

2.7. Dissolution of metals from columbite-tantalite

Uranium typically occurs as tetravalent (U4+) species in tantalum ores with extensive crystal lattice substitutions (Venter and Boylett, 2009; Lunt et al., 2007). This makes the treatement of these oxide materials more complicated. Thus, the uranium leaching technique from oxides requires aggressive chemicals, typically digestion in hot hydrofluoric or sulfuric acids or a combination of both, followed by safe removal of U and Th (Nete et al., 2014b;

McMaster et al., 2015). Fine grinding and prolonged leaching time at elevated temperatures are

Chapter 2: Literature Review

often taken into consideration to treat tantalite ores bearing uranium. The acid baking is a technique in which the material is mixed with an acid and roasted to obtain a calcine which is partly soluble in dilute acid or water. This method has the advantage of using sulfuric acid and avoiding the use of hydrofluoric acid. The process is suitable for treating ores which cannot be beneficiated to a certain extent, using a physical treatment method (for example fine and/or weathered pyrochlore ores). The acid baking process is generally carried out at temperatures between 150 oC and 400 oC in the presence of concentrated sulfuric acid. Table 2.13 gives a summary of U, Ta, Nb, Th and REE leaching efficiencies under different conditions.

Table 2. 13. Leaching efficiencies of metals from coltan in different media

Reference Cheru et Cheru et al. Nete et al. El-Husaini and EL-Hazek, Gebreyoha El-Hussaini and al. 2018 2018 2014a 2004 -nnes et al. Mahdy, 2002 2017 Ore Columbite Columbite Columbite Columbite tantalite Tantalite Columbite Tantalite tantalite tantalite tantalite Ore grade High High High Low High Low

st nd Lixiviant 97 wt % HF:H2SO4 97 wt % 1 stage: 2 stage: 1 vol% 97 wt % H2SO4 3:1 H2SO4 97 wt % Concentrated KOH H2SO4/68 wt % H2SO4 HF HNO3 S/L (g/L) 1:6 1:25 NR 1:3 1:3 1:50 1:3 T oC 100 150 50 100 80 20 200 Time (h) 3 3 3 4 3 1 min 3 Digestion Autoclave Autoclave Direct Direct leaching Washing Direct leaching leaching Particle size < 100 -100+75 +9.1–11 < 50 < 74 < 74 (μm) Stirring rate 250 250 NR NR NR NR (rpm) LE U 92 96 64 91 U and Th 60 (%) were Th 70.8 85 56 81 completely 76 Ta 4.4 - 0.8 32 removed 94 (no values Nb NR - 3.0 47 reported) 99.5 Ti NR - - 17 -

RE2O3 NR - - 76 -

NR: Not reported; LE. Leaching efficiency

Chapter 2: Literature Review

The boiling point of sulfuric acid is in between 108 oC and 340 oC. Thus, it is preferable to bake the material using temperature varying in the range 250 - 300 oC. A ratio of liquid to ore of between 500 kg/t and 1500 kg/t is also favorable (Jerome et al., 2016). The niobium, tantalum, uranium, thorium and REE quantitavely react with H2SO4 to give soluble sulfates (Zhang et al.,

2016 and Jerome et al., 2006). For instance, the tantalum and niobium reacts with H2SO4 to give tantalum sulfate Ta2O3(SO4)2 and niobium sulfate Nb2O3(SO4)2 (Jerome et al., 2006). It is estimated that 90% of a pyrochlore is attacked during the acid baking. The iron minerals in the ore form iron sulfate Fe2(SO4)3 which enhance the break-up of mineral phases carrying Nb, Ta,

U and REE elements associated with the pyrochlore mineral group. This improved digestion is significant for ores with high levels of iron with Fe/(Ta + Nb) molar ratio greater than 2. The ferrous or ferric iron in the form of oxide or sulfide can be added to the system on demand to improve the baking performance. The baking of ores with a PO4/(Ta + Nb) molar ratio between

2-6 is highly advantageous (Jerome et al., 2006).

The decomposition of the columbite-tantalite concentrate in sulfuric acid at elevated temperature was investigated and reported by Bose and Gupta (2008). The reaction process comprises of two steps. The first step consists of formation of iron, manganese, tantalum and niobium hydroxides. The second step consists of conversion of tantalum and niobium hydroxides into oxysulfate compounds. However, the generation of the hydroxides has resulted in the formation of a shell around the columbite-tantalite retarding further reactions between H2SO4 with metals and/or oxides in the minerals. This causes a decrease in dissolution efficiency of elements (Agulyansky, 2004). A possible solution to this problem is the reduction of the particle size as proposed by Welham (2001) who found that an increase of powder milled time from 2 to

50 hours increases the dissolution rate by 4500 times. This showed that the dissolution rate of elements strongly depends on the mineral surface area and crystalline size. Welham (2001)

Chapter 2: Literature Review

recommended also the use of a strong sulfuric or hydrofluoric acids to dissolve columbite- tantalite at a higher rate. The sulfation consists of the reaction of a gas and the material to be roasted. The gas can be SO3 formed either by oxidation of SO2 or by decomposition of H2SO4.

The SO2 or H2SO4 or at least one of the alkali metal pyrosulphate (M2S2O7), where M = Na, K, and the like are directly used to enhance the decomposition of the mineral species. Generally, the sulfation process is carried out at temperatures between 450 oC and 550 oC, preferably,

(Jerome et al., 2006) to form sulfates as indicated in chemical reactions in Table 2.14.

Table 2. 14. Chemical reactions for acid baking of tantalum and niobium ores

Equations No Acid baking of Ta and Nb ore in sulfuric acid References 12 Nb2O5 + 2H2SO4 = Nb2O3(SO4)2 + 2H2O Jerome et al. (2006) 13 Ta2O5 + 2H2SO4 = Ta2O3(SO4)2 + 2H2O 14 ThO2 + 2H2SO4 = Th(SO4)2 + 2H2O Zhang et al. (2016) 15 UO2 + 2H2SO4 = U(SO4)2 + 2H2O

Figure 2.4 shows a flowchart of the dissolution and recovery of at least Ta, Nb, U and

REE after sulfuric acid baking or sulfation followed by leaching. Eventually, the Ta, Nb, U, Th and REE present in the feed material are converted to salts which are soluble in the concentrated ferric sulfate medium. The presence of iron in the ore is advantageous during the baking.

Chapter 2: Literature Review

Figure 2. 4. Dissolution and recovery of Ta, Nb, U and RE after sulfuric acid baking/sulfation and leaching (Jerome et al., 2006)

2.8. Potential-pH and solubility diagrams

The factors influencing the solubility of uranium in aqueous solution includes the solution Eh and pH, oxidation state of U, crystallinity of the mineral and the levels of associated impurities (Bhargava et al., 2015; Lottering et al., 2008). Figure 2.5 shows the effect of pH on U and Th solubility in 0.5 M NaCl at 25 oC (Casas et al., 1998). According to Sole et al. (2011) the

Chapter 2: Literature Review

solubility of UO2 in acidic solution is higher than that of ThO2 in the pH range 0-2.5. In the pH range above 2.5, however, the solubility of Th4+ is higher than that for U4+. Consequently, more

Th4+ would dissolve in the high pH environment in comparison to U4+.

1E+00

1E-01 UO2 ThO2 1E-02

1E-03 ] (mol/L) ] 2 1E-04

1E-05

] or ][ThO 1E-06 2

1E-07 [UO 1E-08 0 1 2 3 4 5 6 7 8 9 10 pH

o -3 -3 Figure 2. 5. Solubility data at 25 C for UO2 in 8.10 mol dm sodium perchlorate (Casas et al., 1998); ThO2 at I = 0.1 M and 0.5 M (HCl/NaCl) (Neck et al., 2003)

The solubility of uranium is also significantly enhanced by the presence of anions such

- 2− - - as OH , SO4 , Cl and F under oxidising conditions (Guillamont et al., 2003; Deblonde et al.,

2019). The stability of uranium and thorium sulfate complexes as a function of potential (Eh) and pH are shown in Figures 2.6 and 2.7, respectively. In the presence of excess sulfate ions, uranium in the higher oxidation state of U6+ preferentially complexes with the sulfate ions

2- 4- resulting in the formation of UO2(SO4)(aq), [UO2(SO4)2] and [UO2(SO4)3] species. These sulfate complexes are very stable in solutions of pH < 4. At higher pH, however, the formation of uranyl hydroxide species dominates (Bhargava et al., 2015; Sole et al., 2011). Similar to uranium, thorium also forms a complex species Th(SO4)2(aq) in excess sulfate ions that is stable under very acidic environment (~ pH 0) as shown in Figure 2.7 (Kim and Osseo-Asare, 2012).

Chapter 2: Literature Review

Figure 2. 6. Eh–pH diagram for U–S–H2O system at 25 °C (Lottering et al., 2008)

−3 Figure 2. 7. Eh–pH diagrams for Th–SO4–H2O system at 25 °C, {Th}=10 M, {SO4} = 1.0 M (Kim and Osseso-Assare, 2012)

Generally, pentavalent Ta(V) and Nb(V) are stable compounds in acidic solutions.

However, Nb(V) is relatively more reactive and can be reduced to its lower oxidation state of

Nb(III) (Zhu and Cheng, 2011). The Nb(III) ion can exist in concentrated solutions of

3- 3- hydrochloric acid or sulfuric acid, where complex anions (NbCl6) or [Nb(SO4)3] are formed,

Chapter 2: Literature Review

respectively (Rodriquez et al., 2015; Toure et al., 2018). Reduced Nb is unstable and is readily oxidized to Nb(V) by atmospheric oxygen. Figure 2.8 summarises the effect of pH on solubility of Ta and Nb collected from the literature which may be compared with the measured concentration of Ta and Nb in solution in the present study.

1E+00

Ta Deblonde et al. 2015 1E-01 Nb Deblonde et al. 2015 Nb Babko et al. 1963 1E-02 Ta Babko et al. 1963

1E-03

[Nb][Ta] or (mol/L) 1E-04

1E-05

1E-06 1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH o - Figure 2. 8. Solubility data for Nb2O5.nH2O(s) and Ta2O5.nH2O(s), temp = 25 C, I = 0.12 mol L 1 (NaCl/NaCO3/NaOH) (Deblonde et al., 2015); Solubility data of Nb2O5.nH2O(s) and o -1 Ta2O5.nH2O(s), temp = 19 C, I = 1 mol L (KNO3) (Bakbo et al., 1963).

The distribution of soluble chemical species of Nb(V) and Ta(V) in acidic solutions is shown in Figure 2.9. Lines 1 and 3 present the hydrolysis of Nb ions. When the pH is low (< -

1), the cationic form Nb(OH)4+ has a high relative content (> 80%). As the pH increases, the amount in the cationic form decreases, resulting in an increase in neutral species Nb(OH)5. At pH -0.6, the cationic and neutral Nb(V) complexes are at 50% each. In solution of pH above 1, only the neutral Nb(V) species is present. Niobium can be leached in either strongly acidic

+ solution as Nb(OH)4 or in strongly alkaline solutions as Nb(OH)5. Similar data have been obtained for Ta(V), but at higher pH levels. Lines 2 and 4 represent the neutral, Ta(OH)5 and

Chapter 2: Literature Review

+ cationic form, Ta(OH)4 , respectively. When the pH level is below 1, the relative content of the cationic Ta(V) is high while when the pH is above 1, the neutral Ta(V) compound is predominant.

Figure 2. 9. Distribution of hydrolysed Nb(V) and Ta(V) species in acidic solution (Zhu et al., 2011)

Although, tantalum and niobium are so similar in their chemistry, it has been found that there are some differences in the reactions of their oxides in the +5 oxidation state (Theron et al.,

2011). A recent study was carried out to investigate the differences in the oxides of tantalum and niobium in terms of their behaviour in acidic and basic solutions. The results indicated that Ta2O5 is more soluble and stable in basic solutions and fluxes than in acidic media which points to its acidic property (Theron et al., 2011). On the contrary, Nb2O5 is relatively more soluble in acids and acidic fluxes suggesting a more basic character for the Nb2O5 (Zhu et al., 2011; Nete et al.,

2010). In general, this difference can be explained in terms of the acid/base properties of the

Chapter 2: Literature Review

metal oxygen bond character of the oxides, with acidic oxides having relatively stronger metal oxygen bonding. The tantalum-oxygen bond in Ta2O5 is more covalent compared to the more ionic nature of the niobium counterpart. This increased covalent character of metal oxygen bond resulting in low oxygen electron density available for hydrogen bonding and therefore decreases the possible interaction with acids (Theron et al., 2011; Bose and Gupta, 2008).

A Pourbaix diagram has been constructed for the corrosion of Nb in concentrated H2SO4

o + and HCl acid solutions at 25, 75 and 95 C. The species niobium tetrahydroxide Nb(OH)4 and

- metaniobate NbO3 were studied. Figure 2.10a presents all three Pourbaix diagrams superimposed for the Nb-H2O system under the three temperatures. It can be seen from the

- Pourbaix diagrams that as the temperature increases the stability of NbO3 increases and the

- chemical equilibrium between Nb2O5 and NbO3 shifts to the more acidic values. The Pourbaix

+ diagram also revealed that the formation of Nb(OH)4 from Nb2O5 is more likely to occur in a strong acid solution. Robin and Rosa (1999), however, proposed that steady state corrosion, which in the case of Nb in concentrated acid solution is observed after roughly 200 h, due to the formation of film on mineral surface, which can explain the low or partial extraction of Ta and

Nb during the leaching process even in extreme acidic media. Figure 2.10b presents the Pourbaix

o + diagrams for the Ta-H2O system at 25 C. In the Ta-O-H system a moderate field of TaO2 occurs at pH (0 – 5) (Figure 2.10b). It is apparent that Ta2O5 is stable over all Eh at higher pH. In fact,

Ta is contained in tantalate-niobite salts; weathering under mildly to more pronounce acidic

+ conditions may release Ta(V) as TaO2 (Pourbaix, 1966).

Chapter 2: Literature Review

Figure 2. 10. Pourbaix diagram for (a) Nb-O-H system at 25, 75 and 95 oC, (b) Eh-pH diagram for Ta-O-H system (●) represents the corrosion potential for the 75 oC polarization tests and (◌) represents the 95 oC. Lines a and b represents hydrogen evolution and oxygen reduction, respectively (Asselin et al., 2007). Assumed activity of dissolved Ta = 10-8 at 25 oC (Pourbaix, 1966).

Chapter 3: Materials and Methods

CHAPTER 3 MATERIALS AND METHODS

3.1. Materials

3.1.1. Concentrates

All samples were mined and supplied by Societé Minière de Bisunzu (SMB) from

Democratic Republic of Congo (DRC) in the form of concentrates. Eight samples of columbite- tantalite concentrates have been collected from eight different mining zones and labelled as indicated in Table 3.1. In addition, three samples of F labelled as heavy (FH), middling (FM) and light (FL) portions were collected after a pegmatite ore (Nyange) was treated by gravity separation method in water at the SMB mining site D2 Bibatama.

Table 3. 1. Columbite tantalite concentrates collected at different SMB mining sites

Samples SMB mining sites A Budjali B D4 Gakombe C Gasasa D Luwowo E Mambuya F Nyange G D2 Bibatama H D3 Bibatama FH Nyange FM FL

Chapter 3: Materials and Methods

Each concentrate sample was homogenised and splitted to obtain a representative sample prior to various characterisation analyses. As the concentrates contained traces of radioactive materials, the samples were stored at room temperature in closed containers and the Geiger

Müller meter was used to measure the ionisation radiation levels. The samples were kept in the room and cabinets at about 5 m from the working place and were locked to prevent any unauthorised access. Protective equipment such as disposable gloves and respiratory dust masks were used during the handling of concentrates and chemicals to avoid any contamination through ingestion or inhalation. The labelled polyethylene plastic bags were used to keep all the minerals, while the solutions containing U and Th were kept in polyethylene sampling tubes.

3.1.2. Chemical reagents

All chemical reagents used in the present study were of analytical grade and no further purification was made on the chemicals. Table 3.2 gives a list of all the chemical reagents used in the experiments in the present project.

Table 3. 2. List of chemical reagents

Reagents Formula Purity (%) Grade Manufacturer Hydrochloric acid HCl 32 AR AJAX Fine Chem Pty Ltd. Sulfuric acid H2SO4 95 AR Ferrous sulfate FeSO4 - AR Sigma-Aldrich Ferric sulfate Fe2(SO4)3 - - Nitric acid HNO3 67-69 AR Hydrogen peroxide H2O2 32 - Manganese dioxide MnO2 - - Platinum wire Pt 99.9 AR Potassium persulfate K2S2O8 99 - Ammonium persulfate (NH4)2S2O8 98 - Methanol CH3OH - - Lithium tetraborate Li2B4O7 > 99 -

Chapter 3: Materials and Methods

3.2. Methods

3.2.1. Sample preparation

Preparation of the concentrate samples for characterisation and subsequent experimentation was achieved by grinding the received concentrate materials using the micronizing and the iron-ring mill. The crushed samples were then subjected to screening to obtain different size fractions –75 μm, –106+75 μm, –150+106 μm, and –250+150 μm. For selected samples labelled as (FH, FM and FL), portions of the concentrates were subjected to magnetic separation to investigate the possibility of upgrading Ta2O5 and Nb2O5 content of the material. The magnetic separation experiments were conducted under the conditions shown in

Table 3.3. The magnetic and non-magnetic particles were separated using the Frantz model magnetic separator at Murdoch University. Figure 3.1 gives the magnetic separation and concentration scheme.

Table 3. 3. Conditions for magnetic separation and concentration

Current (Amp) Voltage (kV) Roll speed (rpm) Particle size (μm) 3 10 150 < 150

Chapter 3: Materials and Methods

Figure 3. 1. Scheme for tantalite concentrate beneficiation using magnetic separation

Chapter 3: Materials and Methods

Portions of about 5.0 g were carefully weighed and spread steadily over a smooth glass surface. The collected magnetic and non-magnetic fractions from each concentrate were carefully weighed to determine the mass of each lot. The magnetic and non-magnetic portions were then quantitatively analysed by XRF. In addition, a mass of 5.0 g from samples labelled as

D and G were carefully weighed and splitted into two portions of (–1000 μm) considered as fines and (–2500+1000 μm) as coarses using a stack of sieves and analysed to investigate the possibilities of upgrading the Ta2O5 and Nb2O5 and removal of uranium and thorium.

The chemical analysis of feed concentrate samples and leach residue solution was achieved using the fusion dissolution technique. In this method, approximately 0.1 g of the solid sample was weighed accurately and placed in a platinum crucible. About 2.0 g of lithium tetraborate fluxing reagent was added to the crucible and thoroughly mixed with the sample before heating. The crucible was then placed in a high-temperature oven and heated to 1100 °C for one hour to produce a glassy melt. The melt was cracked by quenching the hot crucible in cold water bath to facilitate faster dissolution. A volume of 10 mL concentrated H2SO4 was pipetted into the cold melt, followed by the slow addition of approximately 30 mL methanol to prevent the formation of boric acid and to facilitate the volatization of boron as the trimethyl borate. The crucible was then heated to 40 °C for 45 minutes on a hot plate with constant stirring to remove both the boron ester and the excess methanol (Nete et al., 2012). The clear solution was quantitatively transferred to a 100 mL volumetric flask and diluted to the mark with distilled water. Quantitative analysis was performed with the ICP-MS. The whole process was carried out in triplicate to ensure the reproducibility of the results.

Chapter 3: Materials and Methods

3.2. Analytical techniques

3.2.1. X-ray Fluorescence

The X-ray Fluorescence Spectroscopy (XRF) measurements were performed on pressed powders using the ‘Major pellets Programme’ and the ‘Pro-trace Programme’, depending on the relative concentration of the elements in the mineral sample. For the ‘Major pellets Programme’, approximately 8.0 g of the sample was added to 3.0 g Hoechst wax (binder material) and mixed for 30 minutes to obtain a homogeneous sample (approximately 11.0 g). For the ‘Pro-trace

Programme’, approximately 0.08 g of the mineral sample was added to 7.92 g SiO2 and approximately 3.0 g Hoechst wax, and mixed thoroughly for 30 minutes by a flask shaker to ensure a homogeneous mixture. Pellets or disks of 30 mm diameter and 9 mm thickness were used for the analyses. The XRF analysis was performed at Nagrom Laboratory in Perth, Western

Australia.

3.2.2. Geiger-Muller Counter

A Geiger-Müller detector was used to measure the maximum radiation emitted by the materials in units of micro-Sievert per hour (μSv/h). Radioactivities in units of Becquerel per gram (Bq/g) of concentrates were calculated based on the assays and the conversion factors from oxides or elemental composition of uranium and thorium. The conversion factors used are described in a report published in 2016 on a guide to the transport of Niobium (Nb) and Tantalum

(Ta) raw materials that contain naturally occurring radioactive materials (NORM), by the

Tantalum-Niobium International Study Center as shown in Table 3.4. More on the radioactivity concentration levels calculations are discussed in section 4.2 of Chapter 4.

Chapter 3: Materials and Methods

Table 3. 4. Calculating of radioactivity of uranium and thorium from an assay

Assay Conversion factors (Bq/g) 1% U3O8 104 1% ThO2 35.6 1% U 40.6 1% Th 123

Example of radioactivity calculation: [(% U3O8 × 104) + (% ThO2 × 35. 6)] (Bq/g)

3.2.3. Scanning electron microscopy

Scanning electron microscope and energy dispersive spectroscopy (SEM-EDS) analyses were used to determine the micrograph and mineralogy of the solid samples. Roughly 0.1 g of each of the eight concentrates were mounted on the epoxy and embedded in the resin for about

24 h. The samples were recovered and polished using 3 and 1 μm sandpaper to remove any unwanted material from the epoxy surface. The bake and leach residues were mounted on the aluminium studs using the conductive carbon tape. The TESCAN VEGA3 equipped with an

Oxford Instruments X-Max 50 silicon drift EDS system with AZtec and INCA software, was used to analyse the raw feed, bake products, and leach residue samples. The instrument was operated with accelerating voltage of 20 kV and a current of 15 mA. These analyses were conducted at the Centre for Microscopy, Characterisation and Analysis (CMCA) at the

University of Western Australia (UWA).

Chapter 3: Materials and Methods

3.2.4. X-ray diffraction

The X-ray powder diffraction (XRD) analyses of the solid samples were conducted to identify the mineral phases of the raw feed materials, bake products and leach residues. All samples were prepared as described in Section 3.2.1. Each powdered sample was evenly placed onto a flat sample holder plate. The XRD patterns were obtained using an Empyrean high throughput X-ray diffractometer while the mineral identification was achieved using Highscore and match softwares. The instrument was operated at 40 kV and 40 mA and samples were scanned from 5 to 65o 2 θ (o) position with a 1o divergence slit at a scan of 0.02o per minute using graphite-monochromatised CuKα radiation. The analysis was done at the Centre of Macroscopy and Characterisation Analysis (CMCA) facility located at the University of Western Australia as well as at the Murdoch University.

3.2.5. Inductive coupled plasma mass spectrometry (ICP-MS)

Inductive Coupled Plasma Mass Spectrometry (ICP-MS) was used to determine the concentrations of the target elements (Ta, Nb, U and Th) and the minor elements (Li, Cs, Fe,

Mn, Si, Al) in all solution samples. The instrument used was the iCAP Q series manufactured by

Thermo Fisher Scientific.

Chapter 3: Materials and Methods

3.3. Batch baking and leaching experiments

The experiments described in this section are divided into two categories, namely direct sulfuric acid leaching and sulfuric acid baking followed by leaching. For the direct leaching tests, the samples of required solid to liquid ratio were carefully prepared by mixing the concentrate material, calcium fluoride, and sulfuric acid solution. The conditions used to perform the direct leaching experiments are summarized in Tables 3.5 and 3.6. Table 3.7 presents various conditions used to perform the baking and leaching experiments. Leaching tests after acid bake, on the other hand, were carried out at temperatures in the range 20 to 90 oC. Prior to leaching, the mixture of concentrate and acid was baked at temperatures in the range 100 to 400 oC in a muffle furnace for about 2 h. A sample of the concentrate (2.5 _ 7.5 g) was subjected to acid baking in the present study. The concentrated (95-98 wt %) sulfuric acid was used for the baking as a source of sulfate ions to react with cations in the sample. After baking, the product obtained was leached in about 200 mL of 4 M sulfuric acid. A glass stirrer (250 rpm agitation speed) was used to facilitate the dissolution of suspended particles. After all the leaching experiments, the leach liquor was separated from the residue by filtration. Aliquots of 1 mL of the leach liquor was pipetted from the reactor using syringe at various intervals of time, filtered through whatman cellulose nitrate membrane filters with a pore size of 0.22 μm to avoid the solid particles, and analysed for elemental composition. The leaching efficiency of elements, LE, was calculated using the following relationships:

푚 LE = 2 × 100% (1) 푚1

푚 +푚 LE = 2 3 × 100% (2) 푚1

Chapter 3: Materials and Methods

where m2 and m3 are the mass of element of interest in the leach and wash liquors, respectively,

whereas m1 is the mass of element in feed. The concentration (ppb) of U, Th, Ta and Nb in the bake product, leach residue, filtrate and wash solutions were analysed using the ICP-MS at the

Nagrom and TSW analytical commercial Labs as well as at Murdoch University to cross check the results. These three sets of results from Nagrom, TSW and Murdoch laboratories all located in Western Australia are denoted by XRF, ICP-MS 1 and ICP-MS 2, respectively. The leach residues were washed and dried at 50 oC for 24 h prior to XRD and SEM-EDS analyses. The dried leach residues were digested and assayed for elemental mass balance calculations.

Table 3. 5. Direct leaching of U, Th, Ta and Nb under various conditions

T.N. S. I. Acid Conc Reagent Mass Temp Time S/L PSD Agitation (M) (g) (oC) (h) (g/L) (μm) speed (rpm) T 1 C - - KOH 1 90 6 10 < 75 500

T 2 E, H, G & B H2SO4 10 FeSO4 2 90 6 10 < 75 500 MnO2 1 T 3 H, G & C H2SO4 18 - - 100 3 50 < 45 250 T 4 C H2SO4 18 - - 50 3 10 < 75 T 5 C, D, E & F H2SO4 10 Fe2(SO4)3 4 50 24 50 < 75 250 MnO2 1.5 C, D, E , G & F H2SO4 10 Fe2(SO4)3 4 100 24 50 < 75 250 MnO2 1.5 FeSO4 2 T 6 C H2SO4 10 Fe2(SO4)3 20 100 24 10 < 45 250

T 7 C H2SO4 2 Fe2(SO4)3 20 100 24 10 < 45 250

C H2SO4 10.8 - - 100 24 150 250 HNO3 5.30 C H2SO4 10.8 - - 25 24 150 250 HNO3 5.30 T 8 C H2SO4 18 9.7 M H2O2 - 100 12 10 < 75 250

T.N.: Test number, S.I.: Sample identification, PSD: Particle size distribution

Table 3. 6. Direct leaching of U, Ta, Nb and Th using CaF2 in H2SO4

T.N. S.I. Acid Conc Reagent Solid to CaF2 Time Temp S/L PSD Agitation speed (M) ratio (h) (oC) (μm) (rpm) T 24 C H2SO4 4 CaF2 1 : 2 4 90 37.5 < 75 250

T.N.: Test number and S.I.: Sample identification. PSD: Particle size distribution. Test 24 results are reported in Chapter 4, Section 4.3.

Chapter 3: Materials and Methods

Table 3. 7. Sulfuric acid baking and water or acid leaching conditions

T.N. Baking Leaching S/A or S/B PSD Time Temp R/L Agitation Conc S/L Temp (g/L) (μm) (h) (OC) (g/L) speed (rpm) (M) (g/L) (oC) T 9 1 : 5* < 75 2 400 27 250 Water 12.5 40 T 10 625 < 75 2 600 27 250 4 12.5 40 T 11 1250 < 75 2 400 27 250 Water 12.5 40 T 12 625 < 75 2 400 27 250 4 12.5 40 T 13 625 < 75 2 400 27 250 4 12.5 40 T 14 625 < 75 2 400 27 250 4 12.5 90 T 15 625 < 75 0.5 400 27 250 4 12.5 90 1.0 1.5 2.0 T 16 625 < 75 2 400 27 250 4 12.5 90 +75–106 -106+150 –150+250 T 17 2500 < 75 2 400 27 250 4 12.5 90 1250 833 625 T 18 625 < 75 2 400 27 250 4 12.5 20 60 80 90 T 19 625 < 75 2 400 27 250 1 12.5 90 2 3 4 T 20 625 < 75 2 400 27 250 4 37.5 90 25.0 12.5 T 21 625 < 75 2 400 40.5 250 4 12.5 90 27.0 13.5 T 22 625 < 75 2 400 27 0 4 12.5 90 250 T 23 625 < 75 2 100 27 250 4 12.5 90 200 300 400 T 25 625 < 75 2 300 27 250 4 12.5 90

T.N.: Test number; S/B*: (B: KOH), A = acid, S = solid, baking with 18 M H2SO4. Leaching with water or H2SO4, R = reagent (K2S2O8) and time: 3 h, C sample. (Note that Test 13 was conducted on F, H, D, G and B, while Test 14 on: A, B, C, D, E, F, G and H concentrates). PSD: Particle size distribution

Chapter 3: Materials and Methods

3.3.1. Eh - pH measurements

The solution or slurry potential (Eh) during the reaction was continuously monitored and recorded using a platinum electrode combined with a Ag/AgCl/(3.5 M KCl) reference electrode.

It should be noted that the Eh is generally a mixed potential because there are many reactions taking place at the electrode. A pH probe was used to measure the acidity of the aqueous solution.

3.3.2. Characterisation and elemental mass balance

The elemental mass balance has been performed to investigate the reliability of the methods used for the analysis of U, Th, Ta and Nb in solution and solids. Figure 3.2 shows a summary of assay methods used for the elemental mass balance. A mixture of 2.5 g of columbite-

o tantalite concentrate and 7.4 g of H2SO4 was placed in a furnace at 400 C for 2 h. Due to loss of

H2SO4 during the baking, a mass of 3.8 g as a final bake product was obtained. The bake product was transferred to the leaching process. Solid – liquid separation after leaching was by vacuum filtration. The leach residues were washed with deionized water and ethanol and dried in a furnace at 50 oC for 24 h. A sample of (0.1 g) of bake product and leach residue were digested and assayed for elemental mass balance using the method discussed in previous sections. Leach filtrates were analysed by ICP-MS, while the reaction products in bake and leach residues were characterised by a combination of XRF, XRD and SEM-EDS.

Chapter 3: Materials and Methods

Figure 3. 2. Elemental mass balance diagram Section A: Assays based on XRF of solids (FH, FM, FL) and characterisation based on XRF. Section B: Assays based on XRF of solids, ICP-MS for liquors and digested solids as well as characterisation based on XRD and SEM-EDS

Chapter 3: Materials and Methods

Pegmatite ore and concentrates were from the mining sites shown in Figure 1.5, Chapter

1. Various steps for gravity separation and concentration to obtain a concentrate for export are also detailed in Figure 1.7, Chapter 1. Equations 3 – 5 were used to perform the elemental mass balance as shown in Figure 3.2, where mf, mb and mr are the masses of feed, bake product and leach residue, af, ab, ar are elemental assays in feed, bake product and leach residue, respectively, al and awl are elemental assays in leach and wash liquors, respectively, and finally, vl and vwl are the volumes of leach and wash liquors, respectively. A complete elemental analysis was conducted with the solids and liquors described in Test 25 of Table 3.7.

(m ×a )−(푚 ×a ) LE = f f r r × 100% (3) (mf ×af)

(m ×a )−(m ×a ) LE = b b r r × 100% (4) (mb ×ab)

(v ×a ) + ( vw ×aw ) LE = l l l l × 100% (5) [(mr ×ar)+((vl ×al)+(vwl×awl))]

Chapter 4: Results and Discussions

CHAPTER 4 RESULTS AND DISCUSSIONS

4.1. Characterisation of ore and concentrate from SMB deposit

4.1.1. Geology and mineralogy of Bisunzu mines

The SMB mines are located on the western and northern slopes of the Bisunzu mountains,

North Kivu Province, DR Congo. The most characteristic feature of the ore deposits is the pervasive kaolonitization of pegmatite resulting in white-to-pink colour of the Ta and Nb rich ore. All of the ore deposits are covered of approximately 10 - 30 meters of red-to-brown soil.

The top parts of the deposits host quartz veins. The ore is characterised by a complete replacement of feldspars by kaolinite and fine-grained muscovite (sericite). The pegmatite bodies are hosted and have sharp contacts with the metamorphic rocks of various lithology, including mica schists (possibly fyllite), gneiss and amphibolite. Sometimes, mica schists are found as enclaves surrounded by altered pegmatite (Figures 4.1a-f). Based on the mineralogical observations, the pegmatite can be classified as lithium-caesium-tantalum (LCT) family (Černy and Ercit, 2005; Dewaele et al., 2016). The pegmatite hosts Ta, Nb, Sn, and Li mineralization in the form of columbite-tantalite (coltan), cassiterite, and lepidolite with minor spodumene as shown in Figure 4.1a-d. The Ta and Nb mineralization is disseminated throughout the entire body of the pegmatite. No apparent macro or micro zones which would concentrate the Ta-Nb minerals that can be identified in the ore bodies. The grain size of coltan is variable and the majority ranging between 38 μm to about 1 mm. Rare, larger crystals of (Ta-Nb) minerals of approximately 2.8 mm can be found near the contact with the metamorphic envelope and or as

Chapter 4: Results and Discussions

small veins (Figures 4.1c and b). Cassiterite is found in the veins with albite, quartz and tourmaline (Figures 4.1c). At Luwowo deposit, shafts and tunnels of artisanal miners exploiting cassiterite ore are located at the feet of the Luwowo mountain, on the eastern slopes of the mountain ride.

(a) (b)

(e) (f)

Chapter 4: Results and Discussions

There are shafts located in the middle of the slope, at about 2300 meter from the sea level, with a maximum length of the tunnel is approximately 70 m. The pieces of schist can be observed in the tailings. Contact between fyllite and white is at various depth, from 10 m down to 50 m from the surface. The hand size pieces of cassiterite-bearing rocks contain quartz, muscovite and altered feldspar (Figure 4.1b). They remind coarse-grained rocks or greisen. The cassiterite is found in the deeper parts of the tunnels.

Horizon of the disseminated pink lepidolite marks the top of the (Ta-Nb) mineralized zone. The thickness of the Li-zone (of pink/purple colour) is variable with approximately (5-

150) meters. For instance, a maximum thickness of the Li-zone is about 40 m at Luwowo and

D4 Gakombe deposits. Pockets of gem-quality tourmaline were found between D2 Bibatama and

D4 Gakombe (Figure 4.1e). However, the local miners provided no access to the shafts and tunnels in order to assess the relationship between the tourmaline zone and the Nb-Ta-Sn and Li mineralization. The close proximity of shafts mining cassiterite and tourmaline (~ 100 - 200 metres), suggests a genetic link between these two minerals. In addition, Figures 4.1d and f show black needles of tourmaline found in the altered gneiss and in schists at the contact with pegmatite. The Ta and Nb concentrates extracted at the Bisunzu deposit are of association of verities of colours ranging from black to reddish-brownish. Table 4.1 displays colours of ores hosting the same minerals as Bisunzu mines including common associated minerals associated with minerals carrying impurity elements.

Chapter 4: Results and Discussions

Table 4. 1. Names and chemical formulae of minerals associated with columbite-tantalite deposits

No Mineral Name Chemical formula Colour of ore

(i) Cassiterite SnO2

(ii) Columbite (Fe,Mn)(Nb,Ta)2O6

(iii) Feldspar (Na,K)AlSi3O8

(iv) Kaolinite Al2Si2O5(OH)4

(v) Lepidolite K(Li,Al)3(Si,Al)4O10(F,OH)2

(vi) Spodumene LiAlSi2O6

(vii) Tantalite (Mn,Fe)(Ta,Nb)2O6

(viii) Tourmaline (Na,Ca)(Mg,Fe,Ti,Al,Li)3(Al,Mg,Cr,Fe,V)6(BO3)3 Si6O18(F,OH)4

(Mineralogy Database)

Chapter 4: Results and Discussions

4.1.2. Tantalite concentrates of different colours from SMB mines

The concentrates produced are of various colours ranging from black, golden-brown, and yellowish-green to dark-brown. According to Knorring (1958), colours of tantalum and niobium concentrates depend primarily on the amount of Ta, Nb, Mn and Fe present in the sample. In addition, trace elements or other physical factors including the age of geological deposit and the location may have a great influence on colours. Table 4.2 gives details of the main features of columbite tantalite concentres collected in Namibia. In a similar work performed by Knorring

(1958) and Singh (2007) reported the main features of concentrates collected from Namibia which exhibited various colours which led to the conclusion that the features are mainly based on the content of Ta, Nb, Mn and Fe along with other physical and chemical aspects. From these features as well as the common colours of columbite tantalite retrieved from the mineral data base, the concentrate of different colours produced at Bisunzu mines were matched to mineral varieties as presented in Table 4.3.

