i

School of Chemical and Minerals

Engineering

CEMI427: Conceptual design of
uranium extraction plant

G.H. Coetzee 20253362
A. Nel 20311478
D. Postma 20243499
I.S. Scott 20259557

09 November 2009



Engineering
Faculty of
School of Chemical and Minerals Engineering
ii
Declaration

Declaration

We, G.H. Coetzee, A.Nel, D.Postma and I.S. Scott of IPL Engineering hereby declare that
the report entitled:

CEMI427: Conceptual design of uranium extraction plant

submitted for the partial fulfilment of the requirements for the degree B.Eng Chemical
engineering at the North-West University, Potchefstroom campus, is entirely our own work
with external sources referenced in the added reference list.

Sighed at Potchefstroom on the day ______ November 2009.




_______________________________ ________________________________
G.H. Coetzee A.Nel




_______________________________ ________________________________
D. Postma I.S.Scott

School of Chemical and Minerals Engineering
iii
Executive summary

Executive summary

South Africa is the leading country in regards to nuclear power technology, with regards to
the new nuclear power reactor, PBMR, on the horizon. The new technology pushed the
boundaries of Uranium extraction and enrichment, creating new opportunities for design and
optimization. For the South African market, Eskom is planning to increase their electricity
capacity by implementing only nuclear power (NEA & IAEA, 2007:313).

Uranium is a solid at room temperature and is the last discovered natural occurring element
on the periodic table, with an atomic number of 92. The fact that Uranium is a heavy metal,
makes it a highly toxic element for both human and environmental health. Uranium is a
naturally radioactive substance that emits gamma radiation which is very dangerous, but
very useful if used correctly.

Uranium in the enriched form is mainly used for electricity generation and medical
applications. The medical application is used to treat cancer patients, the gamma rays are
used to break the genetic structure of the cancerous cells. There is industrial use for
Uranium in the X-ray sector. The first use of Uranium was in fact for weapons of mass
destruction, but is considered an ethical aspect in todays terms.

The world-wide demand for uranium is increasing rapidly and the current production rate is
not able to satisfy this international demand. Although the South African demand is already
satisfied, the export market still holds great advantages. Therefore several uranium mining
and processing plants are being developed or expanded. AngloGold Ashanti is one of the
companies planning to expand its processing of uranium containing ores.

Problem statement

The South Uranium Plant (SUP) currently processes 240 000 ton of ore per month,
producing 624 ton of ADU per year. AngloGold Ashanti is aiming to increase the processed
ore feed to the SUP to 360 000 ton per month, with the additional 120 000 ton of ore per
month. The equipment on the existing plant should be evaluated to determine if the
equipment can handle the increased feed. The existing solvent extraction equipment is
completely exhausted, therefore this unit should be replaced entirely.

School of Chemical and Minerals Engineering
iv
Executive summary
Process overview

The proposed expansion design is able to handle the additional feed and will result in a
production rate of 1068 ton ADU per annum and an overall uranium recovery of 78%. This
design includes a modified process used for leaching which allow the use of the existing
pachuca tanks. Fixed-bed column ion exchange is used rather than CCIX which allows the
use of existing columns and a new solvent extraction (SX) unit is designed to replace current
SX process. The existing precipitation tanks and equipment have enough capacity to handle
the increased feed.

In the leaching process nitric acid is used as oxidant and sulphuric acid is used as lixiviant.
Due to the faster kinetics of nitric acid leaching, only eleven of the existing pachuca tank are
used for the expanded ore feed. Two of the existing unused pachuca tanks are used as
back-up tanks in case of equipment maintenance or breakdown. Each pachuca is air
agitated to ensure sufficient mixing resulting in adequate contact time. The kinetics of nitric
acid leaching is largely dependent on temperature and therefore the first pachuca tank is
heated with steam to above 30 C with additional steam addition to every fourth tank to
maintain the temperature required. A theoretical analysis on the temperature influence on
the leaching kinetics shows that an increased feed capacity is easily processed by
increasing the leaching temperature or the nitrate addition.

In the counter-current decantation (CCD) process, the leach product is washed to recover
the uranium containing liquids. The existing counter-current decantation equipment are
designed for approximately 8 000 ton ore per day, while an additional 5 000 ton ore per day
capacity is required for the expansion. Therefore an additional counter-current decantation
train is required to process the additional ore feed. Each train consists of six thickeners with
the first thickener in each train used as a clarifier to ensure minimum solids in the solution
sent to ion exchange. The floculant, Magnafloc 90L, is added to the thickeners, except for
the two clarifiers, to promote the sedimentation of the solids. A wash ratio of 1:1 is used to
wash the leach product and results in a uranium recovery of 99.99 %.

The solid slurry product stream from the counter-current decantation is sent to the
neutralization unit where slaked lime is used to increase the pH of the slurry to 10.5 as
required by the gold extraction plant. The neutralization is done in one of the existing
pachuca tanks and after neutralization, the slurry is sent to the gold plant. The liquid product
of the CCD section is sent to ion exchange unit for further processing.

School of Chemical and Minerals Engineering
v
Executive summary
The pregnant leach liquor, over flow from the clarifiers, proceeds to ion exchange where the
concentration of uranyl sulphate complexes is increased and several of the impurities are
removed. This is achieved by the selective adsorption and elution of the uranyl sulphate
complexes onto resin (Ambersep TM400). The ion exchange unit consists out of five fixed-
bed columns in series with an additional regeneration column. The current adsorption
columns are modified for fixed bed column ion exchange. One important modification is to
enclose the top part of the columns to allow for the back-wash process.

The ion exchange process is divided into four steps; adsorption, washing, elution and
regeneration. The resin remains stationary in the adsorption columns throughout the first
three steps of the process, except for the regeneration step where the resin is moved to the
elution column for treatment. The uranium concentration is increased during the elution step
where diluted sulphuric acid is used to remove uranyl sulphate complexes from the resin.
This results in a concentration of 3.7 g U
3
O
8
per litre in the eluate stream.

The eluate stream is sent to the counter-current solvent extraction section where the
uranium containing liquid is processed to further up concentration and remove impurities.
The solvent extraction process consists of four sections, i.e. extraction, scrubbing, stripping,
and regeneration. The organic solvent feed to the eluate feed has a ratio of 1.1:1. The
extraction section consists of three stages with an additional after settler to reduce solvent
loss. The extractant used is Alamine

336 with kerosene as diluent and isodecanol as third

phase modifier.

In the scrubbing section the impurities are washed from the solvent with demineralised water
in three mixer-settlers. The demineralised water is recycled to reduce cost with an
occasional purge to remove possibility of build-up in the system. The feed ratio of organic to
demineralised water is 5:1. The stripping section consists of four stages with an additional
after settler to reduce solvent loss. The first mixer-settler is used to reduce the pH of the
stripping solution which results in better pH control. The OK liquor leaving the stripping
section has a concentration of 12 g U
3
O
8
per litre which proceeds to the precipitation section.
The feed ratio of organic to stripping solution is 3.6:1. The organic solvent is continuously
regenerated in the last mixer-settler using a caustic solution with a feed ratio of 1:1.

In the precipitation section the OK liquor and ammonia is fed to the precipitation reactor
where the solid ammonium diuranate (ADU) product is formed. The OK liquor stream is
heated to temperature to 30 C for adequate reaction kinetics. Two precipitation reactors
are used to acquire efficient precipitation of the product. The density of the slurry product
School of Chemical and Minerals Engineering
vi
Executive summary
from the second precipitation reactor is increased using one thickener where the over flow is
recycled back to the stripping section of the solvent extraction process. The under flow is
washed in a series of centrifuges to ensure the product specification are met. All the existing
equipment for this section is adequate to handle the increased through-put.

Detail design of solvent extraction unit

In the detail design of the solvent extraction unit the number of stages for each step was
optimised using the McCabe-Thiele method and assumed stage efficiency. The mixers were
designed using basic guidelines for the specific impeller used. The settling kinetics is used
to design dimensions for the settlers. Using the settling kinetics a sensitivity analysis is done
on an estimation of capital cost to obtain an optimum settler vessel dimensions. A
mechanical drawing is done to give guidelines for the construction of the equipment.

Economic evaluation

This expansion project offers an impressive return on investment (ROI) rate of 60.5% with a
payback period of only 3.98 years. The total initial capital investment is calculated as R 847
million which consist of a fixed capital investment of R 720 million and working capital of R
126 million. The operating fixed cost is estimated at R 126 million per annum while the
variable operating cost amounts to R 198 000 per ton ADU produced. The revenue was
determined by assuming R 370 000 will be received for a ton ADU and 50% of the revenue
from the gold sales will be received. The internal rate of return (IRR) is calculated to be
352% with a net positive value (NPV) of R 2.36 billion after a period of 20 years.

Safety consideration

AngloGold Ashantis number one values by which the company is driven, is safety and
therefore attention has been dedicated to ensure a safe working environment and minimum
impact on the environment. The general safety design procedure was done by using the
Hazardous and Operability (HAZOP) studies to identify and evaluate the hazards associated
with this uranium extraction plant. The HAZOP studies identified potential risks if accidental
spills of ADU, nitric and sulphuric acid, Magnafloc 90L, kerosene and dust dispersion of the
ore occurs and avoidance measures are suggested. Many of the raw materials are
corrosive and are harmful if humans come into contact with these materials.

School of Chemical and Minerals Engineering
vii
Executive summary
Special attention was dedicated to the development of fire fighting measures and radiation
exposure control. The solvent extraction unit holds the highest fire risks and therefore
additional fire-fighting equipment is installed. A dead zone of 15 m is created around this unit
to minimize the damage of the other units in case of fire. To eliminate all possible ignition
sources, especially static discharge, all the electrical equipment is sent outside the dead
zone and is covered by zener barriers. Another major safety concern is the exposure to
radiation associated with the handling of uranium ores and products.

The greatest impact of processing plants on the environment is the release of chemical
waste into the atmosphere and water resources. The waste from the SUP include the solid
waste slurry containing the gold, which is neutralized and sent back to the gold plant and the
waste water containing nitrates, which is denitrified by using bio-organisms. A rehabilitation
and decommissioning procedure is included to ensure that the environment is eventually
restored and the impact of the plant on the environment is minimized.

Process control

Control objectives for plant wide control are identified and satisfactory control strategies are
developed with special attention to plant safety. For the solvent extraction unit an
ASPENTech

HYSYS simulation of the proposed control system is done. From this

simulation the feasibility of the control strategies are studied. The proposed control
strategies are sufficient to safely control the process disturbances in order to optimise
production.




School of Chemical and Minerals Engineering
viii
Table of contents

Table of contents

DECLARATION ii
EXECUTIVE SUMMARY iii
TABLE OF CONTENTS viii
LIST OF FIGURES xiv
LIST OF TABLES xvii

CHAPTER 1: INTRODUCTION 1

1.1. Motivation 1
1.2. Problem statement 2
1.3. Assumptions and design limitations 2

CHAPTER 2: LITERATURE STUDY 3

2.1. Ore 3
2.2. Uranium 4
2.2.1. Uses 5
2.2.2. Future tendencies 6
2.2.3. Environmental impact 6
2.2.4. Safety considerations 7
2.3. Ammonium diuranate 8
2.4. Market research 9
2.4.1. Global uranium market 9
2.4.1.1. World demand 10
2.4.1.2. World production 11
2.4.1.3. Relationship between production and demand 11
2.4.2. Southern African market 13
2.4.2.1. South African demand 14
2.4.2.2. South African production 15
2.4.2.3. Relationship between demand and production 16
2.5. Process background 16
2.5.1. Leaching 16
2.5.1.1. Leaching process alternatives 17
School of Chemical and Minerals Engineering
ix
Table of contents
2.5.1.2. Leaching raw materials 22
2.5.1.3. Leaching design parameters 24
2.5.1.4. Leaching kinetics 25
2.5.2. Ion exchange 27
2.5.2.1. Ion exchange process alternatives 33
2.5.2.2. Ion exchange raw materials 39
2.5.2.3. Ion exchange design parameters 40
2.5.2.4. Ion exchange kinetics and equilibrium 40
2.5.3. Solvent extraction 43
2.5.3.1. Process alternatives 44
2.5.3.2. Raw materials added 45
2.5.3.3. Design parameters 46
2.5.3.4. Kinetics 47
2.5.4. Precipitation 47
2.5.4.1. Precipitation process alternatives 48
2.5.4.2. Precipitation raw materials 50
2.5.4.3. Precipitation design parameters 52
2.5.4.4. Precipitation kinetics 53
2.6. Economic evaluation 55

CHAPTER 3: PROCESS DEVELOPMENT 57

3.1. Level 0: Input information 57
3.1.1. Reactions and reaction conditions 58
3.1.2. Desired production rate and purity 62
3.1.3. Raw materials 62
3.1.4. Processing constraints 64
3.1.5. Other plant and site data 65
3.1.6. Physical properties of all components 65
3.2. Level 1: Batch versus continuous 66
3.2.1. Process units needed 67
3.2.2. Interconnections among units 68
3.2.3. Estimate the optimum processing conditions 69
3.2.4. Additional conceptual design information 70
3.3. Level 2: Input-output structure of the flowsheet 71
3.3.1. Feed purification 71
3.3.2. Recycle streams 72
School of Chemical and Minerals Engineering
x
Table of contents
3.3.3. Removal and purge streams 72
3.3.4. Number of product streams 72
3.3.5. Preliminary economic potential analysis 74
3.4. Level 3: Recycle structure of the flowsheet 75
3.4.1. Reactor systems required 76
3.4.2. Number of recycle streams 76
3.4.3. Abundance of reactants 78
3.4.4. Operational considerations 78
3.4.5. Recycle economic evaluation 79
3.5. Level 4: Separation systems 80
3.5.1. General structure 80
3.5.2. Vapour recovery system 81
3.5.3. Solid recovery system 81
3.5.4. Liquid recovery system 83
3.5.5. Separation system economic evaluation 85
3.6. Level 5: Heat integration 86
3.7. Equipment design 86
3.8. Mass and energy balance 90
3.8.1. Mass balance 91
3.8.2. Energy balance 91
3.9. Process Flow Diagrams and process description 92
3.9.1. Unit 1 (U01): Leaching, CCD, and neutralization 97
3.9.2. Unit 2 (U02): Ion exchange 99
3.9.3. Unit 3 (U03): Solvent extraction 101
3.9.4. Unit 4 (U04): Precipitation 103
3.10. Innovations 105

CHAPTER 4: DETAIL DESIGN 108

4.1. Choice of system type 108
4.2. Kinetics and thermodynamics 110
4.3. Detail chemical design 112
4.3.1 Number of stages 112
4.3.2 Mixer 113
4.3.3 Settler 114
4.3.3 Pipe sizing 115
4.3.4. Pump sizing 116
School of Chemical and Minerals Engineering
xi
Table of contents
4.3.5. Control valve sizing 117
4.4. Start-up and shut-down procedures 119
4.4.1. Commissioning of process unit 119
4.4.2. Shutdown procedures 121
4.4.3. Emergency shutdown procedures 121
4.5. Mechanical aspects 123

CHAPTER 5: TECHNO-ECONOMIC EVALUATION 130

5.1. Definitions and assumptions 130
5.1.1. Definitions 131
5.1.2. Assumptions 134
5.2. Estimation of capital, operating cost and revenue 134
5.2.1. Capital investment 135
5.2.2. Operating cost 137
5.2.3. Revenue 138
5.3. Cash flow analysis 138
5.4. Economic sensitivity analysis 140
5.5. Recommendation for profitability 142

CHAPTER 6: SAFETY AND ENVIRONMENT 143

6.1. Overall safety specifications 143
6.1.1. Leaching 144
6.1.2. Ion exchange 145
6.1.3. Solvent extraction 145
6.1.4. Precipitation 146
6.1.5. Fire fighting 146
6.1.6. Training and personal safety 148
6.1.7. Danger zones and signs 149
6.1.8. Emergency response plan 150
6.1.9 Radiation control 151
6.2. HAZOP level 1 152
6.2.1. Project definition 153
6.2.2. Process description 154
6.2.3. Assessment of chemical hazards 155
6.2.4. Assessment of chemical interactions 156
School of Chemical and Minerals Engineering
xii
Table of contents
6.2.5. Means of handling 158
6.3. Environmental impact and management 159
6.3.1. Waste block diagram analysis 159
6.3.2. Handling and disposal of wastes 162
6.3.3. Accidental releases of hazardous materials 164
6.3.4. Rehabilitation and decommissioning of plant 164
6.4. Plant layout and positioning 165

CHAPTER 7: PROCESS CONTROL 167

7.1. Plant wide control 167
7.1.1. Unit 1: Leaching, CCD and neutralization 168
7.1.2. Unit 2: Ion exchange 175
7.1.3. Unit 3: Solvent extraction 179
7.1.4. Unit 4: Precipitation 182
7.2. Specific safety considerations for solvent extraction 188
7.3. Detailed process control for solvent extraction 199
7.3.1. Flow control loops 201
7.3.2. Tank level control loops 203
7.3.3. pH control loops 205
7.3.4. Range, alarms and trips 207
7.4. Dynamic control analysis 209
7.4.1. Tuning control loops 210
7.4.2. Variable pairing 216
7.5. Conclusion 218

REFERENCES 219

APPENDIX A: MASS AND ENERGY BALANCE 229

APPENDIX B: EQUIPMENT SIZING 254

APPENDIX C: DETAIL EQUIPMENT CALCULATIONS 264

APPENDIX D: TECHNO-ECONOMIC CALCULATIONS 283

APPENDIX E: PLANT LAYOUT AND POSITIONING 299

APPENDIX F: MSDS INFORMATION 302
School of Chemical and Minerals Engineering
xiii
List of figures

List of figures

CHAPTER 2: LITERATURE STUDY

Figure 2.1: Processed yellow cake 8
Figure 2.2: Global energy consumption 9
Figure 2.3: Distribution of global uranium requirements for 2006 11
Figure 2.4: Cumulative global uranium requirements and demand 13
Figure 2.5: Gold price for 2005 to 2009 14
Figure 2.6: The classification of the various leaching techniques 18
Figure 2.7: In-situ leaching process description 19
Figure 2.8: Illustration of a heap leach process 20
Figure 2.9: Typical adsorption 30
Figure 2.10: Eluate concentration profile 32
Figure 2.11 Fixed bed ion exchange column layout 35
Figure 2.12: Typical solvent extraction process 44

CHAPTER 3: PROCESS DEVELOPMENT

Figure 3.1: Simplified block flow diagram of process 68
Figure 3.2: Input-output structure of overall process 73
Figure 3.3: Block flow diagram for reactor system 76
Figure 3.4: Recycle structure 77
Figure 3.5: General separation structure 81
Figure 3.6: Waste solid recovery system 82
Figure 3.7: Solid recovery system for ADU 83
Figure 3.8: Schematic representation of ion exchange 84
Figure 3.9: Schematic representation of solvent extraction 84

CHAPTER 4: DETAIL DESIGN

Figure 4.1: Representation of mixer-settler equipment 109
Figure 4.2: Stripping loading isotherm with (NH
4
)
2
SO
4
111
Figure 4.3: Back-up pump configuration 117
Figure 4.4: Pumper mixer Impeller 124
School of Chemical and Minerals Engineering
xiv
List of figures
Figure 4.5: pH box for measurement 125

CHAPTER 5: TECHNO-ECONOMIC EVALUATION

Figure 5.1: Cash flow diagram 139
Figure 5.2: Economic sensitivity analysis 141

CHAPTER 6: SAFETY AND ENVIRONMENT

Figure 6.1: Examples of necessary signage 150
Figure 6.2: HAZOP structure for plant 153
Figure 6.3: General waste block diagram structure 159
Figure 6.4: Leaching waste block diagram 160
Figure 6.5: Ion exchange waste block diagram 160
Figure 6.6: Solvent extraction waste block diagram 161
Figure 6.7: Precipitation waste block diagram 161
Figure 6.8: Plant layout 166

CHAPTER 7: PROCESS CONTROL

Figure 7.1: Control schematic of the leaching section 169
Figure 7.2: Control schematic of the counter current decantation section 172
Figure 7.3: Control schematic of the neutralization process 174
Figure 7.4: Control schematic of the adsorption stage of ion exchange 176
Figure 7.5: Control schematic of the back-wash stage of ion exchange 177
Figure 7.6: Control schematic of the elution stage of ion exchange 178
Figure 7.7: Control schematic of the regeneration stage of ion exchange 179
Figure 7.8: Basic control schematic of the solvent extraction unit 180
Figure 7.9: Control schematic of the precipitation reactors in the precipitation unit 183
Figure 7.10: Control schematic of the solid-liquid separation in the precipitation unit 186
Figure 7.11: Extraction simulation flowsheet 211
Figure 7.12: Ratio control results in extraction section 212
Figure 7.13: Scrubbing simulation flowsheet 213
Figure 7.14: The primary control loop results 214
Figure 7.15: The secondary control loop results 215


School of Chemical and Minerals Engineering
xv
List of figures
APPENDIX A: MASS AND ENERGY BALANCES

Figure A.1: Aspen Tech

simulation 230

APPENDIX B: EQUIPMENT SIZING

Figure B.1: Influence of nitrate concentration on reaction kinetics 257
Figure B.2: Influence of temperature on reaction kinetics 258
Figure B.3: Leach tank size evaluation 258

APPENDIX C: DETAIL DESIGN CALCULATIONS

Figure C.1: Modelling of the extraction loading isotherm 264
Figure C.2: Loading isotherm data for the stripping section 265
Figure C.3: Solvent extraction process flow schematic 265
Figure C.4: McCabe-Thiele for the Extraction section 267
Figure C.5: McCabe-Thiele for the Stripping section 268
Figure C.6: Vessel width sensitivity analysis 273
Figure C.7: Dispersion layer velocity sensitivity analysis 274
Figure C.8: Economical sensitivity analysis on the extraction settlers 275
Figure C.9: Dispersion height profile over the length of the extraction vessel 276
Figure C.10: Economical sensitivity analysis on the stripping settlers 278
Figure C.11: Height profiles for internal recycle and without internal recycle 279

APPENDIX D: TECHNO-ECONOMIC CALCULATIONS

Figure D.1: Cost of general-purpose centrifugal pumps 285
Figure D.2: Cost of electric motors 285
Figure D.3: Installed cost for single-compartment thickeners 286

APPENDIX E: PLANT LAYOUT AND POSITIONING

Figure E.1: Google Earth air photo 299
Figure E.2: Existing plant layout of South Uranium Plant with number legend 300
Figure E.3: Existing plant layout of South Uranium Plant with colour legend 301
School of Chemical and Minerals Engineering
xvi
List of tables