Chapter 4: Results and Discussions

Table 4. 2. Feature of columbite-tantalite concentrate in Namibia

No. Mineral name and Formula Description of main features References

(i) Columbite  Dark-brown Knorring, 1958; (Fe,Mn)(Nb,Ta)2O6  Nb > Ta Melcher et al., 2008;  Fe > Mn Singh, 2007 (ii) Manganocolumbite  Red to brown (Mn,Fe)(Nb,Ta)2O6  Nb > Ta or intermediate  Mn > Fe  Occur in varieties of Li- pegmatite (iii) Ferrocolumbite  Black (Fe,Mn)(Nb,Ta)2O6  Nb > Ta and Fe > Mn  Occur in Li - poor pegmatite  Inclusion of Ti, Sn and W  Presence of Sb may be due to presence of tin (iv) Tantalite  Red-brown (Fe,Mn)(Ta,Nb)2O6  Disseminated grain  Intergrown with microlite  Occur in Li pegmatite (v) Manganotantalite  Golden brown to red (Mn,Fe)(Ta,Nb)2O6  Ta > Nb and Mn > Fe  Highly rich in Ta  Occur in highly fractioned Li pegmatite (vi) Tapiolite  Black (Fe)(Ta,Nb)2O6  Domination of Ta and Fe However, Mn rich may be found  Occur in Li-poor pegmatite (vii) Wodginite  Yellowish-brown to brown, (Ta,Nb,Sn,Mn,Fe)16O32 darkrimmed with sphenoidal shape up to 25 mm in size  Yellowish assigned to free Fe and presence of Sn and Mn rich end member  Tantalian cassiterite was observed in lepidolite rather than wodginite (viii) Microlite  Multiplicity of colours (Ca,Na,U)(Ta,Nb,Ti)2(O,OH,F)  Ta high up to 79 wt %  Occur in a small size with irregular form  Dispersed in nature  Camouflaged in the host mineral matrix  Commonly associated with, or replacing Ta

Chapter 4: Results and Discussions

Table 4. 3. Feature of columbite-tantalite concentrate at the Bisunzu mines

Ta and Nb concentrates Description of main features from Table 3.1 (A)  Mixture of dark and reddish brown  Ta may be greater or intermediate to Nb and Mn > Fe  Probably tantalite (Fe) and columbite (Mn) as major phases  It may occur in highly fractioned Li pegmatite  Refer to (ii) in Table 4.1 and (vii) in Table 4.2

(B)  Yellowish to whitish  Probably Ta > Nb and Mn > Fe  Tantalite (Mn) could be the major and microlite as minor phase  Ore may occur in highly fractioned Li pegmatite  Refer to (ii), (v) in Table 4.1 and (vii) in Table 4.2

(C)  Golden brown to red, pink and some dark particles  Probably, Ta > Nb and Mn > Fe  The bold brown to pinkish may be due to high content of Mg  Tantalite (Mn with Mg) as major and microlite as minor  May occur in highly fractioned Li pegmatite  Refer to (ii), (v) in Table 4.1 and (vii) in Table 4.2

(D)  Dark to brown  Probably, Nb greater or intermediate to Ta and Fe > Mn  Presence of Columbite (Fe) and Tantalite (Fe) as an option phase  May occur in Li-poor pegmatite  Refer to (ii) in Table 4.1 and (vii) in Table 4.2

(E)  Dark and red to brown  Ta > or intermediate to Nb and Mn > Fe  Tantalite (Mn) and Columbite (Mn) as major phases  May occur in Li-poor pegmatite  Refer to (ii) in Table 4.1 and (vii) in Table 4.2

(F)  Golden brown to red and some dark particles  Probably, Ta > Nb and Mn > Fe  Tantalite (Mn) as major and microlite as minor as an option  May occur in highly fractioned Li pegmatite  Refer to (ii), (v) in Table 4.1 and (vii) in Table 4.2 (G)  Dark and red to brown  Ta > or intermediate to Nb and Mn > Fe  Tantalite (Mn) and Columbite (Mn) as major phases  May occur in Li-poor pegmatite  Refer to (ii) Table 4.1 and (vii) in Table 4.2 (H)  Yellowish to whitish and dark to brown  Could be that Ta > Nb and Mn > Fe  Probably, tantalite (Mn), columbite (Mn) as major, and microlite and kaolinite as minor phases  May occur in highly fractioned Li pegmatite associated with kaolinite  Refer to (ii), (iv), (v) in Table 4.1 and (vii) in Table 4.2

Chapter 4: Results and Discussions

4.1.3. Chemical composition of ore and concentrate from SMB deposit

Table 4.4 gives the typical chemical composition of the selected pegmatite ore sample at the Bisunzu mine. It can be seen that aluminium (42 wt % Al) is the major element, while 0.3 wt

% Ta, 0.08 wt % Nb and 0.04 wt % Sn are the minor elements. In addition, extremely low levels of 0.001 wt % U and 0.0004 wt % Th in the pegmatite ore were observed. Referring to Table 1.3 regarding the exported sample (ES*) described in Chapter 1, it is clear that the content of U (0.14 wt %) and Th (0.017 wt %) increases with the pegmatite beneficiation process, as shown in Table

4.4.

Table 4. 4. Chemical composition of pegmatite ore and exported concentrate sample at Bisunzu mine

Category Wt % Pegmatite orea Ta Nb Sn U Th Al Ti Fe Mn K P Cs 0.30 0.08 0.04 0.001 0.0004 42 0.005 0.06 0.05 0.6 0.03 0.01 ES*b 27 4 2 0.14 0.017 3 0.58 3 0.18 - - - a. analysis from this work, b. ES*: exported sample (from Table 1.3)

The ICP-MS quantitative analyses revealed that the content of Ta depends on the mineralisation potential of the mining zones with a wide range varying from 13 wt % to 37 wt

% as shown in Table 4.5. The quantitative analysis of exported sample in Table 4.4 is comparable to quantitative results of sample F for Ta, Nb, U and Th in Table 4.5 (ICP-MS) and Table 4.6

(XRF). Interestingly, the average of the quantitative results obtained across the eight concentrates (A to H) is 27 wt % Ta, reflecting the bulk of concentrates after mixing before export by the SMB (Table 4.6). The samples D and E have relatively low tantalum content i.e.,

15 and 16 wt % Ta, respectively. In contrast, sample B was identified to have high tantalum content i.e., 38 wt % Ta (Table 4.6). The concentration of Nb varies in the range of 4 wt % in sample E to 15 wt % in sample G. Table 4.6 shows the XRF analysis of major as well as minor

Chapter 4: Results and Discussions

and trace elements. Mass percentages of Ta and Nb as major elements in Table 4.6 are comparable with the ICP-MS analysis in Table 4.5.

Table 4. 5. Chemical composition of concentrates collected at Bisunzu mine based on ICP-MS

Element Method Samples Major (wt %) ES* A B C D E F G H FH FM FL Ta ICPMS 1 - 18 32 37 13 14 32 36 28 - - - ICPMS 2 - 19 42 32 - 15 31 34 23 - - - Nb ICPMS 1 - 9 6 7 5 3 4 12 6 - - - ICPMS 2 - 4 7 9 - 3 4 7 6 - - -

ES*: exported sample

Table 4. 6. Chemical composition of concentrates collected at Bisunzu mine based on XRF

Elements Samples ES* A B C D E F G H FH FM FL Major (wt %) Ta 27 17 38 37 15 16 31 37 26 37 33 0.4 Nb 4 13 8 9 6 4 5 15 6 6 4 0.1 Sn 2 0.3 1 1.2 13 13 5 4 4 1 0.3 - Fe 3 14 2 4 7 12 4 2 5 4 3 0.7 Mn 0.18 3 2 4 3 1.6 1 8 2 2 9 - Ca - 2 4 3 4 4 4 1 4 4 0.7 0.1 Al 3 4 3 2 2 2 3 1 3 2 2 6 Mg - 3 0.3 1 6 3 1 1 3 1 1 - K - 1 1 1 0.2 0.2 1 0.4 1 1 0.3 0.3 Si 12 8 8 7 9 6 9 3 10 7 7 37 Na - 0.3 2 1 0.4 1 2 1 1 2 0.3 0.2 Ti 0.58 1 0.1 1 1 4 0.3 0.2 1 1 0.2 0.1 P - 0.09 0.01 0.05 0.05 0.10 0.03 0.05 0.04 0.04 0.05 0.01 Minor (g/t) U 1400 45 421 8021 1047 3434 4067 4206 2318 5333 1836 60 Li - 3020 2000 1180 310 160 2490 670 2020 2100 810 5560 Th 170 10.5 722 291 87 301 500 135 436 367 56 5 Cs - 2197 798 305 55 78 954 122 882 717 186 1581 REE (g/t) Dy - 7.0 1.0 3.5 13 80 2.0 2.5 10 2.5 1.5 1.0 Er - 6.0 1.0 3.0 9.0 55 1.0 2.0 6.0 1.3 1.0 0.3 Eu - 1.0 < 0.5 1.0 2.0 11 1.0 0.5 2.0 1.0 < 0.5 1.0 Gd - 4 1 4 11 63 2 3 12 3 2 2 Sc - 38 4 6 12 10 7 2 9 3 3 < 1 Ho - 49 50 4 22 54 91 0.6 2 55 47 5 La - 34 4 21 56 407 47 14 102 25 4 5 Lu - 1.4 0.2 0.6 2.4 7 < 0.2 1 1.4 < 0.2 1 < 0.2 Nd - 28 3 17 54 349 21 12 79 24 4 12 Pr - 8 1 5 15 98 9 4 24 8 1 3 Sm - 5 1 4 10 63 3 3 14 5 1 3 Tb - 1 0.2 0.6 2 11 0.4 0.5 2 0.5 0.2 0.2 Tm - 1 0.2 0.3 1 8 0.2 0.3 1 0.2 0.2 < 0.1 Y - 36 10 21 80 486 10 15 59 11 9 2 Yb - 9 2 3 7 47 1 2 6 1 2 0.2 TREE - 228 79 94 296 1749 196 62 329 141 77 35

ES*: exported sample, TREE: Total rare earth elements

Chapter 4: Results and Discussions

Titanium content of 1 wt % TiO2 in a weathered rutile or ilmenite ore is considered as economically recoverable (Korneliussen et al., 2000). However, the analyses show that the TiO2 is less than 1 wt % in most samples and the quantitative results show that the main elements are

Ta, Nb and Sn. Thus, Ti is not economically recoverable from this particular deposit. Based on the XRF analysis, it is apparent that the increase in the concentration of Ta in 12 samples does not correlate with Nb as shown in Figure 4.2. Overall, the content of Ta is nearly two to five times higher than that of Nb across the eight concentrates (A-H) analysed. Thus, the minerals carrying Ta-Nb varies greatly and appear to be different at the Bisunzu deposits.

Figure 4.3 presents the comparison of uranium and thorium content of 11 samples all from Bisunzu mine. High content of 0.8 wt % U (or 8021 g/t U, 1.07 wt% U3O8) and 0.07 wt %

Th (or 722 g/t Th, 0.075 wt % ThO2) were found in C and B, respectively, as indicated in Table

4.6. From the obtained values, it appears that the minerals carrying U and Th also vary greatly across the mine. In addition, the tested samples contain higher amount of U compared to 0.1 wt

% U3O8 classified as of concern for the purpose of mining activities and handling, while those containing ten times as high are considered a concern for transportation (TIC, 2016; Nete et al.,

2012; Nete et al., 2014b). On the other hand, thorium was not considered as a critical element, since the quantitative results showed low concentration in most of the samples (Table 4.6).

Consequently, it is important to lower the concentration of U in the concentrate to meet the international shipment requirements of tantalite concentrate. The lowest uranium content of 45 g/t was found in sample A. Thus, mixing with A can lower the uranium concentration in the bulk of concentrate.

Chapter 4: Results and Discussions

45 40 Ta Nb 35 30 25 20 15

Ta Nbor content (wt %) 10 5 0 FL D E A H ES* F FM FH G C B

Samples

Figure 4. 2. Grades of Ta and Nb of different samples (ES*: exported sample)

10000 U Th

1000

100

U orTh content (g/t) 10

1 A FL B D ES* FM H E F G FH C Samples

Figure 4. 3. Grades of U and Th of different samples (ES*: exported sample)

Chapter 4: Results and Discussions

Impurity elements of the twelve concentrates were also analysed quantitatively. For instance, Si in sample ES* is as high as 12 wt % in Figure 4.4 indicating large mass of minerals containing silicon resulting in low content of Ta2O5 and Nb2O5 in the concentrate for export. The lowest content of Si was identified in sample G up to 3 wt %. The higher content of silicon in sample FL up to 37 wt % Si as shown in Table 4.6 and Figure 4.4 is due to the fact that sample

FL was collected from the light material considered as waste (sand) to be dumped.

A relatively high content of Fe ranging from 5.4 wt % to 14 wt % was recorded as shown in Table 4.6, across samples, which may point out the presence of ilmenite or rutile minerals at the Bisunzu deposit (Alhassan et al., 2010). Iron is highly concentrated in sample A up to 14 wt

% (Figure 4.5) followed by sample E up to 12 wt %, whereas the lowest was found in sample FL up to 0.7 wt %. Manganese content ranged from 0.18 wt % Mn in sample ES* as the lowest to 9 wt % in sample FM. On the other hand, Mn was not detected in sample FL as presented in Figure

4.6. This may point out that Mn content in the concentrate increases with the pegmatite ore beneficiation. Content of Al varied from 1 wt % in sample G as the lowest to 6 wt % in sample

FL as the highest as shown in Figure 4.7. The content of Ca of about 4 wt % was found to be close to each other in samples B, D, E, F, H and FH, while the lowest Ca content of 0.11 wt % was observed in sample FL. Removing the mass of minerals forming these impurities using physical or chemical treatment may result in the upgrade of Ta2O5 and Nb2O5.

Chapter 4: Results and Discussions

30 Si

Si (wt (wt Si %) 15

0 G E C FH FM A B D F H ES* FL Samples

Figure 4. 4. Grades of Si of different samples (ES*: exported sample)

14 Fe 12

Fe (wt Fe (wt %) 6

0 FL G B ES* FM C FH F H D E A Samples

Figure 4. 5. Grades of Fe of different samples (ES*: exported sample)

Chapter 4: Results and Discussions

10 9 Mn 8 7 6 5

4 Mn (wt Mn %) (wt 3 2 1 0 FL ES* F E B H FH A D C G FM Samples

Figure 4. 6. Grades of Mn of different samples (ES*: exported sample)

6 Al Ca 5

4 (wt (wt %)

Al Al Ca or 2

0 G C D E FH FM ES* B F H A FL Samples Figure 4. 7. Grades of Al and Ca of different samples (ES*: exported sample)

Chapter 4: Results and Discussions

Referring to Table 4.6, the chemical composition results showed that the content Li is high as twice as Cs, specifically, in samples B, C, D, F and G. The high content of 5560 g/t Li was identified in sample FL, while the lowest of 160 g/t was recorded in sample E as indicated in Figure 4.8. The high content of Li in the light material shows a significant importance in regulating the washing practices. It is obvious that the current traditional way of gravity concentration in water ends up in losing the light materials rich in lithium-bearing minerals.

Thus, an establishment of physical processing facility to recover and concentrate both the light and the heavy materials is of importance which will bring additional income to the company.

The presence of trace rare earth elements (REE) in Table 4.6 were also identified in some samples with high content in sample C along with phosphate. This point out probably the presence of monazite mineral at the Bisunzu mine. The Li and Cs content in the exported sample (ES*) was not analysed.

Samples D and E were identified to possess high content of 13 wt % Sn each as presented in Figure 4.9, with its relatively high content across the other samples. Despite, the unknown tin volume and its mining life span at the Bisunzu deposit, however, it appears that tin can bring additional income to the company.

Chapter 4: Results and Discussions

6000

Li Cs 5000

4000 (g/t) 3000

Li Li Cs or 2000

1000

0 ES* E D G FM C B H FH F A FL Samples

Figure 4. 8. Grades of Li and Cs of different samples (ES*: exported sample)

12 Sn

6 Sn (wt Sn (wt %)

0 FL A FM B FH C ES* G H F D E Samples

Figure 4. 9. Grades of Sn of different samples (ES*: exported sample)

Chapter 4: Results and Discussions

Commonly, a concentrate with 25 wt % Ta2O5 is upgraded to 50 wt % in a clean-up plant.

The presence of Nb2O5 in the present ore is important. According to Burt (1996), concentrate suitable for further processing to recover tantalum are generally required to exceed 25 wt %

Ta2O5, with 50 wt % combined Ta2O5 and Nb2O5 (Adetunji et al., 2005) using the methods discussed in Table 2.9, Chapter 2. The samples B, C, F, G and H all have Ta2O5 and Nb2O5 combined content of about 57.5 wt % to 67 wt %, with the Ta2O5 content of over 25 wt %. Thus, these concentrates can be subjected to clean-up without passing through the pre-concentration or primary concentration for removal of associated minerals followed by chemical treatment

(Bludssus et al., 1993). The Ta2O5 content in samples A and E are 21 wt % and 20 wt %, respectively, which is reasonably high to justify a primary concentration stage. Sample D has 18 wt % Ta2O5 while samples D and E have a high content of SnO2 of about 16.5 wt % each. Thus, it will be appropriate to treat the concentrates primarily for tin recovery and then secondary treatment for tantalum and niobium oxides (Adetunji et al., 2005).

Figures 4.10a, b, c and d show the correlation between Ta, Nb, U and Th analyses obtained from ICP-MS and XRF (Table 4.6), respectively. It is clear that a good linear relationship with a correlation coefficients between 0.97 and 1.00 indicates extremely good agreement between the quantitative results obtained using the two analytical methods. This indicates that the tantalite concentrates successfully dissolved during the flux digestion with a little or no loss of the major element. The disagreement between the ICP-MS and XRF for Ta in samples B and H may be due to the inhomogeneity. Table 4.7 shows the values of R2 and slopes of correlation between Ta, Nb, U and Th analyses based on Figures 4.10a, b, c and d, respectively. Due to the sensitivity of the XRF to major and minor elements and its agreement with the ICP-MS analysis, the XRF quantitative results were used for further analyses in the present study.

Chapter 4: Results and Discussions

a) b) 45 16

40 B G 14 Nb Ta C G 35 12

30 F H 10 A 25 8 C 20 H y = 1.03x 6 15 A B y = 0.78x D E R² = 0.97 F D R² = 0.97

XRF (wtanalayis%) XRF 10 4 XRF (wtanalysis%) XRF E 5 2

0 0 0 5 10 15 20 25 30 35 40 45 0 2 4 6 8 10 12 14 16 18 ICP-MS analysis 1 (wt %) ICP-MS analyis 1 (wt %) c) d) 14000 1600

C 1400 12000 U Th B 1200 10000 1000 F 8000 F 800 H 6000 E G 600 C E y = 1.46x XRF (g/t) analyis XRF 4000 H 400 y = 1.67x

R² = 1.00 (g/t) analyis XRF R² = 0.99 2000 G D 200 D A A 0 B 0 0 2000 4000 6000 8000 10000 0 100 200 300 400 500 600 700 800 900 ICP-MS analyis 1 (g/t) ICP-MS analysis 1 (g/t)

Figure 4. 10. Correlation between chemical assays based on ICP-MS 1 and XRF (a) Ta, (b) Nb, (c) U and (d) Th

Table 4. 7. Correlation parameters of Ta, Nb, U and Th analyses from ICP-MS 1 and XRF

Component Coefficient of determination (R2) Slope (m) Ta 0.97 1.03 Nb 0.97 0.78 U 1.00 1.46 Th 0.99 1.67

Foot wrote: based on Figure 4.10.

Chapter 4: Results and Discussions

4.1.4. XRD analysis of SMB ore and concentrates

Figure 4.11 shows the XRD pattern of the SMB pegmatite ore revealing mineral components. The results show that the pegmatite ore is mainly dominated by kaolinite and muscovite, while mangotantalite is the minor mineral component. Lepidolite was also identified.

From the XRD results, it is clear that the pegmatite ore hosts lepidolite as Li-source and probably the Ta-Nb bearing mineral is hosted in the kaolinite zone.

Figure 4. 11. XRD pattern of pegmatite ore at Bisunzu mine

Chapter 4: Results and Discussions

Table 4.8 summarises the mineral identification of eight concentrates collected at the

Bisunzu mine using XRD and SEM-EDS. The results obtained using XRD analyses of the concentrates showed that all samples are similar with respect to the mineralogical composition with the exception of trace minerals such as diopside, hematite, augite and olivine. The XRD patterns of samples C and G are presented in Figures 4.12 and 4.13, respectively, as a typical example, whereas XRD patterns for samples A, B, D, E, F and H are presented in Appendix A

5, Figures A 5.1a and A5.1b, respectively. Ercit (1986) reported an association of simpsonite, cassiterite, manganocolumbite, rankamaite, plumbomicrolite, microlite and parabariomicrolite from a nearby location at Mumba. It was observed that the minerals identified in the present study are in agreement with those identified in the work accomplished by Ercit (1986). For instance, cassiterite, mangacolumbite, microlite and parabariomicrolite were identified in both projects (Table 4.8). Ercit (1986) indicated that this association originate from spodumene or petalite and, rare-element granitic pegmatites. In addition, Knorring et al. (1969) reported the presence of rankamaite mineral in Mumba, a mining site near Bisunzu Mine. He found that the mineral constitutes a matrix made up of corroded grains of simpsonite, minute crystals of cassiterite and some manganotantalite and muscovite. He suggested that the formation of rankamaite was probably due to the alteration of simpsonite and the source rock was probably a lithium pegmatite.

Chapter 4: Results and Discussions

Table 4. 8. Mineral identification from eight concentrates collected at the Bisunzu mine

Mineral name Chemical Formula Reference codea Sample identification A B C D E F G H

Tantalite (Fe) (Fe,Mn)(Ta,Nb)2O6 04-018-7027 √ ------

Tantalite (Mn) (Mn,Fe)(Ta,Nb)2O6 04-012-9860 - √ √ - √ √ √ √

Columbite (Mn) (Fe,Mn)(Nb,Ta)2O6 98-017-1774 - - - √ - - √ -

Pyrochlore (Na,Ca,U)2(Nb,Ta)2O6(OH,F) 01-084-8280 √ - - - - - √ -

Fluorocalciomicrolite (Ca,Na,U) 2(Ta,Nb)2O6 (OH,F) 04-015-1266 - √ √ √ √ √ √ √

Parabariomicrolite (Ba,U)(Ta,Nb)2(O,OH)7 04-012-1873 - √ - - - √ √ -

Stannomicrolite Sn2Ta2O7 00-023-0599 - - √ √ √ - - -

Cassiterite SnO2 04-005-5929 - - - √ √ - √ -

Muscovite KAl2(Si3Al)O10(OH,F)2 00-058-2036 √ √ √ √ √ √ √ √

Diopside CaMgSi2O6 00-041-1370 √ √ - √ - √ - √ Goethite Fe3+O(OH) 01-084-8280 √ ------

Forsterite Mg2(SiO4) 01-076-0513 √ ------

Quartz SiO2 01-085-0790 √ √ √ √ √ √ √ √

Augite (Ca,Na)(Mg,Fe,Al,Ti)(Si,Al)2O6 00-041-1483 - - √ - √ - - -

Hematite Fe2O3 01-076-4579 - - √ √ √ √ - -

Olivine (Mg,Fe)2SiO4 01-083-1484 - - - √ √ √ - √

Perovskite CaTiO3 01-083-0191 - - - - - √ - √

Kaolinite Al2Si2O5(OH)4 00-058-2028 - - - - - √ - √

2+ Ilmenite Fe TiO3 00-029-0733 - - - - √ - - -

Rare earth mineral (Ca,La,Y,Th)PO4 - - √ - - - - -

Lepidolite K(Li,Al)3(Si,Al)4O10(F,OH)2 - √ - √ √ √ √ - Topaz Al₂SiO₄(F,OH)₂ - √ - - √ √ √ -

2+ 3+ Amphibole Ca2(Mg,Fe ,Fe ,Al)5(Si,Al)8O22(OH)2 - - √ - - - √ √ -

Diopside CaMgSi2O6 - - √ √ √ √ √ - √ Zircon ZrSiO₄ - - - √ - √ √ √ -

Columbite (Fe,Mn)(Nb,Ta)2O6 - - - √ - - - - -

Simpsonite Al4(Ta,Nb)3O13(OH,F) Ercit (1986)

Rankamaite (Na,K,Pb,Li)3(Ta,Nb,Al)11(O,OH)30 Knorring et al. (1969) a. Note that minerals without reference codes were identified specifically by SEM-EDS shown in Figures 4.14 and 4.17, and Appendix A 6.

Chapter 4: Results and Discussions

Figure 4. 12. XRD pattern of tantalite concentrate of sample C

Figure 4. 13. XRD pattern of tantalite concentrate of sample G

Chapter 4: Results and Discussions

4.1.5. SEM observations and EDS analysis of sample C

The micrograph analyses using the SEM are presented in Figures 4.14, 4.15 and 4.16 for

C and 4.17 for G as selected samples. The SEM-EDS micrograph definitions of samples A, B,

D, E, F and H are shown in Appendix A 6. The backscattered electron (BSE) images of the selected parent samples revealed that most of the Ta-Nb bearing minerals at the Bisunzu mines belong to the columbite group, where the brightness is caused by Ta and Nb rich end-members.

The EDS analysis has also shown that elements Ta and Nb are hosted in the major components tantalite (Fe), tantalite (Mn) and columbite (Mn) as well as in minor components including pyrochlore, fluorocalciomicrolite, parabariomicrolite and stannomicrolite as presented in Table

4.8. The presence of uranium was identified in minerals including fluorocalciomicrolite and parabariomicrolite.

The presence of quartz, hematite, and augite associated with amphibole, muscovite and rare earth minerals were also identified in the concentrate sample C (Figure 4.14 and Table 4.8).

Most of the coltan grains are zoned. Zoning can be oscillatory, with multiple Ta-rich zones

(bright), and some Nb-rich (dark) zones, but not often. Bands and irregular patchy zoning can be found in the ore as can be seen in Figure 4.14. Uranium was found in the microlite subgroup of the pyrochlore group of mineral. In addition, there is > 78 wt % Ta in the sample, whereas Nb is approximately 7 wt % being hosted in the microlite phase in the sample (Figure 4.15). Calcium is the most abundant (5 wt %) cation followed by sodium which encountered an average of about

2 wt % in the microlite (Figure 4.16c). Fluorine in some cases was observed with an overall of 3 wt % (Figure 4.16c). Yttrium was not detected with the SEM-EDS. Since, the elemental composition of Ca > Na and F > Y as indicated in Figures 4.15 and 4.16, the mineral is identified as fluorocalciomicrolite (Andrade et al., 2013). The minerals identified using the EDS were consistent with those identified using XRD, with the exception of the rare earth minerals,

Chapter 4: Results and Discussions

columbite, augite and zircon. Based on the XRF, ICP-MS and the SEM-EDS analyses, Ta and

Nb in sample C is mainly hosted by manganotantalite (Mn,Fe)(Ta,Nb)2O6 of the columbite- tantalite group.

Mineral identification in Figure 4.14 1. Fluorocalciomicrolite 2. Manganotantalite 3. Quartz 4. Hematite 5. Augite associated with amphibole 6. Muscovite 7. Rare earth mineral 8. Columbite 9. Zircon

Figure 4. 14. Backscattered SEM micrograph of tantalite sample C from Bisunzu mine

The topography of the elemental peaks of sample C using the EDS are presented in Figure A 6.1,

Appendix A 6. The inclusions of Si, Al and Fe elements in microlite (Figures 4.15 and 4.16) may be due to the fact that columbite-tantalite has been altered along fractures due to the crystallisation of new minerals which is common for columbite-tantalite pegmatite ore (Alfonso et al., 2016).

Chapter 4: Results and Discussions

(a)

(b)

Figure 4. 15. Backscattered SEM images and their correspondent EDS elemental analysis of tantalite concentrate from Bisunzu mine sample C: (a) Fluorocalciomicrolite (10); (b) Silica, aluminium and tantalum associated with uranium in black spot (11)

Chapter 4: Results and Discussions

Figure 4. 16. Backscattered SEM image and their correspondent EDS elemental analysis of tantalite concentrate from Bisunzu mine sample C: Fluorocalciomicrolite associated with aluminium (12)

Figure 4.14 shows that the distribution of Ta and Nb in the minerals is uniform. However, some elements are irregularly distributed giving place to minerals with zoned textures (Figures

4.15a-b). As a result, the optimal design of the liberation-milling operations during the processing lies in the structural characterisation of these minerals in order to obtain a relatively high degree of Ta2O5 richness of the concentrate (Adetunji et al., 2005).

Chapter 4: Results and Discussions

4.1.6. SEM observations and EDS analysis of sample G

Figure 4.17 shows the backscattered electron image (BSE) of sample G. The SEM image shows the interlocking of minerals including the subgroup of microlite carrying U with manganotantalite Ta rich end-member. Impurities including quartz and muscovite are present in the sample. Zircon was also identified in the present concentrate which may point out the isomorphism of U4+ with Zr (De Moura, 2009). The EDS analysis indicated tantalite (Mn) and columbite (Mn) as major phases, fluorocalciomicrolite and parabariomicrolite as minor phases, while quartz and muscovite as well as augite associated with amphibole as impurities. Most of the grains of Ta-Nb bearing mineral belong to the columbite-tantalite solid solution series. Their composition is mainly manganotantalite, and in minor amounts of pyrochlore, microlite and manganocolumbite.

The backscattered images (spots 1 & 2) of the bright grain point out the association of

Ta, Nb, U, among others forming fluorocalciomicrolite. For some cases, the microlite grains were locked into columbite and tantalite as well as cassiterite in the present sample. The microlite particles (< 1000 μm) are usually associated with columbite and tantalite, and thus, in most cases they are locked in these minerals (Gonzaléz et al., 2017). Silica is dominant in the dark grain of minerals (Figure 4.17). The rare earth elements (REE), yttrium (Y), cerium (Ce), neodymium

(Nd) and lanthanum (La) are hosted in grey grains, probably forming monazite in association with Zircon. The EDS elemental analysis confirmed the minerals identified using XRD, except the parabariomicrolite, augite, zircon, topaz and the olivine identified using EDS. Cassiterite was found to contain some tantalum and niobium where tantalum in cassiterite may be due to the inclusion of microcrystals of columbite group minerals and microlite and, probably, as a result of substitution of Sn by Nb and Ta (Ghorbani et al., 2017).

Chapter 4: Results and Discussions

Mineral identification in Figure 4.17 1. Fluorocalciomicrolite 2. Parabariomicrolite 3. Manganotantalite 4. Cassiterite 5. Quartz 6. Muscovite 7. Augite 8. Topaz 9. Zircon 10. Olivine

Figure 4. 17. Backscattered SEM image of tantalite sample G from Bisunzu mine

The topography of the elemental peaks of sample G using the EDS are presented in

Appendix A 6, Figure A 6.2. The SEM micrograph and EDS mineral identification for samples

A, B, D, E, F and H are shown in Appendix A 6, Figures A 6.3, A 6.4, A 6.5, A 6.6, A 6.7 and

A 6.8, respectively. Thorium may be hosted in monazite mineral [(Ce,La,Nd,Th)PO4] since it was sourced from a pegmatite ore. It may also be hosted in euxenite

(Y,Er,Ca,Ce,U,Pb,Th)(Nb,Ta,Ti)2(O,OH)6, which is often associated with columbite-tantalite or pyrochlore (Knorring, 1958; Jerome et al., 2006). However, no mineral phases carrying Th were identified in the present study. It may be due to the low Th content in the concentrate.

It is clear from the XRD and SEM-EDS results that the Ta-Nb rich end-members were mainly manganotantalite and manganocolumbite. In addition, members of manganotantalite series with high content of Ta and Mn compared to Nb and Fe has been observed. The colour of manganotantalite varies mainly from golden-brown to red (Table 4.1). The colours can be

Chapter 4: Results and Discussions

attributed primarily to the contents of Nb and Fe in the concentrate (Knorring, 1958). However, the presence of other minerals such as kaolinite, tourmaline, beryl, feldspar, muscovite, quartz among others can influence the colours of concentrates (Amuda et al., 2007). It was also found that manganotantalite is more widespread than other minerals and it may be regarded as a type of Ta mineralisation in the poor fractionated Li pegmatite. Fluorocalciomicrolite was found to accommodate a large number of elements with a high content of Ta (~ 78 wt %). This may be due to the fact that microlite belongs to (F,OH)-series, the last minerals to crystallise and therefore is noted in the latest micaceous replacing units, and frequently closely associated or replacing the quartz core of the pegmatites (Knorring, 1958 and Andrade et al., 2013). In the related work Knorring (1958) and Andrade et al. (2013) reported a chemical composition close to 79 wt % Ta in a concentrate sourced in Namibia and Volta Grande pegmatite, Minas Gervais in Brazil, which is in the same range of Ta content observed in the present concentrate.

4.1.7. Correlation of elemental assays

Figures 4.18a-d show the correlation between elemental assays referring to Table 4.6.

The linear correlation between Al and K, Al and Li as well as Li and Cs in Figure 4.18 show that these elements occur in the same mineral lepidolite which was also identified in the pegmatite ore using XRD as shown in Figure 4.11. The values of slopes and R2 are listed in Table 4.9.

Lepidolite with a general formula K(Li,Al)3(Al,Si)4O10(F,OH)2 can be enriched in the pegmatite rock with columbite-tantalite, and in addition to lithium, it can contain potassium, caesium and rubidium (Wang et al., 2004; Bradley et al., 2017).

Chapter 4: Results and Discussions

a) b) 4.0 4.0 A 3.5 H A 3.5 y = 12.49x H 3.0 y = 2.54x F 3.0 R² = 0.88 F R² = 0.99 2.5 B 2.5 B 2.0 D 2.0 C C

(wt%) D (wt%) 1.5 G 1.5

E Al Al Al G 1.0 1.0 E 0.5 0.5 0.0 0.0 0.0 0.3 0.5 0.8 1.0 1.3 1.5 0.00 0.10 0.20 0.30 0.40 K (wt %) Li (wt %) c) d) 0.60 3.0 y = 2.67x y = 35.06x 0.50 R² = 0.93 2.5 R² = 0.97 B 0.40 2.0 F

0.30 F A 1.5

(wt%) H

C (wt%) 0.20 B & H Na 1.0 Li G C E 0.10 0.5 A G D D E 0.00 0.0 0.00 0.05 0.10 0.15 0.20 0.25 0.00 0.02 0.04 0.06 0.08 Cs (wt %) Th (wt %)

Figure 4. 18. Correlation between of chemical assays (a) Al and K, (b) Al and Li, (c) Li and Cs, (d) Na and Th

A linear correlation of Na and Th assays has been observed for most of the samples as can be seen in Figure 4.18d with a correlation coefficient R2 = 0.97. The presence of Th was not detected in any of the samples using different characterisation methods. However, loparite

(Na,Ce,Ca,Sr,Th)(Ti,Nb,Fe)O3 may be present in a small amount which may host Na and Ta.

Monazite (Ce,La,Nd,Na,Th)(PO4,SiO4), euxenite (Y,Ca,Er,La,Ce,U,Na,Th)(Nb,Ta,Ti)2O6 and pyrochlore (Na,Th,Ca,U)2(Nb,Ta)2O6(OH,F) can also host Na and Th (Popova et al., 2014;

Lumpkin and Ewing, 1995; Demol et al., 2018). These minerals are commonly associated with columbite-tantalite in the pegmatite ores (Linnen et al., 2014; BGS, 2011).

Chapter 4: Results and Discussions

Table 4. 9. Correlation of elemental assays of different samples

Component correlation Coefficient of determination (R2) Slope (m) Al and K 0.99 2.54 Al and Li 0.88 12.49 Li and Cs 0.93 2.67 Na and Th 0.97 35.06

Based on Figure 4.18 for samples (A – H).

4.1.8. Comparison between different ores

A comparison between the minerals associated with the ore from the Bisunzu mine with those from Mozambique and Nigeria shows a very distinct types of minerals that contain Ta and

Nb as presented in Table 4.10. It can be seen that the minerals carrying Ta-Nb from the three locations are quite similar. Table 4.11 compares the chemical composition of columbite-tantalite concentrate from the three deposits. As shown in Table 4.11, the columbite-tantalite in Nigeria has substantially higher contents of Ta and Nb ranging between 6–60 wt % Ta2O5 and 16–38 wt

% Nb2O5, while the content of Ta2O5 and Nb2O5 from Bisunzu lies in between those in Nigeria and Mozambique samples. On the other hand, relatively high contents of 0.54 wt % ThO2 and

2.81 wt % U3O8 makes the columbite-tantalite concentrate of that particular deposit of

Mozambique very critical compared to the samples from Nigeria and Bisunzu. High content of

33.4 wt % TiO2 in samples from Nigeria makes the ore a source for rutile mineral compared to other deposits. However, high content of 16.5 wt % SnO2 makes Bisunzu mine an important source for cassiterite mineral compared to the low contents of 8 wt % and 0.44 wt % SnO2 in other two deposits. The tantalite ores from DR Congo and Mozambique are mainly associated with iron, while that from Nigeria is associated with titanium as impurities which would affect the beneficiation process. In contrast, the columbite-tantalite concentrate from Mozambique is

Chapter 4: Results and Discussions

very critical for handling and transportation due to high content of radioactive minerals which appears to be a complicated matrix for beneficiation purposes.

Table 4. 10. Mineralogical of columbite-tantalite deposits from Nigeria, Mozambique and DR Congo

Country of origin Nigeriaa Mozambiqueb DR Congoc Minerals Tantalite Manganotantalite Manganotantalite Columbite Ferrotantalite Ferrotantalite - Microlite Fluorocalciomicrolite - Manganocolumbite Parabariomicrolite - Ferrocolumbite Manganocolumbite - Wodginite - a. Adetunji et al. (2005); b. Knorring (1958); c. This work.

Table 4. 11. Chemical composition of columbite-tantalite from various locations

Oxides Chemical composition of columbite-tantalite (wt %) from different locations Nigeriaa Mozambiqueb DR Congoc Ta2O5 6-60 33 25-46 Nb2O5 16-38 27 5-25 U3O8 0.01-1.58 2.81 0.17-1.57 ThO2 0.01-0.13 0.54 0.01-0.34 TiO2 33.4 2 7 SnO2 8.0 0.44 16 Fe2O3 11 15 20 MnO 10 19 11 ZrO2 0.42 0.39 2.0 a. Adetunji et al. (2005); b. Knorring (1958); c. This work.

Chapter 4: Results and Discussions

4.1.9. Summary of characterisation

The Bisunzu pegmatite ore hosts Ta, Nb, Sn, and Li mineralisation in the form of columbite-tantalite (Coltan), cassiterite, and lepidolite with minor spodumene, respectively, as shown in Figure 4.1a-f. The Ta and Nb mineralization is disseminated throughout the entire body of the pegmatite (Figure 4.1a-b) with a particle size distribution in the range of –38 μm to about

2.8 mm. Cassiterite is found in the veins with albite, quartz and tourmaline (Figure 4.1c). Horizon of the disseminated pink lepidolite marks the top of the (Ta-Nb) mineralised zone. The thickness of the Li-zone of pink/purple colour is variable; maximum thickness of the Li-zone is about 40 m, specifically, in Nyange and Luwowo deposits (Figure 4.1f). Pockets of gem-quality tourmaline were found between D2 and D4 Bibatama mines. Based on the mineralogical observations, the present pegmatite belongs to lithium-caesium-tantalum (LCT) family (Cẽrný and Ercit, 2005; Schulz, 2017).

The chemical analysis of the tantalum concentrates sourced from eight different prominent deposits have revealed that Ta2O5 content is larger than Nb2O5 content in all samples as shown in Tables 4.4 - 4.6. Thus, pre-concentration plant is of importance to concentrate the heavy minerals by eliminating the light materials, in series with the primary concentration to separate the heavy materials. The presence of TiO2, MnO, and Fe2O3 suggest the possibility of using magnetic separation, while the heavy minerals such as cassiterite, ilmenite, tantalite, niobate and wolframite can be electrostatically recovered leaving the non-conducting zircon in the gangue.

The Ta-Nb rich end-members were identified in various phases including tantalite (Mn), columbite (Mn) and tantalite (Fe), whereas the presence of U was identified in fluorocalciomicrolite and parabariomicrolite. The presence of quartz, muscovite and cassiterite

Chapter 4: Results and Discussions

was also identified using the XRD and the SEM-EDS. In the case of microlite group, they are few microns in size locked in the columbite-tantalite minerals, and sometimes in cassiterite.

Columbite and cassiterite occur mainly in coarse particles of diameter ~ 0.1 mm, while tantalite is in contact with or included in cassiterite. At SMB, some grains are heterogeneous containing various elements, while some other grains are homogeneous and mostly zoned. Zoning can be oscillatory, with multiple bright zones of the main component with inclusion of U and REE in some cases. The grey grains host the Nb-rich bands and irregular patchy zoning as has shown in Figures 4.14 and 4.17. This zoning is common for pegmatite bearing tantalum ore (Alfonso et al., 2016). The silicon, oxygen, iron, zircon and hafnium mainly hosted in dark grains. Among the eight samples, the high content of Ta2O5 and Nb2O5 was identified in sample G, whereas the lowest was encountered in samples A and E, respectively. High amount of U and Th was found in samples C and B, respectively, whereas the lowest amount of these elements were found in sample A. Large quantities of Sn were found in samples D and E, while relatively high amount of Li was found in the FL sample considered as the light materials or gangue (Table 4.6).

4.2. Radioactivity of tantalite samples collected at Bisunzu

Table 4.12 summarises the measured radiation levels and calculated radioactivity of eleven samples. The conversion factors used are found in Tantalum-Niobium International Study

Center report released in 2016 as described in Chapter 3. For instance, for a material containing

0.01 wt % U3O8 and 0.00 wt % ThO2, the radioactivity would be: (0.01 wt % × 104) + (0.00 wt

% × 35.6) = 1.04 Bq/g, i.e., < 10 Bq/g, which allows the exemption for transportation. Material of high radioactivity is subjected to transport regulations and a total radioactivity of the package has to be calculated to obtain a certificate for shipment.

Chapter 4: Results and Discussions

Radiation levels were determined from a 0.5 g of concentrate of each sample minus the radiation level of the background of 0.09 μSv/h. It is clear from the results obtained that the radiation levels for all samples are below 5 μSv/h. Thus, the materials can be transported.

However, for large quantities, the radiation may increase to high levels. The radioactivities reported in this study are comparable to those found at different mining sites including Bisunzu mine conducted by Mustapha et al. (2007).

Table 4. 12. Radioactivity due to uranium and thorium oxides in coltan concentrates

Samples U3O8 ThO2 Radiation Radioactivity Total Observation (wt %) (wt %) Levels (Bq/g) radioactivity (μSv/h) (Bq) A 0.01 0.00 0.13 1.04 - E.L. B 0.05 0.08 0.16 8.05 - E.L. C 0.96 0.03 0.10 101 50.5 A.E.L. D 0.13 0.01 0.14 14.0 7.0 E.L. E 0.41 0.03 0.12 44.0 21.8 A.E.L. F 0.49 0.05 0.14 53.4 26.7 A.E.L G 0.51 0.01 0.12 31.0 15.4 A.E.L H 0.28 0.05 0.17 52.7 26.3 A.E.L FH 0.64 0.04 0.16 68.0 34.0 A.E.L FL 0.22 0.01 0.09 23.2 11.6 A.E.L FM 0.01 0.00 0.12 1.04 - E.L

E.L.: exemption level A.E.L.: above exemption level

Referring to Table 4.12, some of the concentrates such as E, F, G, H and FH exhibit radioactivity which exceed the recommended exemption levels according to the reports published by the International Atomic Energy Agency in 2004 (TIC, 2016) and the European

Commission in 2002 on transporting Ta-Nb raw materials that are NORM. Thus, according to the international specifications coltan mining should be subjected to stringent regulations in

Chapter 4: Results and Discussions

order to keep doses as low as reasonably achievable. The most likely potential risk of high occupational exposures to naturally occurring radioactive minerals during coltan mining activities is inhalation as coltan mineral produce dust while crushing, grinding and sieving in the mills. It is therefore recommended that columbite-tantalite mills should be adequately ventilated and the workers should wear appropriate protective clothes including gloves and dust-masks as a short-term solution. The sustainable solution on the other hand is to remove the uranium and thorium from the concentrates using physical separation or chemical treatment to meet the transportation regulations.

4.3. Physical separation and concentration

4.3.1. Sieving to upgrade tantalum and niobium oxides and remove uranium and thorium

Samples of the two concentrates D and G were screened into two fractions fine (–1000

μm) and coarse (–2800+1000 μm) to investigate the possibilities of upgrading Ta2O5 and Nb2O5 as well as lowering the concentration of U and Th. Results are summarised in Figures 4.19 and

4.20. For sample D, the low content 18 wt % Ta2O5 in the feed changed to 30 wt % Ta2O5 in the coarse fractions compared to 16 wt % Ta2O5 in the fine fractions. In sample G, on the other hand, the content of Ta2O5 in coarse and fine size fractions did not change. This may be due to low mass fraction of mineral containing Fe or Ti in the concentrate. Figure 4.19b shows a high content of 21 wt % Nb2O5 in the coarse fraction compared to 8 wt % Nb2O5 in the fine fraction in sample D. In sample G, Nb2O5 content in the feed (22 wt %) has increased to 30 wt % in the coarse fraction. As shown in Figures 4.19c, d and e, significant amounts of Fe2O3, TiO2 and SiO2 were retained in fine fractions allowing the improvement of the content of Ta2O5 and Nb2O5 in coarse fractions.

Chapter 4: Results and Discussions

From the quantitative results obtained, it is clear that sieving can be considered as a suitable method to upgrade the content of Ta2O5 and Nb2O5 in the concentrate, which in turn can bring a high revenue to the mining company or minors. According to Burt (1984), the tantalum minerals are friable. Consequently, Ta-Nb concentration decreases with the decrease of particle size, allowing the liberation of the valued elements from the impurities to maximise the benefit.

a) b)

50 35 46 45 47 45 30 D G 30 D G 40 35 25 22 30 21 20 30 20

25 %) (wt (wt %) (wt 5 15 5 20 18

16 O

O 2

2 9

15 10 8 Nb Ta 10 5 5 0 0 Feed Fine Coarse Feed Fine Coarse Samples Samples

c) d) 12 2.5 D G 10 10 D G 2.0 2 10 2.0

7 1.5

(wt %) (wt

3 2

6 1 O

2 1.0

4 TiO Fe 4 3 0.5 0.4 0.3 2 1 0.06 0 0.0 Feed Fine Coarse Feed Fine Coarse Samples Samples e) f) 25 16 23 D G D G 14 20 14 20 12 11 10

10 9 (wt %) (wt

(wt %) (wt 15 2 8

SiO 10 8 MnO 6 7 4 4 4 5 4 2 2

0 0 Feed Fine Coarse Feed Fine coarse Samples Samples

Figure 4. 19. Sieving beneficiation of samples D and G (a) Ta2O5, (b) Nb2O5, (c) Fe2O3, (d) TiO2, (e) SiO2 and (f) MnO

Chapter 4: Results and Discussions

Interestingly, the quantitative results of the fine and coarse size fractions have also revealed an approach to deal with the radioactive materials (i.e. U and Th). It can be seen from

Figure 4.20a that the U content in fine fractions is approximately 2 to 4 times higher compared to coarse fractions in samples D and G, respectively. The same trend can be seen for Th in Figure

4.20b with high content in fine compared to coarse size fractions of about 9 and 3 times in samples D and G, respectively. For instance, in Figure 4.20a, the content of U in fine fraction of sample D is 1023 g/t, whereas the content of U is 670 g/t in coarse fraction. Regarding the concentrate G, the content of U in fine fraction was found to be as high as 4844 g/t compared to

1132 g/t in coarse fractions. The quantitative results of sample D have also revealed a Th content as high as 0.9 g/t in fine fraction compared to 0.1 g/t in coarse fraction as shown in Figure 4.20b.

a) b) 6000 160 135 135 D G 4844 140 5000 D G

4206 ) 120 (g/t) 4000 100 87

3000 80 Uranium Uranium Thorium Thorium (g/t 60 51 2000 1047 1023 1132 40 1000 670 20 0.9 0.1 0 0 Feed Fine Coarse Feed Fine Coarse Samples Samples Figure 4. 20. Sieving removal of (a) U and (b) Th from samples D and G

The high content of U in fine fractions point out the fact that U is definitely hosted by the microlite subgroup of mineral. The microlite subgroup of mineral was reported by Knorring

(1958) as a mineral associated with columbite-tantalite in a pegmatite ore in Namibia and he found that microlite carrying U was hosted in small grains of diameter < 1 mm. Furthermore,

Andrade et al. (2013) also reported the presence of microlite in Minas Gervais, Brazil containing

U in a small size fraction of < 0.1mm diameter. These findings are in agreement with the present

Chapter 4: Results and Discussions

results of U to be hosted in a mineral particle of small size of < 0.1 mm. These results suggest that the sieving process can be employed to minimise the content of radioactive elements in the concentrate, which in turn may eliminate the use of chemical treatment to dissolve the U and Th from the concentrate, specifically for coarse particles.

4.3.2. Magnetic separation and concentration of Ta-Nb oxides

A comparison of Ta2O5, Nb2O5, Fe2O3 and TiO2 content in magnetic and non-magnetic fractions of samples FH, FM and FL is presented in Figures 4.21a-d. The results from chemical composition analysis of magnetic and non-magnetic portions are provided in Appendix A 4,

Tables A 4.5a and A 4.5b as oxides and elements, respectively. Sample FH, with relatively higher

Fe and Ti content (Figure 4.21 feed), gave high content of Ta2O5 in the non-magnetic portion.

For instance, the Ta2O5 content of 45 wt % in FH changed to 9 wt % and 50 wt %, respectively, in magnetic and non-magnetic fractions (Figure 4.21a). Likewise, the Nb2O5 content in non- magnetic fraction of FM is 24 wt % compared to 6 wt % in magnetic fraction (Figure 4.21b).

It is clear from the results in Figures 4.21c and 4.21d that the high content of Fe and Ti indicates higher magnetic susceptibility of the concentrates. The analysis (Figure 4.21a) of sample FL show lower magnetism indicating approximately the same concentrations of 0.7 wt

% Ta2O5 and 0.4 wt % Ta2O5 in magnetic and non-magnetic portions, respectively. The low recovery of Ta2O5 and Nb2O5 in non-magnetic portion is due to low content of these oxides in sample FL.

Chapter 4: Results and Discussions

a) b) 60 30 Feed M NM Feed M NM 50 50 25 24 45 21 40 40

40 20

( wt%)

( wt%) (

O O

2 30 15

Ta Nb 9 9 20 10 15 6 5 9 10 5

0.4 0.7 0.4 0.12 0.33 0.06 0 0 FH FM FL FH FM FL Samples Samples

c) d)

60 9 51 Feed M NM 8 Feed M NM 8 50 7 6

(wt %) (wt 40 6

( wt %) ( wt

2 O

2 5

30 Fe 25 TiO 4 20 3 2 8 2 10 7 1 1 1 1 4 3.4 1 1.4 1 1 0.09 0.1 0 0 FH FM FL FH FM FL Samples Samples

Figure 4. 21. Composition of samples FH, FM and FL before and after magnetic separation (a) Ta2O5, (b) Nb2O5, (c) Fe2O3 and (d) TiO2

The results show a high content of 51 wt %, 25 wt % and 8 wt % of Fe2O3 in samples

FH, FM and FL, respectively, indicating that a total of 86% Fe2O3 ended up in the magnetic portion as evident from Figure 4.21c. Similarly, high recovery of approximately 16% of TiO2 was achieved in magnetic portion (Figure 4.21d). These results indicate a relatively high separation of mineral mass containing Fe and Ti into the magnetic fraction which allowed an increase of Ta2O5 and Nb2O5 in the non-magnetic fraction. Nete et al. (2014a) reported a removal of 64.1 wt % and 72.0 wt % of the total Fe and Ti, respectively, and approximately 2 wt % each of Nb2O5 and Ta2O5 from a columbite-tantalite concentrate collected in Mozambique using magnetic separation method. These results for the two concentrates (Mozambique and D R

100

Chapter 4: Results and Discussions

Congo) are in good agreement indicating high recovery of Fe and Ti in magnetic portions, as shown in Figure 4.21c-d.

Titanium(IV) is the most stable oxidation state of Ti, which is a d0 species, which would not be influenced by external magnetic fields (Nete et al., 2014a). In addition, Ta(V) and Nb(V) are most stable oxidation states (d0). Thus, Ta and Nb oxides should not exhibit paramagnetic properties. The presence of Ti, Fe, Ta and Nb in the magnetic fractions shown in Figure 4.21 may be explained as a result of the association of Ti, Nb and Fe in the form of ilmenorutile

2+ 3+ (Ti,Nb,Fe )3O6 as well as struverite (Ti,Ta,Fe )3O6 (Nete, 2013). The close association of Ti,

Ta, Nb and Fe in these two forms can also be considered as the basis of Ti, Ta and Nb ending up in the magnetic portion during the magnetic separation. Another explanation for the presence of

Ti in magnetic portion may be the presence of ilmenite FeTiO3 in the tantalite deposit. For instance, ilmenite was identified in sample E (Table 4.8). Finally, the grain size of minerals would still be too coarse to release Ti, Ta and Nb from the paramagnetic material and would subsequently be separated with the magnetite material (Nete et al. 2014a) as shown in Figure

4.21a-b.

A high content of uranium and thorium was observed in non-magnetic materials as indicated in Figures 4.22a and b, respectively, for all samples. For instance, non-magnetic and magnetic material of sample FH contain 5717 and 154 g/t U, respectively (Figure 4.22a). In sample FM, the uranium content in non-magnetic portion is 1944 g/t compared to 172 g/t in magnetic portion. Referring to Figure 4.21a, Ta2O5 is highly concentrated in the non-magnetic portions. These results prove the association of Ta, Nb and U in the same minerals as was also proved by SEM-EDS (Figure 4.14). The same trend of high content of Th as U in non-magnetic was also observed in Figure 4.22b indicating little magnetism of minerals bearing Th.

101

Chapter 4: Results and Discussions

a) b) 7000 400 Feed M NM 367 335 Feed M NM 6000 5717 350 5333 300

5000 (g/t)

(g/t) 250 4000 200

3000 Thorium

Uranium Uranium 150 1836 1944 2000 100 56 54 1000 50 20 154 172 60 50 50 11 5 9 9 0 0 FH FM FL FH FM FL Samples Samples

Figure 4. 22. Composition of samples FH, FM and FL before and after magnetic separation (a) uranium, (b) thorium

In the present study, concentrates of black, golden, yellowish and brownish grain colours were strongly attracted by the magnetic field, whereas the white and colourless grains were greatly diamagnetic indicating that the magnetic susceptibility of concentrates can be influenced by the mineralogical composition. The white grains were predominantly made up of quartz mineral. Therefore, it is important to separate grains based on colours in the future studies to determine the best range of magnetic field intensity for effective separation. In a related study by Raslan and Fawzy (2017) on Nb-Ta oxide minerals from a rare-metal pegmatite, collected in

Eastern Desert and Sinai, Egypt reported that the magnetic susceptibility of the Nb-Ta oxide minerals was influenced by the chemistry of these minerals besides grain shape and intensity of grain colour. The columbite-tantalite placer ore in Nigeria is treated through magnetic separation to remove magnetite, ilmenite, columbite, monazite and magnetic cassiterite. The columbite and magnetic cassiterite are separated using flotation, whereas columbite and monazite are separated on the basis of electrostatic properties (Miller, 1995). The present study gives a general understanding on concentrating and upgrading process of the main economic elements and

102

Chapter 4: Results and Discussions

strategic Ta-Nb oxides by physical methods, which could be greatly advantageous in reducing the leaching costs.

4.3.3. Summary of physical separation and concentration

Figures 4.19a and b show high content of Ta2O5 and Nb2O5 in coarse (–2800+1000 μm) fractions compared to that in fine (–1000 μm) fractions. The fine fractions host minerals containing high grade of U and Th compared to coarse fractions as shown in Figures 4.20a and b, respectively. Thus, the coarse grains of low U and Th content can be exported without chemical treatement. Unfortunately, chemical treatment was not performed on the present samples to evaluate its effectiveness on the removal of U and Th compared to samples which were not subjected to magnetic separation. The results obtained in this study show that the magnetic separation is successful for those samples with high content of Fe and Ti as shown in

Figures 4.19c and d. The selective removal of the magnetic material indicates the upgrading of the two main oxides Ta2O5 and Nb2O5 in sample FH compared to samples FM and FL as shown in Figures 4.20a and b, respectively. The chemical analysis shows that the magnetic separation performance is a result of the high iron and titanium content within the concentrates. Sample FH, with higher Fe and Ti concentration gave a non-magnetic fraction of 50 wt % Ta2O5 (Figure

4.21a). It was also found that the minerals of high U and Th content were retained in the non- magnetic portions (Figures 4.22a-b), indicating that the mineral particle hosting U are locked into tantalum mineral particles. The grains of tantalum minerals are of small particle size. Thus, it is worthwhile to consider the magnetic separation and concentration instead of the washing technique which will reduce the loss of tantalum in fine fractions during the gravity concentration using water.

103

Chapter 4: Results and Discussions

4.4. Leaching to remove uranium and thorium from tantalum and niobium concentrate

4.4.1. Summary of leaching results and elemental mass balance

First and foremost, the elemental leaching efficiencies (LE) of targeted metals were determined according to the mathematical relationships as shown in equations 1 and 2, to check the reliability of the leaching results and analytical method. For this purpose, some leaching experiments were repeated under identical conditions. Table 4.13 shows the comparison between elemental leaching efficiencies in Test 25 using equation 1 without considering the wash liquor and the repeated Test 25 R1 based on equation 2 in which wash liquor was taken into account,

m LE = 2 × 100% (1) m1

m + m LE = 2 3 × 100% (2) m1

where, m2 and m3 are the masses of dissolved elements in leach and wash solutions, respectively,

while m1 is the total mass of the given element in the solid feed. Referring to Table 4.13, the leaching efficiencies obtained using the two equations are quite similar. However, for further leaching experiments, washing liquors were taken into consideration. Under the same conditions described in Table 4.13, Test 25 was repeated several times to check the reliability of the leaching efficiencies and the results are shown in Table 4.14. Referring to Tables 4.13 and 4.14, it is clear that the leaching efficiencies for some elements including U, Th, Ta and Nb are quite similar, while Fe, Mn and Al shows a disagreement in their leaching efficiencies. This may due to the inhomogeneity of the concentrate samples used in T25 R4.

104

Chapter 4: Results and Discussions

Table 4. 13. Comparison of elemental leaching efficiencies

T.N. S.I. Mass of Leach Wash Element Grade in Category Solution Leaching feed liquor liquor solid feed of liquors concentrations efficiencies (%) (g) (mL) (mL) (wt %) (ppb) T25 T25 R1 25 C 2.5 200 100 U 0.800 Leach 94554 95 97 Wash 4823 Th 0.029 Leach 1203 33 35 Wash 109 Ta 37.00 Leach 186690 4 4 Wash 9711 Nb 9.00 Leach 121631 11 11 Wash 5210 Fe 3.560 Leach 138599 31 32 Wash 10476 Mn 4.000 Leach 164029 34 34 Wash 12234 Zr 0.300 Leach 16769 45 46 Wash 645 Ti 0.590 Leach 9576 13 15 Wash 1380 Li 0.118 Leach 9514 64 67 Wash 795 Cs 0.030 Leach 2878 77 78 Wash 129 Al 1.770 Leach 136685 62 64 Wash 9639

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, time: 2 h, o temperature: 400 C and acid concentration: 18 M H2SO4 for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 1110 mV - 1140 mV, time: 3 h, solid to liquid volume ratio: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

Table 4. 14. Repeat test results of elemental leaching efficiencies

S.I. Mass of Leach liquor Wash liquor Elements Grade in the Leaching efficiencies (%) the feed volume volume solid (wt %) T25 R2 T25 R3 T25 R4 (g) (mL) (mL) C 2.5 200 100 U 0.080 93 99 100 Th 0.029 33 38 41 Ta 37.00 4 4 7 Nb 9.000 11 10 10 Fe 3.560 31 34 43 Mn 4.000 33 30 35 Zr 0.300 11 11 27 Ti 0.590 14 12 12 Li 0.110 69 61 62 Cs 0.030 77 69 85 Al 1.770 64 68 78

T25 R: Test 25 at various repeats

105

Chapter 4: Results and Discussions

As shown in Table 4.15, low content of 0.05 wt % U and 0.016 wt% Th were observed in the leach residue indicating almost complete removal of these radioactive elements along with low dissolution of Ta and Nb. The leaching efficiencies in Table 4.15 show the reliability of elemental mass balance of U, Th, Ta and Nb from the tantalite concentrate. Relatively low Ta and Nb dissolution observed (< 10 wt %) in this study is in agreement with other studies conducted on similar feed material reported by previous researchers (Cheru et al. 2018; Nete et al. 2014a; El-Hussaini and Mahdy, 2002).

Tables 4.16 - 4.19 present a summary of leaching efficiencies of U, Th, Ta and Nb and other minor elements Fe, Mn, Al, Li and Cs obtained under various conditions of direct leaching or baking followed by leaching process. The leach solutions were collected for quantitative analysis after the first 3 h of leaching process. In Test 24, samples were collected at different time intervals (0.5 – 4 h) to determine the leaching efficiencies as a function of time. Results also show that significant amount of Li leaching as well as complete dissolution of Cs under the same conditions was achieved. Further discussions of results are presented in the next few sections.

106

Chapter 4: Results and Discussions

Table 4. 15. Elemental mass balance for bake – leach results of sample C, T 25

Element Elemental composition in solid (wt %) Mass (g) Elemental composition Elemental composition in Elemental leaching efficiencies (%) in residue (g/L) solution (g/L)

F BP LR F BP LR LR LL WL BP to LR F to LR LL and WL to LR Ta 37.00 23.00 37.00 0.9250 0.8740 0.99900 4.99500 0.3940 0.0130 14 8 8 Nb 9.000 7.000 8.000 0.2250 0.2660 0.29700 1.48500 0.0900 0.0080 19 4 6 U 0.800 0.560 0.050 0.0200 0.0200 0.00137 0.00685 0.1195 0.0075 94 93 95 Th 0.029 0.020 0.016 0.0007 0.0008 0.000432 0.00216 n.d n.d 44 40 - Fe 3.560 2.400 0.600 0.0890 0.0912 0.01620 0.08100 0.4330 0.0520 82 82 86 Mn 4.000 2.500 0.500 0.1000 0.0950 0.01350 0.06750 0.4810 0.0280 86 87 88 Zr 0.300 0.060 0.003 0.0075 0.0023 0.000081 0.000405 0.0150 0.0010 96 99 98 Li 0.110 0.080 0.060 0.0027 0.0030 0.00162 0.00810 0.0100 n.d 47 41 - Cs 0.030 0.020 0.010 0.0007 0.0080 0.00027 0.00135 0.0030 n.d 64 64 - Al 1.770 1.150 0.575 0.0443 0.0437 0.01550 0.07750 0.1580 0.010 64 65 68

o Solid to acid ratio: 625 g/L, 18 M H2SO4, temperature: 400 C and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 1110 mV- 1140 mV, time: 3 h, S/L: 12.5 g/L, acid concentration: 4 M H2SO4 and K2S2O8 to liquid ratio: 27 g/L for leaching. Mass of feed = 2.5g, mass of bake product = 3.8 g and mass of leach residue = 2.7 g.

F: feed, BP: bake product, LR: leach residue, LL: leach liquor and WL: wash liquor

107

Chapter 4: Results and Discussions

Table 4. 16. Leaching efficiencies under direct leaching with base, acid and oxidants

T.N. S. I. Acid Conc Reagent Mass Temp Time S/L PSD Agitation speed Elemental leaching efficiencies (%) (M) (g) (oC) (h) (g/L) (μm) (rpm) U Th Ta Nb T 1 C - - KOH 1 90 6 10 < 75 500 n.d n.d n.d n.d

T 2 E, H, G & B H2SO4 10 FeSO4 2 90 6 10 < 75 500 n.d n.d n.d n.d MnO2 1

T 3 H, G & C H2SO4 18 - - 100 3 50 < 45 250 n.d n.d n.d n.d T 4 C H2SO4 18 - - 50 3 10 < 75 n.d n.d n.d n.d T 5 C H2SO4 10 Fe2(SO4)3 4.0 50 24 50 < 75 250 n.d n.d n.d n.d

D MnO2 1.5 5 50 n.d n.d E Fe2(SO4)3 2.0 n.d n.d n.d n.d F n.d n.d n.d n.d T 6 C H2SO4 10 Fe2(SO4)3 20 100 24 10 < 45 250 n.d n.d n.d n.d

T 7 C H2SO4 2 Fe2(SO4)3 20 100 24 10 < 45 250 n.d n.d n.d n.d C H2SO4 10.8 - - 100 24 150 250 n.d n.d n.d n.d HNO3 5.30 C H2SO4 10.8 - - 25 24 150 250 n.d n.d n.d n.d HNO3 5.30 T 8 C H2SO4 18 9.7 M H2O2 - 100 12 10 < 75 250 8 78 2 8

T.N: Test number, S.I.: Sample identification, 3 h of leaching process, n.d.: not detected

Table 4. 17. Leaching efficiencies under direct leaching with acid and fluoride

T.N. S.I. Acid Conc (M) Reagent Solid : CaF2 Temp S/L (g/L) PSD (μm) Agitation speed (rpm) Time (h) Elemental leaching efficiencies (%) (oC) U Th Ta Nb Mn Si T 24 C H2SO4 4 CaF2 1 : 2 90 37.5 < 75 250 0.5 76 70 48 26 15 81 1 96 90 62 35 19 90 2 97 90 66 37 20 88 3 106 87 70 42 22 82 4 102 69 81 43 24 86

T.N: Test number, S.I.: Sample identification, 4 h of leaching process

108

Chapter 4: Results and Discussions

Table 4. 18. Leaching efficiencies after acid or base baking and leaching with acid and oxidants (part I)

T.N. S.I. Baking Leaching Elemental leaching efficiencies (%) S/A (g/L) PSD Time Temp R/L Agitation Conc S/L Temp U Th Ta Nb Fe Mn Al Li Cs or S/B (g/g) (μm) (h) (OC) (g/L) speed (rpm) (M) (g/L) (oC) T 9 C 1 : 5* < 75 2 400 27 250 Water 12.5 40 n.d n.d 4.5 29 - - - - - T 10 C 625 < 75 2 600 27 250 4 12.5 40 n.d n.d n.d n.d - - - - - T 11 C 1250 < 75 2 400 27 250 Water 12.5 40 25 n.d n.d n.d 12 1 10 2 3 T 12 C 625 < 75 2 400 27 250 4 12.5 40 31 38 9 9 38 79 38 47 73 T 13 B 625 < 75 2 400 27 250 4 12.5 40 54 26 1 5 16 20 24 30 43 D 100 59 11 13 58 43 49 - 44 F 100 38 5 6 21 61 19 22 23 G 68 38 3 6 60 33 32 14 30 H 100 38 5 6 29 48 17 13 11 T 14 A 625 < 75 2 400 27 250 4 12.5 90 61 46 9 34 91 57 45 53 45 B 100 56 3 6 63 32 30 31 23 C 87 28 6 10 91 96 50 59 60 D 67 36 36 36 87 100 36 61 56 E 100 67 9 11 77 57 24 49 100 F 100 35 2 7 80 66 25 27 38 G 100 56 4 13 73 49 40 49 44 H 100 48 4 7 85 82 26 23 21 T 15 C 625 < 75 0.5 400 27 250 4 12.5 90 90 42 4 8 52 78 40 44 56 1.0 72 33 4 6 45 73 37 27 35 1.5 85 40 4 5 49 85 50 28 38 2.0 87 28 6 10 91 96 40 59 60 T 16 C 625 < 75 2 400 27 250 4 12.5 90 87 28 6 10 91 96 40 59 60 + 75 – 106 98 43 6 0.2 99 95 45 82 81 -106 + 150 95 36 4 8 94 80 63 72 71 – 150 + 250 61 24 3 5 68 54 73 53 50 T 17 C 2500 < 75 2 400 27 250 4 12.5 90 85 40 0.4 0.7 33 69 44 35 41 1250 100 42 1 2 40 85 43 45 44 833 94 47 2 3 40 76 46 37 41 625 87 28 6 10 91 96 40 59 60

T.N: Test number, S.I.: Sample identification, *S/B (B = KOH), R = reagent (K2S2O8), 3 h of leaching process

109

Chapter 4: Results and Discussions

Table 4. 19. Leaching efficiencies after acid baking and leaching with acids and oxidants (part II)

T.N. S.I. Baking Leaching Elemental leaching efficiencies (%) S/A PSD Time Temp R/L Agitation Conc S/L Temp U Th Ta Nb Fe Mn Al Li Cs (g/L) (μm) (h) (OC) (g/L) speed (rpm) (M) (g/L) (oC) T 18 C 625 < 75 2 400 27 250 4 12.5 20 58 22 9 8 52 77 37 26 30 60 79 24 9 16 74 92 42 60 56 80 85 29 6 8 85 93 41 68 62 90 87 28 6 10 91 96 40 59 60 T 19 C 625 < 75 2 400 27 250 1 12.5 90 85 40 0.4 0.7 61 34 44 34 38 2 100 42 1 2 82 66 44 29 35 3 94 47 2 3 75 58 54 36 40 4 87 28 6 10 91 96 40 59 60 T 20 C 625 < 75 2 400 27 250 4 37.5 90 100 31 4 4 79 86 39 56 56 25.0 84 31 4 4 79 86 39 56 56 12.5 87 28 6 10 91 96 40 59 60 T 21 C 625 < 75 2 400 40.5 250 4 12.5 90 79 27 3 9 93 88 60 48 52 27.0 87 28 6 10 91 96 40 59 60 13.5 75 30 10 10 83 83 50 57 59 T 22 C 625 < 75 2 400 27 0 4 12.5 90 78 30 6 8 100 95 58 51 60 250 87 28 6 10 91 96 40 59 60 T 23 C 625 < 75 2 100 27 250 4 12.5 90 2 1 n.d 1 2 90 40 43 100 200 20 11 1 12 47 64 37 30 31 300 100 62 6 8 64 87 50 55 69 400 87 28 6 10 91 96 40 59 60

T.N: Test number, S.I.: Sample identification, R = reagent (K2S2O8), 3 h of leaching process

110

Chapter 4: Results and Discussions

4.4.2. Direct leaching with H2SO4

Various direct leaching tests were performed to investigate the dissolution efficiencies of

U and Th as detailed in Chapter 3 (Test 1 – Test 7). The influence of the leaching conditions on

Ta and Nb dissolution as the main elements was also taken into consideration. In general, the analytical results of the solutions under direct leaching of sample D have shown an extremely low dissolution efficiency of 5% U, except for Th, which reached 50% after 24 h leaching in 10

o M H2SO4 at 50 C, with 4 g of Fe2(SO4)3 and 2 g of MnO2 (Test 5, Table 4.16). The main chemical reactions of uranium in the presence of ferric sulfate and hydrogen peroxide are shown in chemical equations 1 – 5 (Sole et al., 2011). However, Ta, Nb and U leaching efficiencies were negligible in Test 5. The thermodynamic feasibility of chemical reactions for uranium and thorium leaching is shown in Equation 6 and 7 in Table 4.20.

3+ 2+ 2+ UO2 + 2Fe = UO2 + 2Fe (1) 2+ + 3+ 2+ 2Fe + MnO2 + 4H = 2Fe + Mn + 2H2O (2)

+ 2+ UO2 + H2O2 + 2H = UO2 + 2H2O (3)

UO2 + 2H2SO4 = U(SO4)2 + 2H2O (4)

ThO2 + 2H2SO4 = Th(SO4)2 + 2H2O (5)

Previous studies show that at ordinary temperature, thorium can form the readily soluble sulfate salt which in turn forms the octahydrate [Th(SO4)2.8H2O] in solution of high concentration of H2SO4 (Mathur and Tandon, 1986). It was also reported by Bielecki et al. (1991) that at temperatures in the range of 40 to 100 oC, most of the thorium was dissolved by leaching the ore in solutions of 0.5 to 5.0 M sulfuric acid. Low dissolution of 1% Ta and Nb each was observed in most cases. Long leaching time as well as low extraction efficiencies of U and Th has led us to investigate other possibilities to dissolve both U and Th. Equations (3) – (7) gives

111

Chapter 4: Results and Discussions

the chemical reactions of UO2 or ThO2 along with ΔGo values presented in Table 4.20. A mixture of H2O2 and H2SO4 significantly increased the dissolution of Th to 78%, while U leaching efficiency was low at 8% even after 12 h, as shown in Table 4.21 indicating the decomposition of H2O2 to H2O and O2 at high temperature. The relatively low extraction of U (8%) can also be related to its occurrence in refractory fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6(F,OH) mineral. The low dissolution of 2% Ta and 8% Nb may be due to formation of protective coating of insoluble tantalum and niobium oxides on the mineral surface during the leaching process (El-

Hussaini and Mahdy, 2002). In the present study, no minerals were identified as host minerals of thorium. This may due to its low content (5 – 700 g/t Th) in the concentrate as can be seen in

Table 4.6. Thus, it was difficult to specifically determine which mineral the Th has dissolved from. However, Kim and Osseo-Assare (2012) reported that Th can dissolve as Th(SO4)2 in concentrated H2SO4 (at pH < 4) as shown in the Eh-pH diagram, Figure 4.23 and Table 4.20 (Eq.

7).

Table 4. 20. Chemical reactions of UO2 and ThO2 with H2SO4

Chemical reactions Temp (oC) ΔGo (kJmol-1) Reference Eq.

a UO2 + 2H2SO4 = U(SO4)2 + 2H2O 100 -141.60 This work 6 ThO2 + 2H2SO4 = Th(SO4)2 + 2H2O - 229.80 7 a. H2O2 was not included in HSC software package due to the decomposition to H2O and O2 at 100 oC.

Table 4. 21. Solution concentrations and leaching efficiencies

T.N. S.I. Acids and H2SO4/ PSD Temp Leachi S/L Agitation Elements (ppb) O 8 C oxidant H2O2 (μm) ( C) ng time (g/L) speed U Th Ta Nb ratio (h) (rpm) 18 M H2SO4 1:1 < 75 100 12 10 250 6734 2251 75600 74623 9.8 M H2O2 Leaching efficiencies (%) 8 78 2 8

T.N: Test number, S.I: Sample identification, Volume of liquid: 200 ml, Mass of solid: 2 g, U= 0.8 wt %, Th = 0.029 wt %, Ta = 37 wt % and Nb = 9 wt %. The elemental assays used in Table 4.21 are the same as the rest of the bake and leach tests.

112

Chapter 4: Results and Discussions

Bock (1979b) also reported the low solubility of U, Th, Ta and Nb in a mixture of 10.8 M H2SO4 and 5.0 M HNO3. However, the increase of the HNO3 concentration in the mixture to 10.8 M

o H2SO4 and 12.7 M HNO3 at 200 C resulted in 70% Th dissolution efficiency. Therefore, referring to Figure 4.23, the ThO2 may have dissolved as Th(SO4)2(a) in strongly acidic solutions.

−3 Figure 4. 23. Eh–pH diagram for Th–SO4–H2O system at 25 °C, {Th}=10 M, {SO4} = 1.0 M (Kim and Osseo-Assare, 2012)

Figure 4.24 compares the solubility data of UO2 and ThO2 collected from literature review (Chapter 2) and the experimental values of the concentrations from this work repeated in

Test 8, Table 4.21. It is clear that the solubility of UO2 and ThO2 decreases with the increase in pH. The concentration of U in solutions of pH below 4 is in the range of 1×10-6 mol/L and 1×10-

4 mol/L, whereas in solutions of pH above 4, the concentration is below 1×10-7 mol/L. The

-4.5 concentration of UO2 in the current study in Test 8 is 1×10 mol/L. It is within the range of the literature data but slightly lower at similar pH (Figure 4.24a). In comparison to uranium, the solubility of thorium appears to be less dependent on pH variation (Figure 4.24b). At the pH of

-6 1.5 in this study, the concentration of ThO2 was about 1×10 mol/L representing an extraction

113

Chapter 4: Results and Discussions

efficiency of 78% (Table 4.21) indicating the refractoriness or the non-amenability of Th to the leaching agent.

a) b) 1E+00 1E+00

1E-01 UO2 UO2* 1E-01 ThO2 ThO2* 1E-02 1E-02 1E-03 1E-03

1E-04 1E-04

] (mol/L) ]

2 2

1E-05 1E-05 [UO 1E-06 [ThO 1E-06 1E-07 1E-07 1E-08 1E-08 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 pH pH

-3 -1 Figure 4. 24a-b. Total uranium in solution for the dissolution of UO2 in 8.10 mol L sodium perchlorate at 25 oC (Casas et al., 1998) and Th in acid solution at I = 0.5 M (NaCl/NaOH) and o o 22 C (Neck et al., 2003). UO2 and ThO2 literature data, *. 400 and 40 C bake and leach temperatures, respectively, in 4 M H2SO4 in this work

o o 4.4.3. Leaching with K2S2O8 at 40 C after roasting at 600 C

Preliminary experiments were performed to investigate the most effective solid mass to acid volume ratio for roasting which lead to a reasonable leaching efficiency of the elements from the concentrate. The tantalite concentrate was roasted at 600 oC for 2 h without adding

o H2SO4 followed by leaching the roast product in 4 M H2SO4 at 40 C. As can be seen from Table

4.22, the elements U, Th, Ta and Nb remained in their solid phases unleached indicating the necessity of the use of the hot acid. Extremely low extraction of minor elements (1-12%) was also observed as shown in Table 4.23.