List of tables

CHAPTER 2: LITERATURE STUDY

Table 2.1: Physical and chemical properties of uranium 4
Table 2.2: Proposed interim water quality guidelines for the radiation dose in
drinking water 7
Table 2.3: Uranium plants in South Africa 15
Table 2.4: Typical composition of Rand leach liquor 27
Table 2.5: Suggested concentrations for eluant choices 31
Table 2.6: Feed and product streams for ion exchange 39
Table 2.7: Comparison of solvent extractor types 45
Table 2.8: Comparison of neutralization reagents 51
Table 2.9: Summary of uranium project survey 56

CHAPTER 3: PROCESS DEVELOPMENT

Table 3.1: Leaching reactions 58
Table 3.2: Reaction information for ion exchange 59
Table 3.3: Reaction information for solvent extraction 60
Table 3.4: Reactions kinetics and conversion 61
Table 3.5: Ore composition 63
Table 3.6: Summary of raw materials 64
Table 3.7: Utilities 65
Table 3.8: Physical properties of unknown compounds 66
Table 3.9: Processing conditions 69
Table 3.10: Product stream classification 74
Table 3.11: Level 2 preliminary mass balance with prices 75
Table 3.12: Capital and operation costs for reactors 79
Table 3.13: Separation systems cost calculations 85
Table 3.14.a: Summary of leaching equipment 87
Table 3.14.b: Summary of counter-current decantation equipment 88
Table 3.14.c: Summary of ion exchange equipment 89
Table 3.14.d: Summary of precipitation equipment 90
Table 3.15: Overall mass balance 91
School of Chemical and Minerals Engineering
xvii
List of tables
Table 3.16: Innovations 106

CHAPTER 4: DETAIL DESIGN

Table 4.1: Extraction loading isotherm constants 113
Table 4.2: Results from McCabe-Thiele method 113
Table 4.3: Mixer box design results 114
Table 4.4: Designed dimensions for the settler vessels 115
Table 4.5: Solvent extraction pipe sizes 116
Table 4.6: Pump sizing results 116
Table 4.7: The calculated pressure drop and valve coefficients 118

CHAPTER 5: TECHNO-ECONOMIC EVALUATION

Table 5.1: Multiplying factors used in the delivered-equipment cost method 136
Table 5.2: Operating costs 137
Table 5.3: Revenue from product and by-product sales 138
Table 5.4: Profitability results from cash flow analysis 139

CHAPTER 6: SAFETY AND ENVIRONMENT

Table 6.1: Examples of radiation dosages 152
Table 6.2: Projected production rate for the upgraded plant 153
Table 6.3: List of chemicals 155
Table 6.4: Chemical hazard data sheet (HS1A) 156
Table 6.5: Chemical interactions data sheet (HS1B) 157
Table 6.6: Means of handling (HS1C) 158
Table 6.7: Waste classification 162
Table 6.8: Accidental releases 164



CHAPTER 7: PROCESS CONTROL

Table 7.1: HAZOP 3 Proforma HS3A form 191
Table 7.2: HAZOP 3 record for Proforma HS3A 192
Table 7.3: Specifications for flow controllers 203
Table 7.4: Specifications for drain system level control 204
School of Chemical and Minerals Engineering
xviii
List of tables
Table 7.5: Specifications for feed flow level control 205
Table 7.6: Specifications for pH control 206
Table 7.7: RAT list for the solvent extraction section 207
Table 7.8: Control loop tuning values 216
Table 7.9: Relative gain array for control on U03-ST07 217

APPENDIX A: MASS AND ENERGY BALANCES

Table A.1: Summary of mass balances for individual units 229
Table A.2: Mass balance for leaching process 231
Table A.3: Mass balance for CCD process 234
Table A.4: Mass balance for neutralization process 237
Table A.5: Mass balance for ion exchange process 240
Table A.6: Mass balance for solvent extraction process 243
Table A.7: Mass balance for precipitation process 246
Table A.8: Leaching pachucas energy balance 249
Table A.9: Counter-current decantation energy balance 251
Table A.10: Neutralisation energy balance 252
Table A.11: Precipitation energy balance 253

APPENDIX B: EQUIPMENT SIZING

Table B.1: Leaching kinetics constants 254

APPENDIX C: DETAIL DESIGN CALCULATIONS

Table C.1: Variables for Equation C-22 and C-23 282

APPENDIX D: TECHNO-ECONOMIC CALCULATIONS

Table D.1: Purchased equipment cost 287
Table D.2: Capital investment calculations 288
Table D.3: Detailed operating cost calculation 290
Table D.4: Revenue calculations 292
Table D.5: Cash flow analysis 292
Table D.6: Chance in variables 298
Table D.7: Economic sensitivity analysis in terms of percentage rise/fall of NPV 298
School of Chemical and Minerals Engineering
1
Chapter 1: Introduction

Chapter 1: Introduction

Uranium is a remarkable element that can be used as an abundant source of concentrated
nuclear energy. Although the radioactivity and possible application in weaponry made the
world sceptical about the uses of uranium, the recent developments in nuclear power
stations and reactors have created an entire new market for uranium.

Uranium was discovered in 1789 by Martin Heinrich Klaproth just after the discovery of the
planet Uranus, from which the name for uranium derives. It appears that uranium was
formed as a decay product of elements with a higher atomic weight, in supernovae about 6.6
billion years ago. Uranium is widely available in most rocks on earth and the radioactive
decay of uranium is a main heat source inside the earth (Cleveland, 2008).

In nature, uranium is mainly found as an oxide such as Triuranium octaoxide (U
3
O
8
) and
uranium dioxide (UO
2
), but can also occur in the form of uranium hexafluoride (UF
6
) and
uranium tetrafluoride (UF
4
). Uranium is also found in the form of a metal which is more
dense, more ductile and softer than steel, with a density of 19 000 kg/m
3
(Cleveland, 2008).

The first use of uranium dates back to 79 A.D., when it was used to produce yellow-coloured
glass. Presently uranium is used mainly to generate energy, in research reactors and can
also be used to create explosives and weaponry. Uranium is essential for the nuclear
enterprise and the rapid development of nuclear reactors and power plants results in an
increasing demand for uranium (Cleveland, 2008).

1.1. Motivation

Uranium is a by-product of gold extraction which ends up in the mine tailings which has an
enormous environmental impact. The pre-leaching of the gold ore removes the uranium and
some of the impurities while increasing the gold recovery. Further more this results in a
great economic potential due to the increased gold recovery and uranium by-product. The
global demand for alternative energy sources has placed strain on the global uranium supply
and stockpiles. Due to the economic and environmental advantages associated with
uranium extraction, it is necessary to conduct a feasibility study on the expansion of the
South Uranium Plant (SUP) at AngloGold Ashanti near Orkney.
School of Chemical and Minerals Engineering
2
Chapter 1: Introduction
1.2. Problem statement

The existing South Uranium Plant (SUP) should be upgraded to handle an additional feed of
120 000 ton ore per month which increases the total feed to 360 000 ton ore per month. The
additional ore feed is received from another gold plant and will have slightly different
specifications as the regular feed. The preliminary economic evaluation showed that the
following is needed:

Upgrade of the leaching section.
Upgraded or new ion exchange section.
New solvent extraction section.
Upgrade of precipitation section.

Information given:

The upgraded equipment will be situated on the existing SUP.
The final product i.e. yellow cake (ADU) at a mass percentage of 35% U
3
O
8
that
should be stored for further shipping.
All utilities and certain raw materials are provided at the existing plant at a fixed cost.
Gas and liquid emissions should adhere to applicable laws and regulations and
should be processed on site, but process equipment does not have to be designed.

1.3. Assumptions and design limitations

All the assumptions made throughout the feasibility study is given in the respective sections.
The assumptions are as far as possible based on literature sources, existing plant data, and
logical reasoning. The aspect falling outside the battery limits for the design project are as
follows:

Crushing and milling of ore received from gold plant.
Gold processing.
The delivery of utilities and raw materials.
The treatment of wastes.
Transport of ADU product to NUFCOR.

School of Chemical and Minerals Engineering
3
Chapter 2: Literature survey

Chapter 2: Literature survey

This chapter gives a brief oversight into the processing of uranium and describes
subsequent effects on the environment, population and economics. In this chapter the
following aspects are addressed:

Ore that is used in the extraction of uranium.
Uranium, its properties and uses.
The product ammonium diuranate.
The market research for uranium.
Background on uranium processing processes.
Economic evaluation for uranium extraction from literature.

Understanding and respecting the above aspects will lead to sustainable development,
which is the most important aspect to consider.

2.1. Ore

The ore used at the AngloGold Ashanti Southern Uranium Plant is mined from the
Witwatersrand basin. The ore is primarily used for gold extraction, with uranium as a main
co-product. The ore from the Witwatersrand Basin in South Africa is a quartz-pebble deposit
type (British Geological Survey, 2007: 5).

Quartz-pebble deposit type is formed from ancient sedimentary deposits buried when the
atmosphere was believed to be less oxidizing than today. From literature it is also believed
that the rapid filling of the basin by rivers isolated the uranium before oxidation could take
place. The typical grade of the uranium in the quartz-pebble deposit type ore is 130 to 1100
ppm uranium (British Geological Survey, 2007: 7).

The ore sent to AngloGold Ashanti South Uranium Plant, consist of the following non
uranium containing minerals (Lottering et al., 2007: 18)



School of Chemical and Minerals Engineering
4
Chapter 2: Literature survey
Quartz (SiO
2
).
Muscovite (K-Al-silicate).
Chlorite (Mg-Fe-Al-silicate).
Pyrophylite (Al
4
(Si
8
O
2
)(OH)
4
).
Pyrite (FeS
2
).
Albite (Na-Al-silicate).

As well as the above minerals, the uranium containing minerals of the ore are (Lottering et
al., 2007: 18):

84.9% Uraninite (UO
2
).
12.9% brannerite ((U,Th,Ca)(Ti,Fe)
2
O
6
).
2% U-phosphate ((U,Cl)PO
4
).
0.2% coffinite (U(SiO)
41-x
(OH)
4x
).

Uraninite is the mineral from which it is the simplest to extract uranium from, and the
extraction chemistry will be based on this mineral.

2.2. Uranium

Uranium is the last natural occurring metal in the periodic table and is ranked 49
th
most
abundant element in the earths crust out of 92 elements. Uranium was discovered in 1789
by Martin Heinrich Klaproth in Berlin, German, while investigating the mineral pitchblende.
Uranium is a gray metal which is chemically reactive. Table 2.1 gives some of uraniums
physical and chemical properties (Enghag, 2004: 1166-1169).

Table 2.1: Physical and chemical properties of uranium
Property Information
Symbol U
Molar weight (g/mol) 238.03
Density (kg/m
3
) 18 950
Melting point (K) 1405.5
Isotope range, all nuclides radioactive 218 to 242

School of Chemical and Minerals Engineering
5
Chapter 2: Literature survey
The history and properties of uranium is important, but will serve no purpose if there is no
use for the element. In the next sub-section the uses of uranium is discussed.

2.2.1. Uses

Uranium has various uses, with its main use in the generation of electricity. It is estimated
that 95% of all uranium is used to generate electricity, with the remaining 5% aimed at other
uses (British Geological Survey, 2007: 15).

To create electricity from uranium, nuclear power stations use enriched uranium-235 as heat
source to convert water into steam. The steam produced from this is used to turn turbines
which generate the electricity. This principle is the same as fossil fuel power stations, but
the fossil fuel is replace by a significantly lower amount of uranium and with little to none
emissions (British Geological Survey, 2007: 15).

Some of the other uses of uranium include the following (British Geological Survey, 2007:
18-19):

Nuclear-powered ships.
Research reactors.
Desalination.
Weapons.

The South African Nuclear Energy Corporation (NECSA) has a research nuclear reactor
situated in Pelendaba, named the Safari-1 reactor. Currently this reactor is the leading
producer of the medical isotope molybdenum-99 in the world. One of the other main uses of
the reactor is the irradiation of silicon semiconductors (NECSA, 2009).

For uranium to be used in all these practices, there needs to be a supply in the future, the
next sub-section will discuss the future tendencies of uranium production in South Africa.






School of Chemical and Minerals Engineering
6
Chapter 2: Literature survey
2.2.2. Future tendencies

With the rise of concern about the global environment and global warming, uranium is set to
replace fossil fuels in the generation of electricity. With the increase in nuclear reactors, the
consumption of uranium will increase, which must lead to an increase in production of
uranium.

In 2009 the production of U
3
O
8
in South Africa is proposed to be 2 800 tons per annum.
Furthermore it is estimated that the production of U
3
O
8
will surpass 5 000 tons per annum in
2011. This value will be reached if all the proposed projects start on schedule with the
proposed production rates (Damarupurshad, 2007: 11).

2.2.3. Environmental impact

Uranium is naturally spread widely throughout the environment and can be found in small
amounts in rock, soil, air and water. Humans release additional uranium metals and
compounds into the environment as a result of milling and mining which causes
environmental and health concerns (Lenntech, 2008). Uranium poses two threads that will
be investigated, its toxicity and the effects of radioactivity.

Uranium itself is radioactive, though with the major isotope U-238 having a half-life equal to
the age of the earth; it is certainly not strongly radioactive (World Nuclear Association, 2008).
Human activities such as mining enhance the transport of radionuclides into the environment
by transferring the ore from underground to tailing dams on the surface. These
radionuclides can separate from the bulk ore and pass into the environment via water
streams, for example, uranium in underground water pumped to the surface, and uranium
leakage from tailing dams resulting from oxidation of pyrite (Wendel, 1998:92).

Uranium is a heavy metal that is also toxic chemically. The toxicity becomes a danger when
it is found in groundwater contaminating soil where crops are grown, and in drinking water.
As water, containing oxygen, passes over rocks and soil, many compounds and minerals,
such as uranium, dissolve and is transferred into the groundwater. Uranium in groundwater
also results from human activities such as mining, combustion of coals and other fuels, the
use of phosphate fertilizers, and nuclear power production (Skipton et al 2008:1). The table
below shows the interim water quality guidelines for the radiation dose in drinking water
proposed by the Department of Water Affairs & Forestry (Kempster, 1999).
School of Chemical and Minerals Engineering
7
Chapter 2: Literature survey
Table 2.2: Proposed interim water quality guidelines for the radiation dose in drinking water
Radiation dose (nSv/a) Suitability
<= 0.1 (WHO reference level) Ideal, suitable for lifetime use
> 0.1 and <= 0.25 Water acceptable for lifetime use, subject to
confirmation of dose.
> 0.25 and <= 1 Water acceptable for short term use. Use in
longer term (lifetime) requires further
investigation.
> 1 and <= 5 Unacceptable for lifetime use
> 5 Unacceptable even for short term use

The health risks associated with uranium include kidney diseases, diseases of the
respiratory tract and cancer. Uranium also binds to biological molecules, is distributed within
the body and builds up in bone and teeth (Lindemann, 2008:5).

2.2.4. Safety considerations

To provide a safe working environment it is important to identify the potential health hazards,
provide the correct first aid measures and accidental release measures, have appropriate
handling and storage procedures and have preventative exposure controls and personal
protection.

The potential hazards identified when working with uranium include acute and chronic health
effects. It can be corrosive and irritant when in it comes in contact with the skin or eyes.
Liquid or spray mist which is ingested or inhaled may produce tissue damage particularly on
mucous membranes of eyes, mouth and respiratory tract. Inhalation may also produce
severe irritation of respiratory tract, characterized by coughing, choking or shortness of
breath. Repeated or prolonged exposure can produce accumulation and damage in target
organs. Severe over-exposure can result in death (Science Lab. 2008).






School of Chemical and Minerals Engineering
8
Chapter 2: Literature survey
There must be first aid measures for eye contact, skin contact, ingestion and inhalation both
minor and serious. If there is a small spill it must be diluted with water, absorb with an inert
dry material and place in an appropriate waste disposal container. If there is a larger spill
the MSDS must be consulted for correct clean-up method. The uranium must be kept in a
dry container and should be stored in a separate safety storage cabinet or room. When
working with uranium personal protection include gloves, boots, face shield, full suit and
approved/certified vapor respirator (Science Lab, 2008).

For additional safety measurements, employees can be monitored to ensure the uranium
levels in their bodies are under the acceptable limit.

2.3. Ammonium diuranate

Yellow cake is the universal name for ammonium diuranate (ADU), which is produced at
most uranium extraction plants. ADUs chemical formula is (NH
4
)
2
U
2
O
7
, which is yellow in
color and has a melting point of 2878 C (Hausen, 2009). Figure 2.1 is a picture of
processed yellow cake (Snooper, 2008).



Figure 2.1: Processed yellow cake

In South Africa, the yellow cake produced is sent to NUFCOR for processing consisting of
filtering, drying and calcining. After processing the yellow cake is shipped to international
countries for use in nuclear reactors and for its other uses.

School of Chemical and Minerals Engineering
9
Chapter 2: Literature survey
2.4. Market research

As with most commodities, the price of uranium is greatly influenced by the supply and
demand. It is therefore very important to determine the current state of the uranium market
which will shed light on the present and future economic viability of uranium production.
Since the extraction of uranium from the gold ore holds recovery benefits for the gold
extraction, the gold market outlook also has an influence on the economic viability of
uranium production. The South African and global markets are investigated and compared
in the following section.

2.4.1. Global uranium market

The climate changes, that are observed globally, cause increasing concern for the release of
greenhouse gasses such as carbon dioxide and methane. For this reason legislation is
implemented to decrease the use of fossil fuels for energy. As mentioned before, uranium is
mainly used in the nuclear energy industry and this is a very attractive alternative fuel
compared to fossil fuels. This results in an increasing global demand for uranium.

Figure 2.2 shows the world energy use from 1980 to 2006 and what type of source the
energy came from as derived from tables produced by the Energy Information Administration
(EIA) (2008).

0
20
40
60
80
100
120
140
G
i
g
a
w
a
t
t

h
o
u
r
s
M
i
l
l
i
o
n
s
Petroleum Dry Natural Gas Coal Net Hydroelectric Power Net Nuclear Electric Power Others

Figure 2.2: Global energy consumption

School of Chemical and Minerals Engineering
10
Chapter 2: Literature survey
As seen in Figure 2.2, the world has become increasingly dependent on nuclear energy as a
source of energy. The worlds energy demand is increasing with a steady fraction of energy
obtained from nuclear power plants.

2.4.1.1. World demand

Since the world nuclear energy consumption is increasing, the global uranium demand is
also increasing. There was however a drop in world uranium usage from 66 500 tU in 2007
to 64 500 tU in 2008. This drop in uranium usage was due to the closure of seven units in
Japans largest nuclear power plant (Kashiwazaki-Kariwa) and other nuclear power plants
were closed for maintenance and repairs. It is reported by the Department: Minerals and
Energy of South Africa (2008:52) that in 2007, uranium was used as fuel in 439 nuclear
power plants around the world and that there are 36 currently under construction. This
means that the uranium demand will increase when these new nuclear power plants are
finished (Department: Minerals and Energy of South Africa, 2008:52).

There are other factors that may decrease the overall global uranium demand. The
decommissioning of old or first generation nuclear power plants will eventually start causing
a decrease in uranium demand. The efficiency of nuclear power plants is also constantly
improved through new research and technologies, which decreases the uranium demand.
Regardless of these small factors, the global uranium demand is expected to increase since
the world electricity requirements are increasing rapidly (World Nuclear Association, 2008).

Since the planning and construction of nuclear power plants is expensive, developed
countries are the largest uranium consumers. However, nuclear power is a very attractive
source of energy for developing countries with a lack in infrastructure, such as African
countries, since small amounts of fuels is necessary and transport is easier. Figure 2.3
shows the geographical distribution of global uranium requirements in 2006 (NEA & IAEA,
2007:52).

School of Chemical and Minerals Engineering
11
Chapter 2: Literature survey


Figure 2.3: Distribution of global uranium requirements for 2006

As seen from Figure 2.3, the developed world uses the most uranium. The United States of
America is the largest user of uranium followed by France, Japan and then Germany
(Department: Minerals and Energy of South Africa, 2008:53). The United States and
Canada are large producers of uranium and therefore provide for their own demand.

France has many nuclear power plants but virtually no natural uranium resources. Most of
Frances 10 500 tU needed per year for electricity generation is imported from Canada and
Niger. France do supply a part of its own uranium demand through conversion and
enrichment plants that recycle used uranium. This causes 30% more energy produced from
the original uranium and a great reduction in the amount of nuclear waste disposed (World
Nuclear Association, 2009).




School of Chemical and Minerals Engineering
12
Chapter 2: Literature survey
2.4.1.2. World production

In 2008 the world produced 44 000 tU which is quite higher than the 41 300 tU in 2007. The
largest uranium producers are Canada, Kazakhstan and Australia which jointly produced
approximately 60% of the global uranium production in 2008. The forecast global uranium
production for 2009 is 49 400 tU according to the World Nuclear Association (2009). This
sharp increase is expected because of major increases in production of the leading uranium
producing countries.

The methods of uranium production have also changed over the last few years. In situ
leaching makes the extraction of uranium much more economically viable because the ore
does not have to be mined, crushed and milled. This makes the In situ leaching method an
increasingly attractive choice for the production of uranium. Currently 28% of the uranium
produced is produced with the In situ leaching method (World Nuclear Association, 2009).

2.4.1.3. Relationship between production and demand

From the figures in the section above it is seen that in 2008 the uranium production rate only
provided for approximately 70% of the global demand. The rest of the uranium demand is
currently supplied for with uranium stockpiles. These stockpiles were built up during the cold
war when the United States and Russia produced high-enriched uranium (HIU) for the
manufacturing of nuclear weapons. These weapons are now being dismantled and the
uranium used in commercial nuclear reactors. In 2006 the uranium from the military
provided for more than 50% of the nuclear fuel in United States nuclear reactors (Johnson,
2007).

There is concern for when these stockpiles are depleted, because the current production
rate is not enough for the annual uranium demand. This is seen in Figure 2.4 which shows
the cumulative global uranium requirements and demand (NEA & IAEA, 2007:75).






School of Chemical and Minerals Engineering
13
Chapter 2: Literature survey


Figure 2.4: Cumulative global uranium requirements and demand

From the slopes of the end of the graph in Figure 2.4, it is derived that at some point in the
future the global demand will exceed the global supply of uranium. If the global uranium
production increases as it is currently doing, then the uranium supply will be able to sustain
the growing demand. Another obstacle in uranium production is the availability of
enrichment facilities. There are strict regulations for the creation of enrichment facilities to
ensure it is only used to produce low-enriched uranium (LEU) and not HEU which is readily
used to construct nuclear weapons.