114

Chapter 4: Results and Discussions

Table 4. 22. Leaching efficiency after roasting and leaching with acid

T.N. S.I. Roasting Leaching Elements leached (%) Additive Conc Temp Temp Time U Th Ta Nb (M) (oC) (OC) (h) 10 C - - 600 40 0.5 Nd Nd Nd Nd 1 2 3

T.N.: Test number, S.I.: Sample identification. Nd: Not detected. Solid to acid volume ratio: 625 g/L and time: 2 h for roasting. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~850 - 1120 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching

Table 4. 23. Leaching efficiency of minor elements after roasting and leaching with acid

Roasting Leaching Elements leached (%) T.N. S.I. Conc Temp Temp Time (h) Fe Mn Al Li Cs (M) (oC) (oC) T 10 C - 600 40 0.5 5 1 10 2 4 1 7 1 8 4 3 2 10 1 12 2 3 3 12 1 10 2 3

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for roasting. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~850 - 1120 mV, K2S2O8 to liquid volume ratio: 27 g/L, time: 3 h, S/L: 12.5 g/L and 4 M H2SO4 for leaching.

o 4.4.4. Leaching with K2S2O8 at 40 C after alkaline bake

The effect of KOH bake and water leach on dissolution efficiencies of U and Th are presented in Table 4.24. The results indicate that using the KOH bake had a small influence on

Ta compared to that on Nb dissolution, while U and Th remained insoluble. Again, the low dissolution of U is due to the refractory nature of fluorocalciomicrolite. Gebreyohannes et al.

(2017) reported a complete removal of uranium and thorium from a tantalite concentrate through a washing technique using 1% (w/v) KOH and 1 M H2SO4. However, the presence of mineral phases carrying U and Th were not reported in the tantalite concentrate sourced from

Kenticha/Ethiopia used in their work. In the present study, an overall dissolution of 29% Nb was

115

Chapter 4: Results and Discussions

observed over the 3 h leaching process along with 4% Ta (Test 9, Table 4.24). According to

Wang et al. (2017), the KOH bake technique is effective on low grade columbite-tantalite concentrate with Ta2O5 < 10 wt % compared to high grade concentrate of 45 wt % Ta2O5 used in the present study which may be one of the possible reasons for low dissolution of Ta, warranting further studies. The relatively high leaching efficiency of Nb in Table 4.24 can be attributed to the different behaviour of Ta2O5 and Nb2O5 towards the reactions with KOH.

Equations 8-9 present the chemical reactions of Ta2O5 and Nb2O5 with KOH. Wang et al. (2009) and Berhe et al. (2017) reported that the KOH fusion of columbite-tantalite concentrate followed by water leaching resulted in a higher dissolution of Nb due to conversion of Nb2O5 to the soluble

K3NbO4. On the other hand, the reaction between Ta2O5 and KOH resulted in the formation of insoluble KTaO3 as shown in chemical equation 9. Referring to Test 9, the leaching efficiencies of Fe and Mn were extremely low with less than 1% each during the entire leaching process.

Table A 7.1b in Appendix A 7 presents the Ta and Nb concentrations in leach solution.

(Fe,Mn)O(Ta,Nb)2O5 + 8KOH + (n-4)H2O = K8[(Ta,Nb)6O.19nH2O] + 3(Fe,Mn)O (8)

K8[(Ta,Nb)6O19.nH2O] = 6K(Ta,Nb)O3 + 2KOH + (n+1)H2O (9)

Table 4. 24. Leaching efficiency after baking with KOH and leaching with water

T.N. S.I. Baking Leaching Elemental concentration in Elements leached (%) liquor (ppb) Additive Mass Time Temp Time Ta Nb Ta Nb U Th (g) (h) (oC) (h) 9 C KOH 10 2 40 0.5 200968 290974 4.0 25 Nd Nd 1 199464 307741 4.0 25 2 210810 328426 4.0 27 3 209123 330782 4.5 29

T.N.: Test number 9, S.I.: Sample identification, Nd: Not detected. Solid to acid volume ratio: 12.5 g/L and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, Solid to liquid volume ratio: 12.5 g/L and K2S2O8 to liquid volume ratio: 27 g/L for leaching

116

Chapter 4: Results and Discussions

Figure 4.25 is a plot of Ta(V) and Nb(V) solubility data from literature (Deblonde et al.,

2015; Babko et al., 1963) in comparison to that obtained in the current study. It is clear from the plotted literature data that for the pH values in the range 9 – 14, the solubilities of Nb(V) and

Ta(V) are influenced by variation in pH. The concentrations of Ta (~ 1×10-3 mol/L) obtained in our study fell within the range reported by Deblonde et al. (2015) as shown in Figure 4.25a. The lower solubility of Nb(V) in the current study compared to that reported by Deblonde et al.

(2015) is probably due to the differences in chemical composition of the feed materials (Figure

4.25b). For instance, Deblonde et al. (2015) used 99.9 wt % Nb2O5 in their studies compared to tantalite concentrate containing 13 wt % Nb2O5 used in the present study. The presence of other elements in the tantalite concentrate may have affected the solubility of Nb(V). For Ta(V) solubility, however, the value obtained in the present study at pH ~ 13 is comparable to that in literature (Deblonde et al., 2015), as shown in Figure 4.25a.

a) b)

1E+00 1E+00

Ta Deblonde et al. 2015 Nb Deblonde et al. 2015 1E-01 1E-01 Ta This work Nb This work Ta Babko et al. 1963 Nb Babko et al. 1963 1E-02 1E-02

1E-03

1E-03 mol/L)

] ( [Nb [Ta](mol/L) 1E-04 1E-04

1E-05 1E-05

1E-06 1E-06 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH pH

o Figure 4. 25(a)-(b). Solubility data for Nb2O5.nH2O(s) and Ta2O5.nH2O(s) at 25 C and I = 0.12 -1 mol L (NaCl/NaCO3/NaOH) (Deblonde et al., 2015); Solubility data of Nb2O5.nH2O(s) and o -1 Ta2O5.nH2O(s), temp = 19 C, I = 1 mol L (KNO3) (Bakbo et al., 1963); Ta(V) and Nb(V) at o 40 C in 4 M H2SO4 (This work)

117

Chapter 4: Results and Discussions

o o 4.4.5. Leaching with K2S2O8 at 40 C after baking with H2SO4 at 400 C

The concentrate was baked with sulfuric acid at a solid mass to acid volume ratio of 1250 g/L for 2 h and was subjected to leaching process at 40 oC for 3 h. The chemical reactions expected in the dissolution of U, Th, Ta and Nb are shown in Tables 4.20 and 4.25. Uranium, thorium, tantalum and niobium in H2SO4 are converted into soluble sulfates during the acid baking as presented in chemical equations (10-13) where the role of K2S2O8 is to oxidise UO2

(Eq. 10b). The measured concentrations of the dissolved metal ions correspond to slight variations of leaching efficiencies as the time progressed. A highest leaching efficiency of 31%

U is evident in Table 4.26. However, the dissolution of Th was low and not detectable due to the high solid mass to acid volume ratio and the lack of sufficient acid to dissolve the Th content.

A low leaching efficiency of 9% of Ta or Nb was observed as shown in Table 4.26 compared to higher leaching efficiencies (18 – 72%) of minor elements presented in Table 4.27. Appendix A

7, Table A 7.3 presents the solution concentrations of the dissolved elements.

Table 4. 25. Chemical reactions of UO2, ThO2, Ta2O5 and Nb2O5 with H2SO4

Chemical reactions Temp (oC) ΔGo (kJ mol-1) References Eq. a UO2 + 2H2SO4 = U(SO4)2 + 2H2O 90 -95.82 This work 10a UO2 + K2S2O8 = K2UO2(SO4)2 - Proposed 10b a ThO2 + 2H2SO4 = Th(SO4)2 + 2H2O -226.35 This work 11 * Nb2O5 + 2H2SO4 = Nb2O3(SO4)2 + 2H2O - Jerome et al. 12 * Ta2O5 + 2H2SO4 = Ta2O3(SO4)2 + 2H2O (2006) 13

* Note : Nb2O3(SO4)2 and Ta2O3(SO4)2 are salts of niobic oxide (Nb2O5.nH2O) and tantalic oxide (Ta2O5.nH2O) which are insoluble oxides (Deblonde et al., 2019). a. based on HSC data base.

118

Chapter 4: Results and Discussions

Table 4. 26. Leaching efficiency after acid baking and leaching with water

T.N. S.I. Baking Leaching Elements leached (%) Additive Conc Time Temp Time U Th Ta Nb (M) (h) (oC) (h) 11 C H2SO4 18 2 40 0.5 22 Nd 7 6 1 24 7 7 2 31 9 9 3 25 7 8

T.N.: Test number, S.I.: Sample identification, Nd: not detected. Solid to acid volume ratio: 1250 g/L, temperature: 400 oC and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching

Table 4. 27. Leaching efficiency of minor elements after acid baking and leaching with water

Baking Leaching Elements leached (%) T.N. S.I. Conc Temp Temp Time (h) Fe Mn Al Li Cs (M) (oC) (oC) T 12 C 18 400 40 0.5 32 52 18 Nd Nd 1 36 56 25 2 50 72 28 3 43 60 23

The variation of the solid mass to acid volume ratio from 1250 g/L to 625 g/L during baking resulted in an improvement of Th leachability from different samples as indicated in

Table 4.28. The results obtained indicated a leaching efficiency of about 32% U and 41% Th after the first hour of leaching process of sample C. The sulfuric acid baking followed by leaching process did not improve the leaching efficiency of Ta and Nb due to precipitation. The same trend of precipitation of tantalum and niobium sulfate during the leaching was observed by El-

Hazek et al. (2012). The extraction of other elements was relatively low but higher than that of

Ta and Nb, as shown in Tables 4.28 - 4.29. The solid mass to acid volume ratio of 625 g/L for

119

Chapter 4: Results and Discussions

baking was kept constant for further leaching tests. The elemental (Ta, Nb, U and Th) concentrations in leach liquors are presented in Appendix A 7, Tables A 7.4 – A 7.5.

Table 4. 28. Leaching efficiency after acid baking and leaching with acid

T.N. S.I. Baking Leaching Elements leached (%) Additive Conc Temp Temp Time U Th Ta Nb (M) (oC) (oC) (h) T 12 B H2SO4 18 400 40 0.5 33 20 1 3 1 44 26 1 4 2 53 27 2 5 3 54 26 1 5 C 0.5 30 40 8 8 1 32 41 9 8 2 32 41 8 8 3 31 38 9 9 D 0.5 100 76 13 15 1 100 66 12 14 2 100 59 12 14 3 100 59 11 13 F 0.5 44 25 2 5 1 40 24 2 6 2 43 24 2 6 3 45 23 2 6 H 0.5 100 41 5 6 1 100 40 5 6 2 100 37 5 6 3 100 38 5 6 G 0.5 58 55 3 5 1 64 48 3 6 2 63 46 3 6 3 68 38 3 6

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~ 850 - 1120 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching

120

Chapter 4: Results and Discussions

Table 4. 29. Leaching efficiencies of minor elements after acid baking and leaching with acid

Baking Leaching Elements leached (%) T.N. S.I. Conc Temp Temp Time (h) Fe Mn Al Li Cs (M) (oC) (oC) T 12 B 18 400 40 0.5 9 16 14 10 9 1 13 17 17 16 11 2 16 19 24 27 17 3 16 20 24 30 19 C 0.5 61 54 30 39 37 1 70 58 35 43 40 2 79 60 40 46 42 3 79 55 36 45 43 D 0.5 57 48 50 7 61 1 56 44 48 2 57 2 65 50 54 7 65 3 58 45 49 Nd 44 F 0.5 19 59 19 17 20 1 20 58 16 17 20 2 21 61 17 25 26 3 23 61 19 22 23 H 0.5 19 45 17 8 7 1 24 45 13 7 7 2 25 45 15 13 11 3 29 48 17 13 11 G 0.5 59 34 30 11 27 1 68 40 33 18 36 2 75 43 41 21 38 3 60 33 32 14 30

T.N.: Test number, S.I.: Sample identification. Nd: Not detected. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~ 850 - 1120 mV, K2S2O8 to liquid: 27 g/L, time: 3 h, S/L: 12.5 g/L and 4 M H2SO4 for leaching.

o o 4.4.6. Leaching with K2S2O8 at 90 C after baking with H2SO4 at 400 C

Even though, relatively low amounts of U and Th were leached at 40 oC, a complete removal of U was not achieved from some concentrates (Table 4.28). Therefore, the leaching temperature was increased to 90 oC to investigate the elemental dissolution performance. The increase in temperature from 40 to 90 oC while keeping other conditions unchanged was found to be effective in increasing the leaching efficiency of U (50 – 100%) and Th (30-70%) in all samples except in samples A and D as shown in Table 4.30. The solution concentrations of the

121

Chapter 4: Results and Discussions

leached elements are presented in Appendix A 7, Table A 7.7. The leaching efficiencies listed in

Table 4.30 show that the primary refractory minerals carrying U including fluorocalciomicrolite

(Ca,Na,U)2(Ta,Nb)2O6(OH,F), parabariomicrolite (Ba,U)(Ta,Nb)2(O,OH)7 and pyrochlore

(Na,Ca,U)2(Nb,Ta)2O6(OH,F) across the eight concentrates were virtually decomposed during the acid baking stage, allowing high leaching efficiencies at 90 oC. The measured Eh of 0.84 –

2+ 1.12 V reported in Table 4.30 corresponds to the stability region of UO2 (aq) and UO2SO4(aq) as shown in Figures 4.26.

Figure 4. 26. Eh-pH diagram for uranium in sulfate solution at 25 °C, [U] = 4.0 × 10−3 M, [S] = 1.0 M (Nettleton et al., 2015)

122

Chapter 4: Results and Discussions

Table 4. 30. Leaching efficiencies obtained from eight different concentrates after acid baking and leaching with acid

T.N. S.I. Fe/(Nb+Ta) Leaching time Elements leached (%) molar ratio (h) U Th Ta Nb A 1.0 0.25 50 45 9 26 0.5 71 61 9 33 1 51 40 9 28 2 61 43 9 33 3 61 46 9 34 B 0.1 0.25 87 39 3 10 0.5 95 45 3 12 1 98 49 3 9 2 100 49 3 6 3 100 56 3 6 C 0.2 0.25 69 32 5 8 0.5 72 29 5 8 1 80 32 5 8 14 2 84 31 5 9 3 87 28 6 10 D 1.0 0.25 57 32 33 43 0.5 62 31 34 44 1 62 34 35 35 2 64 36 37 37 3 67 36 36 36 E 2.0 0.25 100 65 9 10 0.5 100 71 9 10 1 100 62 9 11 2 100 61 9 11 3 100 67 9 11 F 0.3 0.25 86 27 2 6 0.5 82 27 2 6 1 100 31 2 6 2 100 31 2 6 3 100 35 2 7 G 0.1 0.25 89 48 3 13 0.5 98 49 3 13 1 100 63 4 15 2 100 55 3 12 3 100 56 4 13 H 0.5 0.25 100 48 4 7 0.5 100 50 4 7 1 100 49 4 7 2 100 46 3 6 3 100 48 4 7

T.N.: Test number 14, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, 18 M o H2SO4, time: 2 h and temperature: 400 C for baking. Agitation speed: 250 rpm, PSD: < 75 μm, o Eh: ~ 840 - 1120 mV, time: 3 h, S/L: 12.5 g/L, temperature: 90 C and K2S2O8 to liquid volume ratio: 27 g/L for leaching

123

Chapter 4: Results and Discussions

o Figure 4.27 compares the solubility data of UO2 and ThO2 at 25 C in NaCl solutions from the literature and the measured concentrations of the same components at 90 oC in sulfate solution shown in Appendix A 7, Tables A 7.42 – A 7.43, for the baking tests conducted at 300 oC and 100 oC reflecting high and low leaching efficiencies in the current study. The concentrations of dissolved U and Th obtained in the current study (sulfate solutions) are in agreement with that from the literature review (perchlorate and chloride solutions). Higher solubilities of uranium from the bake product at 300 oC, shown in Figure 4.27a can be related to the beneficial effect of sulfate ions due to formation of UO2SO4(aq) complex species as shown in

Figure 4.26. Zhang et al. (2016) reported that the acid baking of tantalite concentrate at

o temperatures higher than 300 C resulted in the formation of insoluble ThP2O7 which may explain the low solubilities and leaching efficiencies of Th and warrants further studies.

a) b) 1E+00 1E+00

1E-01 UO2 UO2* UO2** 1E-01 ThO2 ThO2** ThO2* 1E-02 1E-02

1E-03 1E-03

1E-04 1E-04

] (mol/L) ]

2 2

1E-05 1E-05

[UO [ThO 1E-06 1E-06

1E-07 1E-07

1E-08 1E-08 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 pH pH

o o ▲ UO2 (perchlorate at 25 C) ♦ ThO2 (chloride at 25 C) * o o ■ UO2 (sulfate at high bake temperature: 300 C) ● ThO2* (sulfate at high bake temperature: 300 C) o o □ UO2** (sulfate at low bake temperature: 100 C) Ο ThO2** (sulfate at low bake temperature: 100 C)

o -3 -3 Figure 4. 27a-b. Solubility data at 25 C for UO2 in 8.10 mol dm sodium perchlorate (Casas o et al., 1998); ThO2 at I = 0.1 M and 0.5 M (HCl/NaCl) (Neck et al., 2003); * bake at 300 C, ** o o bake at 100 C and leach at 90 C in 4 M H2SO4, Test 23

124

Chapter 4: Results and Discussions

High dissolution of about 35% Ta and 10% to 44% Nb may be attributed to high content of Fe in samples A and D (Table 4.30). According to Jerome et al. (2006), sulfuric acid baking followed by leaching of columbite-tantalite with a molar ratio of Fe/(Nb+Ta) ≥ 2 will enhance the dissolution of U, Ta and Nb. However, Fe/(Nb+Ta) molar ratio in samples A, D and E is in the range 1-2 as shown in Table 4.30. Furthermore, no zircon mineral was identified in both samples A and D (Table 4.8) which may have rendered the acid to easily reach the grain hosting

Ta and Nb and cause the Ta2O5 and Nb2O5 to convert to tantalum and niobium sulfates, and high dissolution of these elements during leaching.

In fact, the leaching of the acid bake product in a solution of medium acidity range of 1

+ M ≤ [H ] ≤ 6 M, hydrolysed niobium and tantalum sulfates to niobic oxide (Nb2O5.nH2O) and tantalic oxide (Ta2O5.nH2O), respectively, which are insoluble (Deblonde et al., 2019). Figure

2.9 also indicates that Ta and Nb cations become neutral at pH between 0.5 and 1.5 due to hydrolysis resulting in low dissolution of Ta and Nb species. Furthermore, the Eh-pH diagram in Figure 2.10a shows that Nb2O5 is the stable species in acid media at pH range 0.4 to 6.2. These results warrant further systematic studies on solubility – pH relationships in practical leach solutions.

Demol et al. (2018) used the TG-DSC to monitor the weight loss of sulfuric acid during baking process. As shown in Figure 4.28, at a temperature below 100 oC, a gain of 5.7% weight relative to initial feed was observed which was attributed to an uptake of water by sulfuric acid.

The loss of weight was observed at 123 oC due to evaporation of water and at least a partial reaction of sulfuric acid. A second endothermic event was observed at 170 oC to 260 oC. The majority of the associated mass loss was attributed to the decomposition of excess sulfuric acid with a loss of about 19.3% (Demol et al., 2018). The TG-DSC analysis of 96% (w/w) H2SO4

125

Chapter 4: Results and Discussions

indicates how reaction of various oxides in tantalite with sulfuric acid may have occurred during the acid baking stage at different temperatures until about 300 oC. The evaporation or decomposition of H2SO4 may have resulted in low dissolution of elements, specifically, Th throughout different baking tests performed in the present study at temperatures above 300 oC.

Figure 4. 28. TG-DSC of 96 wt % sulfuric acid in air (heating rate = 5 oC/min, Demol et al., 2018)

From Figures 4.27, it is apparent that the UO2 solubility is higher at low pH and lower at high pH. On the other hand, the thorium solubility is low at low pH but remains higher than uranium at higher pH. The solubility of uranium is probably achieved through its oxidative dissolution and complex formation with sulfate according to the reactions listed in Table 4.25.

In addition, the sulfuric bake product is likely to be in the form of various sulfate complexes which lead to their high solubility depending on pH.

Ekberg and Brown (1989) reported the solubility of uraninite (UO2) in HF solution, at

o o 300 C to 600 C at P = 1 kbar. They found that the UO2 solubility increased and was most

126

Chapter 4: Results and Discussions

probably due to the formation of uranium(IV) fluoride complexes. In our study, a small amount of fluoride was observed in all concentrates in the fluorocalciomicrolite mineral as shown in

Figures 4.15 – 4.16, which may have also contributed to the high dissolution of U warranting further studies. The chemical analysis presented in Table 4.6 has shown that sample C possess high content of 0.8 wt % U. In addition, it was observed that during acid bake at 400 oC, the uranium was not completely dissolved as shown in Table 4.30. Therefore, sample C was used for further studies to investigate the suitable conditions to leach out U and Th from the concentrate as detailed in the following section.

o 4.4.7. Effect of baking conditions on uranium and thorium leaching with K2S2O8 at 90 C

This section focuses on the relative effects of the different baking condition on dissolution of the radioactive elements (i.e., U and Th) in the tantalite concentrate obtained from

SMB in (Test 15 – Test 17). Sample C was baked at 400 oC, varying the time, particle size distribution, and solid-acid ratio. Figures 4.29a-f combine the results of leaching efficiencies of

U and Th under different conditions. Appendices A 7.8 – A 7.13 present the different solution concentrations as well as the dissolution efficiencies of U, Th, Ta and Nb caused by different baking time, particle size distribution and solid to acid ratio. Solid to acid mass ratios of 1.0 :

3.0, 1.0 : 2.0, 1.0 : 1.5 and 1.4 : 1.0 stands for solid mass to acid volume ratios of 625 g/L, 833 g/L, 1250 g/L and 2500 g/L, respectively, in Figures 4.29c and f.

(a) Effect of bake time on uranium leaching

From Figure 4.29a, it can be seen that as the leaching time progresses, the extraction of

U increases to 60-80% and remains relatively unchanged. However, the highest leaching efficiency was obtained when the concentrate was baked for 0.5 h, reaching an efficiency of 94%

127

Chapter 4: Results and Discussions

within the first 2 h of leaching. The extraction of U after baking for 1.5 h increases from 64% within the first 15 minutes and reaches 93% after less than 2 h of leaching time. Results in

Figures 4.29a show that further increase in bake time beyond 0.5 h has a negative effect on leaching efficiency of U. Thus, 0.5 h of bake time could be sufficient for almost a complete digestion of the sample to occur. This indicated that the uranium and thorium sulfates are immediately formed at high temperature during the sulfuric acid bake, which dissolves in the solution during the leaching process afterwards.

(b) Effect of particle size on uranium leaching

Particle size plays a major role in leaching processes. In the present study, particle size fractions of –75 μm, –106+75 μm, –150+106 μm, and –250+150 μm were utilised for baking the

o o concentrate at 400 C for 2 h, and leaching at 90 C in 4 M H2SO4. The effect of the particle size distribution (Figure 4.29b) manifests in the following order; –106+75 μm resulted in the highest leaching of 98% U (after 3 h of leaching), while relatively higher particle size incur lower efficiency due to limited surface area and, hence less accessibility for leaching. However, the relatively low extraction of U from some size fractions may be due to the formation of aluminosilicate in pegmatite ores which may undergo dissolution in H2SO4 consuming acid as presented in equation 14 (Cheru et al., 2018)

(Mg,Ca)O.Al2O3.5SiO2.nH2O + 4H2SO4 = Al2(SO4)3 + 5H2SiO3 + (Mg,Ca)SO4 + (n-1)H2O (14)

In fact, the silicic acid produced in equation (14) forms colloids, which then precipitate as a gel, which obstruct the leaching process. According to Cheru et al. (2018), the formation of the gel is caused by the overgrinding of the concentrate which causes an increase of pulp viscosity and a decrease in leaching rate.

128

Chapter 4: Results and Discussions

Figure 4. 29. Effect of bake conditions on U and Th leaching efficiencies of sample C Bake time effect (T 15), bake particle size effect (T 16) and bake solid to acid volume ratios (T o 17). Time: 2 h, solid to acid volume ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. PSD: < 75 μm, Eh: ̴ 1110 – 1140 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4, temperature: o 90 C, agitation speed: 250 rpm and K2S2O8 to liquid volume ratio: 27 g/L for leaching

129

Chapter 4: Results and Discussions

Referring to Figure 4.29c, the effect of bake solid mass to acid volume ratios on leaching efficiency show that the dissolution of elements decreases with the increase in solid to acid ratio,

U, in particular. The highest leaching efficiency of 87% U was observed using the solid to acid ratio of 1 : 3 (625 g/L), followed by 1 : 2 (833 g/L) with 72% U, while the other ratios were varying in the same range of 65% to 68% U dissolution, after 1 h of leaching reaction. The decrease of U leaching efficiency with the increase of solid to acid ratios may due to the decrease in solubility of uranium sulfate in H2SO4. El-Hussaini and Mahdy (2002) reported the low extraction of U and Th, when the solid to acid ratio was greater than 625 g/L, which was assigned to the refractory mineral davidite (Fe,La,U,Ca)(Ti,Fe,V,Cr)(O,OH). At solid to acid volume ratio of 625 g/L for baking at 200 oC, the leaching efficiency of U reached 70% (El-Hussaini and

Mahdy, 2002) compared to the maximum leachability of 87% U obtained in the present study at solid to acid volume of about 625 g/L and baking at 400 oC. These results warrant further studies at lower bake temperatures discussed later in Section 4.4.10.

(d) Effect of bake conditions on thorium leaching

As shown in Figures 4.29d, e and f, the effect of baking condition on Th leaching follows a similar trend to that of U. For instance, baking time of 0.5 h and 1.5 h give the best leaching efficiency of Th. At 1.5 h an overall dissolution of 42% Th was recorded after 2 h of leaching time. Generally, the leaching efficiency after the first 15 minutes was 37% and reached 47% after

3 h as can be seen in Figures 4.29d, e and f. The main elements Ta and Nb were not much affected with an average of about 5% Ta and nearly 10% Nb as reported in Appendix A7, Table

A 7.9. Thus, it is promising to conclude that the best bake effect which resulted in a significant leaching of U and Th was achieved at solid to acid ratio of 625 g/L. Figures 4.30, 4.31 and 4.32

130

Chapter 4: Results and Discussions

show the effect of bake time on the variation of U and Th dissolution efficiency after the initial

(15 minutes) and the final (3 h) times of leaching, respectively. Results show that 0.5 h bake time at 400 oC give highest leaching efficiencies of U and Th.

100

90 83 80 75 69 70 64 60 50 40

U U leached (%) 30 20 10 0 0.5 1.0 1.5 2.0 Bake time (h)

b) 100 90 90 90 87

80 72 70 60 50 40

U U leached (%) 30 20 10 0 0.5 1.0 1.5 2.0 Bake time (h)

Figure 4. 30. Effect of bake time on leaching efficiency of U at short and long leaching times (a) 15 minutes, (b) 3 h

o Time: 2 h, solid to acid volume ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. o PSD: < 75 μm, Eh: ̴ 1110 – 1140 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4, temperature: 90 C, agitation speed: 250 rpm and K2S2O8 to liquid volume ratio: 27 g/L for leaching (Test 15).

131

Chapter 4: Results and Discussions

100 90 80 70 60 50 36 36 40 32 28 Th Th leached (%) 30 20 10 0 0.5 1.0 1.5 2.0 Bake time (h)

Figure 4. 31. Effect of bake time on leaching efficiency of Th at short leaching time (15 minutes)

o Time: 2 h, solid to acid ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. PSD: < o 75 μm, Eh: ̴ 1140 - 1110 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4, temperature: 90 C, agitation speed: 250 rpm and K2S2O8 to liquid volume ratio: 27 g/L for leaching (Test 15).

100 90 80 70 60

50 42 42 40 33 28 Th Th leached (%) 30 20 10 0 0.5 1.0 1.5 2.0 Bake time (h)

Figure 4. 32. Effect of bake time on leaching efficiency of Th at long leaching time (3 h)

132

Chapter 4: Results and Discussions

4.4.8. Effect of leaching conditions on uranium and thorium leaching with K2S2O8 after baking at 400 oC

After the baking stage at 400 oC, the bake products were leached by varying conditions such as temperature, sulfuric acid concentration, solid to acid ratio, reagent to liquid ratio and the agitation speed. Appendix A 7, Tables A 7.14 to A 7.21 present the solution concentrations as well as the dissolution efficiencies of U, Th, Ta and Nb at different conditions.

(a) Effect of leaching temperature

From Figure 4.33a, it can be seen that the increase of leaching temperature had a significant effect on U dissolution. The highest extraction of 87% U was observed at 90 oC after

3 h of leaching. It is important to mention that a similar extraction of 85% U was obtained at 80 oC after 3 h. Figures 4.34a-b show the variation of the initial (15 minutes) and the final (3 h) leaching efficiencies of U at different leach temperatures indicating the beneficial effect of higher leaching temperatures. The same trend was observed for Th with high leaching efficiencies at

80 oC and 90 oC, where the maximum dissolution (29% and 31%, respectively) was reached after

2 h. As shown in Figure 4.35a, Th was rapidly dissolved during the first 15 minutes, levelling off thereafter. The dependence of the leaching of Th on temperature was somewhat mixed, with an initial increase and a decrease afterwards up to 3 h of leaching time under all temperatures used. In a related study on tantalite concentrate, Cheru et al. (2018) reported a Th dissolution

o efficiency of 85% at 170 C for 3 h with concentrated HF and H2SO4. They reported that the high dissolution of Th was due to the use of HF. Nete et al. (2014b) reported the dissolution of 80%

o Th at 230 C in a solution of 18 M H2SO4 for a leaching time of 1.5 h. It is important to note that the Th concentration of the feed samples under study was too low and widely distributed which may make it difficult for the acid to reach the mineral particle carrying Th. It is clear from these results that it is possible to leach the majority of U and Th in the bake product in a mixture of

133

Chapter 4: Results and Discussions

o o H2SO4 and K2S2O8 as a leaching agent at a temperature between 80 C to 90 C, and in the leaching time range of 2 h to 3 h.

a) b) 100 100 20 °C 40 °C 60 °C 80 °C 90 °C 90 90 80 80 70 70 60 60 50 50 H2SO4: 1 M 40

40 H2SO4: 2 M U leached (%) U leached U (%) U leached 30 30 H2SO4: 3 M H2SO4: 4 M 20 20 10 10 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h) c) d)

100 100 90 90 80 80 70 70 60 S / L: 37.5 g/L 60 S / L: 25.0 g/L R / L= 40.5 g/L 50 50 S / L: 12.5 g/L R / L: 27.0 g/L R / L: 13.5 g/L

40 40

U leached (%) U leached U (%) U leached 30 30 20 20 10 10 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h)

Figure 4. 33. Effect of various leaching conditions on U dissolution of sample C Leaching temperature effect (T 18), leaching acid concentration effect (T 19), leaching solid to liquid ratio (Test 20) and leaching reagent to liquid ratio (T 21). Time: 2 h, solid to acid volume o ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. PSD: < 75 μm, Eh: ̴ 1110 - o 1140 mV, time: 3 h, 4 M H2SO4, temperature: 90 C, agitation speed: 250 rpm and K2S2O8 to liquid ratio: 27 g/L for leaching

134

Chapter 4: Results and Discussions

a) 100 90 80 69 69 70 66

(%) 60 51 50

leached 40

U U 30 20 10 0 20 60 80 90 Leach temperature (oC)

b) 100 90 85 87 79 80 70 58

(%) 60 50

leached 40

U U 30 20 10 0 20 60 80 90 o Leach temperature ( C)

Figure 4. 34. Effect of leaching temperatures on U leaching efficiency at short and long leaching times (a) 15 minutes, (b) 3 h

o Time: 2 h, solid to acid volume ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. o PSD: < 75 μm, Eh: ̴ 1110 - 1140 mV, time: 3 h, 4 M H2SO4, temperature: 90 C, agitation speed: 250 rpm and K2S2O8 to liquid ratio: 27 g/L for leaching (Test 18).

135

Chapter 4: Results and Discussions

a) b)

100 100 90 20 °C 40 °C 60 °C 80 °C 90 °C 90 H2SO4: 1 M H2SO4: 2 M 80 80 H2SO4: 3 M 70 70 H2SO4: 4 M 60 60 50 50

40 40 Th leachedTh (%)

30 leachedTh (%) 30 20 20 10 10 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h)

c) d) 100 100 90 90 80 80 S / L: 37.5 g/L R / L: 40.5 g/L 70 70 S / L: 25.0 g/L R / L: 27.0 g/L 60 S / L: 12.5 g/L 60 R / L: 13.5 g/L 50 50

40 40 Th leachedTh (%) 30 leachedTh (%) 30 20 20 10 10 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h)

Figure 4. 35. Effect of various leaching conditions on Th dissolution of sample C Leaching temperature effect (T 18), leaching acid concentration effect (T 19), leaching solid to liquid ratio effect (Test 20) and leaching reagent to liquid ratio (T 21). Time: 2 h, solid to acid o volume ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. PSD: < 75 μm, Eh: ̴ o 1110 -1140 mV, time: 3 h, 4 M H2SO4, temperature: 90 C, agitation speed: 250 rpm and K2S2O8 to liquid ratio: 27 g/L for leaching.

(b) Effect of acid concentration

From Figure 4.33b, it can be seen that at a low acid concentration of 2 M H2SO4, a leaching efficiency of 98% U was reached after 3 h. The leaching efficiency reached a maximum of 94% after 3 h in the solution of acid concentration 3 M H2SO4. In the presence of 4 M H2SO4,

136

Chapter 4: Results and Discussions

a relatively low leaching efficiency of 87% U was observed. In 1 M H2SO4, the dissolved uranium of 95% after 0.5 h decreased with the increase of time and reached 85% after 3 h. In a related work on pyrochlore concentrate subjected to sulfuric acid baking, Jerome et al. (2006) reported that the dissolution of uranium was > 90%, along with Th, Ta and Nb. Jerome et al.

(2006) stated that the presence of ferric iron during leaching after baking the concentrate at 300 oC improves the leaching of the mineral carrying Ta and Nb, in particular pyrochlore group of mineral which hosts the uranium. The molar ratio Fe/(Ta + Nb) ≥ 2 is suitable for leaching, otherwise, a diluted sulfuric acid is used to improve the elemental leaching efficiencies.

However, the molar ratio of the sample C studied in this work is Fe/(Ta + Nb) = 0.2 which is less than 2 (Table 4.30). Thus, a diluted solution of H2SO4 yield better results for initial dissolution of uranium as shown in Figure 4.33b but further investigation is warranted.

The effect of H2SO4 concentration on the dissolution of Th is presented in Figure 4.35b.

The highest Th extraction of 47% was observed in 3 M H2SO4 after 2 h and the leaching efficiency remained constant for the range of 2 – 3 h. Similar results were obtained for the dissolution with an overall extraction of 42% Th in H2SO4 solutions of 1 – 2 M. The lowest leaching efficiency was 32% Th in 4 M H2SO4 after 1 h, which decreased with the increase of time and reached 28% Th after 3 h. Again, an average of about 5% Ta and 10% Nb were leached in the presence of 4 M H2SO4. Based on the results obtained, it is clear that the leaching efficiency

o of U and Th is higher in H2SO4 solutions in the concentration range of 3 – 4 M at 90 C after 3 h indicating that these are the suitable conditions to treat the present concentrate.

Figures 4.33c and 4.35c show the leaching efficiencies of U and Th, respectively, obtained under different solid mass to liquid volume ratios of 12.5 g/L, 25 g/L and 37.5 g/L

137

Chapter 4: Results and Discussions

keeping other condition unchanged. The highest leaching efficiency of 79% U and fastest rate was observed at a solid mass to liquid volume ratio of 37.5 g/L after 15 minutes and reached a value of 100% U after 2 h (Figure 4.33c). In the present study, it was also recorded that the leaching efficiency of Th was independent of the effect of the S/L ratios (Figure 4.35c). No significant change in Th dissolution (~ 30%) was observed for different S/L ratios.

(d) Effect of oxidant mass to liquid volume ratio

The leaching efficiencies of U and Th were investigated by varying the K2S2O8 reagent mass to liquid volume ratio as 13.5 g/L, 27 g/L and 40.5 g/L as shown in Figures 4.33d and

4.35d, respectively. The K2S2O8 mass to liquid volume (R/L) ratio of 13.5 g/L and 27 g/L have shown to cause a significant effect on leaching efficiency. At a low R/L ratio of 13.5 g/L, the extraction of U and Th reached 67% and 32% within the first 15 minutes, and 87% and 30%, respectively, after 1 h. A higher R/L ratio of 27 g/L has shown similar results with an increase of the leaching efficiency as the time progressed specifically for U. On the other hand, the leaching efficiency of Th has shown a fluctuation with an increase or decrease as the time progressed. A leaching efficiency of 28 - 32% Th was observed within the first 2 - 3 h of leaching.

Based on the results described above, it is clear that the addition of K2S2O8 has a great influence on the dissolution of elements, U in particular, indicating its role in oxidation of UO2 according to equation 10b in Table 4.25. A significant dissolution and high leaching efficiencies of U was observed by employing both 13.5 g/L and 27 g/ L. On the other hand, Th, Ta and Nb showed a higher dissolution at R/L of 40.5 g/L in comparison with other ratios.

138

Chapter 4: Results and Discussions

(e) Effect of agitation

The effect of agitation has a beneficial effect on the leaching efficiencies of U in 4 M

o H2SO4 at 90 C after 0.25 h and 3 h leaching time as shown in Figures 4.36. In a related study,

Cheru et al (2018), reported a change in the dissolution of 69% after 1 h to 89% U after 3 h, at a temperature range of 100 oC to 300 oC, at a stirring speed of 250 rev/min, solid to acid of 625 g/L, in the presence of concentrated HF-H2SO4. Appendix A 7, Tables A 7.22 and A 7.23 present the solution concentrations as well as the dissolution efficiencies for leaching at different agitation speeds showing that the extraction of Th, Ta and Nb was relatively independent of agitation.

a) 100 0.25 h 3 h 90 87 78 80 69 70 60 56 50 40

U U leached (%) 30 20 10 0 0 rpm 250 rpm Agitation speed (rpm)

100 90 0.25 h 3 h 80 70 60 50

40 32 30 28 Th Th leached (%) 30 26 20 10 0 0 rpm 250 rpm Agitation speed (rpm)

Figure 4. 36. Effect of agitation speed on leaching efficiency of (a) U and (b) Th of sample C

o Time: 2 h, solid to acid volume ratio: 625 g/L, 18 M H2SO4 and temperature: 400 C for baking. o PSD: < 75 μm, Eh: ̴ 1110 - 1140 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4, temperature: 90 C and K2S2O8 to liquid ratio: 27 g/L for leaching (Test 22).

139

Chapter 4: Results and Discussions

4.4.9. Effect of leaching conditions on minor elements after baking at 400 oC

During the acid bake, other elements are also converted to soluble sulfates as presented in chemical reactions 15 – 19 in Table 4.31 (Zhang et al., 2016). Figure 4.37 plots the leaching efficiencies of the minor elements such as Fe, Mn, Al, Li, and Cs as a function time, obtained at the best leaching conditions for extraction of U and Th. The figure indicates that these minor elements have moderate to high dissolution efficiencies, ranging from approximately 40% (for

Al) to about 85% (for Mn). The solution concentrations are presented in Appendix A 7, Table

A 7.24. However, as shown in previous Tables and Figures, the elements of interest (i.e., Ta and

Nb) are well preserved in the solid residue.

Table 4. 31. Chemical reactions of minor oxides with H2SO4

Chemical reactions ΔGo (kJ mol-1) ΔGo (kJ mol-1) Reference Equation at 90oC at 400oC No Fe2O3 + 3H2SO4 = Fe2(SO4)3 + 3H2O - 15 9 .8 - 140.6 This work ; 15 MnO + H2SO4 = MnSO4 + H2O -117.5 -128.2 based on 16 Al2O3 + 3H2SO4 = Al2(SO4)3 + 3H2O - 154. 2 - 132.4 HSC data 17 Li2O + H2SO4 = Li2SO4 + H2O -312.2 -303.5 base 18 Cs2O + H2SO4 = Cs2SO4 + H2O -560.1 -552.9 19

Table 4.32 gives the summary of leaching efficiencies for each sample of Bisunzu mine concentrates under study. Fastest rate of leaching and highest efficiencies were observed for Fe and Mn. Generally, more than 50% and 70% Fe dissolution was reached after the first 15 minutes and 3 h, respectively, which may have enhanced high leaching efficiencies of U, accordingly. A similar trend was observed for Mn leaching in most samples. However, leaching efficiencies of sample B did not exceed 60% for Fe and 32% for Mn. This may be due to low concentration of leachable minerals containing these metals in the initial sample as indicated in Tables 4.6 and

4.8, respectively. Quite similar leachability of Li and Cs were observed for different samples.

Generally, leaching efficiencies of 20 - 60% were recorded for Li with a complete dissolution of

140

Chapter 4: Results and Discussions

Cs for sample E (Table 4.32) warranting further studies. On the contrary, the leaching efficiency of Al did not exceed 45% for all samples, even less (30%) for samples B, E and F. It is clear that an aluminosilicate species was formed during baking which hindered the performance.

100 90 80 70 60 50 40 30

Elements leached (%) leached Elements 20 Fe Mn Al Li Cs 10 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h)

Figure 4. 37. Dissolution of minor elements at suitable conditions of sample C

Time: 2 h, solid to acid volume ratio: 625 g/L, acid concentration: 18 M H2SO4 and temperature: 400 oC for baking. Temperature: 90 oC, PSD: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

Generally, the leaching efficiencies of Fe and Mn as well as Li and Cs have shown the same trend of dissolution using various baking and leaching conditions indicating that these elements are being hosted by same minerals. The Fe and Mn are hosted in minerals such as pyrochlore of general formula A2B2O6(O,OH,F), with A= U, RE, Na, Ca, Ba, Th or Bi and B =

Nb, Ta, Ti or Fe (in general), euxenite, samarsakite, perovskite or fergusonite group and their mixtures. Such ores and concentrates comprise of iron. Other minerals carrying Fe and Mn which can be decomposed during the acid baking are microlite and columbite-tantalite group while Li and Cs are hosted in lepidolite.

141

Chapter 4: Results and Discussions

Table 4. 32. Elemental leaching efficiencies of minor elements

T.N. S.I. Leaching time (h) Elements leached (%) Fe Mn Al Li Cs U Th A 0.25 73 53 28 40 30 50 45 0.5 88 58 38 45 30 71 61 1 80 58 32 46 34 51 40 2 93 59 40 50 39 61 43 3 91 57 45 53 45 61 46 B 0.25 23 24 24 23 18 87 39 0.5 28 28 27 25 21 95 45 1 35 28 26 27 21 98 49 2 46 28 30 26 20 100 49 3 63 32 30 31 23 100 56 C 0.25 65 82 39 47 49 69 32 0.5 68 84 39 50 49 72 29 14 1 70 82 45 57 52 80 32 2 83 90 48 57 56 84 31 3 91 96 50 59 60 87 28 D 0.25 73 93 28 52 49 57 32 0.5 78 96 32 53 50 62 31 1 77 92 37 55 55 62 34 2 87 97 44 55 53 64 36 3 87 100 36 61 56 67 36 E 0.25 36 53 21 43 100 100 65 0.5 45 58 19 41 100 100 71 1 59 58 17 46 100 100 62 2 74 59 25 46 100 100 61 3 77 57 24 49 100 100 67 F 0.25 39 50 15 21 30 86 27 0.5 42 53 19 23 29 82 27 1 63 66 21 28 36 100 31 2 67 60 22 24 33 100 31 3 80 66 25 27 38 100 35 G 0.25 58 38 47 21 37 89 48 0.5 68 42 38 42 41 98 49 1 82 53 48 53 43 100 63 2 69 45 39 45 49 100 55 3 73 49 40 49 44 100 56 H 0.25 51 71 22 18 18 100 48 0.5 56 74 22 20 19 100 50 1 67 81 20 23 21 100 49 2 71 74 26 22 21 100 46 3 85 82 26 23 21 100 48

T.N.: Test number 14, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, 18 M o H2SO4, temperature: 400 C and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, o time: 3 h, S/L: 12.5 g/L, Eh: ~ 840-1120 mV, temperature: 90 C and K2S2O8 to liquid volume ratio: 27 g/L for leaching

Tables 4.33 and 4.34 summarise the dissolution efficiencies of the minor elements at varying conditions such as particle size of bake feed and leach temperature. Low extractions of approximately 40% Al with about 60% Li and Cs each, were observed in the case of fine particles

142

Chapter 4: Results and Discussions

(size distribution < 75 μm) as listed in Table 4.33. This can be related to the presence of aluminosilicate which hinder the elemental dissolution as previously described for U and Th.

Table 4. 33. Effect of particle size on leaching of minor elements

T.N. S.I. Baking particle size Leaching time Elements leached (%) (μm) (h) Fe Mn Al Li Cs 0.25 71 65 55 58 57 –250+150 0.5 79 69 58 65 61 1 84 71 80 65 59 2 91 75 69 71 64 3 68 54 73 53 50 16 C 0.25 76 72 47 58 57 –150+106 0.5 79 72 57 63 58 1 84 76 65 77 70 2 92 78 69 73 67 3 94 80 63 72 71 0.25 65 82 47 47 49 –75 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L, o 18 M H2SO4 and temperature: 400 C for baking. Agitation speed: 250 rpm, Eh: ~ 890 - 920 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

The leaching efficiency of the minor elements at 60 oC - 90 oC are nearly in the same range as can be seen in Table 4.34. For instance, the Fe extraction vary from 65% to 85% and

Mn from 82% to 92% at these temperatures throughout the leaching process. Only 40% Al was extracted during the entire leaching period of 3 h. This may be due to the changes of the structure of particles by sintering phenomena occurring during the acid baking. The leaching efficiencies of Li and Cs were also in the same range and increased from 60% at 60 oC to nearly 68% at 80 oC. The solution concentrations at different leaching temperatures are presented in Appendix A

7, Table A 7.31. Nogueira et al. (2014) reported that a lepidolite embodiment calcined at 800 oC was transformed into more reactive species, β - spodumene, and then was digested with H2SO4 at 170 oC and finally leached with water, where 88% Li dissolved within 30 minutes of leaching

143

Chapter 4: Results and Discussions

process. From these results, it is clear that the H2SO4 baking is an effective method to recover Li and Cs from this type of pegmatite ore.

Table 4. 34. Effect of leaching temperature on dissolution of minor elements

T.N. S.I. Leaching temperature Leaching time Elements leached (%) o ( C) (h) Fe Mn Al Li Cs 90 0.25 65 82 47 47 49 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 80 0.25 68 85 34 45 41 0.5 71 88 41 55 51 1 75 93 40 59 52 2 76 92 39 62 57 3 85 93 41 68 62 18 C 60 0.25 66 87 37 46 46 0.5 74 94 37 50 46 1 63 84 42 55 52 2 68 88 41 55 54 3 74 92 42 60 56 40 0.5 5 1 10 2 4 1 7 1 8 4 3 2 10 1 12 2 3 3 12 1 10 2 3 20 0.25 21 74 35 19 24 0.5 26 76 34 21 27 1 38 78 38 22 27 2 49 77 43 27 30 3 52 77 37 26 30

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L, 18 o M HS2O4 and temperature: 400 C for baking. PSD: < 75 μm; agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

4.4.10. Effect of baking at 100 – 300 oC and leaching at 90 oC

(a) Major elements

The acid baking process of the concentrate was performed at different temperatures (100 oC, 200 oC, 300 oC and 400 oC) while other baking as well as leaching conditions were kept unchanged. Appendix A 7, Tables A 7.42 – A 7.43 summarise the solution concentrations for

144

Chapter 4: Results and Discussions

various baking temperature as well as the dissolution efficiencies of U, Th, Ta and Nb. Solution concentrations and leaching efficiencies for minor elements at different bake temperatures are provided in Appendix A7, Tables A 7.4 – A 7.44. The effect of bake temperature on the leaching efficiencies of U, Th, Ta and Nb is shown in Figures 4.38a-d.

a) b)

100 100 90 90 Bake temp= 100 °C Bake temp= 200 °C 80 80 Bake temp= 300 °C 70 70 Bake temp= 400 °C 60 Bake temp= 100 °C 60 50 Bake temp= 200 °C 50 Bake temp= 300 °C

40 Bake temp= 400 °C 40 U U leached (%) 30 leachedTh (%) 30 20 20 10 10 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h) c) d) 25 25 Bake temp= 100 °C Bake temp= 100 °C Bake temp= 200 °C Bake temp= 200 °C 20 Bake temp= 300 °C 20 Bake temp= 300 °C Bake temp= 400 °C Bake temp= 400 °C

15 15

10 10

Taleached (%) Nb leached (%)

5 5

0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Leach time (h) Leach time (h)

Figure 4. 38. Effect of baking temperature of 100 – 400 oC on leaching efficiency of U, Th, Ta and Nb of sample C

Solid to acid volume ratio: 625 g/L, 18 M H2SO4 and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 1120 mV, temperature: 90 oC, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching (Test 23).

145

Chapter 4: Results and Discussions

A bake temperature of 100 oC and leaching at 90 oC has resulted in relatively low dissolution of nearly 2% U and 1% Th. The dissolution of Ta and Nb remained very low (Figures

4.38c-d). Increasing the bake temperature to 200 oC caused a slight increase of leaching efficiencies to 27% U, 16% Th and 2% Ta after 2 h. The increase of bake temperature to 300 oC has resulted further increase of leaching efficiency of U and Th. The dissolution of U increased from 89% within the first 15 minutes to 100% after 30 minutes, and remained unchanged for the entire leaching period. The leaching efficiency of Th increased as the time progressed from 47% at 30 minutes to 62% after 3 h. The high dissolution of U and Th at this temperature may be due to the formation of uranium and thorium sulfates during baking (Tables 4.20, 4.25), subsequently dissolved during the leaching process. The leaching efficiency of Ta and Nb remained constant with a maximum of 7% Ta and nearly 9% Nb, throughout the leaching process.

An increase of bake temperature to 400 oC resulted in a slight decrease of leaching efficiency of U and Th compared to values obtained at 300 oC. Based on the results obtained for all four different temperatures, the leaching efficiency increased sharply to a maximum of 100%

U after 30 minutes and 62% Th after 3 h at 300 oC with a moderate dissolution of Ta and Nb with < 10% each. Thus, bake at 300 oC was the best condition to completely leach out the U and a significant amount of Th from the tantalite sample C. These results are consistent with previous studies by Jerome et al. (2006), where a temperature of 200 oC to 300 oC resulted in virtually complete decomposition of mineral carrying Ta, Nb, U and REE, and more effective on pyrochlore. Using the same procedure, Zhang et al. (2016) reported successful dissolution of U,

Th and REE achieved by baking the ore at temperatures between 180 oC to 300 oC for 2 – 4 h.

146

Chapter 4: Results and Discussions

(b) Minor elements

As shown in Table 4.35, relatively high dissolution of minor elements was observed at

300 oC, where 87% Mn, 64% Fe, 50% Al, 55% Li and 69% Cs were leached after 2 h. However, the leaching efficiency of Fe and Mn after baking at 400 oC and leaching at 90 oC for 3 h is also higher (91% Fe and 96% Mn) compared to baking at 300 oC. Interestingly, Cs was completely dissolved at 100 oC. These trends can be studied in detail in future work.

Table 4. 35. Effect of bake temperature on leaching efficiency of minor elements

T.N. S.I. Bake temperature Leaching time Elements leached (%) o ( C) (h) Fe Mn Al Li CS 400 0.25 65 82 39 47 49 0.5 68 84 39 50 49 1 70 82 45 57 52 2 83 90 48 57 56 3 91 96 50 59 60 23 C 0.25 53 53 43 45 58 300 0.5 62 66 49 50 61 1 59 68 46 46 59 2 68 86 54 50 61 3 64 87 50 55 69 0.25 20 49 29 25 27 0.5 22 60 39 27 29 200 1 24 74 45 30 34 2 24 82 44 41 42 3 47 64 37 30 31 0.25 2 55 22 20 75 100 0.5 2 63 - 25 97 1 2 72 33 30 100 2 2 80 37 34 100 3 2 90 40 43 100

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, 18 M H2SO4 and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 o h, S/L: 12.5 g/L, 4 M H2SO4, temperature: 90 C and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

147

Chapter 4: Results and Discussions

Appendix A 7, Tables A 7.29 – A 7.40 present the solution concentrations as well as the leaching efficiencies under different conditions such as bake solid to acid ratio, leaching acid concentration, solid to liquid ratio, K2S2O8 to liquid ratio and agitation speed of the minor elements. Figure 4.39 summarises the steps followed while conducting the laboratory scale experiments. The recommended best conditions are: acid bake at 300 oC for 2 h, solid to acid ratio of 625 g/L and < 75 μm as particle size for baking, and temperature of 90 oC in solution of

4 M H2SO4 at solid to liquid of 12.5 g/L, leaching time: 3 h, Eh: ~ 1000 mV and K2S2O8 to liquid volume ratio: 27 g/L.

4.4.11. Correlation between leaching efficiencies

The relative leaching behaviour of Nb and Ta can be rationalised by considering the correlation in Figure 4.40. Appendix A, Figures A.8.1a and b show the correlation between Nb and Ta leaching efficiencies at various leaching temperatures, acid concentrations, solid to liquid ratios and K2S2O8 to liquid ratios. A linear correlation of slope close to 1.70 is shown in Figure

4.40. The slopes of the linear relationship of 1.7 at low S/L variation (i.e. excess acid) in Figure

4.40 and Appendix A8 is closer to the expected slope from the atomic masses ratio of Ta/Nb =

1.9 warranting further studies beyond the scope of this thesis. High leaching efficiency of 29%

Nb along with low extraction of 4.5% Ta (Table 4.24 and Figure 4.40) was achieved in the presence of KOH shifting the slope towards Nb axis (Appendix A 8, Figure A 8.1c).This behaviour is consistent with the high solubility of Nb compared to that of Ta shown in Figure

4.25, which also warrants further studies.

148

Chapter 4: Results and Discussions

Figure 4.41 shows the correlation between Li and Cs leaching efficiencies at different conditions including leaching temperatures, acid concentrations, solid to liquid ratios, and

K2S2O8 to liquid ratios. Figure A 8.2 in Appendix A 8 shows the correlation between Li and Cs leaching efficiencies at different conditions where most of the data points for the sample C show a linear correlation of slope 1 in Figure 4.41. These results show that Li and Cs are leached from the same mineral (lepidolite).

Figure 4. 39. Experimental scheme for leaching uranium and thorium from tantalite concentrate

149

Chapter 4: Results and Discussions

LE in KOH 25

20 LE in H2SO4

y = 1.72x

Nb leached (%) 10 R² = 0.98

0 0 2 4 6 8 10 12 Ta leached (%)

Figure 4. 40. Correlation between Ta and Nb leaching efficiencies at different conditions (LE denotes leaching efficiency)

50 y = 1.03x R² = 1.00 40

Li leached leached (%)Li 30

0 0 10 20 30 40 50 60 70 80 Cs leached (%)

Figure 4. 41. Correlation between Li and Cs leaching efficiencies at different conditions

150

Chapter 4: Results and Discussions

Figures 4.42a and b show the correlation between U and Fe as well as the correlation between U and Mn, respectively in 4 M H2SO4. A plot of percentage dissolution of U against that of Fe or Mn resulted in a linear correlation of slope 1.

a) 100 y = 0.96x 90 R² = 0.98 80

U leached (%) leached U 30

0 0 10 20 30 40 50 60 70 80 90 100 Fe leached (%)

b) 100 y = 1.10x 90 R² = 0.98 80

40 U leached (%) leached U 30

0 0 10 20 30 40 50 60 70 80 90 100 Mn leached (%)

Figure 4. 42. Correlation between (a) U and Fe, (b) U and Mn leaching efficiencies at different conditions

151

Chapter 4: Results and Discussions

A slope close to 1 supports the observation that U, Fe, Mn are hosted in the same mineral fluorocalciomicrolite as evident from the SEM-EDS scan shown in Appendix 6, Figure A.6.1.

In addition, manganotantalite (Mn,Fe)(Ta,Nb)2O6 may have also been affected by the acid baking which affect the dissolution of Fe and Mn. However, more work is required to investigate the leaching correlations of Ta, Nb, U, Fe and Mn shown in Figures 4.40 – 4.42.

4.4.12. Characterisation of bake product and leach residue of sample C

The XRD patterns for bake product and leach residue for sample C considered as an example is presented in Figure 4.43; where a1, a2 and a3 stands for feed, bake product and leach residue, respectively. List of identified minerals in the feed sample as well as the samples after baking and leaching is shown in Table 4.36. Fluorocalciomicrolite, stannomicrolite, augite, muscovite and hematite were not identified in the XRD pattern of the bake product as shown in

Figure 4.43a2 showing that these phases were virtually decomposed during the acid baking.

There is no evidence for the formation of tantalum sulfate in the bake product. Lack of formation of tantalum sulfate may be due to the high affinity of tantalum for oxygen compared to that of niobium which promote oxide formation, resulting in structural changes of Ta (Cullis and

Beanland, 1999). Microlite and stannomicrolite reflections were observed in the leach residue.

Baking and leaching of the concentrate resulted in the elemental dissolution of 5% Ta, 9% Nb,

100% U and 62% Th leaching efficiencies in the present study. However, it is important in future work to perform the XRD quantification to determine the contribution of concerned phases in the elemental dissolution.

152

Chapter 4: Results and Discussions

Figure 4. 43. XRD patterns (a1) feed, (a2) bake product and (a3) leach residue of sample C

Table 4. 36. Mineral identification of feed, bake product and leach residue of sample C

Product at various treatment Mineral/compound name Reference codes a1. Feed 1. Fluorocalciomicrolite: (Ca,Na)2(Ta,Nb)2O6(F,OH) 04-015-1266 2. Quartz: SiO2 01-085-0790 3. Manganotantalite: (Mn,Fe)(Ta,Nb)2O6 04-012-9860 4. Hematite: Fe2O3 01-076-4579 5. Stannomicrolite: Sn2Ta2O7 00-023-0599 6. Muscovite: KAll2(Si3Al)O10(OH,F)2 00-0582036 7. Augite: (Ca,Na)(Mg,Fe,Al,Ti)(Si,Al)2O6 00-041-1483 a2. Bake 8. Niobium sulfate: Nb2O3(SO4)2.0.25H2O 96-153-2683 2. Quartz: SiO2 96-101-1160 3. Manganotantalite: (Mn,Fe)(Ta,Nb)2O6 96-900-4110 a3. Leach 1. Microlite: (Ca,Na)Ta2O6 96-901-2069 2. Quartz: SiO2 96-900-5019 3. Manganotantalite: (Mn,Fe)(Ta,Nb)2O6 96-900-4110 4. Hematite: Fe2O3 96-900-0140 5. Stannomicrolite: Sn2Ta2O7 96-900-9948

153

Chapter 4: Results and Discussions

In a related study on the pyrochlore concentrate, previous researchers (Jerome et al.,

2006; Zhang et al., 2016) reported that after sulfuric acid baking at a temperature range of 250 to 300 oC, the pyrochlore mineral carrying niobium, tantalum, uranium and REE were decomposed to form various sulfates of niobium, tantalum, uranium and rare earth, specifically, with about 90% of the pyrochlore being decomposed following the chemical equations in Table

4.25. As expected, hydrated niobium sulfate Nb2O3(SO4)2·0.25H2O mentioned in Table 4.25 was identified in the XRD patterns of solid bake product at 400 oC (Figure 4.43a2 and Table

4.36a2) in the current study.

The SEM images of the bake products and the leach residues appear to be porous and crumbled as can be seen in Figures 4.44a, 4.45a, 4.46a and 4.47a. This appearance is most likely due to the disintegration of the mineral structure under the acid baking or leaching process of minerals containing cations such as U, Th, Ta and Nb. In addition, new phases may precipitate upon reaction between the liberated cations and lixiviant after local supesaturation. In a related work, Cheru et al. (2018) reported porous SEM images of the sulfuric acid treated tantalite sample. They suggested that the porous mineral surface was due to the dissolution of U, Th, Ta and Nb as well as the side reactions which may have led to the precipitation of various species of mineral components in the concentrate.

As expected, Ta, Nb, U, Mn, Fe and Ca elements as well as sulfur (S) were observed by the EDS for bake products of both samples C and G as shown in Figures 4.44b and 4.46b, respectively. The EDS spectrum for all leach residues show the absence of U, and the presence of Ta, Nb, Mn, Fe which can be attributed to the high amount of undissolved manganotantalite in the concentrate as can be compared with the XRD peak in Figure 4.43a3. Table 4.37 presents the comparison of quantitative analysis of the feed, bake product and leach residue for samples

154

Chapter 4: Results and Discussions

C and G based on the EDS. The content of sulfur in sample C decreased from 7.2 wt % in bake product to 1.3 wt % in leach residue in Figures 4.44b and 4.45b, respectively, as shown in Table

4.37. The same trend has been observed in sample G where the sulfur content of 34.7 wt % in bake product decreased to 10.2 wt % in leach residue as illustrated from Figures 4.46b and 4.47b, respectively, and Table 4.37 which may be due to the insoluble sulfate species in the bake product. From Table 4.37, it is clear that the uranium was not detected in the leach residues for both samples C and G indicating the total removal of U from the concentrates.

The elemental composition based on ICP-MS analysis of samples C and G with respect to feed, bake product and leach residue are presented in Table 4.38 after the solid being subjected to baking at 300 oC. It is clear that the Ta and Nb content in the leach residue has reduced by about 5 wt % and 7 wt %, respectively, from 37 wt % to 32 wt % or 31 wt %. The decrease in grade of most of elements after the baking stage compared to the initial feed may be due to formation of various sulfates which suppress their grade in the bake product. However, it is apparent that the concentration of elements by weight in the leach residue were enhanced after the reduction of sulfur content during the leaching process (Tables 4.37 and 4.38). For instance, the contents of 37 wt % Ta and 9 wt% Nb in the feed of sample C reduced to 32 wt % and 6 wt

% in the leach residue (Table 4.38), respectively. These losses are due to their dissolution during the leaching stage. In fact, the average dissolution of 5 wt % Ta and 8 wt % Nb was observed in majority of the leaching experiments performed.

155

Chapter 4: Results and Discussions

Figure 4. 44. (a) Secondary electron (SE), (b) analysis of bake product of sample C. (Note the porous nature of the products. The white denotes the position of EDS analysis)

156

Chapter 4: Results and Discussions

Figure 4. 45. (a) Secondary electron (SE), (b) analysis of leach residue of sample C. (Note the porous nature of the products. The white denotes the position of EDS analysis)

157

Chapter 4: Results and Discussions

Figure 4. 46. (a) Secondary electron (SE), (b) EDS analysis of bake product of sample G. (Note the porous nature of the products. The white denotes the position of EDS analysis)

158

Chapter 4: Results and Discussions

Figure 4. 47. (a) Secondary electron (SE), (b) EDS analysis of leach residue of sample G. (Note the porous nature of the products. The white denotes the position of EDS analysis)

159

Chapter 4: Results and Discussions

Table 4. 37. EDS quantitative analysis of feed, bake product and leach residue of samples C and G

Elements Elemental composition of feed, bake product and leach residue of samples C and G (wt %) C G Feed Bake product Leach residue Feed Bake product Leach residue Ta 65.1 43 58 61.1 32.5 48.8 Nb 3.9 9.1 5.1 6.4 6.6 6.7 U 1.1 0.5 Nd 0.3 Nd Nd O 17.7 36.1 27 16.2 35.5 30.4 Ca 7.7 0.7 - 8.2 0.7 1.3 Na 1.0 0.2 0.3 4.4 2.7 - S - 7.2 1.3 - 34.7 10.2 Mg 0.4 - 3.3 - - - F 2.2 - Nd 3.3 Nd Nd Mn 0.8 0.5 1.2 - 6.0 0.7 K - 2.1 - - - 0.6 Fe - 0.5 0.4 - 2.7 0.9 Y - - - 0.1 - - Ti - - - - - 0.4

o Nd: not detected. Solid to acid ratio: 625 g/L, 18 M H2SO4, temperature: 300 C, and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, temperature: o 90 C, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching

Table 4. 38. Elemental analysis of feed, bake product and leach residue of samples C and G

Elements Samples (wt %) C G Feed Bake product Leach residue Feed Bake product Leach residue Ta 37 26 32 37 28 31 Nb 9 4 6 15 6 4 U 0.8 0.8 0.01 0.42 0.60 0.01 Th 0.03 0.03 < 0.01 0.01 0.01 < 0.01

o Solid to acid ratio: 625 g/L, 18 M H2SO4, temperature: 300 C, and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, temperature: 90 oC, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching

Recall that a content limit of 0.1 wt % U and 0.1 wt % Th is set for export exemption

(Cheru et al., 2018; TIC, 2016). Interestingly, the chemical composition of the leach residues for samples C and G indicated that the uranium and thorium content (≤ 0.01 wt % in Table 4.38) is

160

Chapter 4: Results and Discussions

lower compared to the allowable limits of U and Th in the tantalum/niobium concentrates for export purposes. Thus, the present study succeeded in removing the radioactive materials (i.e.,

U and Th) from the concentrate to the high extent that the concentrate can be transported or further treated by SMB or buyers.

4.4.13. Acid decomposition and leaching of tantalite concentrate with CaF2

A leaching test was performed with a feed to CaF2 mass ratio of 1: 2, in a solution of 4M

o H2SO4 using a solid mass to liquid volume ratio of 37.5 g/L at a temperature of 90 C, using a concentrate of particle size –75 μm. Equations 20-29 in Table 4.39 gives the chemical reactions involved in the leaching process. The concentrations of dissolved metals are presented in

Appendix A 7, Table A7.45. During the leaching process, HF is generated from the reaction between CaF2 and H2SO4 as shown in chemical equation 20. According to Merritt (1971) during mixed acid leaching, the HF decomposes the silicate minerals and react with UO2, while H2SO4 ionise to form the sulfate, bisulfate, and hydrogen ions. These species react with hexavalent uranium (UO3) to produce soluble uranyl sulfate and complex uranyl sulfate anions according to equations listed in Table 4.39.

Table 4. 39. Decomposition reactions of tantalite concentrate in CaF2 and H2SO4

Chemical reactions Temp (oC) ΔGo (kJ mol-1) References number CaF2 + H2SO4 = CaSO4 + 2HF 90 - 9.7500 This work ; 20 UO2 + 4HF = UF4 + 2H2O - 157.22 based on HSC 21 ThO2 + 4HF = ThF4 + 2H2O - 202.25 data base 22 SiO2 + 6HF = H2SiF6 + 2H2O -156.38 23 Ta2O5 + 14HF = 2H2[TaF7] + 5H2O - - Ayanda et al. 24 Nb2O5 + 10HF = 2H2[NbOF5] + 3H2O (2011) 25 + 2+ UO3 + 2H = UO2 + H2O - - Meritt, 1971 26 2+ 2− UO2 + SO4 = UO2SO4 - - 27 2− 2 UO2SO4 + SO4 = [UO2(SO4)2] - - 28 2- 2- 4- [UO2(SO4)2] + SO4 = [UO2(SO4)3] - - 29

161

Chapter 4: Results and Discussions

From Figure 4.48, it is clear that the leaching efficiencies of elements increased rapidly in the first 30 minutes. The leaching efficiency of U reached 76% after 30 minutes and 100% after 3 h. The maximum leaching efficiency of 90% Th was observed after 2 h, which decreased to 86% after 3 h and continued to decrease to 69% after 4 h.

100 90 80 70 60 50 40 30

Elements Elements leached (%) 20 U Th Ta Nb Si 10 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (h)

Figure 4. 48. Leaching efficiency of U, Th, Ta and Nb in CaF2 and H2SO4 of sample C

o Solid to CaF2 mass ratio 1 : 2, acid concentration: 4 M H2SO4, temperature: 90 C, PSD: < 75 μm, time: 4 h, S/L: 37.5 g/L and agitation speed: 250 rpm, Test number 24.

Leaching efficiencies of 96% Al and 90% Si was also observed indicating a relatively high decomposition of silicate mineral in the presence of HF due to decomposition of metal oxygen (M-O) bonds within silicate minerals (Guo et al., 2018). The dissolution of Ta increased from 48% at 0.5 h to 81% after 4 h. Low extraction range of 30% to 43% Nb was observed during the entire leaching period. Low extraction of Nb and non-complete dissolution of Ta may be due to high solid ratio and the lack of sufficient acid or reagent to dissolve these metals in the aqueous

162

Chapter 4: Results and Discussions

media. Figures 4.49a and 4.49b show the comparison between the XRD pattern of the parent concentrate and the leach residue (sample C), respectively. a)

Figure 4. 49. XRD patterns of sample C (a) tantalite concentrate and (b) leach residue

163

Chapter 4: Results and Discussions

The main peaks of SiO2 and (Mn,Fe)(Ta,Nb)2O6 were significantly reduced in Figure

4.49b compared to the XRD pattern of the parent concentrate in Figure 4.49a. This indicates that minerals such as hematite, augite and muscovite completely dissolved forming gypsum indicating that silicate gangue mineral structures were extremely damaged by HF. Thus, high elemental dissolution was achieved specifically for Th as shown in Figure 4.48.

More work is needed to investigate the best conditions to dissolve Ta and Nb to 100% in acid media in the presence of CaF2 which can contribute to the chemical treatement of tantalite concentrate. Cheru et al. (2018) reported the removal of 96% U from a tantalite concentrate obtained from Kenticha in Ethiopia by leaching in a mixture of concentrated HF:H2SO4 with a mass ratio of 3:1 using a particle size of (–100+75 μm), at a temperature of 150 oC, for 3 h leaching time. At the same leaching temperature Th dissolution was compromised. High dissolution of 85% Th was obtained by leaching at 170 oC. Comparing the conditions employed in the current study described above, a complete or significant dissolution of U and Th were observed within the first two hours of leaching process compared to three hours leaching time reported by Cheru et al. (2018). In addition, 4 M H2SO4 was used in this study compared to the concentrated mixture of HF:H2SO4 used by Cheru et al. (2018). Furthermore, the leaching temperature of 90 oC used in the current study is far less compared to 170 oC. Thus, the current method used to dissolve U, Th, Ta and Nb from tantalite concentrate can be taken into consideration to treat these type of minerals at a reduced cost in future studies.

164

Chapter 4: Results and Discussions

4.4.14. Summary

This work aimed for the prospect of selective removal of radioactive elements uranium and thorium from the tantalite concentrate through leaching methods. Various methods were evaluated including direct leaching using alkaline, roasting or acid bake prior to leach with or without the addition of reagents under different conditions as described in Chapter 3. It is clear from the results obtained from the direct leaching, alkaline leaching or roasting in the presence of various reagents to subsequently dissolve U and Th was unsuccessful. The highest leaching efficiency of the elements studied under the direct leaching using the acid did not exceed 5% U, except for Th which reached 50%, while Ta and Nb remained unleached as presented in Test 5,

Table 4.16 for sample D. The same trend of high dissolution for only Th was observed under direct leaching in the mixture of H2SO4 and H2O2 as indicated in Test 8, Table 4.16. The leaching efficiencies of the studied elements were 78% Th, 2% U, 2% Ta and 8% Nb, after 12 h of the leaching process. Under direct leaching process, despite long period of leaching, poor extraction of U was recorded.

The acid baking prior to leaching resulted in high leaching efficiency of U and Th within a short period of reaction due to high rate of dissolution. At acid baking temperature of 300 oC,

Test 23, the quantitative analysis of the filtrate revealed leaching efficiency of an average of 89%

U and 47% Th after 15 minutes and the extraction increased as the time progressed to 100% U and 62% Th, after 0.5 h and 3 h, respectively. An average of 7% Ta and 9% Nb were also extracted under the same conditions (Figures 4.38a-d). Under the same conditions, an overall dissolution of 90% Fe, 90% Mn, 50% Al, 60% Li and 60% Cs were recorded through the acid- bake-leach technique (Table 4.35).

165

Chapter 4: Results and Discussions

The chemical assays of the leach residues were ≤ 0.01 wt % U and Th for samples C and G,

(Table 4.38). These values are lower than 0.1 wt % of U and Th each in the exported columbite- tantalite concentrate. From these results, it can be stated that the present project succeeded in removing the radioactive materials from the concentrate to a higher extent that the concentrate can be transported to international market without any concern of radiation exposure.

166

Chapter 5: Summary, Conclusions and Future Works

CHAPTER 5 SUMMARY, CONCLUSIONS AND FUTURE WORK

5.1. Summary and conclusions

The thesis studied the characterisation of the pegmatite ore and the tantalite concentrate, physical concentration of Ta2O5 and Nb2O5 and chemical dissolution of radioactive materials from the concentrate. Eight columbite-tantalite concentrates labelled A to H were subjected to characterisation for qualitative and quantitative analyses after concentration stages at SMB using gravity and magnetic tweezers. Among them, two samples were chosen for quantitative analysis to check the Ta2O5 and Nb2O5 concentration based on fine and coarse particle size fractions. A pegmatite ore sourced at Nyange mining site was subjected to gravity concentration in water at

D2 Bibatama washing station. Three samples were collected to study the possibilities of separation and concentration using physical methods. Chemical treatment through various methods were assessed for the removal of U and Th from the concentrate. These included direct leaching using alkaline or acid medium and acid baking prior to leach with or without the addition of different reagents. The feed materials, bake products and leach residues were characterised using standard techniques such as XRD and SEM-EDS to investigate the changes in mineral phases.

The results from chemical analysis of the tantalite concentrate revealed a significant variation with a range from 16 wt % to about 38 wt % Ta in samples E and B, and 4 wt % to 15 wt % Nb in samples E and G, respectively. The ratios between Ta/Nb (1.3:1.0, 5:1, 4:1, 2.3:1.0,

4:1, 6:1, 2.5:1.0 and 4:1 for A, B, C, D, E, F, G and H), respectively, shows high domination of

Ta compared to Nb (Tables 4.5 and 4.6). This may reflect that the present deposit is mainly a

167

Chapter 5: Summary, Conclusions and Future Works

tantalite ore (Ghorbani et al., 2017). However, a geochemistry study of the whole rock is required to confirm the assumption. High content of about 0.8 wt % U or (8021 g/t) was found in sample

C compared to 45 g/t U in sample A. All the samples contain low levels of Th (5 – 722 g/t).

Thus, thorium was not considered as a penalty element. Traces of elements including Li and Cs, in particular, were analysed and the results revealed that Li content is generally twice that of Cs.

High content of 1.20 wt % Li2O was identified in the fine residues of sample FL after washing to concentrate Ta and Nb.

Minerals including kaolinite, muscovite, diopside, lepidolite and manganotantalite were identified from a selected pegmatite ore (Figure 4.11). The XRD and SEM-EDS scans of eight concentrates for minerals identification has revealed that at the Bisunzu mine, most grains of Ta-

Nb bearing minerals belong to the columbite-tantalite group. The Ta-Nb rich end-members are manganotantalite, ferrotantalite and manganocolumbite. Fluorocalciomicrolite and parabariomicrolite subgroup and pyrochlore were also identified. Quartz and muscovite minerals were identified as impurities in all samples studied (Table 4.8). Grains bearing Ta and Nb are homogenously distributed. Multiple bands of light zones (Ta-rich) with inclusions of dark pockets (Nb-rich) were observed. At the Bisunzu mine fluorocalciomicrolite, parabariomicrolite and pyrochlore phases are the most common minerals carrying the uranium and mainly occur in small grains of diameter less than < 1000 μm. Calcium is the most abundant cation dominating sodium in the microlite. No relationship was found between the increase of U with the content of Ta itself or Ta/Nb ratio nor Ta/(Nb+Ta), as was expected and has been reported in various publications indicating the refractoriness of the tantalite mineral.

168

Chapter 5: Summary, Conclusions and Future Works

A quantitative analysis conducted on different size fractions showed an increase of Ta2O5 and Nb2O5 content with the increase in particle size. For instance, the content of these two oxides in concentrate feed sample D changed to 16.5 wt % Ta2O5 and 8 wt % Nb2O5 in fine fraction of

–1000 μm to 30 wt % Ta2O5 and 20.8 wt % Nb2O5 in coarse fraction of –2800+1000 μm (Figures

4.19a and 4.19b). On the other hand, Ta2O5 content in sample G did not improve, while Nb2O5 increased to about 30 wt % in coarse fraction compared to 20 wt % in fine fraction. The low content of both Ta2O5 and Nb2O5 in fines was attributed to the fact that tantalum mineral is friable

(Adetunji, 2005).