2.4.2. Southern African market

Only South Africa and Namibia are currently producing uranium in Southern Africa. The
Rossing mine in Namibia is the worlds third largest uranium producing mine and produced 3
400 tU in 2008. The uranium demand of these countries is very low which means that most
of the produced uranium is exported. In South Africa uranium is produced as a by-product of
the gold extraction process and causes an improved gold recovery. Therefore the uranium
extraction from gold ore has an increased economical potential which is greatly dependant
on the gold price.

In 2008 the world economy entered a crisis period which caused many commodity prices,
including that of gold, to fall. In Figure 2.5 the gold price over the last five years is shown
(Goldprice, 2009).
School of Chemical and Minerals Engineering
14
Chapter 2: Literature survey



Figure 2.5: Gold price for 2005 to 2009

The effect of this economy crisis on the gold price is clearly visible in Figure 2.5. However
Figure 2.5 also shows that the gold price did recover and is stable. This provides
economical potential and security for the production of uranium as a by-product of gold
extraction.

2.4.2.1. South African demand

In Southern Africa there are only three nuclear reactors with a very low nuclear fuel
requirement. Two are situated at the Koeberg nuclear power plant near Cape Town, South
Africa. These two reactors consume only 292 tU per year and provide 5% of South Africas
electricity (NEA & IAEA, 2007:313). Since Southern Africa does not have any operating
uranium enrichment facility, all of its enriched uranium fuel has to be import (World Nuclear
Association, 2009).




School of Chemical and Minerals Engineering
15
Chapter 2: Literature survey
Eskom is South Africas leading electricity company, providing 95% of its electricity and 60%
of Africas electricity. In 2008 South Africas electricity demand exceeded the supply and
wide spread power failures became a great concern. This electricity crisis prompted Eskom
to drastically increase its capacity. Eskoms initial plans to increase its nuclear capacity to
20 GWe by 2025 have been halted due to a lack of finances. New projections by Eskom are
that their nuclear capacity might reach 6 GWe by 2025 through the construction of three new
nuclear power plants (NEA & IAEA, 2007:313).

2.4.2.2. South African production

In 2007 South Africa produced 525 tU in 2007 of which Anlogold Ashanti Ltd. produced the
greatest part. This company was the only uranium producer in South Africa for a few years.
However, the recent increase in the uranium price has made the production thereof more
economically viable and more uranium producing plants are being constructed and re-
opened in South Africa. Table 2.3 shows a list of the uranium plants in South Africa that is
currently operating or will be operating in the near future as derived from the World Nuclear
Association (2009).

Table 2.3: Uranium plants in South Africa
Plant/Project Company Projected capacity (tU/yr) Reached in
South Uranium Plant AlgloGold Ashanti Ltd. 760 2012
Dominion Reefs Uranium One 1470 2011
Ryst Kuil UraMin Inc. 1150 2009
Ezulwini First Uranium Corp. 116 2010
Buffelsfontein First Uranium Corp. 330 2010

It is seen from Table 2.3 that the uranium production in South Africa will most probably
increase over the next few years with many expansions and new plants. The Nuclear Fuel
Corporation of South Africa (Nufcor) does the final processing step in the uranium production
process and has a capacity of 3 400 tU per year (NEA & IAEA, 2007:313). From this it is
seen that if all the planned plant do reach full production capacity, Nufcor will not be able to
handle all the uranium at its present capacity. However some of the above mentioned
companies were very optimistic in their projections.
School of Chemical and Minerals Engineering
16
Chapter 2: Literature survey
2.4.2.3. Relationship between demand and production

As mentioned before South Africa currently has very low uranium requirements which may
increase in future as new nuclear power plant is constructed. However South Africa does
not have any enrichment facilities, which means the nuclear reactor fuel is imported. The
Nuclear Energy Corporation of South Africa (Necsa) did enrich uranium for research, military
use and the Koeberg Nuclear power plant, but has seized these operations according to the
agreement under the Non-Proliferation Treaty. Therefore all uranium produced in South
Africa is exported to be enriched and all nuclear reactor fuel is either imported or supplied by
stockpiles from the military (World Nuclear Association, 2009).

2.5. Process background

The process used for the extraction of uranium is described in detail in the form of a
literature study. The processes described are leaching, solid-liquid separation, ion-
exchange, solvent extraction and precipitation. Each step in the process is described in the
following structure:

A brief description of the background theory.
The process alternatives and the safety and environmental aspects.
The raw materials added to the process.
The design parameters.
The kinetics present in the process.

The processes will be evaluated further while new concepts are developed and important
new information can lead to the useless rendering of the process.

2.5.1. Leaching

Leaching is the first hydrometallurgical process in the extraction of uranium, hydrometallurgy
is the process where metals are produced from an aqueous solution (Minerals counsil of
Australia, 2006: 5). Leaching is where the valuable minerals are extracted from grounded
ore by dissolving, preferably selective, the mineral in an aqueous solution. The chemical
used to dissolve the minerals is known as the leaching agent (Lunt & Holden, 2006: 4).

School of Chemical and Minerals Engineering
17
Chapter 2: Literature survey
From the above definitions, it is easy to understand the importance and economical impact
of leaching. Since leaching is the first extraction stage in the processing of minerals, it is
important to selectively dissolve most of the essential minerals. The valuable minerals that
are not dissolved in the leaching section are lost in the tailings. These minerals will not only
be lost as an economical resource but it can also have devastating ecological effects.

It is possible to economically extract more than one valuable metal from the ore, in South
Africa this is done with ore that contains both uranium and gold. In the case of gold and
uranium the different minerals are removed using separate leaching systems. Leaching is
the most expensive key component in the capital cost of an uranium processing plant, the
capital cost of acid leaching equipment can contain up to 49% of the capital cost expenses
of the project (Lunt & Holden, 2006: 4).

2.5.1.1. Leaching process alternatives

The driving force for the development of different leaching techniques is optimization of the
leaching process. The five main factors to take into account when choosing a leaching
process are (Minerals counsil of Australia, 2006: 170):

The degree of recovery for the desired metal.
The selectivity required for the desired metal.
Leaching time required to achieve the desired recovery.
The capital cost required for the leaching equipment.
The operating cost of the reactants.

The optimization of the leaching process has lead to numerous techniques. The leaching
operation can be done in batch, counter current, atmospheric or above atmospheric
pressure and ambient or elevated temperatures. It is therefore important to classify the
techniques and this is illustrated in Figure 2.6 (Gupta,2003: 480):
School of Chemical and Minerals Engineering
18
Chapter 2: Literature survey


Figure 2.6: The classification of the various leaching techniques

Leaching techniques are separated into two main groups, percolation leaching and agitation
leaching. In percolation leaching, the ore is static and the leaching agent is percolated
through the ore to allow dissolution. In agitation leaching, the ore is suspended in the
leaching agent to allow dissolution. The dissolution kinetics for percolation leaching is very
slow but it requires less capital cost when compared to agitation leaching (Gupta,2003: 480).
Percolation leaching is discussed first.

In-situ leaching

The first percolation leaching technique discussed is In-situ leaching. The main use of In-
situ leaching is applied in the extraction of uranium. In-situ leaching has a minimal surface
disturbance and is therefore an environmentally friendly mining process. In-situ leaching
reverses the natural process which deposits the uranium in the porous sandstone (Minerals
counsil of Australia, 2006: 170). Figure 2.7 illustrates the arrangement of an in-situ leaching
operation (NunnGlow, 2008).

School of Chemical and Minerals Engineering
19
Chapter 2: Literature survey



Figure 2.7: In-situ leaching process description

The confined deep aquifer from Figure 2.7 contains the mineral that is leached, the aquifer
should be permeable to the leaching agent. The aquifer is between two clay layers, the clay
layers restrain the leaching agent inside the aquifer and thus preventing the leaching agent
from contaminating the water table. The leaching agent is injected to the aquifer through the
injection wells and the pregnant solution is removed through the recovery well (Minerals
counsil of Australia, 2006: 195).

The biggest advantage of In-situ leaching is the reduction of capital and operating cost. The
reduction results in the economical exploit of low grade deposits which was not possible with
other leaching techniques. There is no need to remove, grind an mil the ore (Mudd, 2000:
528).

One important safety aspect required for In-situ leaching is the drilling of a monitor well. The
monitor well is used during and after mining to test for possible water contamination. The
confined deep aquifer must be allocated below the water table (Gupta,2003: 481).

In-situ leaching is an environmental friendly mining process which has minimal surface
impact (Minerals counsil of Australia, 2006: 195). The biggest environmental impact of In-
situ leaching is underground. The leaking of leaching agent into the underground water
table is frequent. The new research done by the IAEA on the environmental impact the acid
In-situ leaching done in the Soviet block is leading to a wide reviewing of In-situ uranium
mining (Mudd, 2000: 527).



School of Chemical and Minerals Engineering
20
Chapter 2: Literature survey
Heap leaching

The next percolation leaching technique discussed is heap leaching. Heap leaching is
mainly done with waist rock dumps or low grade ore that contain valuable metals. Heap
leaching has been used since the eighteenth century (Gupta, 2003: 482). The life time of a
heap can be up to several years due to the slow dissolution kinetics. Figure 2.8 illustrates a
heap leaching process (Brugess, 2008):


Figure 2.8: Illustration of a heap leach process

The ore is crushed to a size smaller than 25 mm and pilled. The leaching agent is sprayed
at the top of the heap as show in Figure 2.8 and allowed to percolate through the heap. The
leaching agent dissolves the mineral and so it recovers the valuable metals (Gupta, 2003:
482). The pregnant solution is collected in a pond and then processed to retrieve the
valuable metals.

There is no need to for expensive milling and thus the main advantage of the heap leaching
is the reduction in capital and operating cost. There biggest disadvantage is the low
recoveries and the extended period for leaching (Gupta, 2003: 482).

The construction of a leaching pad is a major safety and environmental aspect. The
leaching pad should be impervious to the leaching liquid. The leaching pad acts as a barrier
between the environment and the heap. The leaching agent is a danger to the safety of the
workforce and environment.


School of Chemical and Minerals Engineering
21
Chapter 2: Literature survey
Atmospheric agitation leaching

The first agitation leaching technique discussed is atmospheric agitation leaching. The
leaching is done at atmospheric pressure and usually elevated temperature. The ore is in a
solution with the leaching liquid and is kept suspended, in contrast with the above
percolation techniques (Gupta, 2003: 483). Atmospheric agitation leaching is done in a
cylindrical steel tank, the steel is lined with rubber to protect the steel from corroding (Gupta,
2003: 483). The aggressive leaching conditions that is achieved in atmospheric agitation
leaching results in higher recoveries in shorter time compared to percolation leaching. The
slurry can be heated to the desired temperature using steam and the elevated temperature
will increase the reaction kinetics (Gupta, 2003: 483).

There are two methods to agitate the slurry namely pneumatic and mechanical. Mechanical
agitation is created using a propeller or a turbine mixer driven by a motor. Compressed air is
used for pneumatic agitation. The compressed air is introduced at the bottom of a cylindrical
vessel with a conical bottom.

There is an important safety and environmental hazard concerning the air agitation. The air
can transport poisons material into the environment from the leaching tanks. It is therefore
important to have an air monitoring system (Weber, 1996: 439). The environmental impact
is reduced because the leaching is done in a tank and the system is easily containable.

Pressure leaching

The next agitation leaching technique discussed is pressure leaching. Pressure leaching is
applied when the rate of dissolution is to slow for temperatures below 90C and results in the
recovery of the metal not to be economical. The technology is new but it has found
numerous applications around the world. The biggest drawback of pressure leaching is the
high capital, maintenance and operating cost (Minerals counsil of Australia, 2006: 185).

There exist a lot of advantages in the use of pressure leaching, the prevalent advantage is
the increase in dissolution kinetics. At the Aflease Uranium plant in Klerksdorp the
residence time is reduced from 18 hours to 2 hours (Bateman, 2004: 3).



School of Chemical and Minerals Engineering
22
Chapter 2: Literature survey
There are two types of pressure leaching. The first is pressure leaching without oxygen and
the second is with oxygen. Pressure leaching with oxygen is used for sulphide or uranium
ores. The oxygen is used as an oxidizing reagent in this system. The reaction is done in an
autoclave which is designed to handle the high temperature and pressure (Gupta, 2003:
484).

The slurry is pumped into the autoclave at high pressures. The slurry is then heated with
steam to achieve the desired temperature. Once the temperature is achieved the
exothermic reaction supplies the energy to maintain the temperature in the autoclave. In
some cases it is necessary to control the temperature with cooling water.

The autoclave is determined a pressure vessel and is therefore a great safety hazard if the
proper maintenance is not done on a routinely bases. The environmental impact is reduced
because the leaching is done in a vessel and the system is easily containable.

2.5.1.2. Leaching raw materials

The raw materials used for the leaching process is heavily dependent on the ore (Lurie et al.,
1962: 16). The one important ore characteristic is the numerous minerals present in the ore.
The amount of minerals that contains the valuable metal is important but the choice of
leaching agent is also highly dependent on the gangue minerals present in the ore (Merrit,
1971: 63). The leaching of ore can lead to the dissolution of other metals in the system
which causes impurities in the valuable metal product. It is important not to fully dissolve the
impurities in the ore, the lack of impurities in the leach solution will reduce the extraction cost
of the valuable metals (Lurie et al., 1962: 16). There are two main raw materials used in the
leaching process namely the leaching and oxidizing agent.

The two raw materials change the properties of the ore. The changes of properties are
necessary to ensure that the valuable minerals are dissolved. The properties that are
changed are the pH and the oxidation state of the metal in the mineral.
The pH is controlled using either an acid or an alkaline in which the metal dissolves. The
acid or alkaline is called the leaching agent. Sulphuric acid, nitric acid and hydrochloric acid
are the most widely used leaching agents in the industry. Sodium cyanide or sodium
hydroxide is used in alkaline leaching processes (Minerals counsil of Australia, 2006: 132).
The choice between acid and alkaline leaching is dictated by the ore used and the metal that
is extracted.

School of Chemical and Minerals Engineering
23
Chapter 2: Literature survey
The nature of the host rock is very important. Acid leaching is used for silicone ores and
alkaline leaching is used in high lime content ore. The carbonate minerals will increase the
consumption of acid. The alkaline and acid react first due to the fast reaction kinetics. The
increase of acid consumption leads to greater raw material cost which can lead to the
uneconomical extraction of the metal (Lurie et al., 1962: 16).

The advantages of acid leaching are (Lurie et al., 1962: 16):

Acid leaching is much more efficient than alkaline leaching
The ore does not need to be ground to as fine a mess as with alkaline leaching and
this reduces the cost of milling.

The disadvantages of acid leaching are (Lurie et al., 1962: 16):

The acids used for acid leaching is very corrosive and all the equipment that comes
in contact with the acid should be designed to withstand the corrosive thus increasing
capital cost.
The acid should be neutralised before it can leave the system to reduce
environmental impact.
Acid leaching is not very selective for Uranium and a lot of impurities is soluble in the
acid solution, this produces difficult separation of Uranium from the system.

The advantages of alkaline leaching are (Lurie et al., 1962: 17):

Alkaline leaching is much more selective with regards to Uranium.
The equipment does not need to be designed to resist acid corrosion and this
dramatically reduces the capital cost.

To explain the use of an oxidant in the leaching process, uranium leaching is used. Uranium
has two valence states in Uranium oxide, Uranium(IV) has a positive charge of 4 and form
the Uranium oxide complex namely UO
2

and Uranium(VI) has a positive charge of 6 and
form the Uranium oxide complex namely UO
3
. U(VI) forms Uranium oxide in the tetravalent
state and has a molecular formula of UO
2

and U(VI) forms Uranium oxide in the hexavalent
state that has a molecular formula of UO
3
or UO
2
O. U(VI) is easily soluble in dilute non-
oxidizing acids such as HCl and H2SO
4

where as U(IV) is not easily soluble in dilute non-
oxidizing acids. U(IV) is soluble in dilute oxidizing acids such as HNO
3
(Venter &
Boylett,2009: 445).
School of Chemical and Minerals Engineering
24
Chapter 2: Literature survey
For this reason it is very important to oxidize the U(IV) to U(VI) to reduce the cost of the
reactant used in the leaching process. There are a few common used oxidizing agents that
are easily produced, attained and economically viable for use, just to name a few; MnO
2
,
MnO
2
, NaCl, Fe
2
(SO
4
)
3
and NaNO
3
. There is a down side to using the oxidation reactants,
side reactions will occur with other in-solute metallic complexes and reduce them to become
solute. This reduce the purity of Uranium in the solution and complicates the separation of
the Uranium from the system. The acid leaching efficiency is also dependent on the Fe
2+
concentration. The Fe
2+
is oxidized with the oxidation reagent to Fe
3+
, the Fe
3+
then acts as a
oxidation reagent and U(IV) is oxidized to U(VI) (Lurie et al., 1962: 18).

2.5.1.3. Leaching design parameters

Leaching is a chemical reaction and there is a lot of design parameters to keep in mind. The
effect that each design parameter has is different for each mining application. It is important
to try and design the leaching process keeping most of the parameters in mind. The
important design parameters for uranium extraction are as follows (Merrit, 1971: 63):

The ore that is mined.
The leaching agent concentration.
The temperature of the reaction.

The leaching efficiency is dependent on the ore characteristics. The ore can contain several
different minerals that contain the metal that should be extracted. The dissolution of the
different minerals follows different mechanisms and thus the dissolution kinetics is different.
The different dissolution kinetics can result in the full dissolution of one mineral and only
partial dissolution of a different mineral (Lottering 2007, 1). One important example of this
phenomena is in the extraction of uranium. The maximum leaching efficiency of 90% is
achieved in the Vaal River, South Africa, and this is due to the different minerals present in
the ore. There are two predominant uranium minerals present in the Vaal River, uraninite
and brannerite. Uraninite is in abundance in the ore and is easily leached while brannerite is
the mineral that dissolves slowly. The kinetics of brannerite is so slow that economical
extraction of uranium from it is impossible.



School of Chemical and Minerals Engineering
25
Chapter 2: Literature survey
Acid leaching is heavily dependent on the acid concentration, the requirement of the acid in
the leaching operation is to have enough free acid in the system to attack the Uranium
mineral inside the grounded to extract it, without excessive acid to dissolve the gangue
minerals. The reaction starts with the acid neutralizing the gangue lime minerals, in high lime
percentage ore the amount of acid consume to neutralize the lime minerals is as high as 200
kg per metric tone ore. After the Uranium is dissolved in the liquid, there should still be
enough acid present in the liquid to stop precipitation of the Uranium (Merrit, 1971: 63).

The time and temperature for the leaching process is interdependent of one another, when
the temperature is increased the reaction time is reduced dramatically. There exist an
optimum temperature and time for each type of ore; this optimization is compared to
operating cost and Uranium extraction. The optimum is usually designed for longer reaction
time and lower temperature (Lurie et al., 1962: 71). There are disadvantages with the
regards of increasing the temperature are as follows (Merrit, 1971: 71):

The dissolution of gangue minerals will be much more, this will increase the
complexity of extracting Uranium,
Increased corrosion on the equipment, even on certain stainless steel materials.

2.5.1.4. Leaching kinetics

A chemical reaction has occurred when a number of molecules from one or more species
are converted to form new species (Fogler 2006, 5). Reaction kinetics is the study on how
fast the reaction occurs (Fogler 2006, 4). The most useful scheme to classify chemical
reaction for an engineer is to classify the number of phases. The two classification schemes
are heterogeneous and homogeneous. One phase is present in homogeneous systems and
more than one phase is present for heterogeneous systems (Levenspiel 1999, 2). The
leaching process is a heterogeneous system because solid particles, the ore, is dissolved
with a leaching agent, a liquid.

There are numerous variables that affect the rate of a chemical reaction. The variables
range from easily manipulated variables, such as temperature and pressure, to complex
variables, such as mass and heat transfer. The rate of reaction is a single calculated
variable that takes into account all the variables that affect the rate of chemical reaction
(Levenspiel 1999, 3). The rate of reaction is discussed for the extraction of uranium.

School of Chemical and Minerals Engineering
26
Chapter 2: Literature survey
The fact that more than one phase is present in the system, it is very important to
incorporate the mass transfer term into the usual chemical kinetics term. To easily calculate
the overall rate of reaction, the individual rate steps should be written on the same base
(Levenspiel 1999, 371). The base that is used with uranium leaching is interfacial surface of
the uranium particles. The design equation for the extraction of uranium is displayed in
Equation 2-1 (Ikeda et al., 1995: 267):

=
dC rSN
dt V
(2-1)


When the variables, S, N and V, stay constant for a system and only the leaching agent is
changed, the only change in the design equation is k, the dissolution rate of the uranium.
The variable k is the rate of reaction which is dependent on the leaching agent and the
process conditions. The equation of k for nitric acid as the leaching agent is as given in
Equation 2-2 (Ikeda, 1995: 267)

| | ( )

( +

2.3
a b 3 3
r= k k HNO NO
(2-2)


The reaction rate is dependent on the concentration of the nitrate ion to the power of 2.3.
The value is reported to be 1 if the concentration of nitric acid is larger than 10 M (kinetic of
leaching 1994, 6). HNO
2
is a very important reactant in the leaching of uranium using nitric
acid. HNO2 acts as an auto-catalyst (kinetic leaching, 4). It is not necessary to add HNO2
to the system because it is formed as a by-product in the leaching of uranium with nitric acid.
The variables k
a
and k
b
is temperature dependant and can be rewritten as Equation 2-3
(Ikeda,1995: 271):

(2-3)

The equation of r for sulphuric acid as the leaching agent is given as Equation 2-4 (Fleming,
1980):

(2-4)

4
79500
2.2 10 exp
36800
0.46exp
a
b
k
RT
k
RT
| |
=
|
\ .
| |
=
|
\ .
| |
1 1
8 3 2
2 3
2
61300
6 10 exp r UO Fe Fe
RT

+ +
| |
( ( =
|

\ .
School of Chemical and Minerals Engineering
27
Chapter 2: Literature survey
The important factors in the dissolution of uranium is the ferric and ferrous ions in the
system. The importance of the oxidation of iron is described in the above section. The
dissolution kinetics using nitric acid leaching is faster than the observed dissolution kinetics
of sulphuric acid leaching (Avvaru et al., 2007: 2109).

2.5.2. Ion exchange

The ion exchange process follows the leaching and clarification processes on a uranium
extraction plant. The leaching process is not very selective and produces leach liquor
containing many ion complexes and a low uranium concentration. The typical composition
of Rand leach liquor is reported by Pinkey et al. (1962:30) and shown in Table 2.4.

Table 2.4: Typical composition of Rand leach liquor
Ion complex Concentration Units
UO
2
2+
0.2 to 1.0 g/L
Free aid as H
2
SO
4
3 to 7 g/L
Fe
2+
1 to 4 g/L
Fe
3+
1 to 4 g/L
Mn
2+
5 to 10 g/L
Silica soluble 1 to 2 g/L
SO
4
2-
, HS O
4
-
20 to 40 g/L
NO3
-
0 to 1.5 g/L
Cl
-
0 to 0.6 g/L
S
x
O
6
2-
0 to 0.06 g/L
Co(CN)
6
2-
0 to 4 ppm

Further processes will purify and increase the concentration of the uranium complexes to
eventually effectively produce the uranium product. Ion exchange is very efficient in
selectively extracting uranium complexes from the leach liquor. In an ion exchange process,
anions or cations are adsorbed onto a solid particle, the resin bead, by replacing another ion
of a similar charge. Ion exchange is used in many processes including water softening
where calcium cations in the water replace sodium cations from a resin, thus reducing the
calcium in the water.


School of Chemical and Minerals Engineering
28
Chapter 2: Literature survey
The application of ion exchange processes in the extraction of uranium was one of the most
significant developments in this process technology. Anionic ion exchangers are used to
selectively adsorb uranium complex anions. The uranium complex ions mainly depend on
the type of acid used in the leaching process. In the case where sulphuric acid is used as
leaching agent, uranyl sulphate complexes, such as UO
2
(SO
4
)
2
2-
, will form. The uranium in
the uranium complexes is in the U(VI) state.

Resins

The uranyl complex anions are adsorbed onto a strong base anionic resin which has a high
affinity for uranium complexes. These resins are synthetic, organic-polymers based on
styrene with a cross-linking agent such as divinylbenzene. The degree of cross-linking
depends on the ratio between the styrene and the divinylbenzene in the resin. Ionic
functional groups are chemically added onto the resin to give it ion-exchange properties. In
strong base anionic resins these functional groups are usually quaternary ammonium, or
pyridinium groups for uranium recovery (Merritt, 1971:139). These functional groups have
mobile anions which can be replace by other anions. These ion exchange adsorbtion
reactions for uranyl sulphate complexes are generalized to Equation 2-5:

nRX + [UO
2
(NO
3
)
n+2
]
n-
RnUO
2
(NO
3
)
n+2
+ nX
-
(2-5)

Here R represents the cationic functional group on the resin and X the mobile anion that is
replaced. These resins have different affinities to different ions in the system and bind more
tightly to some. This difference in affinities is quantified in the molar selectivity coefficient,
which describe the equilibrium state of the resin-solution system. It is reported by Merritt
(1971:140) that the affinities towards different monovalent anions for a typical strong base
anionic resin are decreasing in the order of Equation 2-6:

NO
3
-
> CN
-
> HSO
4
-
> Cl
-
> HCO
3
-
> OH
-
> F
-
(2-6)

Resins have a higher selectivity coefficient towards ion with a higher valence. When the
valence of the mobile anion that is being replaced is lower than the adsorbed anions then a
lower concentration in the solution of the adsorbed anion will cause the equilibrium favours
the adsorption. Resins also have a higher affinity towards ions with the smallest solvated
volume, which in general decreases with the increase in atomic number. Uranium has a
very high atomic number, 92, which causes resins to have a very high affinity towards ion
complexes that contain uranium. There are however other ion complexes, such as
polythionates, cobalticyanides and molybdates, that will displace the uranium ion complexes
from the resin.
School of Chemical and Minerals Engineering
29
Chapter 2: Literature survey
Depending on the degree of cross-linkage, the resin particle will swell to a certain size in
water. For resins operating in columns, more than 95% of the wet resin has a particle
diameter between 3.6 and 5.7 mm (20 50 mesh, U.S. standard screens). For resin in pulp
operation the resin beads are usually larger. Smaller resin beads has many structural and
kinetic advantages, but also causes a higher pressure drop over the column and is easily
lost through sieve plates. Merritt (1971:141) reports that in South Africa the pressure drop
over resin was recorded as approximately 22.6 kPa/m resin at a flow rate of 0.08 to 0.16 m/s
through the resin. Structural strength is also an important property for consideration with
resins, since resin attrition causes resin loss and new resin must be loaded from time to time
(Mirritt, 1971:138-142).

The maximum capacity of a resin is measured in equivalents per unit volume (wet) or mass
(dry). These equivalents are equal to the amount of moles of monovalent ions that is mobile
or can be replaced if only monovalent ions are loaded on the resin. Therefore if an anionic
resin has a capacity of one equivalent per gram, then one mole of hydroxyl ions can be
adsorbed per gram of resin. It should also be noted whether the capacity refers to wet or dry
resin. When the capacity is given per mass then it usually refers to dry resin mass and when
given per volume it usually refers to wet resin volume (Seader & Henley, 2006:557).

Adsorption

Since the kinetics of the adsorption is fast, equilibrium is reached very rapidly. When the
resin is loaded into a column with the pregnant leach liquor flowing slow enough through it
from above, the top fraction of the resin reaches equilibrium or saturation before any uranyl
complexes even start to leave the column. This results in the formation of an adsorption
front or section in which the adsorption of the uranyl complexes takes place. Above this
front the resin is saturated with uranyl complexes and below this front the resin has no uranyl
adsorbed because the solution passing over it is stripped from uranyl complexes. This
adsorption front is shown in Figure 2.9.



School of Chemical and Minerals Engineering
30
Chapter 2: Literature survey
R
E
S
I
N

Pregnant leach
liquor (Conc. = C)
S
a
t
u
r
a
t
i
o
n

Resin
concentration
R
e
s
i
n

b
e
d

h
e
i
g
h
t

Fresh resin
Saturated resin















Figure 2.9: Typical adsorption

As indicated in Figure 2.9, the adsorption front moves downwards through the column and
the thickness of the front depend on the kinetics of adsorption and the flow rate of the leach
liquor. When the lower end of the adsorption front reaches the bottom of the column it is
called break through and the first uranyl complexes starts to pass through. When the top
end of the adsorption front reaches the bottom of the column the column is saturated and the
exit and feed solution will have the same uranyl complex concentrations (Minerals Counsil of
Australia, 2006).

The pregnant leach liquor contains several ions as mentioned before. Anions such as
sulphate and hydrogen sulphate compete with the uranyl complex anions for adsorption onto
the resin. In leach liquor where nitric acid is during leaching, nitrate ions will be present and
competing for adsorption like the sulphate complexes. However these anions are necessary
in the pregnant leach liquor to ensure the equilibrium presence of uranyl anion complexes
such as UO
2
(SO
4
)
2
2-
. Therefore the optimal ratio of these competing anions and uranyl
complex anions should be determined to optimise the adsorption of uranium onto the resin.
It is reported by Merritt (1971, 147) that the uranium adsorption increases because of this
effect with an increase in pH, however at a pH value as low as 2.0 precipitation of uranium
will occur because of the presence of ions such as phosphate and arsenate. For this reason
the adsorption cycle is operated in a pH range of 1.5 to 2 (Merritt, 1971:147-148).

School of Chemical and Minerals Engineering
31
Chapter 2: Literature survey
Although the affinity of the resin towards uranyl complexes are relatively high, there are
other ions that may form anion complexes that compete with uranyl complexes, such as
ferric and vanadic ions. Especially ferric anion complexes are present in greater
concentrations than the uranyl complexes. However, given sufficient time, complete
equilibrium is reached where a very small fraction of the resin is occupied by iron. Since the
kinetics of the adsorption is relatively fast, equilibrium is reached quickly leaving little uranyl
complexes to pass through the rest of the resin. This causes the iron to be adsorbed and
then later displaced by uranyl complexes. As a result of this effect an iron break through is
observed before an uranyl break through is observed (Gupta, 2003:548).

Elution

The adsorbed uranyl complexes are removed from the resin in the elution cycle. Here the
thermodynamic equilibrium is manipulated to favour the release of the uranyl complexes
from the resin. This can be done by introducing a solution or eluant with a high
concentration of a certain ion to replace the uranyl complexes, however this method is slow
and requires large volumes of eluant. A more efficient technique is to use ions in the eluant
that alter the uranyl anionic complex as well as displacing it. Dilute nitrate or chloride
solutions are used as eluant and produce non-adsorbable neutral or cationic uranyl
complexes. Sulphuric acid is also a possibility for use in an eluant. Table 2.5 shows these
eluant choices with suggested concentrations (Merritt, 1971:156-161).

Table 2.5: Suggested concentrations for eluant choices
Elution solution Concentration range (molar)
Chloride ion 0.5 1.5
Nitrate ion 0.8 1.2
Sulphuric acid 1.0

It is suggested by Merritt (1971:160) that the resin should be converted to the sulphate form
after elution with a nitrate or chloride ion solution is done. This will improve the adsorption of
new uranyl complexes.

Another elution process uses acidic ammonium nitrate (NH
4
NO
3
) as an eluant. This gives
the advantage that the adsorbed ferric complexes is eluted first and can be separated form
the rest of the elution with the higher uranyl complex concentration. The concentration of the
eluate, exiting the process, is shown for this system in Figure 2.10.
School of Chemical and Minerals Engineering
32
Chapter 2: Literature survey



Figure 2.10: Eluate concentration profile

The x-section of the eluate seen in Figure 2.10 has a high ferric ion concentration and is sent
back to the pregnant leach liquor tank. The y-section of the eluate has a very high uranyl
complex concentration in comparison with the x-section and is sent to the next process. The
last z-section of the eluate is recycled to the eluant feed. This process is effective to remove
the ferric ions from the system and thus reducing the impurities in the system (Gupta,
2003:548-549).

Regeneration

Certain ions in the pregnant leach liquor have a very strong bond when adsorbed onto the
resin. These anion complexes may even replace uranyl complexes and if it is not removed
during elution, it is called poisoning of the resin. Examples are silica, molybdenum,
polythionates, cobalt cyanide complexes and thiocyanide. Basic fouling of the resin is also a
possibility.


School of Chemical and Minerals Engineering
33
Chapter 2: Literature survey
Some of these poisons are temporary poisons which mean that that it is possible to remove
them from the resin through regeneration of the resin. The resins can be regenerated by
contacting it with a caustic soda solution. This removes the temporary poisons, but should
be done in stages to prevent resin damage through a pH shock. Other poisons are not
easily removed from the resins without damaging the resin itself and are called permanent
poisons. Therefore at some point the level of permanent poisoning will necessitate the
replacement of the resin (Merritt, 1971:163-167).

2.5.2.1. Ion exchange process alternatives

The previous sections described certain stages (adsorption, elution and regeneration) in the
ion exchange process. These fundamental stages along with other washing stages are all
incorporated in the processes used in practice. For this reason most of the existing ion
exchange processes are semi-continuous. In the section below the fixed bed column ion
exchange, moving bed column ion exchange and the resin in pulp processes is discussed
shortly.

Fixed bed column ion exchange

The operation of ion exchange through the use of fixed resin beds are commonly applied in
the uranium industry. Fixed resin beds are usually operated with three or four column
systems. Each of these columns goes though the discussed cycles of adsorption, elution
and back-washing. Regeneration of resin is not done as frequently and therefore the resin is
removed from the columns for this step.

These fixed bed columns are operated semi-continuously, which means that the pregnant
leach liquor is processed continuously, but the columns are not continuously in the
adsorption cycle. For the adsorption cycle, at least two fixed resin bed columns are
operated in series. The resin in the first column is saturated with uranyl complexes, which
occurs when the exit effluent has the same uranyl complex concentration as the pregnant
leach liquor fed to the column. This should take place before a breakthrough is observed in
the exit effluent of the second column, in which case uranium loss may occur. The exit
effluent from the adsorption cycle or the barren solution, which has a low pH value and
contains very little uranium, is sent back to the leaching section or an acid recovery plant
(Merritt, 1971:167,170).

School of Chemical and Minerals Engineering
34
Chapter 2: Literature survey
As soon as the resin in the first column of the adsorption cycle is saturated with uranyl
complexes, the column is taken out of the adsorption cycle. Since there are some solid
particles from the pregnant leach liquor that accumulate on the resin, the column is back-
washed using slightly acidified water. If the solid particles are not removed, the pressure
drop over the column will increase and channelling may occur which will reduce the
efficiency of the resin. This back-wash step can also be done after the elution cycle, but
then the density of the resin is lower which limits the effectiveness of the back-washing.
During back-washing the resin bed may expand to twice its usual volume. The spent water
from the back-wash cycle is sent to the feeding tank of the ion exchange columns, since it
contains significant amounts of uranium (Weiss, 1985:24-27).

To remove the uranyl complexes, the saturated resin is next eluted in the elution cycle. The
use of elution medium is kept to a minimum by keeping the flow rate as low as possible
within the available time for elution. This also causes higher uranyl complex concentrations
in the eluate leaving the elution cycle. As mentioned before, split elution is used to increase
the uranyl complex concentration sent to the next process and may in some cases reduce
impurities such as ferric ions. It is suggested that the resin must be converted back to the
sulphate form to increase the recovery at the beginning of the adsorption cycle. This is done
by flushing the resin with a sulphate solution (Weiss, 1985:24-27).

The presence of resin poisons necessitates the regeneration of the resin. This regeneration
is usually done with caustic soda in another tank, therefore the resin is hydraulically pumped
out of the fixed bed column to a regeneration tank. It is important to prevent a physical
shock or resin shock when the resin is transferred from an acidic form to a hydroxyl from in
which it swells significantly. For this reason the resin is first contacted with 0.5% and then
5% caustic soda to remove the temporary poisons and regenerate the resin. Then the resin
is washed with neutral salt and water before converted back to the sulphate form (Weiss,
1985:24-28).

The efficiency of the resin decreases over time because of permanent poisons which is not
removed during regeneration. The capacity of the new resin should compensate for this
decrease and therefore columns with new resin have extra capacity. This extra capacity is
dependant on several factors such as rate of resin poisoning, regeneration frequency,
optimum resin life, variation in ore feed grade and plant amortization cost (Merritt, 1971:170).


School of Chemical and Minerals Engineering
35
Chapter 2: Literature survey
The process equipment for fixed bed column ion exchangers usually consists of cylindrical
pressure vessels with slightly domed ends. A general ion exchange column has the
following dimensions, a diameter of 2m and a height of 3.7 m. It is usually constructed of
steel and lined with hard rubber on the inside to be acid proof and a variety of acid proof
piping is used. At the bottom of the vessel a diffuser piping system serves as an outlet for
the barren liquor and eluate and an inlet for the back-wash water. The resin fills about half
the vessel and lies on graded gravel or sand which supports it. During the backwash cycle
the resin volume expands approximately 100% to take up the whole column.

Figure 2.11 shows the layout for columns in an ion exchange system from Weiss (1985:24-
27). The columns have two piping inlet systems, one situated at the top of the column and
one in the middle. The inlet at the top of the column serves as an inlet for the pregnant
leach liquor and as an outlet for the back-wash water. The eluant is introduced into the
column right above the settled resin, in the middle of the column. During elution the space
above the resin in the column may be filled with either water or air, where the latter causes
least uranium loss due to diffusion of eluate into the water.


Figure 2.11 Fixed-bed ion exchange column layout


School of Chemical and Minerals Engineering
36
Chapter 2: Literature survey
In Figure 2.11 shows the different valve to control the important flow cycles of the column
system. These valves are usually automatic controlled spring-closing, air-opening rubber-
lined diaphragm valves. Figure 2.11 also shows a manhole and windows in the columns,
which is necessary to check on the resin and for maintenance. The valve that allows the
hydraulic removal of the resin is seen near the bottom of the middle column in Figure 2.11.

The control of the fixed bed column ion exchange system is an important aspect for the
efficient performance of the process. The adsorption cycle determines the time available for
the other cycles. Since the loss of uranium must be prevented there must be an eluted resin
column available when a column on the adsorption cycle is saturated. When nitric acid is
used as eluant, the elution cycle is faster and only a total of three ion exchange columns are
necessary. For slower elution cycles, four ion exchange columns are needed to ensure the
continuous adsorption. These cycles are controlled automatically based on the time
required or volume flow required for each cycle. It is important that frequent checks are
done on effluent analysis to ensure that the process is optimally controlled and that no
uranium is lost (Merritt, 1971:172).

Moving bed column ion exchange

In the moving bed column ion exchange process the resin is moved between columns which
each perform a certain cycle similar to the cycles in the fixed bed system. A typical layout of
this system is two set of three columns for adsorption, one set of three columns for elution
and one column for back-washing. The resin is transferred hydraulically from one column to
the next.

There are three adsorption columns in series in the adsorption cycle, and the resin is
removed from the leading column when it is saturated. The adsorption cycle is then
operated with only two columns in series for as long as possible to obtain maximum loading
of the resin. When breakthrough is reached in the leading column, the third column is added
to the series, containing fresh eluted resin.






School of Chemical and Minerals Engineering
37
Chapter 2: Literature survey
Saturated resin from the adsorption cycle is moved to a back-wash column where the resin
is back-washed before it is sent to the last column on the elution cycle. The elution is
different from the fixed bed system since here three columns in series are used of elution
instead of one. Since the fresh eluant is contacted with the first elution column containing
the resin whish has been partially eluted already, this system is a counter current system.
This gives the advantage of a high uranyl complex concentration in the eluate sent to the
next process (Merritt, 1971:174).

Each of the columns used in this process has an outlet for the transferring of the resin which
is usually located about 10 feet from the base. The typical adsorption and elution columns
have diameters of 2.5 m and a height of 4.5 m. The resin beds are nearly as deep as the
column heights since back-washing is not done in them. For this reason the back-wash
column is larger with a typical height of 5 m to accommodate the expanding resin bed during
back-washing.

Less piping is necessary since each column as a specific purpose and therefore the
connections is simpler. This also prevents the mixing of pregnant leach liquor and eluant
due to improper valve operation. The deeper rein beds also give the advantage of better
plant space utilisation and therefore less capital cost. As mentioned before, this system
produces an exit stream with a high uranyl complex concentration to the next process
because of multiple elution columns.

A general disadvantage of this system is the loss of uranium when the resin is not
transferred completely. The remaining resin from the adsorption cycle is loaded with uranyl
complexes which is lost when the column is reintroduced as the last column of the
adsorption cycle. This also decreases the total capacity of the resin.

Resin in pulp

There are several resin in pulp ion exchange processes. In this processes the leach slurry is
directly contacted with resin for the adsorption of uranyl complexes from the leach slurry. If
the solid particles in the leach slurry are small enough, the resin is easily separated from the
leach liquor with sieves. Therefore intensive solid liquid separation is not necessary, and
capital costs are reduced. In the following section the basket and continuous resin in pulp
processes are discussed.


School of Chemical and Minerals Engineering
38
Chapter 2: Literature survey
Basket resin in pulp

In the basket resin in pulp process the resin is situated in baskets constructed of stainless
steel. The stainless steel frame is covered with stainless steel or plastic screen cloth with 5
mm holes. A group of these baskets are called a bank of baskets. This banks moves up
and down in rectangular tanks though which the leach slurry of eluant flows. The leach
slurry or eluant flows from one tank to the next through the use of air lifts of gravimetric flow.
Pumps are used to return leach slurry or eluant to the first tank, with piping from every tank
to the first tank and the next process.

These tanks or banks of resin baskets are operated in circular cycles. The leading tank in
the adsorption section is rinsed once saturated and then becomes the last bank in the
elution cycle. Once the leading bank in the elution cycle is saturated, it is again rinsed and
becomes the last bank in the adsorption cycle. The resin particles used in this system are
larger for effective separation from the leach slurry, 90% of the resin have particle diameters
between 6 mm and 8 mm. The larger particles require more time to reach saturation and
therefore a system of fourteen banks is usually used (Merritt, 1971:176-178).

Continuous resin in pulp ion exchange

In this process the resin and leach slurry or eluant flows in counter currents through a series
of tanks. The resin and leach slurry or eluant mixture that exits each tank is pumped or air
lifted to the top of the next tank. Here the resin is separated from the slurry with a vibrating
screen and falls into the tank below. The leach slurry or eluant from here goes to the tank
preceding the one it came from. This way the counter current flow is achieved. Resin traps
are situated at all exit streams to reduce resin loss to a minimum.

It is reported that six to eight stages are used in the adsorption section and seven to fourteen
stages are used in the elution section. This process requires more elution stages than seen
in basket resin in pulp systems because less efficiency is achieved due to lower eluant flows.
Higher flow rates are not introduced since it will cause low concentrations of uranyl
complexes in the eluate.

Finer resin is used in this operation with particle diameters between 6 mm and 4mm. More
resin attrition is observed because of the air agitation, screening and transport between
stages. A 20 % to 30 % resin loss per year is caused by attrition by this process. However it
uses less resin than the basket resin in pulp system (Merritt, 1971:178-180).
School of Chemical and Minerals Engineering
39
Chapter 2: Literature survey
2.5.2.2. Ion exchange raw materials

In an ion exchange system there are a few reagent and product streams entering and
leaving the system. For the following discussion a fixed bed column ion exchange system is
used as base. In Table 2.6 a list of feed and products is given for this system as well as the
origin and destination of each.

Table 2.6: Feed and product streams for ion exchange
Feed streams
Name Description Purpose Origin/destination
Resin Strong base anionic type
Adsorb uranyl
compexes
Loaded periodically
Leach liquor
Contains low concentration
uranyl complexes
Feeds end product Clarifiers
Back-wash water Water
Washes solid particles
out of resin
Recycled water
Eluant Diluted sulphuric acid
Remove uranyl
complexes from resin
Recycled from
solvent extraction,
return eluant
Flush water Water Rinse eluate from resin Recycled water
Product streams
Barren liquor Uranium removed Acid recovery
Back-wash water Water/solids Clarifiers
Eluate
Uranium loaded diluted nitric
acid
Carry uranyl complexes
Next process,
return eluate tank
Flush water Water Rinsed resin
To return eluate
tank

As seen in Table 2.6, there are certain secondary systems necessary for the operation of
this section. A leach liquor feed tank, water recovery plant, return eluate tank and acid
recovery are required.



School of Chemical and Minerals Engineering
40
Chapter 2: Literature survey
2.5.2.3. Ion exchange design parameters

In the design of a fixed bed column ion exchange process, the configuration and reagent
should be established first. There are certain structural parameters of importance for the
design of a column such as corrosion resistance and physical strength. The main factors
when sizing the column is the flow rates and time. For a given flow rate the time required for
the resin to reach saturation is dependant on the dimensions of the resin bed, the transport
kinetics and the equilibrium of the system. It is then further important to consider the time
required for the elution cycle in establishing the time for saturation of the resin in adsorption
cycle.