High contents of U and Th were found in the fine particles compared to those in coarse particles. For instance, the content of U was 1023 g/t in fines compared to 670 g/t in coarse fractions (Figures 4.20a-b). The same trend of high Th of 0.9 g/t was recorded in fines compared to 0.1 g/t in coarse fractions for sample D. These results point out the existence of microlite and parabariomicrolite phases carrying the U, naturally to occur in the form of a small grains at the

Bisunzu mine.

The results obtained from magnetic separation and concentration of selected samples resulted in an increase of Ta2O5 by 5% compared to the feed of selected sample (Figure 4.21).

However, it was found that majority of grain particles carrying Ta2O5 and Nb2O5 of selected samples were reported to non-magnetic portion. For instance, the content of Ta2O5 and Nb2O5 in magnetic fraction were 9.8 wt % and 4.9 wt % compared to 49.6 wt % and 8.6 wt % in non- magnetic fraction (Figures 4.21a and 4.21b). On the other hand, Fe and Ti were highly concentrated in the magnetic fraction (50.6 wt % and 8.7 wt %), while only 1.44 wt % and 0.13 wt % were recorded in non-magnetic fraction (Figures 4.21c and 4.21d). High contents of U and

Th were also present in non-magnetic fractions. For example, U and Th grades were 5717 g/t

169

Chapter 5: Summary, Conclusions and Future Works

and 335 g/t in the non-magnetic fraction compared to 154 g/t and 20 g/t in the magnetic fraction

(Figures 4.22a and 4.22b). The results obtained in this work point out the successful concentration by separation of magnetic materials and non-magnetic materials with the Ta, Nb,

U and Th enriched in the non-magnetic fractions.

The final step of the objectives of the present study was the selective removal of the radioactive materials from tantalite to obtain a concentrate free of uranium and thorium. It is clear from the results obtained that the direct leaching using alkaline and various acids did not leach out the U and Th to significant extent. A 78% Th and 2% U leaching efficiencies were achieved in sample C in a mixture of H2SO4 and H2O2 after a 12 h leaching time using the conditions 100 oC, 250 rpm, PSD of < 75 μm (Test 8, Table 4.21). The low extraction efficiencies of U in the acid or alkaline media under direct leaching can be attributed to the presence of U in the refractory mineral phases fluorocalciomicrolite and parabariomicrolite.

A combination of both bake and leach in the presence of sulfuric acid proved to be the most effective technique in the removal of the majority of the uranium and a reasonable amount of Th. The concentrate after being baked with sulfuric acid, Nb2O5 was converted into niobium sulfate Nb2O3(SO4)2.0.25H2O and proven by the analysis obtained using XRD and SEM-EDS

(Table 4.36) indicating the possibility of increasing the solubility of various sulfates including uranium and thorium. The use of acid bake at 300 oC for 2 h, solid mass to acid volume ratio of

o 625 g/L and < 75 μm as particle size for bake, and temperature: 90 C, 4 M H2SO4, S/L: 12.5 g/L, time: 3 h, and K2S2O8 mass to liquid volume ratio: 27 g/L for leaching processes were the most suitable conditions for the removal of U and Th. Under these conditions, the removal of approximately 100% U and 62% Th were recorded while an average of 7% Ta and 9% Nb entered

o the leach liquor (Figure 4.38). Low temperature leaching with H2SO4/K2S2O8 at 40 C after acid

170

Chapter 5: Summary, Conclusions and Future Works

bake at 400 oC, on the other hand, resulted a leaching efficiency of 100% U from samples H and

D as well as 41% and 76% Th (Table 4.28). The high dissolution of U and Th at low leaching temperature may be due to the absence of zircon in concentrates H and D.

This project has shown that the concentrate from different mining sites at SMB are associated with various minerals as impurities containing Fe and Ti which inhibit the increase in

Ta and Nb grades. Generally, it can be concluded that the dissolution of tantalite concentrate initially involves the acid bake where the phase carrying Ta, Nb and U is decomposed to form complex oxide sulfate products soluble in water/diluted H2SO4. The low extraction of U in some cases at low temperatures can be attributed to their existence in primary refractory phases. A relatively complete acid decomposition of the columbite-tantalite was achieved through a direct sulfuric acid leaching of unbaked concentrate in the presence of CaF2. Leaching efficiencies of

100% U, 81% Ta, 43% Nb and 90% Th were recorded (Figure 4.48). This means that almost all the minerals carrying U, Th, Ta and Nb as well as silicate minerals within the concentrate were heavily decomposed in the presence of HF.

5.2. Recommendations for future work

It is important to undertake further investigations to define more and in depth mineralogical analysis of samples from other mining sites at SMB. This will help to understand the effect of the mineralogy on the leaching efficiencies to remove radioactive materials. In the

o present study, the leaching efficiency of Th was likely to reach its maximum in H2SO4 at 40 C and KOH media at 100 oC after a long period of reaction compared to uranium using the direct leach. The low loss of Ta and Nb during direct leach warrants further investigations to maximise the recovery of U as well as to minimise the loss of Ta and Nb. Using the acid baking followed

171

Chapter 5: Summary, Conclusions and Future Works

by leaching with acid causes a loss of an average 7% Ta and 9% Nb. Further studies are required to recover those valuable metals from the leachate. Extension of this work to chemical treatment to separate the Ta2O5 and Nb2O5 as well as Ta and Nb as metals is essential to produce the products ready for end-users. Lepidolite, as source of lithium, was identified to exist in SMB deposit, however, further research is needed to determine its economical recoverability. Finally, adjustment of parameters to decompose the columbite-tantalite concentrate in sulfuric acid in the presence of CaF2 warrant further studies to optimise the conditions for elemental leaching efficiency and separation.

172

References

REFERENCES

Adetunji, A., Siyanbola, W.O., Funtua, I.I, Afonja, A.A., Adewoye, O.O., 2005. Assessment of beneficiation Routes of Tantalite Ores from Key Location in Nigeria. Journal of Minerals & Materials characterization & Engineering. Vol. 4, No. 2, pp. 85-93.

Agulyansky, A., 2004. The Chemistry of Tantalum and Niobium Fluoride Compounds. Elsevier Amsterdam, pp. 22-24.

Agulyansky, A., 2004. The Chemistry of Tantalum and Niobium Fluoride Compounds. pp. 4- 13, 25, 51, 84, 256-259, 318.

Agulyansky, A., 2004. The Chemistry of Tantalum and Niobium Fluoride Compounds. Elsevier Amsterdam, pp. 11-21.

Agulyansky, A., Agulyansky L.V., Travkin, F., 2004. Liquid-liquid extraction of tantalum with 2-octanol. Chemical Engineering and Processing 43 (002), pp. 1231-1237.

Alfonso, P., Oliva, J., Garcia-Valles., M., Tomasa, O., Calvo, D., Guasch, E., Anticoi, H., Parcerisa, D., 2016. Mineralogical Characterisation of Tantalum Ores for the Optimization of their Processing. International Journal of Mining, Materials and Metallurgical Engineering, Vol. 2, DOI: TBD.

Alhassan, Y., Tsafe, A.I., Birnin-Yauri, U.A., Okunola, O.J., Yargamji, G. I., Yebpella, G.G., Ndana, M., 2010. EDXRF analysis of tantalite deposit of Mai-Kabanji, North-western Nigeria. Journal of Environmental Chemistry and Ecotoxicology, Vol. 2(7), pp. 117-119.

Amuda, M.O.H., Esezobor, D.E., Lawal, G.I., 2007. Adaptable Technologies for life- Cycle Processing of Tantalum Bearing minerals. Journal of Minerals & Materials characterization & Engineering. Vol. 6, No. 1, pp. 69-77.

Andrade, M.B., Atencio, D., Persiano, A.I.C., Ellena, J., 2013. Fluorocalciomicrolite, (Ca,Na)2Ta2O6F, a new microlite-group mineral from Volta Grande pegmatite, Nazareno, Gervais, Brazil. Mineralogical Magazine, Vol. 77(7), pp. 2989-2996.

Asselin, E., Ahmed, T.M., Alfantazi, A., 2007. Corrosion of niobium in sulphuric and hydrochloric acid solutions at 75 and 95 oC. Corrosion Science 49, 694-710.

Ayanda, O.S., Adekola, F.A., 2011. A Review of Niobium-Tantalum separation in Hydrometallurgy. Journal of Minerals & materials characterisation & Engineering, Vol. 10, No 3, pp. 245-256.

Babko, A.K., Lukachina, V.V., Navibanets, B.I., 1963. Solubility and acid-base properties of tantalum and niobium hydroxides. Russ. J. Inorg. Chem. 8, 957-961.

173

References

Balasubramaniam, A., 2017. Gravity separation in ore dressing. Centre for Advanced Studies in Earth Science. University of Mysore, Mysore. DOI: 10.13140/RG.2.2.15278.31045.

Berhe, G.G., Alberto, V.R., Yimam, A., Woldetinsae, G., Tadesse, B., 2017, Alternative beneficiation of tantalite and removal of radioactive oxides from Ethiopian Kenticha pegmatite–spodumene ores, Int. J. Min. Met. Mater. 24, 727–735.

Bhargava, K.S., Ram, R., Pownceby, M., Grocott, S., Ring, B., Tardio, J., Jones, L., 2015. A review of acid leaching of uraninite. Hydrometallurgy, Vol. 151, pp. 10-24.

Bielecki, E. J., Romberger, K.A., Bakke, B.F., Hobin, M.A., Clark, C.R., 1991. Recovery of metal values and hydrofluoric from tantalum and niobium waste sludge. U.S. Patent 5023059.

Bludssus, W., Eckert, J., 1993. “Process for the recovery and separation of tantalum and niobium”, US Patent 5209910, assigned to Hermann, C. Starck Berlin, Berlin DE.

Bock, R., 1979b. Decomposition methods in analytical chemistry. Translated and revised by L.L. Marr. T. & A. Constable, Edinburgh, Great Britain, p. 196.

Bose, D.K., Gupta, C.K., 2008. Extractive Metallurgy of Tantalum. A Review: Mineral processing and Extractive Metallurgy, Taylor & Francis 22:4-6, 389-412.

Bose, D.K., Gupta, C.K., 2002. Extractive Metallurgy of Tantalum. Taylor &Francis Group, Processing and Extractive Review, 22:4-6, pp. 389-412.

Bradley, D.C., McCauley, A.D., Stillings, L. M., 2017. Mineral –Deposit Model for Lithium- Caesium-Tantalum pegmatites. USGS. Scientific Investigations Report -2010-5070-0.

British Geological survey (BGS), (2011). Niobium-Tantalum. Natural environment research Council. Minerals UK Center for sustainable mineral development.

Bulatovic, M.S., 2010.Handbook of floatation Reagents, Available from: https://books.google.com.au/books?id=ewXC_EtZGhwC&printsec=frontcover&dq=Flo tation+of+Gold,PGM+AND+oXIDE+MINERALS,page+127&hl=en&sa=X&ved=0ah UKEwiLr_nmloDVAhWLjLwKHUxhBHsQ6AEIJjAA#v=onepage&q&f=false ,Accessed on 11th July 2017.

Burt, R.O., 1984. Gravity Concentration Technology. Elsevier Science BV, Amsterdam, p 607.

Burt, R., 1996. “Beneficiation of Tantalum Ore. How it is Achieved and Could it be Better?”; Proceedings of 125th TMS Annual Meeting and Exhibition; Anaheim, CA. 17-22.

Cano, M. E., Cordova-Frage, T., Sosa, M., Bernal-Alvarado, J., Baffa, O., 2008. Understanding the magnetic susceptibility measurements by using an analytical scale. Eur. Phys., Vol. 29, pp. 345-354.

174

References

Cardon, P.B., 1962. Process for recovering niobium and tantalum from ores and ores concentrates containing same, U.S. Patent 3058825.

Casas, I., De Pablo, J., Giménez, J., Torrero, M. E., Bruno, J. Cera, E., Finch, R.J., Ewing, R.C., 1998. The role of Eh, pH, and carbonate on the solubility of UO2 and uraninite under nominally reducing conditions. Geochemical et Cosmochimica Acta, Vol. 62, No. 13, pp. 2223-2231.

Cẽrný, P., Ercit, T.S., 1989. Mineralogy of niobium and tantalum: crystal chemical relationships, paragenetic aspects and their economic implications. Springer-Verlag Berlin Heidelberg.

Cẽrný, P., Ercit, T.S., 2005. The classification of Granites Pegmatites. A review. The Canadian Mineralogist, Vol. 43, pp. 2005-2026.

Cheru, M.S., Del Rosario, A.V., Yiman, A., Tadesse, B., and Berhe, G., 2018. Hydrometallurgical removal of uranium and thorium from Ethiopian tantalite ore. Physicochemical Problems of Mineral Processing.

Costine, A., Nikoloski, A.N., Dacosta, M., Hackl, R., 2013. Uranium extraction from a pure brannerite mineral by acidic ferric sulphate leaching. Mineral Engineering, Vol. 53, pp. 80-90.

Cullis, A.G., Beanland, R., 1999. Microscopy of semiconducting materials. Proceedings of the Institute of Physics Conference held at Oxford.

Damodaran, A.D., Deshpande, S., Majmudar, A.A., Sastri, M.S., Jena, P.K., Bose, D.K., Chandra, C.K., Gupta, A.K., Tanja, 1969. Extraction and utilization of pure niobium and tantalum from Indian ores. Trombay, Bombay 85, Vol. 36, A, No. 5.

De Moura, J.C., 2009. Radioactivity Pegmatites of the Parelhas Region, Rio Grande Do Norte, NorthEast Brazil. Preliminary Investigation on RadioMinerals and Radioactive Minerals. International Nuclear Atlantic Conference-INAC; ISB: 987-85-99414-03-8.

De Paolo, D.J., 1988. Neodymium Isotope Geochemistry. Springer-Verlag, Berlin.

Deblonde, G.J.P., Chagnes, A., Bélair, S., Cote, G., 2015. Solubility of niobium(V) and tantalum(V) under mild alkaline conditions. Hydrometallurgy, Vol. 156, pp. 99-106.

Deblonde, J.P.G., Bengio, D., Denis Beltrami, D., Sarah Bélair, S., Cotea, G., Chagnes, A., 2019. Niobium and tantalum processing in oxalic-nitric media: Nb2O5·nH2O and Ta2O5·nH2O precipitation with oxalates and nitrates recycling. Separation and Purification Technology, Vol. 226, pp. 209-217.

Demol, J.H.E., Senanayake, G., 2018. Sulfuric acid baking and leaching of rare earth elements, thorium and phosphate from a monazite concentrate: Effect of bake temperature from 200 oC to 800 oC. Hydrometallurgy, Vol. 179, pp. 254-267.

175

References

Deveau, C., 2006. Improving Fine Particle Gravity Recovery through Equipment Behaviour Modification. Tantalum Mining Corporation of Canada. 38th Annual Meeting of the Canadian Mineral Processor, Ontario, Canada.

Deveau, C., Young, S.R., 2004. Pushing the limits of Gravity separation. Tantalum Mining Corporation of Canada, Ltd.

Dewaele, S., Hulsbosch, N., Cryns, Y., Boyce, A., Burgess, R., Muchez, P.H., 2016. Geological setting and timing of the world-class Sn, Nb-Ta and Li mineralisation of Manono- Kitotolo (Katanga, Democratic Republic of Congo). Ore Geology reviews, Vol. 72, 373- 390.

Dickin, A., 1995. Radiogenic Isotope Geochemistry. Cambridge University Press.

Ditz, R., Sarbas, B., Schubert, P., Töpper W., 1987. Gmelin Handbook of Inorganic Chemistry. Springer-Verlag Berlin Heidelberg GmbH 1990.

Ekberg C., Brown, P.L., 1989. Hydrolysis of metal ions. WILEY-VCH.

El-Hazek, M.N., Amer, T.E., Abu El-Azm, M.G., Issa, R.M., El- Hady, S.M., 2012, “Liquid- liquid extraction of tantalum and niobium by octanol from sulfate leach liquor.” Arab Journal Chemical, Vol. 5, pp. 31–39. doi: 10.1016/j.arabjc.2010.07.020.

El-Hussaini, O.M., Mahdy, A. M., 2002. Sulfuric acid leaching of Kab Amiri niobium-tantalum bearing minerals, Central Eastern Desert, Egypt. Hydrometallurgy, Vol. 64, pp. 219-229.

El-Hussaini, O. M., 2001. Extraction of niobium and tantalum from nitrate and sulfate media by using MIBK [J]. Mineral Processing and Extractive Metallurgy Review, Vol. 22, pp. 633−650.

El-Hussaini, O.M., Rice, N.M., 2004. Liquid–liquid extraction of niobium and tantalum from aqueous sulphate/fluoride by a tertiary amine. Hydrometallurgy, Vol. 72, pp. 259-267.

El-Hussaini, O.M., El-Hazek, M.N., 2004. Removal of radioactive elements from niobium and tantalum ores. The European Journal of Mineral Processing and Environmental Protection, Vol. 5, No. 1, 1303-0868, pp. 7-16.

Ercit, T.S., 1986. The Simpsonite Paragenesis: The Crystal Chemistry and Geochemistry of Extreme Ta Fractionation. A Phd thesis presented to the University of Manitoba, Canada.

European commission (EC), 2002. Practical use of the concepts of clearance and exemption, Part II, application of the concepts of exemption and clearance of natural radiation sources Radiation Protection. Vol, 122, Part II (Luxembourg: EC).

Falconner, A., 2003. Physical Separation in Science and Engineering, Vol. 12, No. 1, pp. 31–48 Gravity Separation: Old Technique/New Methods.

176

References

Fiedor, N.J., Bostick, D.W., Jarabek, J.R., Farrell, J., 1998. Understanding the mechanism of uranium removal from groundwater by zero-valent Iron using X-ray photoelectric spectroscopy. Environ. Sci, Technol. Vol. 32, pp. 1466-1473.

Frondel, J.W., Fleischer, M., 1955. Glossary uranium-thorium bearing minerals. Third edition. Geological Survey Bulletin 100 - F.

Gaballah, E.A., Djona, M., 1997. Extraction of Tantalum and Niobium from Tin slags by chlorination and carbonisation. Metallurgical and Materials Transactions B, Vol. 28B, p. 359.

Gebreyohannes, B.G., Alberto, V.D., Yiman, A., Woldetinsae, G., Tadesse, B., 2017. Alternative beneficiation of tantalite and removal of radioactive oxides from Ethiopian Kenticha pegmatite-Spodumene ores. International Journal of Minerals, Metallurgy and Materials, Vol. 24, No. 7, pp. 727.

Gherman, G.D., 2010, Organothiophosphoric Ligands-Agents for metal ions separation ions separation, [Phd thesis]. Cluj-Napoca: “abe-olyai” University; 2010 [Cited 2011 Oct 27].

Ghorbani, Y., Fitzpatrick, R., Kinchington, M., Rollinson, G., Hegarty, P., 2017. A Process Mineralogy Approach to Gravity Concentration of Tantalum Bearing minerals. Minerals, Vol. 7, 194; doi: 10.3390/min7100194.

González, T.L., Polonio, F.G., López Moro, F.J., Fernández Fernández, A., Sanz Contreras, J. L.C., Candelas Moro Benito, M., 2017. Tin-tantalum-niobium mineralization in the Penouta deposit (NW Spain): Textural features and mineral chemistry to unravel the genesis and evolution of cassiterite and columbite group minerals in a peraluminous system. Ore Geology Reviews, Vol. 81, pp. 79-95.

Guillamont, R., Fanghäned, J., Neck, V., Fuger, J., Palmer, D. A., Grenthe, I., Rand, M. H., 2003. Chemical Thermodynamics. Vol. 5. An update on the Chemical Thermodynamics of Uranium, Neptunium, Plutonium, Americium and Technetium (OECD, NEA-TDB), Elsevier, North Holland, Amsterdam.

Guo, H., Kuang, G., Wan, H., Yang, Y., Yu, H.Z., Wang, H. D., 2018. Enhanced treatment to extract lithium from the lepidolite with a fluorine–based chemical bond. Hydrometallurgy, Vol. 183, pp. 9-19. Gupta A., Yan, D., 2016. Mineral processing design and operations. An introduction. Second edition. ISBN: 978-0-444-63589-1.

Gupta, C.K., Suri, A.K., 1993. Extractive metallurgy of Niobium. ISBN O-8493-6071-4, pp. 93- 159.

Gupta, C.K., Suri, A.K., 1994. Extractive Metallurgy of Niobium, Tantalum and Vanadium, CRC Press, Boca Raton, Florida, USA.

177

References

Hanson, S.L., Simmons, W.B., Falster, A.U., Food, E.E., Lichte, F.E., 1999. Proposed nomenclature for samarskite-group minerals: new data on ishikawaite and calciosamarskite. Mineral. Mag. 63, pp. 27-36.

International Atomic Energy Agency (IAEA), 1990. Manual on laboratory testing for uranium ore processing. Technical Reports Series No. 313.

International Atomic Energy Agency (IAEA), 1993. Uranium Extraction Technology, Vienna. Technical reports series No. 395.

International Atomic Energy Agency (IAEA), 2001. Manual of acid in situ leach uranium mining technology. IAEA-TECDOC-1239.

International Atomic Energy Agency (IAEA), 2004. Application of the concepts of Exclusion, Exemption and Clearance, Safety Series No. RS-G-A.7 (Vienna: IAEA).

Jerome, A., Antoine, G., Fatima, H., Guillaume L., Jacques T., 2016. Dissolution and recovery of at least one element NB or Ta and of at least one other element U or Rare earth elements from ores and concentrates. W O 2012/093170 A1.

Kabangu, M.J., and Crouse, P.L., 2012. Separation of niobium and tantalum from Mozambican tantalite by ammonium bifluoride digestion and octanol solvent extraction. Hydrometallurgy, pp. 129-130, 151-155.

Kim, E., Osseo-Asare, K., 2012. Aqueous stability of thorium and rare earth metals in monazite hydrometallurgy: Eh-pH diagrams for the systems Th-, Ce-, La-, Nd-, (PO4)-, (SO4)-H2O at 25 oC. Hydrometallurgy.

Knorring, O.V., Vorma, A., Nixon, P.H., 1969. Rankamaite, A New Tantalum Mineral from Kivu, Bull. Geol. Soc. Finland, Vol. 41, pp. 47-56.

Knorring,V.O., 1958. Niobium and tantalum minerals. Communs Geol.Surv. S.W. Afr/Namibia, vol. 1, pp. 85-88.

Korneliussen, A., McEnroe, S.A., Nilsson L.P., Schiellerup H., Gautneb H., Meyer G.B., Størseth, L.R. 2000. An overview of titanium deposits in Norway. Norges geologiske undersøkelse Bulletin, Vol. 436, pp. 27-38.

Küster, D., 2009. Granitoid - hosted Ta mineralisation in the Arabian – Nubian Shield. Ore deposits, tectono-metallogenetic setting and petrogenetic framework. Ore geology reviews, Vol. 35, pp. 68-86.

Linnen R., Trueman, D.L., Burt, R., 2014. Tantalum and niobium. Critical elements Handbook, First edition, John Wiley &Sons, Ltd.

Lottering, M.J., Lorenzen, L., Phala, N.S., Smit, J.T., Schalkwyk, G.A.C., 2008. Mineralogy and uranium leaching response of low grade South Africa ores, Miner. Eng. 21(1), 16-22.

178

References

Lumpkin, G.R., Ewing, R.C., 1995. Geochemical alteration of Pyrochlore group minerals: Pyrochlore subgroup. American Mineralogist, Vol. 80, pp. 732-743.

Lunt, D., Boshoff, P., Boylett, M., El-Ansary, Z., 2007. Uranium Extraction the key Process Drivers, The southern African Institute of mining and Metallurgy. -Namibian Branch- Uranium in Namibia, Vol. 107.

Mackay, D.A.R., Simandl, G.J., 2015. Niobium and tantalum: Geology, markets, and supply chains. British Colombia Geological Survey Paper. Vol. 3, pp. 13-22.

Mathur, H.D., Tandon, O.P., 1986. Chemistry of Rare Elements S. Chand and Company, New Delhi, p. 93.

McMaster, S.A, Ram, R., Charalambous, F., Pownceby, M.I., Tardio, J., Sureh, Bhargava, S.K., 2015. Characterisation and leaching studies on the uranium mineral betafite. Minerals engineering, Vol. 81, pp. 58-70.

Melcher, F., Graupner, T., Gäbler, E.H., Sitnikova, M., Henjes-Kunst, F., Oberthür, T., Gerdes A., Dewaele, S., 2015. Tantalum–(niobium–tin) mineralisation in African pegmatites and rare metal granites: Constraints from Ta–Nb oxide mineralogy, geochemistry and U–Pb geochronology. Ore Geology reviews, Vol. 64, pp. 667-719.

Melcher, F., Graupner, T., Henjes-Kunst, F., Oberthür, T., Sitnikova, M., Gäbler, E., Gerdes, A., Brätz, H., Davis, D., Dewaele, S., 2008. Analytical Fingerprint of Columbite-tantalite (Coltan) mineralisation in Pegmatites-Focus on Africa. Ninth International Congress for Applied Mineralogy, Brisbane, QLD.

Merritt R.C., 1971. The Extractive Metallurgy of Uranium, Colorado School of mines. United States Atomic Energy Commission. First print, Library of Congress Catalog Card No. 71-157076.

Miller, G.L., 1995. Tantalum and Niobium; Butterworths Scientific Publications, London, p. 12.

Mineral data information. Available from: http://webmineral.com/data/Tantalite- (Mn).shtml#.XDf5EVUza71 & http://chinaneolithic.com/en/mineral/photos.asp?ID=455

Murata, K.J., Rose, H.J., Carron, V.I.K., Glass J. J., 1957. Systematic variation of rare-earth elements in cerium-earth minerals. Geochim. Cosmochim. Acta, vol.11, pp. 141- 161.

Mustapha, A.O, Mbuzukongira, P., Mangala, M.J., 20007. Occupational radiation of artisans mining columbite-tantalite in the Democratic Republic of Congo. Online at stacks.iop.org/JRP/27/18.

Neck, V., Altmaier, M., Müller, R., Bauer, A., Fanghänel, Th., Kim, J.I., 2003. Solubility of crystalline thorium dioxide. Radiochim. Acta, Vol. 91, pp. 253-262.

Nete, M., 2013. Separation and purification of niobium and tantalum from synthetic and natural compounds, [Phd thesis]. University of Free State.

179

References

Nete, M., Koko, F., Theron, T., Purcell, W., Nel, T.J., 2014a. Primary beneficiation of tantalite using magnetic separation and acid leaching. International Journal of Minerals, Metallurgy and Materials, Vol. 21, No. 12, p. 1153. DOI: 10.1007/S12613-014-1022-6.

Nete, M., Purcell, W., Nel, J.T., 2014b. Separation and isolation of tantalum and niobium from tantalite using solvent extraction and ion exchange. Hydrometallurgy, Vol. 149, pp. 31- 40.

Nete, M., Purcell, W., Nel, T.J., 2016. Hydrometallurgical Separation of Niobium and Tantalum. JOM, Vol.68, No. 2.

Nete, M., Purcell, W., Snyders, E., J.T. Nel, J.T., Beukes, G., 2012. Characterization and alternative dissolution of tantalite mineral samples from Mozambique. The Journal of the Southern African Institute of Mining and Metallurgy, Vol. 112.

Nete, M., Purcell, W., Snyders, E., Nel, J.T., 2010. “Alternative dissolution methods for analysis of niobium containing samples” S. Afr. J. Chem. Vol. 63, pp. 130-134.

Nettleton, K.C.A., Nikoloski, N.A., Da Costa, M., 2015. The leaching of betafite. Hydrometallurgy, vol. 157, pp. 270-279.

Nikishina, E.E., Drobot D.V., Lebedeva E.N., 2013. Niobium and tantalum. Metallurgy of rare and noble metals. Allerton press, Inc., Vol. 54, No.6, pp. 446-452.

Nikishina, E.E., Drobot D.V., Lebedeva E.N., 2014. Niobium and tantalum. Metallurgy of rare and noble metals. Allerton press, Inc., Vol. 55, No. 2, pp. 130-140.

Nogueira, C.A., Margarido, F., Viecela, N., Durâo, F., Guimarâes, C., 2014. Comparison of Processes for Lithium Recovery from Lepidolite by H2SO4 digestion or HCl leaching. Proceedings of the International Conference on Mining, Material and Metallurgical Engineering, Prague, Czech Republic, paper No. 72.

Othmer, K., 1997. Encyclopedia of Chemical Technology. 4th ed. New Jersey: John Wiley & Sons, Inc. p. 658.

Ovaskainen, R., 1999. The determination of minor isotope abundances in naturally occurring uranium materials. The tracing power of isotopic signatures for uranium. Helsinki. F19900155.

Pappy, F.J., 2012. Niobium (Columbium) and Tantalum advanced release. United States Geological survey. Minerals Handbook.

Partington, G. A., McNaughton N.J., Williams I.S., 1995. Mineralisation, and Geochronology of the Greenbushes pegmatite, Western Australia. A Review of the Geology. Vol. 90, pp. 616-635. Pilbara Minerals Limited Open Report, 2015. Pilbara Reports Update Mineral Resource for Tabba Tabba Tantalum Project, WA. ASX Announcement, CAN 112-425-788.

180

References

Pilbara Minerals Ltd Open Report, 2013. Pilbara Secures Pathway to production with agreement to develop and mine advanced WA Tantalum project.

Pohl,W.L., 2016. Economic Geology of Niobium and Tantalum. Retrieved: https://www.walter- pohl.com/sample.html.

Popova, E.A., Zalessky, V.G., Yakovenchuk, V.N., Krivovichec, S.V., 2014. Loparite-(Ce), a natural ferroelectric with the perovskite-type structure. Applied Physics Letters 105, 082905-1.

Pourbaix, M., 1966. Atlas of Electrochemical Equilibria in Aqueous Solutions, Pergamon Press, New York.

Raslan, M.F., Fawzy, M.M., 2017. Comparative mineralogy and magnetic separation characteristics of Nb-Ta oxide minerals from rare-metal pegmatite and stream sediments, Eastern Desert and Sinai, Egypt. International Journal of Innovative Science, Engineering & Technology, Vol. 4, Issue 4, ISSN (Online), pp. 2348-7968.

Robin, A., Rosa, L.J., 1999. Corrosion behaviour of niobium, tantalum and their alloys in hot hydrochloric and phosphoric acid materials. International Journal of Refractory Metals & Hard materials, Vol. 18, pp. 13-21.

Rodriguez, M.H., Rosales, G.D., Pinna, E.G., Suarez, D.S., 2015. Extraction of niobium and tantalum from ferrocolumbite by hydrofluoric acid pressure leaching. Hydrometallurgy, Vol. 156, pp. 17-20.

Rosenblum, S., Brownfield, I.S. Magnetic Susceptibilities of Minerals. U.S. Geological Survey, Open-File Report 99-529.

Roskill Information, https://roskill.com/market-report/niobium/, 2018.

Roskill information, The Economics of Niobium, 12th ed., Roskill Information Services Ltd., London, 2017.

Schulz, J.K., Nadine M.P., John F.P., 2017. Niobium and Tantalum. Critical Mineral Resources of the United States –Economic and Environmental Geology and Prospects for future Supply. U.S. Geological Survey Reston, Virginia, Professional paper 1802-M.

Selected mineral resources map of the DRC. Retrieved on 21th May 2018. Available from: https://www.google.com.au/search?q=An+updated+mines+map+of+DRC&tbm=isch&s ource=iu&ictx=1&fir=BuIgQeMXe1-_6M%253A%252CZ1EsCARYcHQ- dM%252C_&usg=__sg8KeW6YXxIYUyl8Pcfxst3axC4%3D&sa=X&ved=0ahUKEwit 7Yj49ZXbAhWMUrwKHQf-DOUQ9QEIKTAA#imgrc=FgU-N8rYW1S4vM.

Siderenko, G.A., Doynikova O.A., 2010. To mineralogy of tetravalent uranium. New data on Minerals, Vol. 45.

181

References

Simandl, G.J., Burr, R.O., Trueman, D.L., Paradis, S., 2018. Tantalum and Niobium: Deposits, Resources, Exploration Methods and Market – A Primer for Geoscientists. Journal of the Geological Association of Canada, Vol. 45, pp. 85-95.

Singh, P.K., 2007. Tantalite exploration in “Block A”. Trabajos de Geologίa. Univ. de Oviedo, Vol. 27, pp- 41-69.

Singh, R., Bhattacharyya, K.K., Maulik, S.C., 1997. Gravity concentration of fines and ultrafines. National Metallurgical Laboratory, Jamshedpur- 831007, pp. 40-56.

SMBdepositlocation:https://www.google.com.au/maps/place/NorthKivu,+Democratic+Republi c+of+the+Congo/@-1.4465666,28.672993,11.25z/data=!4m5! 3m4!1s0x19e0865560e56fc3:0x341e4dce5d8e1b94!8m2!3d- 0.7917729!4d29.0459927!5m1! 1e4.

Sole C.K., Cole, P.M., Feather, A.M., Kotze, M.H., 2011. Solvent Extraction and Ion Exchange Applications in Africa’s Resurging Uranium Industry: A Review, 29:5-6, 868-899, DOI: 10.1080/07366299.2011.581101.

Sweetaple, M.T., Holmes, J., Young, J., Grigson, M.W., Barnes, L., Till, S., 2017. The geology and mineralogy of the pilgangoora Lithium-Tantalum Pegmatite Deposit. Australian Ore Deposits Monograph.

Tantalum-Niobium International Study Center (TIC), 2016. www.tanb.org

Tantalum-Niobium International Study Center (TIC), 2017. www.tanb.org

Tantalum-Niobium International Study Center (TIC), 2018. www.tanb.org

The National Academies Press, 2012. Uranium Mining in Virginia: Scientific, Technical, Environmental, Human Health and Safety, and Regulatory Aspects of Uranium. 500 Fifth Street, NW Washington, DC 20001.

Theron, T.A., Nete, M., Venter, J.A., Purcell, W., Nel, J.T., 2011. Dissolution and quantification of tantalum containing compounds: Comparison with niobium. S. Afr. J. Chem., Vol. 64: pp. 173-178.

Toure, M., Arrachat, G., Duhamet, J., Pellet-Rostaing, S., 2018. Tantalum and Niobium Selective Extraction by Alkyl-Acetophenone. Metals, 8, 654, doi: 10.3390/met8090654.

US Geological Survey, 2015. Shift in Global Tantalum Mine Production, Science for a changing world.

US Geological survey, 2016. Mineral Commodity Summaries. Science for a changing world.

US Geological survey, 2017. Critical Mineral Resources of the United States.Economic and Environmental Geology and Prospects for Future Supply.

182

References

US Geological survey, 2018. Mineral Commodity Summaries. Science for a changing world.

US Geological survey, 2019. Mineral commodity Survey. Science for a changing world.

Venter, R., Boylett, M., 2009. The evaluation of various oxidants used in acid leaching of uranium. Hydrometallurgy conference. The Southern African Institute of Mining and Metallurgy.

Viruet, Y.S., Menzie W.D., John F. Papp, F.J., Thomas R. Yager, R.T, 2013. An Exploration in Mineral Supply Chain Mapping Using Tantalum as an Example. US. Geological Survey. Open-File Report 2013–1239.

Wang, R.C., Hu,H., Zhang, A.C., 2004. Pollucite and the cesium-dominant analogue of polylithionite as expressions of extreme Cs enrichment in the Yichun Topaz-Lepidolite Granite, Southern, China. The Canadian Mineralogist, Vol. 42, pp. 883-896.

Wang, X., Zheng, S., Xu, H., Zhang, Y., 2009. Leaching of niobium and tantalum from a low- grade ore using a KOH roast–water leach system. Hydrometallurgy, Vol. 98, pp. 219- 223.

Weil, B., 2012. Uranium Mining and Extraction from ore. Submitted as a course work. Stanford University.

Welham, N.J., 2001. Canadian Metallurgical Quarterly. Vol. 40 (2), pp. 143-154.

Wills, B.A., 1992. Mineral Processing Technology. 5th edition. Pergamon Press.

Yan, D. and Connelly, D., 2008. Implications of Mineralogy on Uranium Ore Processing. ALTA 2008 Uranium Conference, Perth, Western Australia.

Zelikman, A.N., Orekhov, M.A., 1965. Decomposition of tantalum concentrates by sodium and potassium hydroxides at increased temperatures and pressures. Izvestia Akademii Nauk SSSR. Metally, Vol. 6, pp. 38-45.

Zhang, J., Zhao B., Schreiner B., 2016. Separation hydrometallurgy of Rare Earths Elements. Vol. XVI, pp. 104 Illus., 99. Illus. In colour, Hardcover. ISBN: 978-3-319-28233-6.

Zhu, Z., Cheng, C.Y., 2011. Solvent extraction technology for the separation and purification of niobium and tantalum: A review. Hydrometallurgy, Vol. 107, pp. 1-12.

183

Appendix

APPENDIX

Appendix A 1

Table A 1.1. Specification for Nb2O5, Nb metal and Ta2O5 grades and impurities (%) for various applications.

Nb2O5 Nb metal Ta2O5 Ceramics Optical High FeNb Reactor Standard Optical Element Grades Grade purity Grade (%) Grade Nb-Zr Grade (%) Grade (%) (%) (%) Grade alloy (%) (%) Nb > 99.9 > 99.9 > 99.9 min.60 - 0.30 0.008 Ta 0.02 0.02 0.002 0.50 0.1 > 99.5 > 99.99 Al 0.0005 0.0005 0.0003 2.00 0.002 0.015 0.001 Ca 0.001 0.0002 - - - - - Co 0.0005 0.0002 - - 0.002 - - Cr 0.0005 0.0005 0.0003 0.10 0.002 - 0.001 Cu 0.0005 0.0003 - - - - 0.001 F 0.03 0.0005 - - - 0.15 0.015 Fe 0.0005 0.0005 0.0003 - 0.005 0.03 0.001 Mn 0.0005 0.0005 0.0003 - - 0.005 0.0005 Hf 0.001 - - - 0.02 - - Ni 0.0005 0.0005 0.0003 - 0.005 - 0.001 Zr - - - - 0.8 to 1.2 Si 0.005 0.005 0.001 2.50 0.005 0.05 0.005 Ti 0.0005 0.0005 0.0003 0.10 - 0.03 0.0005 C - - - 0.15 0.01 - - P - - - 0.20 - - - S - - - 0.50 - - - Mo - - - - 0.02 - 0.001 W - - - - 0.03 - 0.001 Average 0.6-1.0 m 1.0-1.6 m 1.0-3.0 m - - - - particle Size x 20.6

Viruet et al., 2013; Agulyansky, 2004

184

Appendix

Table A 1. 2: Most important properties for tantalum and niobium metals

Property Ta Nb Lattice type Body centered Body centered cubic cubic Atomic volume (cm3. mol-1) 10.90 10.87 Atomic radius, pm 147 146 Density, g·cm-3 (at 20 oC) 16.62 8.57 Melting Point ( oC) 2996 2477 Boiling point (oC) 5560 4744 Specific heat capacity, J g-1 K-1 0.15 0.26 Thermal conductivity, W m-1 K-1 (at 20 oC) 54.4 53.7 Electrical conductivity, S m-1 8.1 x 106 6.6 x 106 Hardness, (Mohs scale) 6.5 6.0 Electronegativity, (Pauling Scale) 1.50 1.6 Thermal Neutron Absorption Cross Section, Barns/Atom 21.3 1.1 Electron configuration [Xe] 4f14 5d3 6s2 [Kr] 4d4 5s1 Ionization energies (kJ·mol−1): 1st 761 652.1 2nd 1500 1381.7 3rd n.a 2416.0 Electrical resistivity at 25 oC (nΩ m) 134 144 Ionic radius (Six–fold coordination) (pm) 64 64

British Geological Survey, 2011; Bose and Gupta, 2002.

185

Appendix

Appendix A 2

Table A 2.1. Tantalum products, applications and outcomes

Tantalum product Application Technical attributes /Outcomes Tantalum carbide Cutting tools Increased high temperature deformation, control of grain growth

Tantalum oxide - Lenses for spectacles, digital cameras and -Ta2O5 provides a high index of mobile phones refraction so lenses for a given focal - X-ray film strength can be thinner and smaller - Ink jet printers -Yttrium tantalate phosphor reduces X-ray exposure and enhances image quality -Wear resistance characteristics. Integrated capacitors in integrated circuits (ICs) Tantalum powder Tantalum capacitors for electronic circuits in: High reliability characteristics and - medical appliances such as hearing aids and low failure rates, operation over a pacemakers; wide temperature range from -55 to - automotive components such as ABS, airbag +200°C, can withstand severe activation, engine management modules, GPS; vibrational forces, small size per - portable electronics microfarad rating/electrical storage e.g. Laptop, computers, cellular/mobile phones, capability video cameras, digital still cameras; - other equipment such as DVD players, flat screen TVs, games consoles, battery chargers, power rectifiers, cellular/mobile phone signal masts, oil well probes Tantalum fabricated - Prosthetic devices for humans - hip joints, Attack by body fluids is non- sheets, plates, rods, skull plates, mesh to repair bone removed after existent; highly bio-compatible wires damage by cancer, suture clips, stents for blood vessels Tantalum fabricated -High temperature furnace parts Melting point is 2996°C although sheets, plates, rods, protective atmosphere or high wires vacuum required Tantalum fabricated - High temperature alloys for: Alloy compositions containing 3- sheets, plates, rods, -air and land based turbines (e.g. jet engine 11% tantalum offer resistance to wires discs, blades and vanes) corrosion by hot gases, allow higher - rocket nozzles operating temperatures and thus efficiency and fuel economy Tantalum ingot Computer hard drive discs An alloy containing 6% tantalum has shape memory properties Tantalum ingot Explosively Formed Projectile for TOW-2 Balance of density and formability missile allow for a lighter and more efficient system

TIC, 2017; Nikishina et al., 2013; Freeman, 2017.

186

Appendix

Table A 2.2. Niobium products, applications and outcomes

Niobium product Application Technical attributes/outcomes HSLA Ferro-Niobium Niobium additive to high strength low Impart a doubling of strength and (~60%Nb) alloy (HSLA) steel and stainless steel for toughness due to gain refining oil and gas pipelines, car and truck bodies, weight reduction. architectural requirements, tools steel, ships, hulls, railroad tracks. Niobium oxide Manufacturing lithium niobate for surface High index of refraction. acoustic wave filters. Camera lenses. High dielectric constant. Coating on glass for computer screens. Increase light transmittance. Ceramic capacitors Niobium carbide Cutting tool compositions. High temperature deformation, control grain growth. Niobium powder Niobium capacitors for electronic circuits. High dielectric constant, stability of oxide dielectric. Niobium metals plates, Cathode protection system for large steel Corrosion resistance, formation of sheets, wire, rod, tubing structures. oxide and nitrite films. Increase in Chemical processing equipment. high temperature resistance and corrosion resistance, oxidation resistance, improved creep resistance, reduction erosion at high temperatures. Niobium – titanium alloy Superconducting magnetic coils in Electrical resistance of alloy wire Niobium – tin alloy magnetic resonance imagery (MRI), drops to virtually zero at or below magneto encephalography, magnetic temperature of liquid helium (- levitation transport systems, particle 268.8 °C). physics experiments. Niobium-1% Zirconium Sodium vapor lamps. Corrosion resistance, fixation of alloy Chemical processing equipment oxygen, resistance to embrittlement. Vacuum-grade ferroniobium Superalloy additions for turbine blade Increasing in high temperature and nickel-niobium application in jet engines and land-based resistance and corrosion resistance, turbines. Inconel family of alloys, improved creep resistance, reduced superalloys. erosion at high temperature.

TIC (2017); Nikishina et al. (2013); Freeman (2017)

187

Appendix

Appendix A 3

Figure A 3. Generalised flow diagram for production of Ta-Nb oxides (Agulyansky, 2004). The above flow diagram may have some adjustments depending of the physical and chemical characteristics of the ore.

188

Appendix

Table A 3. Solvent extraction of niobium and tantalum from aqueous solution in the absence of HF by various extractants

Types of Extractant Aqueous Extraction, % Reference extractant Amines & TOA/TDA/TDDA Ta: 4.5.10-3 Ta, Nb > 99 Djordjevic et al. (1966) Acid (M) O/A = 1/1 Nb: 2.3.10-3 Ta & Nb (g/L) H2C2O4: 0.2 H2SO4: 0.5 DOAE/DOAP Ta: 1.9.10-3 Ta, Nb > 99 Djordjevic et al. (1969) O/A = 1/1 Nb: 0.93.10-3 H2C2O4: 0.2 Alamine NbF5: 5 336/Aliquat 336 TaF5: 5 O/A = 1/1 HClO4: 7 Ta: 53-55; Nb: 63-72 HNO3: 0-10 Ta, Nb: 0-20 Ungerer et al. (2014) HCl: 9 Ta, Nb: 75-79 H2SO4: 8 Ta: 55-60; Nb: 5-10

Neutral 1/2/3-OCL NbF5: 5 extractants TTA TaF5: 5 O/A = 1/1 HClO4: 7 Ta < 60; Nb: 65 – 69 HNO3: 10 Ta, Nb < 20 HCl: 9 Ta < 20; Nb: 54 – 59 El-Hazek et al. (2012) 2-OCL H2SO4: 0-15 Ta, Nb ≈ 0 O/A = 1/1 Sulfate Ta: 100, Nb: ~ 0 solution, Ph: 2 Acid PA Nb2O5: 17.52 extractants O/A = 1/1 T2O5: 1.58 D2EHPA NH4TaF6: 0.1 Ta: ~ 100; Nb: ~ 20 De Beer et al. (2014) -2 O/A = 1/1 NH4NBF6: 10 H2SO4: 8-10 Ta: ~ 100; Nb: ~ 10 PA/D2EHPA Nb2O5: 17.52 O/A =1/1 T2O5: 1.58 HclO4: 7 Ta, Nb: 91-92 HNO3: 10 Ta: 73-86, Nb: 63-72 Vin and khopkar (1991) HCl: 9 Ta: 84-87, Nb: 96-97 H2SO4: 15/12 Ta: 97; Nb: 85 (PA – H2SO4: 15) Ta: 81; Nb: 59 (D2EHPA – H2SO4: 12) DEHPA Nb2O5: 1 Ta ≈ Nb ≈ 100 O/A = 1/1 Ta2O5: 1 K2O7S2: 50 Vin and khopkar (1991) (NH4)2C2O4: 1 HCL: 1-2

189

Appendix

Appendix A 4

Table A 4.1. Chemical composition of tantalite concentrates collected at SMB mines

Oxides/Elements Samples (wt %) A B C D E F G H FH FM FL Al2O3 6.73 4.73 3.33 3.22 2.8 5.9 2.3 6.07 4.4 2.8 10.5 As2O3 0.19 <0.001 0.02 0.02 0.02 0.02 <0.001 <0.001 0.02 <0.001 <0.001 BaO 0.04 0.17 0.32 0.04 0.13 0.25 0.09 0.09 0.22 0.05 <0.001 CaO 2.16 6.2 3.6 6 5.6 6 1.35 5.9 5 1.08 0.169 Cl 0.09 <0.001 0.0 0.01 0.01 0.02 <0.001 <0.001 0.08 0.05 0.02 CoO 0.00 <0.001 <0.001 0.00 0.00 <0.001 0.05 0.04 <0.001 0.00 <0.001 Cr2O3 0.06 <0.001 0.02 0.15 0.63 0.01 0.10 0.18 0.00 0.00 <0.001 CuO 0.04 0.05 0.05 0.03 0.02 0.06 0.00 <0.001 0.07 0.06 <0.001 Fe2O3 20.4 2.13 5.08 9.5 16.6 5.3 3.5 7.8 7 4 0.8 K2O 1.5 1.2 0.8 0.2 0.2 1.5 0.4 1.2 1.2 0.6 3 MgO 4.096 0.472 1.675 9.518 5.574 2.318 0.842 4.274 0.940 1.074 0.087 MnO 3.787 2.576 5.170 4.098 2.056 1.197 10.799 2.201 2.719 10.936 0.091 MoO3 0.071 <0.001 0.055 0.018 0.008 0.024 0.002 0.008 0.024 0.008 <0.001 Na2O 0.383 3.212 1.491 0.568 0.986 2.411 0.586 1.868 2.376 0.402 0.303 Nb2O5 17.98 11.134 13.021 9.194 5.444 7.481 22.077 8.380 8.703 21.418 0.123 NiO 0.022 0.009 0.016 0.034 0.032 0.022 0.112 0.070 0.012 0.019 0.002 P2O5 0.200 0.029 0107 0.122 0.220 0.064 0.118 0.102 0.094 0.119 0.019 PbO 0.026 0.237 0.380 0.118 0.292 0.550 0.169 0.329 0.867 0.070 0.020 Sb2O3 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.002 0.043 <0.001 <0.001 <0.001 SiO2 17.328 16.890 14.178 19.706 13.125 19.909 6.995 21.810 15.660 13.793 80.152 SnO2 0.391 1.273 1.553 16.321 16.225 6.191 5.014 4.860 1.456 0.410 0.016 SO3 0.661 0.184 0.182 0.144 0.144 0.081 0.052 0.015 0.724 0.703 0.555 SrO 0.014 0.138 0.147 0.047 0.128 0.387 0.101 0.136 0.294 0.049 0.101 Ta2O5 20.912 46.690 44.813 18.401 19.685 37.509 45.716 31.207 44.789 39.573 0.439 TiO2 1.357 0.223 0.991 1.579 7.073 0.576 0.353 1.620 1.065 0.329 0.139 V2O5 0.017 0.009 0.002 0.028 0.018 0.006 <0.001 0.018 0.034 <0.001 <0.001 WO3 0.071 0.128 0.102 0.079 0.055 0.104 0.080 0.096 0.080 0.128 <0.001 ZnO2 0.022 0.024 0.011 0.021 0.022 0.006 0.023 0.009 0.017 0.013 0.006 ZrO2 0.042 0.466 0.408 0.239 2.133 0.404 0.327 0.524 0.203 0.367 0.016 LOI1000 1.49 1.12 1.40 0.24 0.12 1.55 0.73 0.99 1.27 0.90 2.12 Elements (g/t) Bi 3.6 13.6 20.8 156.2 28.2 15.2 70.1 25.2 17.1 4.1 1.9 Ce 57.0 11.0 52.0 127.0 869.0 105.0 43.0 231.0 102.0 10.0 35.0 Cs 2197.0 797.5 305.0 54.5 78.0 954.0 122 882 717.0 186.0 1581.0 Dy 7.0 1.0 3.5 12.5 80.0 2.0 2.5 10.0 2.5 1.5 1.0 Er 6.3 0.9 2.6 9.1 55.1 1.0 1.7 6.1 1.3 1.0 0.3 Eu 1.0 <0.5 1.0 2.0 10.5 1.0 0.5 2.0 1.0 <0.5 1.0 Gd 4.0 1.0 3.5 10.5 63.0 2.0 2.5 11.5 3.0 1.5 2.0 Hf 49.0 50.0 4.0 22.0 54.0 91.0 13.0 16.0 55.0 47.0 5.0 Ho 49.0 50.0 4.0 22.0 54.0 91.0 0.6 2.3 55.0 47.0 5.0 La 34.0 4.0 21.0 56.0 407.0 47.0 14.0 102.0 25.0 4.0 10.0 Li 3020 2000 1180 310 160 2490 670 2020 2100 810 5560 Lu 1.4 0.2 0.6 2.4 6.8 <0.2 0.8 1.4 <0.2 1.2 <0.2 Nd 28.0 3.0 17.0 54.0 349.0 21.0 12.0 79.0 24.0 4.0 12.0 Pr 8.0 1.0 5.0 15.0 98.0 8.5 3.5 23.5 7.5 1.0 3.0 Sc 38.0 4.0 6.0 12.0 10.0 7.0 2.0 9.0 3.0 3.0 <1 Sm 4.5 0.5 3.5 10.0 62.5 3.0 2.5 14.0 4.5 0.5 2.5 Tb 0.8 0.2 0.6 1.8 11.1 0.4 0.5 1.9 0.5 0.2 0.2 Th 10.5 722.0 291.0 87.0 301.0 500.0 135.0 436.0 366.5 56.0 4.5 Tm 1.1 0.2 0.3 1.2 7.9 0.2 0.3 1.1 0.2 0.2 <0.1 U 45.0 421.0 8021.0 1047.0 3434.0 4067.0 4206.0 2318.0 5333.0 1836.0 60.0 Y 35.5 9.5 21.0 80.0 487.5 9.5 15.0 58.5 10.5 9.0 2.0 Yb 8.6 1.6 2.9 7.1 46.5 1.2 1.6 5.8 1.2 1.6 0.2

Based on XRF; Analysed at Nagrom Laboratories

190

Appendix

Table A 4.2. Chemical composition of tantalite concentrates collected at SMB mines

Elements Samples A B C D E F G H Minor (wt %) As 0.14 <0.001 0.01 0.02 0.02 0.02 <0.001 <0.001 Ba 0.04 0.16 0.29 0.04 0.012 0.23 0.08 0.09 Cl 0.096 <0.001 0.008 0.016 0.016 0.024 <0.001 <0.001 Co 0.01 <0.001 <0.001 0.01 0.01 <0.001 0.04 0.04 Cr 0.04 <0.001 0.01 011 0.43 0.01 0.07 0.12 Cu 0.03 0.05 0.05 0.02 0.02 0.05 0.00 <0.001 Mo 0.05 <0.001 0.04 0.01 0.01 0.02 0.00 0.01 Ni 0.02 0.01 0.01 0.03 0.03 0.02 0.09 0.06 P 0.09 0.01 0.05 0.05 0.10 0.03 0.05 0.04 Pb 0.02 0.21 0.33 0.10 0.25 0.48 0.15 0.28 Sb <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.00 0.04 S 0.26 0.07 0.07 0.06 0.06 0.03 0.02 0.01 Sr 0.01 0.12 0.12 0.04 0.11 0.33 0.09 0.12 V 0.00 0.01 0.00 0.02 0.01 0.00 <0.001 0.00 W 0.06 0.10 0.08 0.06 0.04 0.08 0.06 0.08 Zn 0.02 0.02 0.01 0.02 0.02 0.00 0.02 0.01 Zr 0.03 0.34 0.30 0.18 1.58 0.30 0.24 0.39 LOI1000 1.49 1.12 1.40 0.24 0.12 1.55 0.73 0.99 Trace (g/t) Bi 3.6 13.6 20.8 156.2 28.2 15.2 70.1 25.2 Hf 49 50 4 22 54 91 13 16 La 34 4 21 56 407 47 14 102 Y 35.5 9.5 21 80 487.5 9.5 15 58.5 Sc 38 4 6 12 10 7 2 9 REE (g/t) Ce 57 11 52 127 869 105 43 231 Lu 1.4 0.2 0.6 2.4 6.8 <0.2 0.8 1.4 Dy 7 1 3.5 12.5 80 2 2.5 10 Er 6.3 0.9 2.6 9.1 55.1 1 1.7 6.1 Eu 1 <0.5 1 2 10.5 1 0.5 2 Gd 4 1 3.5 10.5 63.0 2 2.5 11.5 Ho 49 50 4 22 54 91 0.6 2.3 Nd 28 3 17 54 349 21 12 79 Pr 8 1 5 15 98 8.5 3.5 23.5 Sm 4.5 0.5 3.5 10 62.5 3 2.5 14 Tb 0.8 0.2 0.6 1.8 11.1 0.4 0.5 1.9 Tm 1.1 0.2 0.3 1.2 7.9 0.2 0.3 1.1 Yb 8.6 1.6 2.9 7.1 46.5 1.2 1.6 5.8 Total of REE 176.7 70.6 96.5 274.6 1713.4 236.3 72 389.6

Based on ICP-MS; Analysed at Nagrom Laboratories

191

Appendix

Table A 4.3. Chemical composition of tantalite concentrates collected at SMB mines

Elements Samples A B C E F G H Minor (wt %) As 0.0 0.00 0.01 0.0 0.01 0.01 0.01 Ba 0.1 0.07 0.04 0.1 0.04 0.02 0.00 Cl 1.8 5.12 5.43 4.6 7.86 5.00 3.99 Co 0.0 0.00 0.00 0.0 0.00 0.00 0.00 Cr 0.0 0.02 0.02 0.3 0.04 0.05 0.04 Cu 0.0 0.02 0.03 0.0 0.01 0.01 0.00 Mo 0.0 0.00 0.00 0.0 0.01 0.00 0.01 Ni 0.0 0.01 0.01 0.0 0.01 0.01 0.01 P 0.1 0.04 0.11 0.1 0.04 0.05 0.03 Pb 0.2 0.18 0.15 0.3 0.23 0.10 0.06 Sb 0.0 0.02 0.01 0.0 0.02 0.02 0.02 Sr 0.1 0.15 0.12 0.1 0.30 0.10 0.05 V 0.0 0.00 0.00 0.0 0.01 0.01 0.01 W 0.0 0.18 0.11 0.0 0.13 0.10 0.10 Zn 0.0 0.02 0.02 0.0 0.02 0.02 0.00 Zr 0.3 0.55 0.33 1.5 0.32 0.33 0.25

Based on ICP-MS; Analysed at Murdoch Laboratories

192

Appendix

Table A 4.4. Chemical composition of minor elements in coarse and fine fractions after sieving

Oxides/Elements Samples D G Feed –1000 μm +2800–1000 μm Feed –1000 μm +2800–1000 μm Minor (wt %) As2O3 0.024 0.02 0.03 <0.001 0.1 < 0.01 BaO 0.040 0 0 0.088 0.07 0.01 Cl 0.016 0.1 0.1 <0.001 0.02 0.14 CoO 0.008 0.003 <0.001 0.052 <0.001 < 0.001 Cr2O3 0.155 0.2 0.03 0.102 0.03 0.03 CuO 0.028 0.01 0.03 0.002 0.05 0.04 MoO3 0.018 0.01 0 0.002 0.04 0.03 NiO 0.034 0.05 0.02 0.112 0.02 0.02 P2O5 0.122 0.13 0.3 0.118 0.01 0.03 PbO 0.118 0.09 0.26 0.169 0.2 0.06 Sb2O3 <0.001 <0.001 <0.001 0.002 0 0 SO3 0.144 0.07 0.14 0.052 0.08 0.1 SrO 0.047 0.06 0.04 0.101 0.13 0.03 V2O5 0.028 0 0 <0.001 0.01 0.01 WO3 0.079 0.04 0.16 0.080 0.06 0.08 ZnO 0.021 0.01 0.02 0.023 0.01 0.01 ZrO2 0.239 0.25 0.1 0.327 0.3 0.1 LOI1000 0.24 0.2 0.78 0.73 0.93 0.29 Elements Minor (g/t) Bi 156.2 105.6 2041.2 70.1 99.6 113.9 Hf 22.0 165 182 13.0 511 245 La 56.0 59 27 14.0 13 2 Sc 12.0 10 5 2.0 3 1 REE (g/t) Ce 127.0 124 49 43.0 43 21 Dy 12.5 13 3 2.5 3 1 Er 9.1 7.6 1.5 1.7 2 0.4 Eu 2.0 2 1 0.5 0.5 <0.5 Gd 10.5 11 4 2.5 2.5 1 Ho 22.0 2.5 0.5 0.6 0.6 0.2 Lu 2.4 1 0.2 0.8 0.4 <0.2 Nd 54.0 58 18 12.0 13 3 Pr 15.0 15 5 3.5 3 0.5 Sm 10.0 11 3.5 2.5 3 0.5 Tb 1.8 1.8 0.4 0.5 0.3 < 0.1 Tm 1.2 74.5 39 0.3 0.2 <0.1 Y 80.0 65 9.5 15.0 16 5 Yb 7.1 5.9 1.4 1.6 2.5 0.7

Based on XRF; Analysed at Nagrom Laboratories

193

Appendix

Table A 4.5a. Chemical composition of oxides before and after magnetic separation

Categories Oxides Samples (wt %) FH FM FL Feed M NM Feed M NM Feed M NM Major SnO2 2.0 - 2.0 0.4 0.2 0.4 - 0.1 - Fe2O3 7.0 51 1.0 4.0 25 3.0 1.0 8.0 1.0 MnO 3.0 3.0 2.0 11 4.0 12 0.1 - - CaO 5.0 4.0 5.0 1.0 7.0 1.0 0.2 3.0 0.1 Al2O3 4.0 5.0 5.0 3.0 6.0 3.0 11 23 13 MgO 1.0 6.0 1.0 1.0 8.0 1.0 0.1 2.0 0.1 SiO2 16 11 17 14 22 13 80 49 79 Minor Na2O 2.0 0.3 3.0 0.4 0.5 0.3 0.3 0.4 0.3 K2O 1.2 0.3 1.0 0.6 0.5 1.0 3.0 5.0 4.0 TiO2 1.0 9.0 0.1 0.3 5.0 0.2 0.1 2.0 -

Based on XRF; FH: Heavy fraction, FM: Middling fraction, FL: Light fraction, M: magnetic and NM: Non-magnetic.

Table A 4.5b. Chemical composition of elements before and after magnetic separation

Categories Elements Samples (g/t) FH FM FL Feed M NM Feed M NM Feed M NM Trace Li 2100 170 2200 810 650 260 5560 8930 6330 Cs 717 35 741 186 41 197 1581 2693 1799 Bi 17 15 19 4 10 5 2 6 2 Hf 55 36 182 47 596 15 5 15 11 La 25 70 24 4 27 54 10 54 6 REE Ce 102 139 85 10 100 7 35 126 26 Y 11 19 9 9 18 9 2 12 12 Sc 3 14 3 3 15 3 <1 9 <1 Dy 3 5 3 2 4 2 1 4 1 Er 1.0 2.0 1.0 1.0 2.0 1.0 0.3 2.0 <0.1 Eu 1.0 2.0 1.0 <0.5 <0.5 2.0 1.0 2.0 <0.5 Gd 3 6 3 2 2 6 2 6 2 Ho 55 1.0 0.3 47 0.3 1.0 5.0 1.0 <0.1 Lu <0.2 <0.2 <0.2 1.2 0.2 0.2 <0.2 <0.2 <0.2 Nd 24 57 19 4.0 45 2.0 12 43 8.0 Pr 8.0 16 6.0 1.0 13 <0.5 3.0 12 2.0 Sm 5.0 10 4.0 0.5 7.0 0.5 3.0 8.0 2.0 Tb 0.5 1.0 0.2 0.2 1.0 0.1 0.2 1.0 1.0 Tm 0.2 0.1 <0.1 0.2 0.1 <0.1 <0.1 <0.1 <0.1 Yb 1.0 1.0 1.0 2.0 2.0 2.0 0.2 1.0 1.0 Total REE 101 98 38 60 75 15 27 78 16

Based on XRF; FH: Heavy fraction, FM: Middling fraction, FL: Light fraction, M: magnetic and NM: Non-magnetic. Analysed at Nagrom Laboratories

194

Appendix

Appendix A 5

(A) (B)

(D) (E)

Figure A 5.1a. XRD patterns of tantalite concentrates collected at SMB mines.

195

Appendix

(F) (H)

Figure A 5.1b. XRD patterns of tantalite concentrates collected at SMB mines.

196

Appendix

Table A 5. 1: Mineral identification of tantalite concentrates collected at SMB mines

S.I. Mineral names Chemical formulas A Ferrotantalite (Fe,Mn)(Nb,Ta)2O6 Pyrochlore (Ca,Na)(Nb,Ta)(O,OH,F) Quartz SiO2 Muscovite KAl2(Si3Al)O10(OH,F)2 Goethite Fe3+O(OH) Diopside CaMgSi2O6 Forsterite Mg2(SiO4) B Fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6 (OH,F) Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Parabariomicrolite (Ba,U)(Ta,Nb)2(O,OH)7 Quartz SiO2 Muscovite KAl2(Si3Al)O10(OH,F)2 Diopside CaMgSi2O6 D Manganocolumbite (Fe,Mn)(Nb,Ta)2O6 Fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6 (OH,F) Olivine (Mg,Fe)2SiO4 Stannomicrolite Sn2Ta2O7 Diopside CaMgSi2O6 Hematite Fe2O3 Muscovite KAl2(Si3Al)O10(OH,F)2 Quartz SiO2 E Fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6 (OH,F) Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Cassiterite SnO2 Stannomicrolite Sn2Ta2O7 Hematite Fe2O3 2+ Ilmenite Fe TiO3 Augite (Ca,Na)(Mg,Fe,Al,Ti)(Si,Al)2O6 Muscovite KAl2(Si3Al)O10(OH,F)2 Quartz SiO2 F Fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6 (OH,F) Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Parabariomicrolite (Ba,U)(Ta,Nb)2(O,OH)7 Muscovite KAl2(Si3Al)O10(OH,F)2 Perovskite CaTiO3 Quartz SiO2 Kaolinite Al2Si2O5(OH)4 Olivine (Mg,Fe)2SiO4 H Fluorocalciomicrolite (Ca,Na,U)2(Ta,Nb)2O6 (OH,F) Manganotantalite (Mn,Fe)(Ta,Nb)2O6 Cassiterite SnO2 Muscovite KAl2(Si3Al)O10(OH,F)2 Quartz SiO2 Hematite Fe2O3 Perovskite CaTiO3 Kaolinite Al2Si2O5(OH)4 Olivine (Mg,Fe)2SiO4 Diopside CaMgSi2O6 Goethite Fe3+O(OH)

Based on XDR; S.I.: sample identification

197

Appendix

Appendix A 6

(1) Fluorocalciomicrolite (2) Manganotantalite

(3) Quartz (4) Hematite

(5) Augite associated with amphibole (6) Muscovite

(7) Rare earth element

Figure A 6.1: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample C)

198

Appendix

(1) Fluorocalciomicrolite (2) Parabariomicrolite

(3) Manganotantalite (4) Cassiterite

(5) Quartz (6) Muscovite

Figure A 6.2a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample G continued).

199

Appendix

(7) Topaz (8) Lepidolite

(9) Zircon (10) Olivine

Figure A 6.2b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample G continued).

200

Appendix

(a1) (a2)

(1) Manganotantalite (2) Manganocolumbite

(3) Lepidolite (4) Hematite

Figure A 6.3: (a1-2) SEM backscattered images, and EDS in (numbers) elemental analysis of tantalite concentrate collected at Bisunzu mine (sample A).

201

Appendix

Figure A 6.4: SEM backscattered image of tantalite concentrate collected at Bisunzu mine (sample B).

(1) Fluorocalciomicrolite (2) Parabariomicrolite

Figure A 6.4a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample B).

202

Appendix

(3) Manganotantalite (4) Quartz

(5) Muscovite (6) Diopside associated with amphibole

(7) Lepidolite (8) Amphibole

Figure A 6.4b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample B continued).

203

Appendix

(9) Topaz (10) Olivine

(11) Augite

Figure A 6.4c: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample B continued).

204

Appendix

Figure A 6.5: SEM backscattered image of tantalite concentrate collected at Bisunzu mine (sample D).

(1) Fluorocalciomicrolite (2) Manganotantalite

Figure A 6.5a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample D).

205

Appendix

(3) Cassiterite (4) Lepidolite

(5) Muscovite (6) Olivine

(7) Diopside associated with amphibole (8) Topaz

(9) Quartz (10) Hematite

Figure A 6.5b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample D continued).

206

Appendix

Figure A 6.6: SEM backscattered image of tantalite concentrate collected at Bisunzu mine (sample E).

(1) Fluorocalciomicrolite (2) Manganotantalite

Figure A 6.6a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample E).

207

Appendix

(3) Cassiterite (4) Hematite

(5) Muscovite (6) Augite

(7) Olivine (8) Ilmenite

Figure A 6.6b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample E continued).

208

Appendix

(9) Diopside associated with amphibole (10) Columbite

(11) Zircon (12) Topaz

Figure A 6.6c: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample E continued).

209

Appendix

Figure A 6.7: SEM backscattered image of tantalite concentrate collected at Bisunzu mine (sample F).

(1) Fluorocalciomicrolite (2) Parabariomicrolite

Figure A 6.7a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample F)

210

Appendix

(3) Manganotantalite (4) Quartz

(5) (6)

(7) Hematite (8) Diopside associated with amphibole

Figure A 6.7b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample F continued)

211

Appendix

(9) Olivine (10) Amphibole

(11) Topaz

Figure A 6.7c: EDS elemental analysis of tantalite concentrate collected a Bisunzu mine (sample F continued).

212

Appendix

Figure A 6.8: SEM backscattered image of tantalite concentrate collected at Bisunzu mine (sample H).

(1) Fluorocalciomicrolite (2) Manganotanatalite

Figure A 6.8a: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample H).

213

Appendix

(3) Quartz (4) Olivine

(5) Diopside associated with amphibole (6) Cassiterite

(7) Lepidolite (8) Topaz

(9) Muscovite (10) Kaolinite asociated with perovskite

Figure A 6.8b: EDS elemental analysis of tantalite concentrate collected at Bisunzu mine (sample H continued).

214

Appendix

Appendix A 7

Table A 7.1a. Concentration of liquors produced after H2SO4/H2O2 leaching

T.N. S.I. Acid Oxidant PSD Temp Leach Agitation Elemental concentration in (μm) (oC) time (h) speed (rpm) solution (ppb) 8 C U Th Ta Nb 18 M 9.8 M < 75 100 12 250 6734 2251 75600 74623 H2SO4 H2O2 (H2SO4 to H2O2 ratio = 1:1)

Table A 7.1b. Concentration of liquors produced after KOH baking and water leaching

T.N. S.I. Bake Leach Elemental concentration in solution (ppb) Time (h) K2S2O8 to Additive Mass Time liquid U Th Ta Nb (g) (h) volume ratio (g/L) 0.5 27 Nd Nd 200968 290974 9 C KOH 10 2 1 199464 307741 2 210810 328426 3 209123 330782

T.N.: Test number, S.I.: Sample identification, Nd: not detected. Solid to KOH mass ratio: 1:10 for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV and S/L: 12.5 g/L for leaching.

Table A 7.2. Concentration of liquors produced after roasting and H2SO4 leaching

T.N. S.I. Bake Leach Elemental concentration in solution (ppb) Additive S/A Temp Time (h) (g/L) (oC) U Th Ta Nb 10 C H2SO4 625 600 0.5 1 Nd Nd Nd Nd 2 3

T.N.: Test number, S.I.: Sample identification. Nd: not detected. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~850 - 1120 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

215

Appendix

Table A 7.3. Concentration of liquors produced after H2SO4 baking and water leaching at 40 oC

T.N. S.I. Bake Leach Elemental concentration in solution (ppb) Additive S/A Temp Time (h) U Th Ta Nb (g/L) (oC) 11 C H2SO4 1250 400 0.5 22330 Nd 302693 67831 1 23892 314417 74856 2 31282 397458 100913 3 25480 333610 85855

T.N.: Test number, S.I.: Sample identification, Nd: not detected. Temperature: 400 oC, time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, S/L: o 12.5 g/L and K2S2O8 to liquid: 27 g/L and temperature: 40 C for leaching.

Table A 7.4. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 40 oC

T.N. S.I. Bake Leach Elemental concentration in solution (ppb) Additive S/A Temp (oC) Time U Th Ta Nb (g/L) (h) 12 C H2SO4 625 400 0.5 29527 1471 363701 85872 1 32385 1512 398729 93296 2 32515 1509 383465 96103 3 30702 1397 388475 97623

T.N.: Test number, S.I.: Sample identification. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ~ 850 - 1120 mV, time: 3 h, S/L: 12.5 g/L, 4 M H2SO4 and K2S2O8 to liquid: 27 g/ L and temperature: 40 oC for leaching.

216

Appendix

Table A 7.5. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 40 oC: Effect of different samples and leach time

Elemental concentration in solution (ppb) T.N. S.I. Time (h) U Th Ta Nb 0.5 21806 1588 81066 35538 1 19938 1494 74942 36017 F 2 21403 1504 78721 38650 3 22712 1418 78284 39712 0.5 45019 2326 159573 50988 1 45264 2180 158262 47910 H 2 47103 1952 155952 51278 3 51715 2101 165455 58249 0.5 20348 861 245988 123738 13 1 18733 775 235498 117383 D 2 21499 832 253066 126990 3 18794 687 226841 107596 0.5 1967 2221 64305 672070 1 2411 2415 69503 844917 B 2 2866 2583 78448 1001070 3 3188 2631 79605 1065582 0.5 30448 906 134951 104917 1 33744 799 122052 106586 G 2 33249 752 111602 112133 3 35683 621 109966 122487

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, time: 2 h and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 o mV, time: 3 h, S/L: 12.5 g/L, K2S2O4 to liquid: 27 g/L and temperature: 40 C for leaching.

Table A 7.6. Elemental leaching efficiencies (based on Table A 7.5)

T.N. S.I. Leach time (h) Elements leached (%) U Th Ta Nb F 0.5 44 25 2 5 1 40 24 2 5.5 2 43 24 2 6 3 45 23 2 6 H 0.5 100 41 5 6 1 100 40 5 6 13 2 100 37 5 6 3 100 38 5 6 D 0.5 100 76 13 15 1 100 66 12 14 2 100 59 12 14 3 100 59 11 13 B 0.5 33 20 1 3 1 44 26 1 4 2 53 27 2 5 3 54 26 1 5 G 0.5 58 54 3 6 1 64 47 3 6 2 63 45 2 6 3 68 37 2 7

217

Appendix

Table A 7.7. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of different samples and leach time

Elemental concentration in solution (ppb) T.N. S.I. Leach time (h) U Th Ta Nb 0.25 279 58 205548 406643 0.5 397 80 266395 513026 A 1 285 52 225166 442413 2 343 56 260011 513860 3 342 60 264940 534676 0.25 4563 3488 119783 95093 0.5 4988 4020 120784 115005 B 1 5172 4379 87032 82989 2 5338 4384 76661 55486 3 5726 5021 89393 61488 0.25 7463 344 625751 344628 0.5 8126 337 642699 357601 D 1 8115 370 654444 361671 2 8441 391 691391 382304 3 8831 396 683348 380137 0.25 63022 2429 176934 46810 0.5 62990 2654 183517 47832 14 E 1 68093 2350 181733 50683 2 69291 2293 187994 50458 3 68816 2507 188848 50413 0.25 43749 1691 85046 40212 0.5 41506 1658 73333 36400 F 1 52171 1919 90487 41785 2 46895 1957 84760 37781 3 51408 2171 94042 43491 0.25 46915 812 144024 249159 0.5 51603 820 153038 248977 G 1 65913 1065 194436 293248 2 55102 931 160413 240167 3 59024 943 170345 255839 0.25 79405 2642 113268 48695 0.5 85291 2751 114265 49544 H 1 87292 2688 117726 50561 2 86321 2488 111997 47532 3 85041 2597 123470 50335

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, time: 2 h and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 840 - 1120 mV, reaction time: 3 h, S/L: 12.5 g/L and K2S2O4 to liquid: 27 g/L for leaching.