2.5.2.4. Ion exchange kinetics and equilibrium

The ion exchange process is written as a reaction in order to understand the equilibrium
behaviour. The reaction where anion A
y-
replaces anion B
z-
from the resin is written as
Equation 2-7:

( ) ( )
z y y z
y B zA z A yB

+ = + (2-7)

Here the ion in brackets is the adsorbed ion while the other one is in solution. In the reaction
equation, y is the electron charge of the A ion and z the electron charge of the B ion.

The equilibrium for this reaction system is estimated with Equation 2-8:

z y
A B
A,B y Z
B A
q c
K
q c
=
(2-8)

Here q is the molar concentrations of the species in the adsorbed state and c the molar
concentration in the solution state. K
A,B
is the molar selectivity coefficient and is a constant
for dilute solutions and a certain ion pair and particular resin with a level of cross-linkage.
The molar concentration in the resin phase is taken as the number of equivalents per unit
bed volume or unit mass. In the solution phase the molar concentration is taken as the
equivalents per unit volume of solution.



School of Chemical and Minerals Engineering
41
Chapter 2: Literature survey
In a system where the electron charge of the competing ions differs, the concentration of the
ion originally in solution has a great influence on the equilibrium. Manipulation of the
equilibrium equations shows that if the ion introduced in the solution has a higher electron
charge and a lower concentration, the equilibrium will favour the adsorption reaction (Seader
& Henley, 2006:565-567).

In a system with more than two competing ions, it may be assumed that the rest of the ions
do not affect the equilibrium of a certain pair. Then the system can be solved by satisfying
the different equilibrium equations for each pair of ion, and the mass balance of the system.
The equilibrium is important to show the available capacity of the resin.
According to Seader & Henley (2006:568) there are four steps that can influence the rate of
adsorption when looking at ion exchange. These four steps are:

Liquid film diffusion (external mass transfer)
Pore diffusion (internal mass transfer)
Surface diffusion
Chemical adsorption reaction

The step that is the slowest in a series of steps is assumed to be the rate determining step,
which is plausible if the specific rate is much slower than the rest. For ion exchange
systems only the external and internal mass transfer steps is considered to be rate
determining since the chemical reaction rate is fast. Either or both the external and internal
mass transfer steps can be rate determining. In general the external mass transfer step is
rate determining for systems with a low exchange-ion concentration, below 0.01N, and the
internal mass transfer step is rate determining for higher concentrations, above 0.1 N. In a
system with a very high selectivity for the adsorbing ion relative to other ions in the system it
has been observed that the external mass transfer step is rate determining. Another
observation is that divalent ions diffuse appreciably slower than monovalent ions through the
pores of the resin.

The kinetic behaviour of the internal mass transfer step is given by Equation 2-8.

(2-8)

2
e 2
c 2 c q
D
r r r t
| |
+ =
|

\ .
School of Chemical and Minerals Engineering
42
Chapter 2: Literature survey
In Equation 2-8 D
e
is the effective diffusivity, r the distance from the centre of the resin bead
and t the time.

The kinetic behaviour for the external mass transfer step is given by Equation 2-9.

(2-9)

In Equation 2-9 the first term is the rate of molecular transport through the liquid film
surrounding the resin beads, k
c
is the transport coefficient, A is the area of the outer surface
of the resin bead, c
bi
is the concentration of component i in the bulk solution and c
si
is the
concentration of component i at the surface of the resin bead. It is important to note that the
transport coefficient is dependant on certain properties of the solution such as its flow rate
and viscosity. This dependence is shown in Equation 2-10.

(2-10)

Equation 2-10 shows that the transport coefficient k
ci
of component i is dependant on the
diffusivity of component i in the mixture (D
i
), the resin bead diameter(D
p
), the fluid mass
velocity (G), the viscosity of the solution () and the density of the solution (). In the
application of Equation 2-9, the concentration of component i at the surface of the resin bead
will be taken as the equilibrium concentration under the specific conditions. Therefore the
resin bead is taken as a pseudo particle with an even distribution of ions which is in a state
of equilibrium (Seader & Henley, 2006:568-572).

There are some short-cut methods which gives the kinetic profile of the adsorption cycle of
ion exchange in a fixed bed of resin. Figure 2.9 shows that there are two equilibrium phases
before and after the adsorption front. If the thickness and the downward velocity of the
adsorption front are known, then most of the design parameters can be determined. As
mentioned before the thickness and downward velocity of the adsorption front is dependant
on the flow rate of the pregnant leach liquor and the kinetics.

The Minerals Counsel of Australia (2006) gives such a short cut method to estimate an
approximate adsorption front thickness and downward velocity. This is given in Equations 2-
11 and 2-12.
( )
i i
c b s
dN
k A c c
dt
=
i
1/3 0.6
p
i
c
p i
D G
D
k 2 1.1
D D
(
| |
| | | |
( = +
|
| |
|

(
\ . \ .
\ .

School of Chemical and Minerals Engineering
43
Chapter 2: Literature survey
(2-11)

In Equation 2-11 H is the thickness of the adsorption front, V is the solution throughput, C is
the feed concentration, A is the cross sectional area of the resin bed, q
e
is the capacity of the
resin for the specific component at equilibrium and e is the fraction of the bed which is void.

The downward velocity of the adsorption front, U
e
, is given by Equation 2-12.

(2-12)

Here U
l
is the superficial velocity of the solution over the resin beads.

The kinetics and equilibrium parameters of the systems are important to determine in order
to produce a viable design for an ion exchange process. The equations given in the section
above is not totally accurate, but it is the most accurate conceptual approaches to the
problems.

2.5.3. Solvent extraction

Solvent extraction is the commonly used name for liquid-liquid extraction. Perry (1997: 15-4)
gives a definitive definition for liquid-liquid extraction:

Liquid-liquid extraction is a process for separating components in solution by their
distribution between two immiscible liquid phases.

The principle of solvent extraction is the same as that of ion exchange, but instead of a solid
resin, a non-aqueous solvent is used. This process is mostly used in hydrometallurgy,
where an organic phase is contacted with the aqueous solution containing the desired metal
that needs to be purified (Gupta, 2003: 510).

There are generally three stages in the solvent extraction process; extraction, scrubbing and
stripping. These three stages are represented in Figure 2.12 (Gupta, 2003: 511).


( )
e
VC
H q eC
A
= +
( )
l
e
e
UC
U
q eC
=
+
School of Chemical and Minerals Engineering
44
Chapter 2: Literature survey



Figure 2.12: Typical solvent extraction process

The first stage is the extraction stage; here the metal is transferred from the aqueous
solution to the organic phase by bringing the two phases in contact. The products of the
extraction stage are the organic solvent loaded with the desired metal and the aqueous
extraction raffinate (Gupta, 2003: 510).
The organic product from the extraction stage is sent to the scrubbing stage where it is
treated with a fresh aqueous solution to remove any impurities present in the solvent. From
the scrubbing stage the organic solvent is then stripped to remove the solute. After the
stripping stage the organic solvent is regenerated and recycled back into the extraction
stage. The purpose of these three stages are to produce two aqueous solutions, one
containing most of the impurities, and the other most of the desired metal ions (Gupta, 2003:
510).

2.5.3.1. Process alternatives

The process through which solvent extraction takes place cannot change radically, there is
however different solvents and contactors that can be used.

The different solvents that can be used depend on the chemicals used in the previous
sections of the plant and the degree of separation required from the solvent extraction unit.
There are different processes concerned with extracting uranium complexes from sulphate
solutions. The two processes most commonly used are the Amex and Dapex processes,
where different extractant chemicals are used.
School of Chemical and Minerals Engineering
45
Chapter 2: Literature survey
The main difference in solvent extraction processes, besides the composition of the solvent,
is the different types of extractors used. The different types of extractors that can be used
are (Madhavan, 2008):

Mixer-settlers.
Centrifugal devices.
Column contactors (static).
Column contactors (agitated).

There are different factors when selecting an extractor type; these factors are stage
requirements, fluid properties and operational considerations. Table 2.7 gives a comparison
between the different extractor types (Madhavan, 2008).

Table 2.7: Comparison of solvent extractor types
Property Mixer-settler
Centrifugal
extractor
Static columns
Agitated
columns
Number of
stages
Low Low Moderate High
Flow rate High Low Moderate Moderate
Interfacial
tension
Moderate to
high
Low to
moderate
Low to
moderate
Moderate to
high
Viscosity Low to high Low to
moderate
Low to
moderate
Low to high
Density
difference
Low to high Low to
moderate
Low to
moderate
Low to high
Floor space High Moderate Low Low

When choosing an extractor type, these properties need to be taken into account, and then
an extractor type can be chosen for the desired process.

2.5.3.2. Raw materials added

The raw materials added to the solvent extraction unit are the organic compounds that make
up the organic solvent, the solvent components consist of the following (Gupta, 2003: 512):


School of Chemical and Minerals Engineering
46
Chapter 2: Literature survey
Extractant.
Diluents.
Modifier.

The extractant is the active component in the solvent and is responsible for extracting the
metal from the aqueous phase. The extractant needed for a sulphate system is an anion
exchanger. The most common anionic extractant is long-chain alkylaminies (alamines),
alamines can be classified as being of primary (RNH
2
), secondary (R
2
NH), and tertiary (R
3
N)
nature. Tertiary alamine is used for the sulphate system (Gupta, 2003: 513).

The diluent is the carrier component in the solvent, and has the purpose of lowering the
viscosity of the extractant to simplify transport. The widest used diluent is kerosene, due to
its low cost and high flash point. During the extracting phase, there may form a third phase
when the organic phase splits into two. To solve this problem, a modifier is added. For the
purpose of this project, isodecanol (C
10
H
22
O) is used as the modifier added to the solvent
(Gupta, 2003: 515).

A summary of the components in the organic solvent are:

Tertiary alkyl amine (Alamine

336).
Kerosene.
Isodecanol (C
10
H
22
O).

To design a solvent extraction section and to decide which solvents to use, there are certain
design parameters that needs to be considered, these parameters are discussed in the next
section.

2.5.3.3. Design parameters

There is one function of a solvent extraction unit, and that is to mix two liquid phases, form
and maintain droplets of dispersed phase and later separate the phases again. The two
design parameters that are the most important are mixing and settling.



School of Chemical and Minerals Engineering
47
Chapter 2: Literature survey
Both mixing and settling is determined by the physical properties of the phases present in
the system. The amount of mixing is very important, if too little mixing is provided large
droplets will form and decreases the interfacial area which reduces mass transfer and
decrease stage efficiency. More mixing will minimize mass transfer resistance during
reactions and extraction (Madhavan, 2008).

Apart from the physical properties, settling is determined by the amount of mixing. The most
problems in settling occur when the phases are mixed too much, this forms an emulsion
which is difficult to separate. If an emulsion is formed, it needs to be settled over an
extensive period of time (Madhavan, 2008).

The other design parameters present in solvent extraction are the selection of the
components in the organic solvent used to extract the metal from the aqueous solution and
the pH of each stage in the solvent extraction process.

2.5.3.4. Kinetics

The two sections where kinetics plays an important role are the extraction and stripping
stages. The solvent extraction of uranyl sulphate, by using tertiary amines as extractant, is
characterized as having extremely rapid reaction kinetics for both the extraction and
stripping stages. Due to this rapid kinetics short residence times is expected in the mixers,
the residence times range between 45 seconds and 2 minute (Kordosky et al.: 10)

2.5.4. Precipitation

The precipitation of uranium is one of the final steps in producing the ADU product. This
process is necessary to obtain a solid product that meets certain grade and purity
specifications. These specifications are important as there are penalty schedules in place,
which means that the revenue will decrease if the specifications are not met.






School of Chemical and Minerals Engineering
48
Chapter 2: Literature survey
Precipitation of metal ions from aqueous solution is widely used in hydrometallurgical
processes as a means of separation, purification and disposal. The precipitate is formed by
changing solution pH, solvent concentration, solution temperature, or by the addition of a
reagent to the solution, also known as reactive crystallization. By adding a reagent to a
metal ion-containing solution, a compound will form which has a very low solubility and so it
readily precipitates. If the reagent is something other than water, it is known as ionic
precipitation and the added reagent functions by contributing anionic species. The
interaction of these anions with the cationic metal in the solution forms the compound
(Gupta, 2003:535).

The precipitation of uranium diuranate (ADU) is largely depended on the purity and
composition of the uranyl solution entering the process. If the solution is a product of ion
exchange, the iron concentration might be too high and an additional precipitation step is
introduced to precipitate the iron and other impurities. It is also important to note that the
ammonium diuranate (ADU) that forms has a very high concentration of uranium. This
uranium is in a complex form which means that it is dissolvable and can therefore be
absorbed into the body. This will cause heavy metal poisoning. Special care must be taken
when working with this process to make sure that there is no leakage of this poisonous
precipitate to the environment through the air.

Manipulating the techniques and conditions of the precipitation process, the selectivity and
physical characteristics of the precipitate can be controlled. There are several alternatives
for precipitation, which all have their specific application (Merritt, 1971:240).

2.5.4.1. Precipitation process alternatives

The product of precipitation, ADU, is ultimately converted to UO
2
powder for the use in
nuclear power plants. It is thus important to obtain a precipitation product with satisfactory
physical characteristics, such as settling and filtering characteristics, and which also meet all
required product specifications for uranium and impurities content (Merritt, 1971:240).

There are two common precipitation methods used, the first involving direct neutralization
with a base such as lime or ammonia, and the second, direct precipitation from acid solution
with hydrogen peroxide (Merritt, 1971: 240-247).



School of Chemical and Minerals Engineering
49
Chapter 2: Literature survey
Neutralization

The neutralization process used for precipitation is involved with the lowing of the pH of the
homogeneous solution to the point at which the metal complexes begin to break down and
cations are released from the soluble complexes. This is done in solutions containing the
desired anions to precipitate the product. Different reagents can be used to decrease the pH
of the solution, depending on the desired precipitate (Cartwright et al, 1967:667).

The precipitation of uranium will take place at pH numbers of 6.5 to 8, but the uranium
solution entering the precipitation process usually has a pH level of approximately 1.4 to 4.
This means that the solution must be neutralized to this level and can be done using caustic
soda, magnesia or ammonia. The reagent most widely used is gaseous ammonia, which
can be introduced into the uranyl sulphate solution and will cause the precipitation of
uranium in the form of ammonium diuranate (ADU) according to the overall reaction:

2 4 4 4 2 2 7 4 2 4 2
2UO SO + 6NH OH = (NH ) U O + 4(NH ) SO + 3H O

The characteristics of the uranium precipitate are dependent on the conditions maintained
during the precipitation process, including temperature, pH, feed solutions and the rate of
precipitation.

Careful pH control is necessary to ensure that a dense and readily filterable precipitate is
produced. A general method for this type of precipitation is to increase the pH value of the
solution gradually in separate tanks as a continuous process, which means that the solution
is pump form tank to tank. This causes a crystalline precipitate to form, which is desired
because this precipitate is easy to handle and wash (Merritt, 1971: 240 246).

Precipitation with hydrogen peroxide

A second division of cation release precipitation method is the release of cations at a
constant pH. This method involves the adjustment of the pH of the complex, and then the
complex is slowly destroyed by boiling with hydrogen peroxide to produce a dense
precipitate (Cartwright et al, 1967:667).


School of Chemical and Minerals Engineering
50
Chapter 2: Literature survey
This method can be applied to precipitate uranium as the complex from which it is
precipitated is stable, the uranium precipitate is insoluble and catalytic breakdown of
hydrogen peroxide is possible by the precipitate particles. The precipitation process to
precipitate uranium is carried out by adding the hydrogen peroxide as an aqueous solution
slowly to the uranyl sulphate solution. The precipitate that forms is uranium peroxide,
UO
4
xH
2
O which is a crystalline product that is easily handled. The reaction is shown below
(Brown, 1982):

+
+ +
2+
2 2 2 4
UO H O UO 2H

The temperature and pH level during this process must be controlled to achieve the desired
product characteristics. The temperature is kept between 30C and 65C and the pH is kept
at 2.8 for the optimum formation of a crystalline product with minimum contaminant.

The cost for using this process is somewhat higher than the conventional method of
neutralization because the raw materials are more expensive. But a higher purity product is
obtained which might save costs if penalty charges are imposed (Merritt, 1971: 247 248).

2.5.4.2. Precipitation raw materials

A two stage precipitation process is usually used, the first stage to precipitate impurities such
as iron, aluminium, titanium and thorium and the second stage to adjust the pH to a value
that will precipitate the desired product.

For the first stage lime is used to precipitate the impurities, which is also an economical
advantage since the quantity of the other, more expensive, reagents needed for
neutralization is decreased.

The raw materials for the second stage are dependent on the method chosen for
precipitation. If the neutralization method is chosen the reagents necessary can be caustic
acid, magnesia or ammonia. The total consumption of the neutralizing reagents will range
from 0.09 to 0.18 kg per kg of U
3
O
8
in the precipitate. The choice between these three
reagents depends on several factors shown in Table 2.8.


School of Chemical and Minerals Engineering
51
Chapter 2: Literature survey
Table2.8: Comparison of neutralization reagents
Cost Product
characteristics
Reaction
speed
Ease of
control
Product
contamination
Caustic
soda
R 4 200/ton Slimy product,
difficult to
handle
Satisfactory Suitable for
automatic
control
Exceed sodium
specifications
Magnesia R 5 000/ton More crystalline
product
Slow Not suitable for
automatic
control
Below
specifications
Ammonia R 15 000/ton Slimy product Satisfactory Suitable for
automatic
control
Below
specifications

If caustic soda is used, it is usually introduced into the process as a 10% NaOH solution. If
ammonia is used, it is vaporized to a gas by heating just prior to use and as it is important to
have good dispersion of the gas, the ammonia is mixed with 2 to 4 parts of air before
addition.

Considering the precipitation process using hydrogen peroxide, a suitable base such as
ammonium hydroxide is also added to the solution to maintain the desired pH by neutralizing
the acid formed in the precipitation reaction. The 30% hydrogen peroxide solution is added
in at least stoichiometric quantity, but if the peroxide is in excess the precipitation is more
likely to be complete. The neutralization of the acid formed also requires a minimum of the
stroichiometric quantity of the suitable base.

The precipitation process using urea (also known as carbamide, (NH
2
)
2
CO) only needs this
one reagent. About a kilogram of urea is added to a kilogram of uranium as a filtered
solution which generates enough ammonia to precipitate the desired product (Merritt, 1971:
240 246).


School of Chemical and Minerals Engineering
52
Chapter 2: Literature survey
2.5.4.3. Precipitation design parameters

There are two major considerations that have to be kept in mind when evaluating the product
that is produced by precipitation. The first consideration is the grade and recovery of the
product and the second is the physical characteristics which make the product easy to
handle and wash. These factors can be controlled by certain design parameters which
include the pH of the solution, the addition of reagent and the temperature at which
precipitation takes place.

pH

The chemical and physical properties of the precipitate product are affected by the changes
of pH of the precipitation solution. The size of ADU crystallites and agglomerates is
decreased with an increase of solution pH. As the agglomerate decrease the filterability of
the ADU slurries also decrease (Janov et al, 1971:1). It has been reported that fine
sediments can be produced at pH above 6, which yields a high-density pellet, favorable for
nuclear energy plants. But at pH values below 5 coarse sediments are produced that settle
rapidly and are easily filtered (Wilson, 1996:146). In practice, different procedures are used
to control the pH at which precipitation should take place. A general procedure is to
increase the pH value gradually in separate tanks as a continuous process (Merritt,
1971:242).

Addition of reagent

When precipitation of a product is done by adding a reagent to the solution, a reaction takes
place to produce another chemical that is almost insoluble in the resulting solution. This
reaction produces a large degree of supersaturation, which is a very important factor in the
precipitation process. The extent of supersaturation is dependent on the ionic
concentrations in the reagent and solution before mixing. This supersaturation causes
primary nucleation, which are very small particles that are formed with a crystalline structure.
The extent of supersaturation influence the time it takes for precipitation to begin, the
number of particles formed per unit volume solution and the particle growth rate (Seader &
Henley, 2006:671-672).




School of Chemical and Minerals Engineering
53
Chapter 2: Literature survey
Temperature

The temperature at which precipitation takes places has an influence on the rate of
precipitation settling and the precipitation time. The precipitation time shows an exponential
dependence on the temperature of precipitation and thus the time can be optimized to
reduce the quantity of precipitant but still maintaining the product specifications. Low
precipitation rates produce a product of a certain required sinterability. At these low
precipitation rates low temperatures favours the dispersion of precipitate particles while high
temperatures favour the agglomeration of the precipitate, which is the desired effect (Murty
et al, 2001).

A study of the temperature effects on the precipitation process using hydrogen peroxide
showed that the grade and recovery of the product decreased with an increase in
temperature, because of the faster decomposition of hydrogen peroxide with increase in
temperature (Gupta et al, 2004).

2.5.4.4. Precipitation kinetics

The kinetics of the precipitation process is complex, involving nucleation and crystal growth,
with supersaturation as the driving force for these steps. Precipitation occurs when two
reacting solutions forms a solid product with low solubility. This solid product is formed by
fast crystallization resulting in large numbers of very small crystals.

Precipitation will only take place a certain time after the development of supersaturation,
called the induction period, because of the slow growth of small particles. When the
supersaturated concentration of the solute is high, spontaneous nucleation of very small
crystals occur, lowing the concentration of the solute. As the concentration of the solute
decreases a metastable region is reached where crystals can grow but cannot nucleate, thus
if no crystals are present, none can be formed. At a certain point, equilibrium is reached
between the saturated solution and crystals that is formed. If the concentration of the solute
further decreases, the solution will be unsaturated and crystals of all sizes will dissolve.





School of Chemical and Minerals Engineering
54
Chapter 2: Literature survey
As mentioned, crystallization consists out of two steps, nucleation and crystal growth.
Nucleation is the formation (birth) of crystals and can be primary or secondary. Secondary
nucleation is mainly observed in industrial crystallizers where crystalline surfaces are
present and large crystals are desirable. The mechanism which is typically encountered in
industrial applications for secondary nucleation is referred to as contact nucleation and
occurs when crystal collide with each other and the crystals collide with metal surfaces such
as the vessel wall or agitator blades. An empirical power-law function is widely used to
describe the secondary nucleation and is given in Equation 2-13 (Seader & Henley, 2006:
659).

=
0 b j r
N T
B k s M N
(2-13)

In Equation 2-9, B
0
is the rate of homogeneous primary nucleation (number of nuclei), s is
the relative supersaturation, M
T
is the mass of crystals per volume of magma, N is the
agitation rate and the constants k
N
,b,j, and r are determined from experimental data.

The crystal growth can be explained by a two-step theory, referred to as the diffusion-
reaction theory. The first step of which is the mass transfer of solute from the bulk solution
to the crystal-solution interface and the given by Equation 2-14 (Seader & Henley, 2006:
659).

( ) =
c i
dm
k A c c
dt
(2-14)

With: dm/dt = rate of mass deposited on the crystal surface
A = surface area of the crystal
k
c
= mass-transfer coefficient
c = mass solute concentration in the bulk supersaturated solution
c
i
= supersaturated concentration at the interface

The second step is assumed to occur at the crystal-solution interface, in which solute
molecules are integrated into the crystal-lattice structure and this step is given in Equation 2-
15 (Seader and Henley, 2006: 659).

( ) =
i s
dm
k A c c
dt
(2-15)
School of Chemical and Minerals Engineering
55
Chapter 2: Literature survey
With: c
s
= Mass solute concentration in the solution at saturation
k
i
= Kinetic coefficient

The combined kinetics of these two-step theory is given by Equation 2-16 (Seader and
Henley, 2006: 659).

( )
=
+
s
c i
A c c
dm
1 1
dt
k k
(2-16)

The mass-transfer coefficient will be depended on the velocity of the solution, which means
at low velocities the growth rate will be controlled by the first step. The second step will be
important if the mass-transfer coefficient is large compared to the kinetic coefficient (Seader
& Henley, 2006: 658-660).

The growth rate of particles can also influence the number of particles formed. If rapid
growth is observed, co-precipitation may occur which make it difficult to obtain a pure
precipitate product. The growth rate can be controlled by the mass transfer of the ions to the
particle surface and/or integration of ions into the particle crystalline structure (Seader &
Henley, 2006: 671-672).

Using the kinetics supplied above, the volume and residence time of a precipitation vessel
can be calculated using the rate at which nucleation and crystal growth occurs.

2.6. Economic evaluation

The last few years showed a significant interest in the uranium market, encouraged by rapid
raising prices. The growing uranium market resulted in various uranium mining projects,
including new mining projects and expansion projects. A uranium project survey was done
in 2007 to investigate a few of these uranium projects. This survey provides a good
indication of capital and operating costs for these various plants and a summary of this
survey is given in Table 2.9.




School of Chemical and Minerals Engineering
56
Chapter 2: Literature survey
Table 2.9: Summary of uranium project survey (Damarupurshad, 2007: 6-7)
Project Company Capital costs
(R million)
Operating costs
Dominion uranium SXR Uranium One 1 036.64 R 98.89/lb of U
3
O
8

Ezulwini uranium &
gold
First Uranium
Corporation
3 219 R 419.58/ton ore
milled
Buffelsfontein
uranium & gold
First Uranium
Corporation
2 486.4 R 18.28/ ton ore
milled


The Ezulwini uranium and gold project investment and cost are explored in more detail. The
initial capital investment consists out of the mine life capital, including contingency, which is
R 2 072 million and the pre production capital which is R 1 147 million and is expended over
three and a half years. The average operating cost over 19 years is R 419.58 per ton of ore
milled. The net positive value for this project was calculated to be R 1 909.2 million using a
gold price of $ 500/oz and a uranium price of $ 40/lb with a rand to dollar exchange rate of R
7.40. The underground mining capacity of the Ezulwini project is up to 200 000 tons per
month and it was expected that at the end of 2008 the uranium production will reach an
output of 130 000 tons/month (Damarupurshad, 2007: 7).

Considering the economic evaluation of the Ezulwini project, it can be seen that a uranium
extraction project is feasible, yield a high net positive value (NPV). This high NPV is also an
indication that the capital investment can be sold at the end of the project lifetime to earn
back the money invested.



School of Chemical and Minerals Engineering
57
Chapter 3: Process development

Chapter 3: Process development

The proposed project requires the upgrade of a uranium extraction plant. In order to meet
the requirements given for this project it is necessary to understand the processes that are
used to extract the uranium. These processes are best understood if a certain design
procedure is followed to develop the process units. This section provides a detailed design
procedure that is given by Douglas (1988) which follows a systematic approach using certain
hierarchy of decisions listed below.

Level 0: Input information
Level 1: Batch versus continuous
Level 2: Input-output structure of the flowsheet
Level 3: Recycle structure of the flowsheet
Level 4: Separation system
o 4a: Vapor recovery
o 4b: Liquid recovery
o 4c: Solid recovery
Level 5: Heat integration

Each of these decision levels will now be discussed in order to understand the different
processes necessary to extract uranium. These levels will also allow continuous economic
evaluations during the design procedure to test economic feasibility. After these levels of
design are discussed, a detailed process description is given.

3.1. Level 0: Input information

The basic information required for the development of a process is given in the Level 0 of the
Douglas design procedure. Without the information gathered for Level 0, it is impossible to
start a conceptual design.




School of Chemical and Minerals Engineering
58
Chapter 3: Process development
3.1.1. Reactions and reaction conditions

The extraction of uranium usually requires four sections with each section consisting of their
own reactions and operating conditions. The most important operating conditions are the
temperature and pH. The extraction of uranium consists of leaching, counter-current
decantation, ion exchange, solvent extraction and precipitation sections. Leaching is the first
section in the extraction of uranium and is considered the most important. The operating
conditions for leaching are a pH of 2 and a temperature of 30 C throughout the leaching
section.

Table 3.1: Leaching reactions
Reaction Stoigiometry
Sulphuric acid
dissociation
H
2
SO
4
+ H
2
O HSO
4
-
+ H
3
O
+

Sulphuric acid
dissociation
HSO
4
-
+ H
2
O SO
4
2-
+ H
3
O
+

Nitric acid
dissociation
HNO
3
+ H
2
O NO
3
-
+ H
3
O
+

Muscovite leaching KAl
3
Si
3
O
10
(OH)
2
+ 10H
3
O
+
16H
2
O + K
2+
+ 3AL
3+
+ 3SiO
2

Chlorite leaching
Mg
2
Al
4
Fe
2
Si
2
O
10
(OH)
8
+ 20H
3
O
+
2Mg
2+
+ 2Fe
2+
+ 4Al
3+
+ 2SiO
2
+
34H
2
O
Pyrophylite
leaching
Al
2
Si
4
O
10
(OH)
2
+ 6H
3
O
+
2Al
3+
+ 4SiO
2
+ 10H
2
O
Pyrite leaching 6FeS
2
+ 30HNO
3
+ 3SO
4
2-
6Fe(SO
4
)
2
-
+ 3H
2
SO
4
+ 30NO + 12H
2
O
Albite leaching NaAlSi
3
O
8
+ 4H
3
O
+
Na
+
+ 2Al
3+
+ 3SiO
2
+ 6H
2
O
Uraninite leaching UO
2
+ 2NO
3
-
+ 4H
3
O
+
UO
2
2+
+ 2NO
2
+ 6H
2
O
Brannerite
leaching
UTi
2
O
6
+ 2NO
3
-
+ 4H
3
O
+
UO
2
2+
+ 2NO
2
+ 6H
2
O + 2TiO
2

U-phosphate
leaching
UO
2
ClPO
4
+ 2NO
3
-
+ 4H
3
O
+
UO
2
2+
+ 2NO
2
+ 6H
2
O + Cl
-
+ PO
4
2-

Coffinite leaching UO
2
2H
2
O + 2NO
3
-
+ 4H
3
O
+
UO
2
2+
+ 2NO
2
+ 8H
2
O
Coffinite leaching UO
2
SiO
2
+ 2NO
3
-
+ 4 H
3
O
+
UO
2
2+
+ 2NO
2
+ 6H
2
O + SIO
2

Uraninite leaching UO
2
+ 4NO
2
+ 2H
3
O
+
2HNO
3
+ UO
2
2+
+ 2NO + 2H
2
O
School of Chemical and Minerals Engineering
59
Chapter 3: Process development
Uranyl sulphate
complex formation
UO
2
2+
+ 3SO
4
2-
UO
2
(SO
4
)
3
4-

Silicon dioxide
dissolve
SiO
2
+ 2H
2
O SiO
2
2H
2
O

The second section in the extraction of uranium is the ion exchange process. In the ion
exchange section uranyl sulphate complexes are selectively adsorbed onto a resin and
desorbed again; this allows separation and upgrade of concentration. Table 3.2 contains the
reactions, and reaction conditions for the ion exchange section.

Table 3.2: Reaction information for ion exchange
Reaction Stoigiometry
Temperature
(C)
pH
Uranyl sulphate
complex formation
UO
2
2+
+ 3SO
4
2-
UO
2
(SO
4
)
2
4-
25 2
Iron sulphate
complex formation
Fe
3+
+ 2SO
4
2-
Fe(SO
4
)
2
-
25 2
Uranyl sulphate
adsorption
2R
2
-SO
4
+ UO
2
(SO
4
)
3
4-
R
4
-UO
2
(SO
4
)
3
+ 2SO
4
2-
25 2
Iron sulphate
adsorption
R
2
-SO
4
+ 2Fe(SO
4
)
2
-
2R-Fe(SO
4
)
2
+ SO
4
2-
25 2
Nitric acid
dissociation
HNO
3
+ H
2
O NO
3
-
+ H
3
O
+
25 2
Uranyl sulphate
elution
R
4
-UO
2
(SO
4
)
3
+ 2SO
4
2-
2R
2
-SO
4
+ UO
2
(SO
4
)
3
4-
25 1.5
Iron sulphate
elution
2R-Fe(SO
4
)
2
+ SO
4
2-
R
2
-SO
4
+ 2Fe(SO
4
)
2
-
25 1.5
Silicon dioxide
precipitation
SiO
2
2H
2
O SiO
2
+ 2H
2
O 25 1.5
Silicon dioxide
dissolve
SiO
2
+ 2H
2
O SiO
2
2H
2
O 25 1.5

The next section is the solvent extraction section; here the uranyl sulphate complexes are
removed from the aqueous phase into an organic phase. The up-concentration of uranyl
sulphate complexes is then done by stripping the complexes from the organic phase into a
School of Chemical and Minerals Engineering
60
Chapter 3: Process development
clean aqueous phase. Table 3.3 contains the reactions, and reaction conditions for the
solvent extraction section.

Table 3.3: Reaction information for solvent extraction
Reaction Stoigiometry
Temperature
(C)
pH
Sulphuric acid
dissosiation
H
2
SO
4
+ H
2
O HSO
4
-
+ H
3
O
+
25 1.5
Sulfuric acid
dissosiation
HSO
4
- + H
2
O SO
4
2-
+ H
3
O
+
25 1.5
Uranyl sulphate
complex formation
UO
2
2+
+ 3SO
4
2-
UO
2
(SO
4
)
3
4-
25 1.5
Solvent
preparation
2R
3
-N + H
2
SO
4
(R
3
NH)
2
SO
4
25

Solvent extraction UO
2
(SO
4
)
3
4-
+ 2(R
3
NH)
2
SO
4
(R
3
NH)
4
UO
2
(SO
4
)
3
+ 2SO
4
2-
25

Solvent stripping
(R
3
NH)
4
UO
2
(SO
4
)
3
+ (NH
4
)
2
SO
4
(R
3
NH)
2
SO
4
+
(NH
4
)
2
UO
2
(SO
4
)
2
+ (NH
4
)
2
SO
4

25 2.5 to7.5

The last section in the extraction process is precipitation; here the pH is manipulated to
precipitate a solid ADU product. The solid precipitate allows for more economical separation
of the product which is sent to NUFCOR which should comply to external specifications.
The reaction conditions for the precipitation are a pH of 7.5 and a temperature of 30 C. The
reaction for the precipitation is given in Equation 3-1:

2UO
2
SO
4
+ 6NH
4
OH (NH
4
)
2
U
2
O
7
+ 2(NH
4
)
2
SO
4
+ 3H
2
O (3-1)

It is important to know the kinetics and conversion of each reaction which is needed to size
the chosen processing equipment. The lack of information for the kinetics of mineral
processes limits the design process, and therefore assumptions are made. Most of these
assumptions are made from literature and is based on the conversion of reaction. The
available reaction kinetics and conversion assumptions are given in Table 3.4.

School of Chemical and Minerals Engineering
61
Chapter 3: Process development
Table 3.4: Reactions kinetics and conversion
Reaction Reaction rate Conversion Source
Sulphuric acid
dissociation
Equilibrium 100% of H
2
SO
4
-
Nitric acid dissociation Equilibrium 100% of HNO
3
-
Muscovite leaching - 6.2% of Muscovite Snll & Liljefors
Chlorite leaching - 60% of Chlorite Snll & Liljefors
Pyrophylite leaching - 5% of Pyrophyllite Lottering et al.
Pyrite leaching - 3% of Pyrite Karaca et al.
Albite leaching - 1% of Albite Snll & Liljefors
Uraninite leaching
| |

| | | | | |
( +
| | |
\ . \ . \ .
2.3
4
2 3
79500 36800
2.2 10 exp 0.46exp HNO NO
RT RT

99.99% of Uraninite Zhao & Chen
Brannerite leaching - 10% of Brannerite -
U-phosphate leaching - 1% of U-phosphate -
Coffinite leaching - 90% of Coffinite Lottering et al.
Uranyl sulphate
complex formation
- 100% of UO
2
2+
-
Silicon dioxide dissolve - 0.001% of SiO
2
-
Iron sulphate complex
formation
- 100% of Fe
3+
-
Uranyl sulphate
adsorption
-
40% of Resin
capacity
Merrit et al.
Iron sulphate adsorption -
1% of Resin
capacity
Merrit et al.
Sulphate adsorption -
50% of Resin
capacity
Merrit et al.
Uranyl sulphate elution -
100% of Uranyl
sulphate adsorbed
-
Iron sulphate elution -
100% of Iron
sulphate adsorbed
-
Sulphate elution -
100% of Sulfate
adsorbed
-
Silicon dioxide
precipitation
-
100% of dissolved
Silicon dioxide
-
Solvent preparation - 100% of R
3
-N Rydberg et al.
School of Chemical and Minerals Engineering
62
Chapter 3: Process development
Solvent extraction Equilibrium
99.99% of Uranyl
sulphate complex
Rydberg et al.
Solvent stripping Equilibrium
99.99% of loaded
solvent
Rydberg et al.
Precipitation -
98.49% of Uranyl
sulphate complex
Mellah et al.

The values from Table 3.4 are used for the mass and energy balance, process design and
detail design.

3.1.2. Desired production rate and purity

The plant is designed to process 360 000 ton ore per month, at a uranium recovery of 78%.
The desired product is ADU in a slurry phase which consist of 35 mass% U
3
O
8
. Penalties
are paid for product which does not meet product specifications. Using the above
information, the amount of ADU produced per month is approximately 89 ton.

3.1.3. Raw materials

As stated in Douglas (1988: 104), laboratory studies are usually carried out with pure
chemical reagents. However, this is not the case for industrial applications, thus it is
important to know which impurities are present. The impurities can cause unwanted side
reactions that should be monitored.

The raw materials needed for leaching includes the ore feed, nitric acid, sulphuric acid,
Magnafloc 90L and potable water. Concentrated nitric acid with a 68 mass% is used which
is a standard product purity from a distillation column, due to an azeotrope formed with
water. A 98 mass% sulphuric acid is used which is already available on the existing plant.
Magnafloc 90L is used as a clarifier in the counter-current decantation section. Potable
water is used to dilute the acids and to achieve a slurry with a SG of 1.6 for the ore feed.
The ore composition is given in Table 3.5 (Lottering et al., 2007:18).






School of Chemical and Minerals Engineering
63
Chapter 3: Process development
Table 3.5: Ore composition
Mineral
Mass %
Quartz (SiO
2
)
70.200
Muscovite (KAl
3
Si
3
O
10
(OH)
2
)
10.100
Chlorite (Mg
2
Al
4
Fe
2
Si
2
O
10
(OH)
8
)
2.000
Pyrophylite (Al
2
Si
4
O
10
(OH)
2
)
9.700
Pyrite (FeS
2
)
1.300
Albite (NaAlSi
3
O
8
)
4.800
Uraninite (UO
2
)
0.014
Brannerite (UTi
2
O
6
)
0.013
U-phosphate (UO
2
ClPO
4
)
0.001
Coffinite (U(SiO)
41-x
(OH)
4x
)
0.002

Resin, dilute sulphuric acid, caustic soda and potable water are the raw materials used for
the ion exchange section. The resin used is a strong base anion exchange resin, Ambersep
TM 400 SO
4
which has quaternary ammonium as functional group. Due to resin poisoning,
caustic soda is used to regenerate the resin. The caustic soda is received in the aqueous
phase and potable water is used to wash the solid impurities from the resin and to dilute
concentrated sulphuric acid.

The raw materials needed for the solvent extraction section are kerosene, isodecanol, alkyl
amine and demineralised water. Kerosene is a collection of organic substances which are
inert to the system, and only acts as a carrier fluid. Isodecanol and alkyl amine is pure
organic chemical substances acting as the third phase modifier and extractant respectively.
Precipitation raw materials include ammonia gas and demineralised water. A summary of all
the raw materials used is given in Table 3.6.


School of Chemical and Minerals Engineering
64
Chapter 3: Process development
Table 3.6: Summary of raw materials
Raw material Purity Description Price
Alamine (Alkyl amine) Pure Solvent extractant R 57 260 per ton
Ammonia Pure gas Precipitation reactant R 15 000 per ton
Calcium oxide Pure Neutralization R 139 per ton
Isodecanol Pure Solvent modifier R 22 410 per ton
Kerosene Pure Solvent diluent R 5 000 per m
3

Magnafloc 90L Pure Clarifying agent R 22 500 per ton
Nitric acid 68 mass% HNO
3
Oxidation agent R 1 500 per ton
Potable water Municipal Multiple uses R 7 per m
3

Resin (Ambersep TM 400) Pure Ion exchange resin R 57 000 per ton
Sodium carbonate Pure Solvent regeneration R 900 per ton
Sodium hydroxide Diluted in water Regeneration agent R 4 200 per ton
Sulphuric acid 98 mass% H
2
SO
4
Leaching agent R 500 per ton

3.1.4. Processing constraints

It is important to identify and understand the hazardous processing conditions to avoid
equipment damage as well as an unsafe working environment. These process conditions
are known as the processing constraints and should play and integral role in the design of
the plant.

In the leaching section concentrated sulphuric and nitric acid is used which are highly
corrosive. The choice of materials for construction must be resistant to these conditions in
order to improve the economic potential of the project. The dissolution of acids in water is
exothermic and needs to be monitored. The leaching of uranium through a nitric acid
system is autocatalytic and exothermic and this combination may result in hazardous
operating conditions.

An unwanted increase in the pH of the pregnant leach liquor will result in the precipitation of
dissolved impurities and causes fouling of the resin. The resin used in the ion exchange
process is also sensitive the sharp changes in the pH of the system which results in swelling
of the resin. The selectivity of the resin towards uranium is the key aspect in designing ion
exchange equipment.


School of Chemical and Minerals Engineering
65
Chapter 3: Process development
The organic compounds used in solvent extraction are extremely flammable and cause an
explosion hazard in enclosed vessels. High temperatures will increase the possibility of a
fire or explosion due to the low ignition point of the organic compounds. One of the
important operating conditions is the pH of the system. A high enough pH (above 7) will
result in precipitation of the uranyl sulphate complexes.

The pH of the precipitation section is the main design parameter for the efficient operation of
this section. Small deviations from the optimum pH will result in precipitation of impurities
and product that does not meet specifications.

3.1.5. Other plant and site data

The battery limits and cost of certain facilities is important for the design of a new process.
The availability and capacity of the utilities on a plant needs to be sufficient to supply the
demand. The existing South Uranium Plant already has the required facilities to provide all
the utilities. The utilities used for the entire plant is listed in Table 3.7.

Table 3.7: Utilities
Utility Conditions Price
Steam 175 C at 12 bar R 50 per ton
Water Demineralised water R 7 per m
3

Electricity - 40c per kWh
Compressed air > 4.5 bar -

The waste disposal facilities needed for the tailings from each process unit are tailing dams
and a neutralization plant which complies with international legislation for the protection of
the environment.

3.1.6. Physical properties of all components

The ASPENTech

chemical database is used to create a databank for the physical

properties of all the components used. However not all the components are in the
ASPENTech

chemical database, thus the properties need to be gathered from other

sources which included MSDS data and chemical compound databases on the internet. The
unknown physical properties of the compounds are given in Table 3.8.

School of Chemical and Minerals Engineering
66
Chapter 3: Process development
Table 3.8: Physical properties of unknown compounds
Compound
Specific heat
capacity (J.kg
-1
.K
-1
)
Vapor
pressure (bar)
Molar volume
(cm
3
/mol)
Heat of
formation
(J/mol)
ADU - Solid - 845200 cal/mol
Amine 366 - Stay liquid - -
Brannerite 213 667 Solid 72
-673.119
kcal/mol
Chlorite 561 063.618 Solid 207
-377.081
kcal/mol
Coffinite 136 969 Solid 53
-454.704
kcal/mol
HNO
2
- Solid - -87 000
Iron complex resin - Solid 8 000 -
Isodecanol - 0.0013 - -
Kerosene 2 010 0.00689 - -24 149
Loaded Amine 366 - Stay liquid - -
Resin - Solid 7 110 -
U-phosphate 149 105 Solid 76
-465.728
kcal/mol
Uranyl complex resin - Solid 14 489 -

3.2. Level 1: Batch versus continuous

There are three guidelines for the decision between whether a process is batch or
continuous. These guidelines are:

The production rate of the process per year.
The market force of the produced material.
Operational problems that might occur.





School of Chemical and Minerals Engineering
67
Chapter 3: Process development
To make an informed decision, it is important to consider the influence of all three the
guidelines. The ore capacity for the plant is 4 320 000 ton per year, and according to
Douglas (1988) a capacity greater than 4500 ton per year results in a continuous process.
The product (ADU) is not a seasonal product and is used worldwide which classifies the
process as a continuous process.