218

Appendix

Table A 7.8. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of bake and leach time

T.N. S.I. Bake time (h) Leach Time (h) Elemental concentration in solution (ppb) U Th Ta Nb 2.0 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 1.5 0.25 64077 1126 160086 57773 0.5 70932 1050 159513 47836 1 78046 1188 170576 49466 15 C 2 92543 1540 210804 63423 3 85138 1448 188064 52792 1.0 0.25 74957 1016 160086 80003 0.5 84899 1119 159513 76034 1 78624 1123 170576 69983 2 75703 1058 210804 63622 3 71536 1183 188064 66089 0.5 0.25 82945 1308 172684 99607 0.5 84499 1315 184990 87973 1 84245 1321 176632 87719 2 94135 1604 207383 94593 3 89615 1545 180637 87938

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 1100 - 1140 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.9. Elemental leaching efficiencies (based on Table A 7.8)

T.N. S.I. Bake time (h) Leach time (h) Elements leached (%) U Th Ta Nb 0.25 69 32 5 8 2.0 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 0.25 64 31 3 5 1.5 0.5 71 29 3 4 15 C 1 78 33 4 4 2 93 42 5 6 3 85 40 4 5 1.0 0.25 75 28 3 7 0.5 85 31 3 7 1 79 31 4 6 2 76 29 5 6 3 72 33 4 6 0.5 0.25 83 36 4 9 0.5 84 36 4 8 1 84 36 4 8 2 94 44 5 8 3 90 42 4 8

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 1110 - 1140 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

219

Appendix

Table A 7.10. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of particle size and leach time

T.N. S.I. Bake particle size (μm) Leach time (h) Elemental concentration in solution (ppb) U Th Ta Nb 0.25 72367 1134 119386 73436 –250+150 0.5 80866 1113 128082 73467 1 83598 1118 137573 75303 2 85019 1160 148389 77775

3 61120 859 116049 60504

0.25 82606 1241 168168 78638 –150+106 0.5 84341 1247 168787 78667 16 C 1 86584 1190 172899 82027 2 92847 1279 194890 84839 3 94678 1315 197146 89113 0.25 74381 1245 234777 2481 –106+75 0.5 81919 1366 244767 2561 1 91418 1442 270481 2730 2 92200 1414 271554 2933 3 97794 1550 297220 3086 0.25 68850 1149 211376 89951 –75 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688

T.N.: Test number, S.I.: sample identification. Time: 2 h, temperature: 400 oC and solid to Liquid: 625 g/L for baking. Agitation speed: 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, temperature: 90 oC, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.11. Elemental leaching efficiencies (based on Table A 7. 10)

T.N. S.I. Bake particle size (μm) Leach time (h) Elements leached (%) U Th Ta Nb 0.25 72 31 2.6 6 –250+150 0.5 81 31 2.8 6 1 84 31 3 7 2 85 32 3 7 3 61 24 2.5 5 0.25 83 34 3.7 7 16 C –150+106 0.5 84 34 3.7 7 1 87 33 3.8 7 2 93 35 4 7 3 95 36 4 8 0.25 74 34 5 0.2 –106+75 0.5 82 38 5 0.2 1 91 40 6 0.2 2 92 39 6 0.2 3 98 43 6 0.2 0.25 69 32 5 8 –75 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10

T.N.: Test number, S.I.: sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Agitation speed: 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, temperature: o 90 C, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

220

Appendix

Table A 7.12. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of solid mass to acid volume ratio and leach time

T.N. S.I. Bake solid to acid volume Leach Time Elemental concentration in solution (ppb) ratios (g/L) (h) U Th Ta Nb 625 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 833 0.25 64866 800 162893 86073 0.5 70715 800 166881 83694 1 71748 834 166257 82395 17 C 2 71095 900 178252 85973 3 66451 879 164243 77637 1250 0.25 52213 591 164055 98497 0.5 50345 538 156730 78329 1 68373 754 200490 104361 2 60826 673 171810 83965 3 59695 747 179610 90614 2500 0.25 65507 823 128713 78198 0.5 60708 822 122499 68290 1 65133 794 126207 68700 2 63778 832 120696 68417 3 60000 827 119287 69677

T.N.: Test number, S.I.: Sample identification. Time: 2 h and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 1040 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid volume ratio: 27 g/L for leaching.

Table A 7. 13. Elemental leaching efficiencies (based on Table A 7. 12)

T.N. S.I. Bake solid to acid volume Leach time (h) Elements leached (%) ratios (g/L) U Th Ta Nb 625 0.25 69 32 5 8 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 0.25 65 22 4 8 0.5 71 22 4 7 17 C 833 1 72 23 4 7 2 71 25 4 8 3 66 24 4 7 0.25 52 16 4 9 1250 0.5 50 15 3 7 1 68 21 4 9 2 61 18 4 7 3 60 21 4 8 0.25 66 23 3 7 0.5 61 23 3 6 2500 1 65 22 3 6 2 64 23 3 6 3 60 23 3 6

221

Appendix

Table A 7. 14. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of leaching temperature and leaching time

T.N. S.I. Leach temperature (oC) Leach time (h) Elemental concentration in solution (ppb) U Th Ta Nb 90 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 80 0.25 68839 1147 241896 126928 0.5 75942 1127 243177 95959 18 C 1 77259 987 250870 104404 2 83064 1046 260209 89217 3 85440 1045 273408 95691 60 0.25 66404 1034 437805 183140 0.5 73966 1059 504978 200924 1 65506 834 433170 185168 2 71638 835 434506 188408 3 78719 869 431213 179081 20 0.25 50678 1109 386708 77851 0.5 53253 991 381980 78545 1 53118 900 412615 83387 2 56547 853 382538 83525 3 57928 811 395650 87218

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, time: 0.5 h - 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.15. Elemental leaching efficiencies (based on Table A 7. 14)

T.N. S.I. Leach temperature Leach time Elements leached (%) O ( C) (h) U Th Ta Nb 90 0.25 69 32 5 8 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 0.25 69 32 5 11 0.5 76 31 5 8 18 C 1 77 27 5 9 80 2 83 29 6 8 3 85 29 6 8 0.25 66 28 10 16 0.5 74 29 11 18 1 66 23 9 16 60 2 72 23 9 17 3 79 24 9 16 0.25 51 30 8 7 0.5 53 27 8 7 20 1 53 25 9 7 2 57 23 8 7 3 58 22 9 8

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, time: 0.5 - 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

222

Appendix

Table A 7.16. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of leaching acid concentration and leaching time

T.N. S.I. Bake temp Leach acid Leach time Elemental concentration in solution (ppb) (oC) concentration (M) (h) U Th Ta Nb 4 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 3 0.25 77190 1319 72995 36269 0.5 82371 1443 75291 32330 1 84228 1525 81024 32370 19 C 400 2 91061 1707 86087 34113 3 94283 1723 93381 34989 2 0.25 77171 1340 31705 12871 0.5 76166 1303 32578 12300 1 83303 1339 36421 13711 2 85692 1526 105223 14617 3 104736 1891 54001 20229 1 0.25 81585 1107 23795 6938 0.5 92648 1291 22042 7092 1 93076 1303 19894 7328 2 91215 1451 17488 7515 3 84903 1472 19902 8118

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 840 - 890 mV, time: 3 o h, S/L: 12.5 g/L, temperature: 90 C and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.17. Elemental leaching efficiencies (based on Table A 7. 16)

T.N. S.I. Leach acid concentration (M) Leach time (h) Elements leached (%) U Th Ta Nb 0.25 69 32 5 8 4 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 0.25 77 36 2 3 3 0.5 82 40 2 3 19 C 1 84 42 2 3 2 91 47 2 3 3 94 47 2 3 0.25 77 37 1 1 2 0.5 76 36 1 1 1 83 37 1 1 2 86 42 2 1 3 100 42 1 2 0.25 82 30 0.5 0.6 1 0.5 93 35 0.5 0.6 1 93 36 0.4 0.6 2 91 40 0.4 0.7 3 85 40 0.4 0.7

223

Appendix

Table A 7.18. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of leach solid mass to liquid volume ratio and leaching time

T.N. S.I. Bake temp Leach S/L Leach time Elemental concentration in solution (ppb) (oC) volume ratios (h) U Th Ta Nb (g/L) 12.5 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 25.0 0.25 - - - - 20 C 400 0.5 143051 1742 299766 61999 1 162542 1890 312745 72177 2 177296 2248 330425 78094 3 168827 2232 324245 87634 37.5 0.25 237752 3347 498802 95732 0.5 262958 3274 505953 101460 1 275843 3095 460844 114692 2 301944 3295 513033 131042 3 323071 3550 564281 147560

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to liquid: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 850 - 900 mV, o time: 3 h, acid concentration (4 M), temperature: 90 C and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.19. Elemental leaching efficiencies (based on Table A 7. 18)

T.N. S.I. Leaching S/L volume ratios (g/L) Leach time (h) Elements leached (%)

U Th Ta Nb 12.5 0.25 69 32 5 8 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 25.0 0.25 - - - - 20 C 0.5 72 24 3 3 1 81 26 3 3 2 89 31 4 3 3 84 31 4 4 37.5 0.25 79 31 4 3 0.5 88 30 4 3 1 92 28 3 3 2 100 30 4 4 3 100 31 4 4

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L – 1875 g/L and temperature: 400 oC for baking. PSD: < 75 μm; agitation speed: 250 rpm, Eh: ̴ o 900 - 1140 mV, time: 3 h, acid concentration (4 M), temperature: 90 C and K2S2O8 to liquid: 27 g/L for leaching.

224

Appendix

Table A 7.20. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 o C: Effect of K2S2O8 to liquid volume ratio and leaching time

T.N. S.I. Bake temp K2S2O8 to liquid Leach time (h) Elemental concentration in solution (ppb) (oC) volume ratios U Th Ta Nb (g/L) 40.5 0.25 61235 1288 377380 102942 0.5 65600 1323 409732 108216 1 66024 1217 395871 103207 2 71371 1088 428281 113450 3 74739 1090 436877 119329 27.0 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 21 C 400 1 80439 1162 229764 93518 2 83556 1115 242398 97074 3 87447 1021 260996 115688 13.5 0.25 67179 1150 102590 82001 0.5 68037 967 100477 77948 1 87218 1097 139270 100985 2 83284 1018 148833 100710 3 79374 997 152270 99069

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ~ 900 – 1140 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M) and temperature: 90 oC for leaching.

Table A 7.21. Elemental leaching efficiencies (based on Table A 7. 20)

T.N. S.I. Leach K2S2O8 g to Leach time (h) Elements leached (%) liquid volume U Th Ta Nb ratios (g/L)

40.5 0.25 61 35 8 9 0.5 66 36 9 10 1 66 33 9 10 2 71 30 9 10 21 C 3 75 30 10 10 27.0 0.25 69 32 5 8 0.5 72 29 5 8 1 80 32 5 8 2 84 31 5 9 3 87 28 6 10 13.5 0.25 67 32 2 7 0.5 67 27 2 7 1 87 30 3 9 2 83 28 3 9 3 79 27 3 9

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 250 rpm, Eh: ~ 1140 - 1190 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M) and temperature: 90 oC for leaching.

225

Appendix

Table A 7.22. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of agitation speed and leaching time

T.N. S.I. Bake temp Leach Leach Elemental concentration in solution (ppb) (oC) agitation time (h) U Th Ta Nb speed (rpm) 250 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518 22 C 400 2 83556 1115 242398 97074 3 87447 1021 260996 115688 0 0.25 55566 953 180791 61045 0.5 60326 964 194935 65576 1 73299 1159 236034 79315 2 70166 1022 225758 78577 3 78418 1101 264705 89707

T.N.: Test number, S.I.: Sample identification; time: 2 h and solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: < 75 μm, agitation speed: 0 rpm & 250 rpm, Eh: ̴ 1140 - 11901 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M), temperature: 90 oC and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.23. Elemental leaching efficiencies (based on Table A 7. 22)

T.N. S.I. Leach agitation Leach time (h) Elements leached (%) speed (rpm) U Th Ta Nb 0.25 69 32 5 8 250 0.5 72 29 5 8 1 80 32 5 8 22 C 2 84 31 5 9 3 87 28 6 10 0.25 56 26 4 5 0 0.5 60 27 4 6 1 73 32 5 7 2 70 28 5 7 3 78 30 6 8

T.N.: Test number, S.I.: Sample identification; time: 2 h and solid to liquid volume ratio: 625 g/L and temperature: 400 oC for baking. PSD: <75 μm, agitation speed: 0 rpm & 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M), temperature: 90 oC and K2S2O8 to liquid: 27 g/L for leaching.

226

Appendix

Table A 7.24. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of different samples and leaching time

T.N. S.I. Leach time (h) Elemental concentration in solution (ppb) Fe Mn Al Li Cs 25 A 0.25 1304860 278325 126990 15252 8221 0.5 1564959 341127 170830 16836 8321 1 1421926 308308 144330 17378 9331 2 1658602 349403 180917 18735 10607 3 1628302 338207 200567 19986 12370 B 0.25 42577 60627 73713 5659 1837 0.5 52977 68931 84409 6342 2065 1 66045 68837 82479 6649 2094 2 85870 69096 92775 6471 2016 3 116760 79825 93484 7656 2314 D 0.25 601887 368250 70422 2032 334 0.5 648969 378576 78789 2036 338 1 641962 365852 91702 2148 377 2 720003 385892 110219 2126 361 3 717857 395964 90924 2364 381 E 0.25 523131 106221 39472 854 1049 0.5 646918 115356 35605 810 994 1 857486 115733 30966 922 1108 2 1079398 116538 46222 910 1116 3 1124616 113112 45757 977 1133 F 0.25 181402 58401 56193 6607 3578 0.5 194408 61400 72546 7029 3484 1 292282 76934 78778 8611 4263 2 310165 69190 83204 7501 3912 3 371860 76362 94790 8539 4459 G 0.25 176962 397061 58363 1718 570 0.5 206808 435705 47622 1927 623 1 250780 554112 60123 2207 653 2 210994 474096 49210 2330 745 3 222147 508747 50147 2310 675 H 0.25 346855 150685 81054 4652 2004 0.5 383308 158117 81599 5059 2074 1 456711 172639 75115 5880 2348 2 484587 157572 98540 5550 2306 3 578827 174782 97529 5733 2272

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L, temperature: 400 oC and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ o 840 – 1120 mV, temperature: 90 C, time: 3h, S/L: 12.5 g/L and K2S2O4 to liquid: 27 g/ L for leaching.

227

Appendix

Table A 7.25. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of baking and leaching time

T.N. S.I. Bake time (h) Leach time Elemental concentration in solution (ppb) (h) Fe Mn Al Li Cs 2.0 0.25 288691 408470 98416 6873 1865 0.5 303094 420298 97112 7432 1887 1 311087 408453 112276 8355 1968 2 369935 448843 119196 8446 2116 3 404492 481341 125246 8703 2303 1.5 0.25 219828 189856 90472 3141 1161 0.5 262637 211315 88657 3512 1191 1 291065 221297 101663 3667 1232 15 C 2 396061 266946 129637 4176 1422 3 380404 245145 114522 4199 1461 1.0 0.25 238857 223214 102115 3637 1182 0.5 271177 248399 100810 4365 1349 1 288943 236929 108401 4012 1232 2 302214 224311 102937 4074 1322 3 323655 224866 107215 4038 1352 0.5 0.25 239489 240312 101755 4768 1578 0.5 273415 251921 101158 5151 1717 1 273582 236582 108767 5009 1691 2 351188 271563 117945 5423 1829 3 346097 258377 110989 6503 2117

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 840 - 1120 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.26. Elemental leaching efficiencies of minor elements (based on Table A 7. 25)

T.N. S.I. Bake time (h) Leach time (h) Elements leached (%) Fe Mn Al Li CS 0.25 65 82 47 47 49 0.5 68 84 38 50 49 2.0 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 0.25 38 49 43 21 30 0.5 42 59 49 24 31 1 44 65 46 25 32 15 C 2 53 89 54 28 37 1.5 3 49 85 50 28 38 0.25 45 54 29 25 31 0.5 50 61 39 30 35 1 47 65 45 27 32 2 45 68 44 28 35 1.0 3 45 73 37 27 35 0.25 48 54 22 32 41 0.5 50 61 99 35 45 1 47 61 33 34 44 2 54 79 37 37 48 0.5 3 52 78 40 44 56

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 0.5 - 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 840 - 1120 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

228

Appendix

Table A 7.27. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of particle size and leaching time

T.N. S.I. Bake particle size (μm) Leach time Elemental concentration in solution (ppb) (h) Fe Mn Al Li Cs 0.25 317541 326706 128754 8496 2165 –250+150 0.5 353753 347492 142436 9535 2344 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 0.25 338285 361616 116934 8612 2192 –150+106 0.5 352687 358255 142765 9274 2230 1 375747 378314 162559 11288 2655 16 C 2 409458 389508 172159 10813 2564 3 420075 397517 158152 10634 2721 0.25 330898 394667 136507 9215 2481 –106+75 0.5 342295 404577 144564 9856 2561 1 391101 445802 198980 10828 2730 2 391069 434559 173102 11349 2933 3 441930 472700 182522 12090 3086 0.25 288691 408470 98416 6873 1865 –75 0.5 303094 420298 97112 7432 1887 1 311087 408453 112276 8355 1968 2 369935 448843 119196 8446 2116 3 404492 481341 125246 8703 2303

T.N.: Test number, S.I.: sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L for baking. Agitation speed: 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, temperature: 90 oC, S/L: 12.5 g/L, K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.28. Elemental leaching efficiencies of minor elements (based on Table A 7. 27)

T.N. S.I. Bake particle size (μm) Leach time (h) Elements leached (%) Fe Mn Al Li CS 0.25 71 65 55 58 57 –250+150 0.5 79 69 58 65 61 1 84 71 80 65 59 2 91 75 69 71 64 3 68 54 73 53 50 0.25 76 72 47 58 57 –150+106 0.5 79 72 57 63 58 1 84 76 65 77 70 2 92 78 69 73 67 3 94 80 63 72 71 16 C 0.25 74 79 52 62 65 –106+75 0.5 77 81 57 67 67 1 88 89 50 73 72 2 88 87 63 77 77 3 99 95 45 82 81 0.25 65 82 47 47 49 –75 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60

T.N.: Test number, S.I.: sample identification. Time: 2 h, temperature: 400 oC and solid to acid volume ratio: 625 g/L for baking. Agitation speed: 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, o temperature: 90 C, S/L: 12.5 g/L, K2S2O8 to liquid: 27 g/L for leaching.

229

Appendix

Table A 7.29. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of bake solid to acid volume ratio and leaching time

T.N. S.I. Bake solid to acid ratio Leach Elemental concentration in solution (ppb) (g/L) time (h) Fe Mn Al Li Cs 625 0.25 288691 408470 98416 6873 1865 0.5 303094 420298 97112 7432 1887 1 311087 408453 112276 8355 1968 2 369935 448843 119196 8446 2116 3 404492 481341 125246 8703 2303 833 0.25 238986 197029 85106 4922 1308 0.5 273403 210575 102436 5328 1359 1 288218 212697 98876 5689 1468 17 C 2 341069 224671 98503 5927 1593 3 338698 202393 103249 5528 1575 1250 0.25 229775 172450 92780 5203 1233 0.5 248784 169662 93013 5308 1247 1 365071 227248 104757 5288 1317 2 334355 190959 102368 5801 1487 3 378900 199502 106026 6577 1675 2500 0.25 240392 172678 88186 4507 1208 0.5 255368 177019 86130 4968 1349 1 269764 173131 95476 5088 1369 2 292112 160637 106337 4841 1407 3 309211 163223 91488 5212 1557

T.N.: Test number, S.I.: Sample identification. Time: 2 h, particle size: < 75 μm and temperature: 400 oC for baking. Agitation speed: 250 rpm, Eh: ̴ 890 - 920 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.30. Elemental leaching efficiencies of minor elements (based on Table A 7. 29)

T.N. S.I. Bake solid to acid volume ratio (g/L) Leach time (h) Elements leached (%) Fe Mn Al Li CS 0.25 65 82 47 47 49 625 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 0.25 39 54 36 33 34 17 C 833 0.5 42 61 35 36 36 1 43 65 41 39 38 2 45 77 52 40 42 3 40 76 46 37 41 1250 0.25 34 52 41 35 32 0.5 34 56 40 36 33 1 45 82 43 36 35 2 38 75 41 39 39 3 40 85 43 45 44 2500 0.25 35 54 41 31 32 0.5 35 57 40 34 35 1 35 61 44 34 36 2 32 66 47 33 37 3 33 69 44 35 41

T.N.: Test number, S.I.: Sample identification. Time: 2 h, particle size: < 75 μm and temperature: 400 oC. Agitation speed: 250 rpm, Eh: ̴ 890 - 920 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

230

Appendix

Table A 7.31. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of leaching temperature and leaching time

T.N. S.I. Leach temperature Leach time Elemental concentration in solution (ppb) (oC) (h) Fe Mn Al Li Cs 0.25 317541 326706 128754 8496 2165 0.5 353753 347492 142436 9535 2344 90 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 0.25 302219 424287 173931 6661 1556 80 0.5 314042 439045 165198 8098 1948 18 C 1 332860 466168 165477 8759 1998 2 336709 459372 155734 9073 2173 3 378816 464731 178003 10074 2374 0.25 293689 434983 106181 6755 1768 0.5 331045 471551 170193 7356 1758 60 1 280373 420922 145976 8142 1983 2 303115 439167 141881 8148 2064 3 327902 458538 149291 8914 2137 0.25 91509 370413 61546 2861 925 0.5 117713 379607 82697 3164 1039 20 1 168359 389348 91123 3177 1026 2 217680 384112 145574 4024 1146 3 231719 384552 120335 3902 1139

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid: 625 g/L, temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, time: o o 0.5 - 3 h, S/L: 12.5 g/L, temperature: 20 C - 90 C and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.32. Elemental leaching efficiencies of minor elements (based on Table A 7. 31)

T.N. S.I. Leach temperature (OC) Leach time (h) Elements leached (%) Fe Mn Al Li CS 0.25 65 82 47 47 49 90 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60

0.25 68 85 34 45 41 18 C 0.5 71 88 41 55 51 80 1 75 93 40 59 52 2 76 92 39 62 57 3 85 93 41 68 62 0.25 66 87 37 46 46 0.5 74 94 37 50 46 60 1 63 84 42 55 52 2 68 88 41 55 54 3 74 92 42 60 56 0.25 21 74 35 19 24 20 0.5 26 76 34 21 27 1 38 78 38 22 27 2 49 77 43 27 30 3 52 77 37 26 30

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L, temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 890 - 1040 mV, time: 0.5 h - 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

231

Appendix

Table A 7.33. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of leach acid concentration and leaching time

T.N. S.I. Leach acid concentration Leach time Elemental concentration in solution (ppb) (M) (h) Fe Mn Al Li Cs 0.25 317541 326706 128754 8496 2165 0.5 353753 347492 142436 9535 2344 4 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 0.25 213997 245847 103564 5626 1365 3 0.5 226294 238199 79897 4542 1328 1 253176 253416 103653 5343 1340 19 C 2 302115 271273 137922 5467 1497 3 335177 288670 135435 5334 1534 0.25 210900 249788 77678 4239 1255 0.5 234286 247645 98046 4072 1255 2 1 252213 263234 99259 4394 1281 2 284430 278995 143105 4630 1384 3 364369 330507 111106 4321 1341 0.25 219766 262037 119896 5200 1496 0.5 259350 283947 102004 5652 1538 1 1 269112 292220 106730 5376 1539 2 280943 296323 121260 5327 1526 3 283666 288334 109061 4980 1448

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 250 rpm, Eh: ̴ 840 - 890 mV, time: 3 h, S/L: 12.5 g/L, o acid concentration (1 M - 4 M), temperature: 90 C and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.34. Elemental leaching efficiencies of minor elements (based on Table A 7. 33)

T.N. S.I. Leach acid concentration (M) Leach Elements leached (%) time (h) Fe Mn Al Li CS 0.25 65 82 47 47 49 4 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 0.25 48 49 41 38 36 3 0.5 51 48 32 31 35 19 C 1 57 51 41 36 35 2 68 54 55 37 39 3 75 58 54 36 40 0.25 47 50 31 29 33 2 0.5 53 50 39 28 33 1 57 53 40 30 34 2 64 56 57 31 36 3 82 66 44 29 35 0.25 49 35 48 35 39 1 0.5 58 38 41 38 40 1 60 36 43 36 40 2 63 36 49 36 40 3 64 34 44 34 38

232

Appendix

Table A 7.35. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of leach solid to liquid volume ratio and leaching time

T.N. S.I. Leach S/L volume Leach Elemental concentration in solution (ppb) ratios (g/L) time (h) Fe Mn Al Li Cs 0.25 317541 326706 128754 8496 2165 0.5 353753 347492 142436 9535 2344 12.5 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 0.25 1836 138 - 163 18 25.0 0.5 513804 728919 191631 13468 3411 20 C 1 663950 883060 220008 17336 4080 2 704948 903002 251289 17194 4114 3 703366 862765 242257 16417 4308 0.25 942805 1426506 327635 21906 6213 0.5 994721 1521832 372502 23416 6229 37.5 1 1015476 1553800 384015 25837 6812 2 1154878 1682535 438275 27579 7071 3 1256784 1736210 415891 30817 8151

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Particle size: < 75 μm; agitation speed: 250 rpm, Eh: ̴ 1140 – 1110 mV, time: 3 h, S/L: 12.5 g/L – 37.5 g/L, acid concentration (4 M), temperature: o 90 C and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.36. Elemental leaching efficiencies of minor elements (based on Table A 7. 35)

T.N. S.I. Leach S/L volume Leach time (h) Elements leached (%) ratios (g/mL) Fe Mn Al Li CS 12.5 0.25 65 82 47 47 49 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 25.0 0.25 - - - - - 20 C 0.5 58 73 31 46 45 1 75 88 35 59 54 2 79 90 40 58 54 3 79 86 39 56 56 37.5 0.25 71 95 35 50 54 0.5 75 100 40 53 54 1 76 100 41 58 60 2 87 100 47 62 62 3 94 100 44 70 71

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Particle size: < 75 μm; agitation speed: 250 rpm, Eh: ̴ o 1110 – 1140 mV, time: 3 h, acid concentration: 4 M, temperature: 90 C and K2S2O8: 5.4 g for leaching.

233

Appendix

Table A 7.37. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of leaching K2S2O8 to liquid volume ratio and leaching time

T.N. S.I. Leach K2S2O8 to Leach Elemental concentration in solution (ppb) liquid volume time (h) Fe Mn Al Li Cs ratios (g/L) 0.25 270452 364084 147004 5481 1628 0.5 313952 410256 148620 6572 1810 40.5 1 330243 396959 153719 6562 1766 2 370534 453274 151736 7652 1894 3 416044 437954 149685 7042 1985 0.25 317541 326706 128754 8496 2165 27.0 0.5 353753 347492 142436 9535 2344 21 C 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 0.25 299926 364820 98586 7025 1717 0.5 280410 350098 137647 7561 1806 13.5 1 402767 430814 144941 7648 1893 2 422105 445665 174450 8547 2110 3 369043 413826 126212 8389 2254

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L, temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 250 rpm, Eh: ~ 1140 - 1190 mV, time: 3 h, S/L: 12.5 g/L, acid concentration: 4 M and temperature: 90 oC for leaching.

Table A 7.38. Elemental leaching efficiencies of minor elements (based on Table A 7. 37)

T.N. S.I. Bake K2S2O8 to liquid Leach time (h) Elements leached (%) volume ratios (g/L) Fe Mn Al Li CS

40.5 0.25 61 73 59 37 43 0.5 71 82 59 45 47 1 74 79 61 44 46 2 83 91 61 52 50 3 93 88 60 48 52 27.0 0.25 65 82 47 47 49 21 C 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 13.5 0.25 67 73 39 48 45 0.5 63 70 55 51 47 1 91 86 58 52 50 2 95 89 70 58 55 3 83 83 50 57 59

T.N.: Test number, S.I.: Sample identification. Time: 2 h and solid to acid volume ratio: 625 g/L, temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 250 rpm, Eh: ~ 1140 - 1190 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M) and temperature: 90 oC for leaching.

234

Appendix

Table A 7.39. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of agitation speed and leaching time

T.N. S.I. Leaching agitation Leach time Elemental concentration in solution (ppb) speed (rpm) (h) Fe Mn Al Li Cs 0.25 317541 326706 128754 8496 2165 0.5 353753 347492 142436 9535 2344 250 1 374424 353232 124169 9643 2268 2 406750 376225 157295 10499 2458 3 300675 269874 111860 7855 1900 22 C 0.25 249234 364561 105336 6779 1478 0 0.5 308806 403787 118940 5269 1655 1 405911 477703 148251 7822 1944 2 406705 457302 141396 8948 2012 3 474040 475508 145266 7473 2295

T.N.: Test number, S.I.: Sample identification; time: 2 h and solid to acid volume ratio: 625 g/L, temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 0 rpm & 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M), temperature: 90 oC and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7. 40. Elemental leaching efficiencies of minor elements (based on Table A 7. 39)

T.N. S.I. Leach agitation Leach time (h) Elements leached (%) speed (rpm) Fe Mn Al Li CS 0.25 65 82 47 47 49 250 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 22 C 3 91 96 40 59 60 0.25 56 73 42 46 39 0.5 69 81 48 36 43 1 91 96 59 53 51 0 2 91 91 57 61 53 3 100 95 58 51 60

T.N.: Test number, S.I.: Sample identification. Time: 2 h, solid to acid volume ratio: 625 g/L and temperature: 400 oC for baking. Particle size: < 75 μm, agitation speed: 0 rpm & 250 rpm, Eh: ̴ 1110 - 1140 mV, time: 3 h, S/L: 12.5 g/L, acid concentration (4 M), temperature: 90 oC and K2S2O8 to liquid: 27 g/L for leaching.

235

Appendix

Table A 7.41. Concentration of liquors produced after H2SO4 baking and H2SO4 leaching at 90 oC: Effect of baking temperature and leaching time

T.N. S.I. Bake temp (OC) Leach Time (h) Elemental concentration in solution (ppb)

U Th Ta Nb 400 0.25 68850 1149 211376 89951 0.5 72166 1066 216763 91575 1 80439 1162 229764 93518

2 83556 1115 242398 97074

3 87447 1021 260996 115688

300 0.25 89283 1721 254329 86443 0.5 102818 1948 289145 97956 1 99089 1848 267050 88297 23 C 2 111537 2216 315959 99944 3 105104 2248 298072 96593 200 0.25 23608 542 68969 182026 0.5 24654 569 71156 176323 1 26480 590 77109 195836 2 26766 568 72974 176863 3 20244 405 54483 133865 100 0.25 1330 78 3565 6039 0.5 1419 67 701 5931 1 1523 44 1285 6765 2 1527 28 986 6208 3 1783 20 1264 7477

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 850 - 900 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.42. Elemental leaching efficiencies (based on Table A 7. 41)

T.N. S.I. Bake temperature Leach time (h) Elements leached (%) o ( C) U Th Ta Nb

400 0.25 69 32 5 8

0.5 72 29 5 8

1 80 32 5 8

2 84 31 5 9

3 87 28 6 10

0.25 89 47 6 8

300 0.5 100 54 6 9

1 100 51 6 8 23 C 2 100 61 7 9 3 100 62 6 8 0.25 24 15 2 16 200 0.5 25 16 2 16 1 26 16 2 17 2 27 16 2 16 3 20 11 1 12 0.25 1.3 2 0 0.5 0.5 1.4 2 0 0.5 100 1 1.5 1 0 0.6 2 1.5 1 0 0.5 3 1.8 1 0 0.7

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 890 – 920 mV, time: 3 h, S/L: o 12.5 g/L, temperature: 90 C and K2S2O8 to liquid: 27 g/L for leaching.

236

Appendix

Table A 7.43. Concentration of minor elements in liquor produced after H2SO4 baking and o H2SO4 leaching at 90 C: Effect of baking temperature and leaching time

T.N. S.I. Bake temp (oC) Leach time Elemental concentration in solution (ppb) (h) Fe Mn Al Li Cs 0.25 288691 408470 98416 6873 1865 400 0.5 303094 420298 97112 7432 1887 1 311087 408453 112276 8355 1968 2 369935 448843 119196 8446 2116 3 404492 481341 125246 8703 2303 0.25 235297 264733 107954 6672 2226 300 0.5 291789 308594 121276 7330 2309 1 304157 292980 115973 6831 2258 23 C 2 384603 340168 134832 7431 2325 3 388265 322469 125965 8140 2617 0.25 219116 102247 73029 3675 1044 200 0.5 265871 109499 97736 3931 1113 1 331130 120045 111776 4422 1307 2 366529 117740 110871 6053 1609 3 286275 82659 93332 4480 1178 0.25 244291 7590 55890 2878 2878 100 0.5 280749 11349 247885 3686 3686 1 318829 9207 83511 4364 4364 2 354129 9138 93691 5020 5020 3 400526 10240 99536 6409 6409

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 890 - 920 mV, time: 3 h, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

Table A 7.44. Elemental leaching efficiencies of minor elements (based on A Table A 7. 43)

T.N. S.I. Bake temperature (oC) Leach time (h) Elements leached (%) Fe Mn Al Li CS 0.25 65 82 47 47 49 400 0.5 68 84 38 50 49 1 70 82 48 57 52 2 83 90 39 57 56 3 91 96 40 59 60 0.25 53 53 43 45 58 300 0.5 62 66 49 50 61 23 C 1 59 68 46 46 59 2 68 86 54 50 61 3 64 87 50 55 69 0.25 20 49 29 25 27 0.5 22 60 39 27 29 200 1 24 74 45 30 34 2 24 82 44 41 42 3 47 64 37 30 31 0.25 2 55 22 20 75 100 0.5 2 63 - 25 97 1 2 72 33 30 100 2 2 80 37 34 100 3 2 90 40 43 100

T.N.: Test number, S.I.: Sample identification. Solid to acid volume ratio: 625 g/L and time: 2 h for baking. Agitation speed: 250 rpm, PSD: < 75 μm, Eh: ̴ 1110 - 1140 mV, time: 3 h, o temperature: 90 C, S/L: 12.5 g/L and K2S2O8 to liquid: 27 g/L for leaching.

237

Appendix

Table A 7.45. Concentration of liquors produced after CaF2 and H2SO4 leaching

T.N. S.I. Leach time (h) Element concentrations in solution (ppb) U Th Ta Nb Si 0.5 228486 7672 6643123 885722 2121355 1 289373 9760 8548176 1188062 2363127 24 C 2 292032 9812 9057645 1255919 2313094 3 317567 9429 9598578 1422276 2161490 4 307000 7572 11083762 1462823 2259844

T.N.: Test number, S.I.: Sample identification. Leaching conditions: Solid to CaF2 mass ratio: o 1:2, 4 M H2SO4, time: 4 h, temperature: 90 C, solid to liquid ratio: 37.5 g/L, PSD: < 75 μm and agitation speed: 250 rpm.

Table A 7.46. Elemental leaching efficiencies (based on Table A 7. 45)

T.N. S.I. Leach time Elements leached (%) (h) U Th Ta Nb Si 24 C 0.5 76 70 48 26 81 1 96 90 62 35 90 2 97 90 66 37 88 3 100 87 70 42 82 4 100 69 81 43 86

T.N.: Test number, S.I.: Sample identification. Leaching conditions: Solid to CaF2 ratio: 1:2, 4 o M H2SO4, time: 4 h, temperature: 90 C, solid to liquid ratio: 37.5 g/L, PSD: < 75 μm and agitation speed: 250 rpm.

238

Appendix

Appendix A 8 a) b) 20 12 y = 1.69x + 0.33 y = 1.66x - 0.01 y = 0.96x + 0.05 y = 0.90x + 0.06 y = 1.07x + 0.13 18 R² = 0.98 R² = 0.99 R² = 0.99 R² = 0.95 R² = 0.99 10 16 y = 1.66x - 0.01 y = 1.66x - 0.01 R² = 0.99 R² = 0.99 14 Leach temp = 90° C 8 12 Leach temp= 80° C 10 Leach temp= 60° C 6 Leach acid con= 4 M 8

Nb leached (%) leached Nb Leach acid conc= 3 M Nb leached (%) leached Nb 4 Leach temp= 40° C 6 Leach acid conc= 2 M Leach temp= 20° C 4 Leach acid conc= 1 M Leach bake product to liquid= 25.0 g/L 2 Leaching bake product to liquid ratio= 12.5 g/L Leach bake prodcut to liquid= 37.5 g/L 2 Leaching reagent to liquid ratio= 27.0 g/L Leach reagent to liquid ratio= 40.5 g/L 0 0 0 2.5 5 7.5 10 12.5 15 0 2.5 5 7.5 10 12.5 Ta leached (%) Ta leached (%)

35 y = 1.72x y = 1.65x y = 1.65x y = 1.65x Leach temp= 90 °C R² = 0.98 R² = 0.99 R² = 0.99 R² = 0.99 Leach temp= 80 °C 30 Leach temp= 60 °C Leach acid conc= 4 M Leach acid conc= 3 M 25 Leach acid conc= 2 M Leach acid conc= 1 M Leach bake product to liquid ratio= 12.5 g/L 20 Leaching reagent to liquid= 27.0 g/L

15 Nb leached (%)

0 0 2 4 6 8 10 12 Ta leached (%)

Figure A 8.1(a), (b) and (c). Correlation between Ta and Nb leaching efficiencies at different conditions

239

Appendix

y = 0.91x + 0.03 y = 0.98x + 0.38 y = 1.02x + 0.57 70 y = 0.98x + 0.38 R² = 1.00 R² = 0.99 R² = 1.00 R² = 0.99

Leach temp= 90° C 50 Leach temp= 80° C Leach temp= 60° C Leach temp= 20° C 40 Leach acid conc= 4 M Leach acid conc= 3 M Leach acid conc= 2 M 30 Li leachedLi (%) Leach acid conc= 1 M Leach bake product to liquid ratio= 12. 5 g/L 20 Leach bake product to liquid ratio= 37.5 g/L Leach reagent to liquid ratio= 13.5 g/L Leach reagent to liquid ratio= 27.0 g/L 10 Leach reagent to liquid ratio= 40.5 g/L

0 0 10 20 30 40 50 60 70 80 Cs leached (%)

Figure A 8.2. Correlation between Li and Cs leaching efficiencies at different conditions

100 y = 0.96x + 0.12 y = 0.98x - 1.04 y = 0.98x - 1.04 y = 1.07x - 1.86 R² = 0.98 90 R² = 0.99 R² = 0.98 R² = 0.98

80 Leach temp= 80° C Leach temp= 60° C 70 Leach acid conc= 4 M Leach bake product to liquid ratio= 12.5 g/L 60 Leach bake product to liquid ratio= 25.0 g/L Leach reagent to liquid ratio= 13.5 g/L 50 Leach reagent to liquid ratio= 40.5 g/L

40 U leached (%) leached U

0 0 10 20 30 40 50 60 70 80 90 100 Fe leached (%) Figure A. 8.3. Correlation between U and Fe leaching efficiencies at different conditions

240

Appendix

100

90 y = 1.09x + 1.34 80 R² = 0.98 70 Leach temp= 90°C 60 Leach temp= 80°C leach temp= 60°C Leach temp= 20°C 50 Leach acid conc= 4 M Leach bake product to liquid ratio= 12.5 g/L 40 Leach bake product to liquid ratio= 25.0 g/L Leach bake product to liquid ratio= 37.5 g/L U leached (%) leached U 30 Leach reagent to liquid ratio= 13.5 g/L Leach reagent to liquid ratio= 27.0 g/L 20 Leach reagent to liquid ratio= 40.5 g/L 10 0 0 10 20 30 40 50 60 70 80 90 100 110 Mn leached (%)

Figure A.8.4. Correlation between U and Mn leaching efficiencies at different conditions

241