The operational problems which occur in the ion exchange process are resolved by
operating the process as semi-batch. Ion exchange is an adsorption and desorption process
and therefore it is operated as a semi-batch process, but this does not affect the continuity of
the entire plant. From this it is concluded that the plant is operated as a continuous process
and the following aspects will help to develop a conceptual design for a continuous process:

Process units needed
Interconnections among units
Estimate the optimum processing conditions
Which units should be batch or continuous
Single vessels versus individual vessels for each step

3.2.1. Process units needed

The ore is processed to a slurry by the gold mining section and is the raw material from
which the uranium is extracted. Low grade South African ore is used, with a low uranium
concentration. For this reason the uranium must first be liberated from the ore and then
concentrated to allow easy recovery of the desired product. The four process units required
to recover the desired product are leaching, ion exchange, solvent extraction and
precipitation.

Acid leaching allows the liberation of uranium from the uranium containing minerals in the
ores. Acid is added to the slurry in order to decrease the pH to a desired level at which
uranium is dissolved from the ore. An oxidation agent is also required to oxidize uranium to
the desired oxidation state. The slurry is sent to counter-current decantation where the
liquids and solids are separated. The underflow is sent to the neutralization plant where it is
prepared for the gold extraction process.


School of Chemical and Minerals Engineering
68
Chapter 3: Process development
The leach liquor has a uranium concentration of 0.15 g U
3
O
8
/L which is increased to 4 g
U
3
O
8
/L in the ion exchange section. This is achieved by the highly selective adsorption of
uranyl sulphate complexes onto a strong base resin. The uranyl sulphate complexes are
recovered from the resin by elution or desorption with the eluant, diluted nitric acid.

The purpose of the solvent extraction unit is to upgrade the uranium concentration as well
removing impurities. The extraction stage is where the selective dissolving of uranium from
the aqueous phase into the organic phase occurs. Further processing of the organic phase
with water allows the removal of impurities such as iron sulphate complexes. The uranium is
stripped from the organic phase with sulphate ions while the pH is controlled with caustic
soda. The final process unit is precipitation where the solid yellow cake product, ammonium
diuranate, is precipitated. In this unit the specifications of the final product, which include 35
mass% ADU, should be reached.

3.2.2. Interconnections among units

The conceptual design is based on the upgrading of the current South Uranium Plant
situated near Orkney in the North-West Province. The same processing units are used as in
the existing flowsheet and are illustrated as a simplified block flow diagram in Figure 3.1.




Figure 3.1: Simplified block flow diagram of process

The ore is fed to the leaching stage which includes the two counter-current decantation
trains with five thickeners and a clarifier in each train. The leaching slurry is separated into
the pregnant leach liquor and solid slurry which is sent for further processing at the gold
extraction plant. The pregnant leach liquor, which contains less than 50 ppm solids, is sent
to the ion exchange stage.




School of Chemical and Minerals Engineering
69
Chapter 3: Process development
The concentration of the uranyl complexes in the leach liquor is upgraded and partially
cleansed of impurities in the ion exchange stage. The eluate from the ion exchange is sent
to the solvent extraction unit, where further up concentration and cleansing of the uranyl
complexes takes place. The OK liquor is sent to the precipitation unit where the ADU
product is formed.

3.2.3. Estimate the optimum processing conditions

The optimum process conditions will ensure balance between favourable economic potential
and efficient overall production of the ADU product. These operating conditions are acquired
from the current South Uranium Plant and external literature. The processing conditions for
each stage are given in Table 3.9.

Table 3.9: Processing conditions
Process conditions Value
Leaching
pH (-) 2
Temperature (C) 30
Final underflow SG (-) 1.55 to 1.6
Ion exchange
pH (-) 1 to 2
Temperature (C) 25
Solid content (ppm) < 50
Solvent extraction
pH (-) 2 to 5
Temperature (C) 25
Precipitation
pH (-) 7.5
Temperature (C) 30

Table 3.9 is the optimum conditions reported from literature, but it is not always possible to
operate the entire plant at these conditions.


School of Chemical and Minerals Engineering
70
Chapter 3: Process development
3.2.4. Additional conceptual design information

As discussed in the above sections, the overall process is considered to be continuous.
However, some of the processing units are operated as semi-batch units. This section
discusses additional process information regarding the continuity and vessel configuration of
the processing units.

Leaching is a continuous process which requires numerous pachuca tanks in series to
ensure adequate residence time. The vessels in series allow for better mean residence time
and mixing of the entire stream, and when combined ensure optimum conversion.
Manufacturing and designing a single leaching vessel with the capabilities to process the
entire stream is not economically viable. Therefore tanks in series are the optimum
configuration. Counter-current decantation is a solid-liquid separation unit which is usually
done using separation stages. Separate thickeners are used to ensure optimum recovery of
the liquid.

In the ion exchange process the resin remains stationary in fixed-bed columns while the
leach liquor flows through it in the adsorption stage and the eluant flows through it in the
elution stage. Since this is a semi-batch process, only three or four fixed-bed columns are
required. Usually the elution or wash stage is conducted in one column while adsorption
continues in the other columns. Since each column must carry out adsorption, washing and
elution, intensive piping is required.

Solvent extraction is a liquid-liquid separation unit which is operated continuously. The
liquid-liquid separation requires different sections which include extraction, scrubbing,
stripping, and regeneration. Each stage requires a certain amount of stages depending on
the through-put of the system. Precipitation is a fairly simple process which only requires a
vessel that provides enough residence time for adequate precipitation and seeding. It is
possible to uses vessels in series to increase the residence time. The precipitated slurry is
sent to a thickener from which the overflow is sent back to solvent extraction and the
underflow is sent to a centrifuge. The centrifuge continuously delivers a product with a high
concentration ADU by removing liquids.


School of Chemical and Minerals Engineering
71
Chapter 3: Process development
3.3. Level 2: Input-output structure of the flowsheet

It is important to account for all the chemical compounds in the system to ensure that there
is no build-up or unnecessary reagent and product losses. To ensure accountability, an
input-output structure for the entire process is done. In Douglas (1988) Level 2, the following
questions are used as a guideline to develop the block flow diagram (Douglas, 1988: 118).

Should the feed stream be purified?
Should a reversible by-product be removed or recycled?
Should a gas recycle and purge stream be used?
Should the reactants be recovered and recycled?
How many product streams will there be?
What are the design variables for the input-output structure, and what economic
trade-offs are associated with these variables?

These questions are answered in the following sections.

3.3.1. Feed purification

The major feed impurity exist in the ore feed and the impurities include pyrite, chlorite,
quarts, muscovite, etc. The above impurities are not inert and consume a large amount of
raw materials. It is impossible to economical remove these impurities from the feed stream.
The processing of uranium is of such a nature that the impurities are effectively and
economically removed throughout the process.

All the other raw material feed streams do not include significant amounts of impurities.
Most of the raw material feed streams are diluted with potable water which is considered as
an inert, but serves a purpose in the processing of uranium. Therefore there is no
purification of the feed.





School of Chemical and Minerals Engineering
72
Chapter 3: Process development
3.3.2. Recycle streams

In Douglas (1988) it is stated that 99% or more of the valuable materials should be
recovered which is achieved by recycle streams or by-product removal. The recycling of
valuable reactants will result in a decrease of raw material cost and reduce the amount of
chemicals introduces into the environment.

The important recycles are nitrate and sulphate ions, solvents and wash solution. Nitrate
ions are the oxidizing agent in the leaching process while the sulphate ions are used as
lixiviant in the leaching process and a complexing agent for the uranyl ions.

It is very important to recycle more than 99% of the organic solvent due to the negative
economic impact. The wash solution is used as a replacement for potable water in the
counter-current decantation section which will recycle un-reacted uranium leading to an
increase in the overall recovery of uranium.

3.3.3. Removal and purge streams

Removal and purge streams are necessary to prevent build-up of impurities in the system.
The three important waste by-products are nitrogen oxide gas, iron sulphate and silicon
dioxide. The full removal of these compounds will prevent a build-up in the system. It is
important to remove these substances in an environmentally friendly and economically
viable manner.

Nitrogen oxide produced in the leaching process is classified as a light component according
to Douglas (1988) which should be purged from the system. Iron sulphate is present in the
aqueous phase and if not removed will affect the product purity. Silicon dioxide is a fouling
agent for the resin used in the ion exchange section which builds-up if not removed.

3.3.4. Number of product streams

It is important to keep track of all the components in the system to prevent impurity build-up
and this is done with an overall input-output structure. The overall input-output structure for
the process is given in Figure 3.2.

School of Chemical and Minerals Engineering
73
Chapter 3: Process development


Figure 3.2: Input-output structure of overall process

Due to the numerous amounts of components in the system, it is irrelevant to list all the
species in the product and waste streams. Table 3.10 shows the product streams from
Figure 3.2.


School of Chemical and Minerals Engineering
74
Chapter 3: Process development
Table 3.10: Product stream classification
Stream Destination
ADU Primary product
Slurry to gold plant Valuable by-product
Nitrogen oxide Purge
Iron sulphide Waste
Silicon dioxide Waste
Nitrate ions Recycle
Sulphate ions Recycle
Solvent Recycle
Wash solution Recycle

The ADU and the slurry to the gold plant streams are the only streams that hold economical
value. The final product in the ADU stream is sent to NUFCOR for further processing while
the gold is extracted from the slurry at the gold plant. The waste is sent to the waste
treatment facilities.

3.3.5. Preliminary economic potential analysis

According to Douglas (1988) the economic potential 1 (EP-1) and EP-2 should be done to
ensure that the process is economically viable to proceed to the next design steps. The
EPs are based on a preliminary mass balance with the following assumptions made:

H
2
SO
4
is used to control pH in leach section.
Enough HNO
3
is used to oxidize minerals.
Reactions occur in series.
Loss of 1.1% of solvent per hour.
Overall product recovery of 76%.

The preliminary mass balance is based on the input-output structure displayed in Figure 3.2.
Table 3.11 consist of the summarised mass balance which includes reagent and product
prices.


School of Chemical and Minerals Engineering
75
Chapter 3: Process development
Table 3.11: Level 2 preliminary mass balance with prices
Component Amount (ton/annum) Price per unit (R/ton) Total cost (R/annum)
Raw materials
Alamine (Alkyl amine) 0.8 57 260 45 808
Ammonia 140 15 000 2 100 000
Caustic soda 18 4 200 75 600
Isodecanol 0.3 22 410 6 723
Kerosene 410 m
3
5 000 per m
3
2 050 000
Magnafloc 90L 86 22 500 1 935 000
Nitric acid 19 000 1 500 28 500 000
Ore 4 320 000 - -
Potable water 8 600 m
3
7 per m
3
60 200
Resin (Ambersep TM 400) 62 57 000 3 534 000
Sulphuric acid 54 300 500 27 150 000
Products
Uranium 1068 743 786.41 794 363 885.88
Gold 2.16 254 552 770 549 833 983.2

The EP-1 analysis is the income from the product minus the expense of the raw materials
used per annum. The EP-1 analysis gives a positive value of R 728 million per annum. The
EP-2 analysis is the income from the product and by-product minus the expense of the raw
materials. It is assumed that 0.5 g gold per ton of ore feed is produced. The EP-2 analysis
gives a positive value of R 1 278 million per annum. The EP-1 and EP-2 both gives large
positive values, thus the design process can continue to Douglas level 3.

3.4. Level 3: Recycle structure of the flowsheet

The degree of recycling plays an important role in the economical potential, sizing and
operational cost of equipment. In Douglas (1988) it is stated that a number of decisions
need to be made concerning the recycle structure of the flowsheet. These decisions are
described in the following sections.


School of Chemical and Minerals Engineering
76
Chapter 3: Process development
3.4.1. Reactor systems required

The process of extracting uranium consists of four sections, with an additional two sections
for the counter-current decantation after the leaching and precipitation sections. In this
section, the leaching and precipitation is regarded as the only reactor systems. The ion
exchange and solvent extraction sections have a high selectivity and recovery of the desired
product complexes and for this reason are regarded as efficient separation systems. Figure
3.3 shows the reactors as white blocks and separation systems as black.




Figure 3.3: Block flow diagram for reactor system

In Figure 3.3 the white blocks represents the reactor systems while the black blocks
represents the separations system as discussed above. The product streams of the various
systems contain valuable reagents which can be recycled to reduce raw material costs.

3.4.2. Number of recycle streams

The valuable reagents which leave the process include streams with high levels of nitrate,
sulphate, and ammonium ions. These streams are recycled to specific process units where
it will most effectively reduce the raw material cost. These recycle structures are shown in
Figure 3.4.







School of Chemical and Minerals Engineering
77
Chapter 3: Process development



Figure 3.4: Recycle structure

In Figure 3.4 three recycle streams are indicated which hold economical and environmental
advantages for the process. The first recycle stream is the barren liquor form the ion
exchange section which is stripped of uranyl sulphate complexes. The stream contains
large amounts of water and is therefore recycled to the counter-current decantation section
as wash solution. Make-up wash water is still required, but in much less quantities to
effectively remove the uranyl sulphate complexes from the remaining solid slurry. The
reduction in make-up wash water results in less potable water being contaminated.

The second recycle stream is the barren eluate from the ion exchange section. At the ion
exchange section sulphate ions in the eluant are used for desorption of uranyl sulphate
complexes from the resin. An excess of sulphate ions are fed to ensure complete removable
of the uranyl sulphate complexes due to the equilibrium of the system. The excess sulphate
ions are removed in the extraction section of solvent extraction and are recycled back to the
leaching process which will reduce the amount of lixiviant needed.

The third recycle stream is the aqueous solution from the solid-liquid separation stage at the
precipitation section. This stream contains ammonium and sulphate ions which is required
as raw materials for the precipitation and solvent extraction respectively. These three
recycle streams play an important role in the economical potential of the process.


School of Chemical and Minerals Engineering
78
Chapter 3: Process development
3.4.3. Abundance of reactants

An abundance of reactants may cause virtually full conversion of the desired reactant, but
will have a negative impact on the economic potential analysis. There are numerous side
reactions for the hydronium and nitrate ions which forms unwanted by-products. An
abundance of the required reactant is supplied to the leaching section to ensure a maximum
recovery of the uranium while sustaining the unwanted side reactions.

The ore feed consist of many impurities which cannot be controlled or separated. These
impurities consume the reactants and form unwanted products. Hydronium ions are used to
dissolve most of the ore minerals while nitrate ions acts as an oxidising agent for the
uranium containing minerals and pyrite. Since the dissolution of uranium bearing minerals
are the main objective, the oxidising agent is fed in excess. The hydronium ions are
controlled at a certain level to ensure the desired pH in the leaching system.

The hydronium ions are mainly supplied by the addition of sulphuric acid which is more
economical compared to nitric acid. This results in an excess of sulphate ions in the
leaching section which is required for the adequate formation of uranyl sulphate complexes.

3.4.4. Operational considerations

Specific utilities for reactors are cooling water, steam, electricity and pressurised air.
Although pressurised air and steam are required at the reactors in the uranium extraction
process. Pressurised air and steam are available from the gold extraction plant which is
located near the uranium plant and therefore there is no need for compressors or boilers on
the South Uranium Plant. The ammonium gas required for the precipitation section is
assumed to be supplied under sufficient pressure. The pressure gradient is used to
transport the ammonium gas to the precipitation section.

In most cases the leaching and precipitation reactors are heated adiabatically to increase the
reaction rates. The immense quantities of water, which is a heat carrier, absorb the small
amounts of energy that is released from the reactions. The temperature difference between
night and day time as well as seasons should be taken into consideration for the operation of
the reactors.


School of Chemical and Minerals Engineering
79
Chapter 3: Process development
3.4.5. Recycle economic evaluation

The economic potential of the project should be continuously evaluated to ensure that the
project does not reach a negative value. The third level of evaluation, EP-3, is done by
subtracting the reactors costs from the EP-2 value calculated (Douglas, 1988: 158). The
reactors in this process include the leaching and precipitation tanks. The total costs of the
reactors should consist of the capital cost for the equipment and the cost of the power
needed to operate the reactors. The operations needed for these reactors are air agitation
and steam heating.

As mentioned this is an upgrade project and therefore existing equipment is used for both
leaching and precipitation therefore the only capital cost required is the purchasing of pumps
and storage tanks for the nitric acid used in the leaching process. Steam is necessary to
heat the pachuca tanks in the leaching unit and the feed stream to the precipitation unit to a
temperature of 30 C. The steam costs are calculated using a utility cost of R 50/ton and the
results are shown in Table 3.12 together with the capital costs of the equipment.

Table 3.12: Capital and operation costs for reactors
Unit Capital cost (R) Operation cost (R) Total (R)
Leaching 34 681 483.46 10 306 799.20 44 988 282.66
Precipitation 15 770 282.59 10 005 643.27 25 775 925.86
Total: 70 764 208.52

Using the capital and steam costs calculated in Table 3.12, the EP-3 is determined to be
R 1 207 million, which is an indication of a positive economic potential for this stage of the
feasibility study. This economic evaluation shows positive results, therefore the study can
continue to the next phase.
School of Chemical and Minerals Engineering
80
Chapter 3: Process development
3.5. Level 4: Separation systems

The usual Douglas (1988) design hierarchy is followed for Levels 0 through 3 for the mineral
extraction process. The usual Douglas (1988) design hierarchy is mainly used for
petrochemical plants; however the main difference in mineral extraction is the presence of
solids in the system. For this reason the Rossiter and Douglas (1988: 408) Level 4 is
followed for the solids system. The questions that need to be answered for this level are
(Douglas, 1988: 410):

How can the primary product be recovered?
What type of solids recovery systems is used?
How should the waste-solid separation be accomplished?
Are any liquid-liquid separations required?
Locations of separation units (purge or recycle streams or both)?

For mineral extraction processes three physical phases are present, and each phase has its
own recovery system. Level 4 can by divided into the following sections:

General structure.
Vapor recovery system.
Solid recovery system.
Liquid recovery system.

3.5.1. General structure

The first separation unit occurs in the leaching section, where two products streams leave
the system. One stream is the solid-liquid stream which contains the desired product and
by-product. The second product steam is a result of the air agitation used and may contain
nitrogen oxide gas. The amount of nitrogen oxide formed is completely soluble in the
aqueous phase. The solid-liquid product stream is processed in separate separation units
and a general separation structure is given in Figure 3.5.





School of Chemical and Minerals Engineering
81
Chapter 3: Process development


Figure 3.5: General separation structure

The solid-liquid product leaving the leaching section proceeds to the counter-current
decantation section where the liquids are recovered from the leaching product stream. The
liquid stream from the counter-current decantation section is sent to the ion exchange
system where the uranyl sulphate ion complexes are selectively removed to the eluate. The
eluate then flows to the solvent extraction unit where impurities are removed using liquid-
liquid separation. The stripping liquor from the solvent extraction section is sent to the
precipitation section. The slurry from the precipitation section is sent to a solid-liquid
separation unit where the ADU product is recovered.

3.5.2. Vapour recovery system

Vapour recovery systems are used to recover valuable gasses from a purge or vent stream.
No vapour recovery system is necessary because small amounts of valuable gasses are
present in the vent. It is not economically viable to recover these small amounts of the
valuable gasses. The vent stream will mainly consist of air due to air agitation.

3.5.3. Solid recovery system

From Figure 3.5 it is noted that there is two solid recovery systems in the uranium extraction
plant. The first is the waste solid recovery which is used to remove the un-reacted ore
containing the valuable gold by-product. The second solid recovery is where the ADU
product is separated from the aqueous phase.

Counter-current decantation is used in the waste solid recovery which is done in two
identical thickener trains. The waste solid recovery system is shown in Figure 3.6.
School of Chemical and Minerals Engineering
82
Chapter 3: Process development



Figure 3.6: Waste solid recovery system

The first thickeners in each train serves as a clarifier to ensure low solid content in the
overflow sent to ion exchange. The floculant, Magnafloc 90L, is distributed evenly over the
five stages of the solid recovery system for the gold, but is not added to the first thickener.
This counter-current decantation system ensures a uranium recovery of 99.99%. The slurry
sent to the gold extraction plant is washed with the barren leach liquor from the ion
exchange section.

The second solid recovery system is the ADU precipitation unit and is illustrated in Figure
3.7. The OK liquor from solvent extraction is sent to the two tanks in series where
precipitation of the ADU takes place. The solution containing the precipitated product is sent
to a thickener to settle the ADU cake and decant the ammonium sulphate which is sent back
to the solvent extraction stripping section. The thickener underflow, containing ADU, is sent
to two centrifuge stages. Inside the first centrifuge stage the slurry is sent to two centrifuges
in parallel where the ADU is washed with demineralised water spray. The solids are sent to
the second stage centrifuge while the liquids are pumped to the centrate tank so that it can
be recycled to the ADU thickener. The solids from the second centrifuge stage flows to the
final product storage tanks for transport to NUFCOR while the liquid is sent to the waste
water treatment facility.
School of Chemical and Minerals Engineering
83
Chapter 3: Process development


Figure 3.7: Solid recovery system for ADU

3.5.4. Liquid recovery system

Figure 3.5 show that there are two liquid recovery systems in the uranium extraction process
with the main purpose to purify and increase the concentration of the U
3
O
8
. The increase in
concentration and purification allows more economical recovery of the desired product. Ion
exchange is the first process used for the purification and up-concentration of the product,
and the second is the solvent extraction process.

The leach liquor contains low concentrations of uranyl sulphate complexes with a high
volumetric flow. The ion exchange process is a liquid recovery system where the uranyl
sulphate ions are transferred from the leach liquor to the eluate which has a much lower
volumetric flow and a higher uranyl sulphate concentration. This is achieved by adsorption
of these ions onto resin particles and then eluted from these particles into the eluate. The
adsorption and elution of the resin takes place in the same columns at different stages. This
concept is shown in Figure 3.8.






School of Chemical and Minerals Engineering
84
Chapter 3: Process development



Figure 3.8: Schematic representation of ion exchange

Figure 3.8 is a schematic representation of the operation of fixed-bed column ion exchange
where column 1 and 3 is in the adsorption stage while column 2 is being eluted. Before
each elution stage, the resin is washed with water to remove solid particles. This setup
usually contains a total of three or four columns.

The second liquid recovery system is the solvent extraction section. The solvent extraction
process requires four different sections to ensure efficient separation between the aqueous
feed and the solvent added. These sections are extraction, scrubbing, stripping and
regeneration; the sections are displayed in Figure 3.9.



Figure 3.9: Schematic representation of solvent extraction




School of Chemical and Minerals Engineering
85
Chapter 3: Process development
The extraction section is responsible for the extraction of the uranyl sulphate complex from
the aqueous phase into the organic phase which will include entrainment of iron sulphate
complex. The scrubbing section removes the entrained iron sulphate complex by washing
the solution with demineralized water. The last section is where the uranyl sulphate complex
is stripped from the organic solvent with ammonia sulphate and up-concentrated in the
aqueous phase. The solvent is regenerated in one stage with caustic soda and sodium
carbonate to remove any impurity build-up. Each section contains multiple stages to
increase the recovery of U
3
O
8
.

3.5.5. Separation system economic evaluation

The final economic evaluation level in the design procedure is level four (EP-4) in which
purge losses and separation unit costs are considered (Douglas, 1988:189). In the designed
process no products are purged, only reagents. The cost of these reagent losses is
considered in the raw materials cost estimations thus no purge loss calculations is
necessary. The separation systems needed for the process, as mentioned above, is
counter-current decantation, ion exchange, solvent extraction and the ADU solid separation
system. The capital and operation cost for the equipment and pumps of the separation
systems are calculated and given in Table 3.13.

Table 3.13: Separation systems cost calculations
Unit Capital cost (R) Operation cost (R) Amount (R)
CCD 69 999 662.07 4 003 586 74 003 248.07
Ion exchange 6 504 779.59 10 294 450 16 799 229.59
Solvent extraction 26 134 497.60 200 215.98 26 334 713.58
Total: 117 137 191.24

The final EP-4 is calculated by subtracting the total capital and operating costs of the
separation units from the EP-3 calculated value. The EP-4 value obtained is R 1 089 million,
which gives a positive economic potential indication. This is the final economic evaluation
which is a good indication of the economic feasibility and therefore the study can continue to
the next phase.

School of Chemical and Minerals Engineering
86
Chapter 3: Process development
3.6. Level 5: Heat integration

Energy conservation is an important aspect that is considered in the design of a process and
can be done by means of energy integration. During energy integration, heat is transferred
from a hot stream that needs to be cooled, to a cold stream that needs to be heated. The
heat integration is done by analysing the minimum heating and cooling requirements for the
heat-exchanger network (Douglas, 1988:216).

Investigating the current uranium extraction process, energy is only necessary for the
heating of the leaching pachuca tanks as well as the OK liquor stream to 30 C. The heating
of the leaching feed streams is done with steam and no cooling is needed. Thus heat
integration will be redundant because there are only two heating requirements. Therefore
only one heat exchanger is used at the precipitation unit and no heat integration network is
created.

3.7. Equipment design

In an expansion project it is vital to determine the capacity needed after expansion, which
will allow the evaluation of current equipment re-sizing. The sizing methods followed to size
each of the discussed equipment below is fully described in Appendix B. Since the detail
design of the solvent extraction unit is done in Chapter 4, the solvent extraction equipment is
not designed in this section.

The first process unit, leaching, is sized using reaction kinetics and sizing methods
discussed by Minerals Council of Australia (2006). From this it is calculated that a
conversion of 93% is achieved, utilising 11 out of the 14 existing air agitated pachuca tanks.
The operating temperature for the leach process is optimised at 30 C and a nitrate
concentration of 0.0618 mol/L , which allows for optimum energy and reagent consumption.

The counter-current decantation section is sized according to the existing equipment. The
unit area is calculated for the existing equipment and falls in the range given by Merritt
(1971). Using the unit area, a second train of 6 thickeners is sized to process the extra feed
capacity. The first thickener in the new train has a diameter of 50 m and the rest have a
diameter of 45 m.

School of Chemical and Minerals Engineering
87
Chapter 3: Process development
Fixed-bed ion exchange columns are preferred to the existing continuous counter-current ion
exchange process. The capacity of the existing columns is determined using basic mass
balance combined with resin specifications for Ambersep TM400. It is found that the existing
columns have adequate capacity to process the increased feed. Four of the existing
adsorption columns are used while the remaining two are back-up columns. The existing
elution column is used for regeneration of the resin.

The precipitation reactor is sized using kinetics from Table 3.4. The two existing
precipitation reactors are adequate to achieve high efficiency thus it is assumed that the
tanks and centrifuges is large enough to handle the increased capacity. The precipitation
thickener is sized using the same method as for the counter-current decantation. The
existing reactors, thickener, tanks and centrifuges have sufficient capacity to process the
increased feed. Table 3.14 gives a summary of the numbers and sizes of the equipment
used for the plant.

Table 3.14.a: Summary of leaching equipment
Specifications Description
Construction material Rubber lined stainless steel
Diameter per tank 10 m
Feed rate 554
Number of tanks 11
Residence time per tank 1 hour and 21 minutes
Tank alignment Series
Volume per tank 750 m
3



School of Chemical and Minerals Engineering
88
Chapter 3: Process development

Table 3.14.b: Summary of counter-current decantation equipment
Specifications Description
Construction material Rubber lined cement
Trains 2
Diameter per tank Thickener 1
a
: 60 m
Thickener 2
a
6
a
: 55 m
Thickener 1
b
: 50 m
Thickener 2
b
6
b
: 45 m
* With subscripts a and b being train 1 and train
2 respectively
Feed rate Train 1: 340.7 m
3
/hr
Train 2: 212.9 m
3
/hr
Number of stages 6 per train
Residence time per stage Thickener 1
a
: 33 hr and 10 min
Thickener 2
a
6
a
: 28 hr
Thickener 1
b
: 35 hr and 20 min
Thickener 2
b
6
b
: 28 min and 30 min
* With subscripts a and b being train 1 and train
2 respectively
Volume per tank Thickener 1
a
: 11 300 m
3

Thickener 2
a
6
a
: 9 500 m
3

Thickener 1
b
: 7 500 m
3
Thickener 2
b
6
b
: 6 000 m
3

* With subscripts a and b being train 1 and train
2 respectively


School of Chemical and Minerals Engineering
89
Chapter 3: Process development

Table 3.14.c: Summary of ion exchange equipment
Specifications Description
Adsorption time 20 hr
Construction material Glass reinforced plastic
Diameter per tank
Adsorption columns: 3 m
Regeneration columns: 3 m
Elution bed volumes 7
Elution time 7 hr and 45 min
Feed rate 570 m
3
/hr
Number of tanks
Adsorption columns: 6
Regeneration columns: 6
Residence time per tank
Adsorption: 2 min and 14 sec
Elution: 20 min
Volume per tank 141 m
3




School of Chemical and Minerals Engineering
90
Chapter 3: Process development

Table 3.14.d: Summary of precipitation equipment
Specifications Description
Reactor
Construction material Stainless steel 316L
Diameter per tank 1.5 m
Feed rate 7.8 m
3
/nr
Number of tanks 2
Residence time per tank 12 min and 20 sec
Volume per tank 2.667 m
3

Thickener
Construction material Stainless steel 316L
Diameter per tank 15 m
Feed rate 7.8 m
3
/hr
Number of stages 1
Residence time per stage 78 hr
Volume per tank 610 m
3



3.8. Mass and energy balance

The mass and energy balance over the entire plant is required in order to validate the
technical feasibility of the process. The mass and energy balance are in some cases
interdependent, since the energy flow is calculated from the mass flow and the mass
conversion rate are dependent on the energy of the system.





School of Chemical and Minerals Engineering
91
Chapter 3: Process development
3.8.1. Mass balance

The mass balance for this process is done to account for all the materials in the process.
The law of conservation states that mass can not be created nor destroyed. An overall mass
balance is given in this section while more detailed mass balances are given in Appendix A.
Table 3.15 is the result obtained from the overall mass balance. This overall mass balance
does not consider internal recycles for the resin and solvent.

Table 3.15: Overall mass balance
IN OUT
Stream kg/hr Stream kg/hr
Ore feed 880 000 Solid slurry to gold plant 992 487.8
HNO
3
feed 2 497.44 Removed silicone from resin 63.781
H
2
SO
4
feed 15 320.9 Spent demin. water 5 610
Wash water make-up 0.998 Precipitation recycle bleed 401.49
Slaked lime 67 584.13 Spend liquid from centrifuges 1 297.82
Eluant 27 507.32 ADU product 374.012
Regeneration feed 25.352
Demin. water 5 600
Centrifuge wash water 1 410.02
Ammonia 85.73
Total 1 000 036 Total 1 000 239

The total amount of materials that entered the process adds up to 1 000 036 kg/hr while the
materials leaving the process is 1 000 239 kg/hr. Comparing these two values, an error of
0.02% is found. This error occurs due to the simulation program used to simulate the
extraction process, ASPENTech

, which uses tolerances to simulate the mass balance and

therefore this error is acceptable.

3.8.2. Energy balance

The energy balance gives an indication of the thermodynamical feasibility of the process
indicating energy requirements, energy generation and energy losses with detailed energy
balances supplied in Appendix A. In this section the energy balance is done with the
following assumptions:

School of Chemical and Minerals Engineering
92
Chapter 3: Process development

Heat capacity of solids: 0.82 kJ/kg/K
Kinetic energy is negligible
Heat of reaction and heat of dissolution are negligible

There are only two processes on the plant that will require thermal energy; the leaching
pachucas and the precipitation reactors. Energy loss to the environment generally occurs at
the leaching pachucas, and thickeners at leaching and precipitation. The reactions in all the
processes take place in small amounts due to diluted mixtures and therefore are not taken
into account for energy generation or consumption. However the reactions in the
neutralisation process may have a significant effect on the temperature of the mixture (see
Appendix A).

The energy required for this plant operations are 10.8 MW at the leaching section and 0.75
MW at the precipitation section. From calculations it is found that the energy generation
caused by reactions in the neutralisation and precipitation processes has no significant effect
on the temperature of the product streams. It is concluded that it is impractical to implement
heat integration systems on this plant.

3.9. Process Flow Diagrams and process description

To conclude the process design procedure, a detailed process desctription is provided. The
project entails the expansion of the South Uranium Plant (SUP) of AngloGold Ashanti to
process an increased ore feed from 240 000 to 360 000 ton per month. The operating time
for this plant is planned at 8 150 hours per year and the production rate for ADU is
calculated to be 1 068 ton per annum. This upgrade of the SUP is done by evaluating the
existing equipment and if necessary, design new equipment with the capacity to process the
new feed. The following process units are defined and illustrated in detail on the process
flow diagrams (PFDs) and discussed below.

Unit 1 (U01): Leaching, counter-current decantation and neutralisation
Unit 2 (U02): Ion exchange
Unit 3 (U03): Solvent extraction
Unit 4 (U04): Precipitation



School of Chemical and Minerals Engineering
93
Chapter 3: Process development









PFD unit 1: Leaching and CCD

School of Chemical and Minerals Engineering
94
Chapter 3: Process development










PFD unit 2: Ion exchange

School of Chemical and Minerals Engineering
95
Chapter 3: Process development










PFD unit 3: Solvent extraction

School of Chemical and Minerals Engineering
96
Chapter 3: Process development










PFD unit 4: Precipitation

School of Chemical and Minerals Engineering
97
Chapter 3: Process development
3.9.1. Unit 1 (U01): Leaching, CCD, and neutralisation

The initial stage in the extraction of uranium is the leaching stage followed by counter-
current decantation (CCD) of the leach liquor. The ore feed is received from the crushing
and milling circuit with an added 330 000 kg/hr of water to give the ore feed a SG of 1.6 and
total mass flow of 880 000 kg/hr. The ore is fed to the first leaching tank in the series to
extract the uranium from the ore. The leach product is sent to the CCD to recover the
uranium containing liquids and to settle out the solids containing gold which is sent back to
the gold plant after neutralization. The liquid overflow from CCD contains the valuable
uranium and is sent to the ion exchange unit for further processing. Unit 1 can be divided
into three sections which are discussed below.

Leaching
CCD
Neutralization

Leaching (U01-P01 to U01-P13)

The leaching process is done in a series of 11 air-agitated and steam heated leaching
pachucas (U01-P01 to U01-P11) where an oxidant and lixiviant is added to liberate the
uranium from the ore. The air agitation is done to ensure efficient mixing of the solution in
the pachucas and the steam is added to every fourth pachuca. The temperature of the first
pachuca (U01-P01) is raised to 30C, using steam addition. It is very important to ensure
that the temperature throughout the leaching section is controlled correctly to ensure
optimum leaching kinetics. The existing pachucas is 750 m
3
which provide an adequate
capacity to process the upgraded feed due to the change in oxidant used. A residence time
of 1.5 hours is achieved for each pachuca tank. The uranium conversion in the leaching
section is 96% for the mineral uraninite. This conversion is reached due to the faster
leaching kinetics of the nitric acid leaching system.

The oxidant used is nitric acid which oxidises the uranium ions from U
4+
to U
6+
. U
4+
is
insoluble in aqueous systems but U
6+
is soluble. An added benefit of the nitric acid is that
multiple reactions occur which create an auto-catalytic environment resulting in faster
kinetics and less oxidant needed. Sulphuric acid is used as lixiviant to control the pH of the
leach product at 2. Sulphuric acid is used to control the pH, rather than nitric acid, because
two hydronium ions are formed per molecule of sulphuric acid which is reacted. The ore is
School of Chemical and Minerals Engineering
98
Chapter 3: Process development
fed to either U01-P01 or U01-P02 using a splitter box, depending on the operatability of
UO1-P01.

It is important to control the nitrate ion concentration and temperature of the solutions in the
pachucas to ensure optimal process conditions and efficient leaching. The reagent flow to
the puchacas should be controlled by the amount of ore that is received to prevent abundant
used of expensive reagents when the ore feed. It is suggested that the level of agitation is
regulated to ensure efficient mixing of the solution in the pachucas to ensure adequate
leaching efficiency.

CCD (U01-TH01A to U01-TH06A & U01-TH01B to U01-TH06B)

After the liberation of the uranium in U01-P13 the leach product flows to the thickener
capacitance tank (U01-DM01). U01-DM01 has a volume of 750 m
3
and is also air agitated
to ensure that no settling occurs in this vessel. From U01-DM01 two streams are pumped to
separate thickener trains, Train A (U01-TH01A to U01-TH06A) and Train B (U01-TH01B to
U01-TH06B). Raked gravity thickeners are used to separate the solids, which are sent to
the gold plant, from the liquids which are sent for further processing at ion exchange. In
each train the first thickener (U01-TH01A and U01-TH01B) serves as a clarifier to ensure no
solids is sent to the ion exchange. The clarifiers for Train A and B are 60 and 55 meters in
diameter respectively while the rest of the thickeners in the trains are 50 and 45 meters
respectively.

The leach product (U01-S17) that is fed to Train A is 62 wt% of the entire stream while Train
B is 38 wt%. The thickeners are operated in a counter-current configuration as displayed in
Page 1 of 4 of the PFDs. The flocuant (Magnafloc 90L) is added to the thickeners to
promote faster settling of the solids to achieve effective separation. The floculant is
distributed between the five thickeners in each train, but no floculant is added to the
clarifying thickeners. The wash solution added to the thickeners is recycled from the
adsorption columns at ion exchange together with a potable water make-up stream. A wash
ratio of 1:1 is used to ensure that less than 0.01 wt% of the uranium is lost. The overflow of
U01-TH01A and U01-TH01B, known as the pregnant leach liquor, is sent to the ion
exchange unit (U02), while the underflow of U01-TH06A and U01-TH06B, containing the
gold, is pumped to the final pachuca (U01-P14) for neutralization of the slurry..

The efficiency of the thickeners depends on the residence time and amount of wash solution
added to the thickeners. The solid content of the pregnant leach liquor should not be more
School of Chemical and Minerals Engineering
99
Chapter 3: Process development
than the prescribed 50 ppm as the solids can poison the resin used in ion exchange. It is
suggested that the height of the rake inside the thickeners should also be controlled
because the thickness of the slurry might damage the equipment.

Neutralization (U01-P14)

The neutralization facility is needed to ensure that the pH of the slurry, which is sent to the
gold plant, is maintained at 10.5 by the addition of slaked lime. The neutralization is done in
an air-agitated pachuca to ensure efficient mixing and neutralization of the slurry. Not all of
the existing pachucas are used for the leaching process therefore the last existing pachuca
(U01-P14) is used for neutralization.

3.9.2. Unit 2 (U02): Ion exchange

The liquid product from U01 contains the valuable uranium ions and other impurities. The
ion exchange process is used to remove these impurities and increase the concentration of
uranium ions from 0.2 to 4 g U
3
O
8
per litre. In this process uranyl sulphate ions are
adsorbed onto resin, the loaded resin is washed to remove impurities and finally the uranyl
sulphate ions are stripped from the resin during elution. All of these processes, except for
the regeneration stage, are done in the same column, leaving the resin stationary inside the
column. The resin attrition in this system is reduced, since the resin is not moved frequently.
This unit can be divided into four steps listed below.

Adsorption
Wash
Elution
Regeneration

Adsorption (U02-AC01 to U02-AC05)

The pregnant liquor stream (U02-PS01) entering U02 is sent to the first adsorption column
(U02-AC01) from where it is pumped through the rest of the columns (U02-AC02 to U02-
AC05) in the adsorption train. Each of these columns has a capacity of 145 m
3
. Due to the
selectivitys of the sulphate ions and uranyl- and iron sulphate complexes, these complexes
are mainly absorbed onto the resin (Ambersep TM400).

School of Chemical and Minerals Engineering
100
Chapter 3: Process development
The selectivitys allow for 60% of the resin capacity to be occupied by the uranyl sulphate
complexes, 10% is occupied by iron sulphate complexes and the remaining 30% by
sulphates. The adsorption front as discussed in Section 2.5.2 moves downward through the
leading adsorption column. When uranyl sulphate complex content of the exit stream of a
adsorption column is equal to that of the feed stream, the resin in the column is saturated
with uranium and the column is isolated from the adsorption train. The stream exiting the
adsorption column train is the barren liquor, which is recycled back to CCD as wash solution.

The use of the fixed-bed ion exchange process means that smaller columns is necessary to
achieve the desired adsorption, therefore five of the existing ion exchange columns will be
used. These columns will be modified by covering the columns with a lid to allow the back-
wash process to be done in these columns. It is calculated that each adsorption will initially
take approximately 30 hours to reach saturation.

Wash (U02-AC01 to U02-AC05)

Once the column has been isolated, the wash process is started. In the wash process any
impurities and unwanted solids are removed by fluidising the resin with potable water which
is fed from the bottom of the column. Care must be taken to avoid blowing out the resin with
the wash water. The wash water is recycled back to the pregnant leach liquor feed tank to
recover any pregnant leach liquor and therefore uranium from the saturated resin.

Elution (U02-AC01 to U02-AC05)

The elution process commences after the wash process is completed. This process is
necessary for desorption of the uranyl sulphate complexes from the resin into the aqueous
phase. The eluant used for desorption is sulphuric acid which is diluted in U02-DM01. The
eluant is pumped from the top of the column, exiting as the uranium rich eluate. This eluate
is sent to solvent extraction for further processing. The eluate flow required is determined by
the volume required to reach the desired uranyl sulphate concentration. It is estimated that
eleven bed volumes of eluate is used to elute each resin column with a residence time of
approximately 15 minutes.





School of Chemical and Minerals Engineering
101
Chapter 3: Process development
Regeneration (U02-RC01)

The leach liquor contains certain ions that forms a strong bond to the resin and are not
removed during the wash stage. These ions are referred to as poisons but can be removed
by regenerating the resin as discussed in Section 2.5.2. Regeneration is done once every
two days by removing the one column from the train, draining the resin out of the column
and pumping the resin to the regeneration column (U02-RC01).

The regeneration step is complex and consists of several internal steps to protect the resin
from the damage caused by pH-shock. The resin from the acidic system is first contacted
with potable water, then with diluted caustic solution and finally caustic solution. This is
done to ensure a gradual pH increase and prevent pH-shock. Before the resin is returned to
the adsorption columns, the pH is gradually decreased by reversing the above procedure.
Some permanent poisons are not removed during regeneration and the resin capacity
decreases over long periods. For this reason the resin inventory is replace with new resin
every three years as suggested by Merritt (1971).

3.9.3. Unit 3 (U03): Solvent extraction

The solvent extraction unit (U03) is included in the extraction process to increase the
concentration of uranyl sulphate ions and remove the iron sulphate complexes which allows
for more efficient precipitation. Solvent extraction is a form of liquid-liquid separation in
which an organic and aqueous phase is used. There are four steps in the solvent extraction
process which are each discussed below.

Extraction
Scrubbing
Stripping
Regeneration







School of Chemical and Minerals Engineering
102
Chapter 3: Process development
Extraction (U03-MSE01 to U03-MSE03 & U03-AS01)

The eluate received from U02 is stored in a 600 m
3
tank (U03-ST01), from where it can be
continuously fed to the first mixer-settler (U03-MSE01) in the extraction section. The
aqueous phase is pumped to the mixer from below from where the impeller pumps the
solution further. The mixer-settlers are operated in a counter-current configuration with the
organic phase fed to the last mixer-settler (U03-MSE03) in a ratio of 1.1:1 (V
org
:V
aq
). The
after-settler (U03-AS01) is used to reduce solvent loss, thus no organic is fed to the mixer-
settlers.

The uranyl sulphate complexes are extracted from the aqueous phase into the organic
phase. In addition to the uranyl sulphate extraction, iron sulphate complexes are entrained
in the solvent as an impurity. The raffinate (barren aqueous solution) from U03-AS01 is
recycled to leaching since it contains a large amount of sulphate and hydronium ions.

Scrubbing (U03-MSS01 to U03-MSS03)

The scrubbing section is necessary to remove any impurities contained in the solvent. The
scrubbing is done with demineralised water in a counter-current configuration. The organic
solvent from U03-MSE03 is fed to U03-MSS01 while the demineralised water is fed to
U03-MSS03 from a demineralised water storage tank (U03-ST04). In this section the ion
sulphate complexes entrained in the solvent, is washed out with demineralised water.

Stripping (U03-MST01 to U03-MST04 & U03-AS01)

The uranyl sulphate complexes are desorbed from the organic phase back to the aqueous
phase in the stripping section using ammonia sulphate which is recycled from the
precipitation unit (U04). If the pH in this section increases, the product will begin to
precipitate therefore the pH must be controlled between 2.5 and 5 by the addition of caustic
soda. The ammonia sulphate entering the stripping section has a high pH which makes it
difficult to control the pH in U03-MST01. Therefore the pH is only controlled in U03-MST02
to U02-MST04 and U03-MST01 is an extra mixer-settler used to increase the efficiency of
the stripping section. The after-settler (U03-AS02) is used to reduce solvent loss, thus no
organic is fed to the mixer-settlers. The aqueous phase, containing the uranyl sulphate
complexes, exiting U03-AS02 is called the OK liquor. The OK-liquor is sent to the
precipitation for the final extraction.

School of Chemical and Minerals Engineering
103
Chapter 3: Process development
Regeneration (U03-MSR01)

The solvent exiting U03-MST01 proceeds to U03-MSR01 where it is regenerated to ensure
maximum efficiency of the solvent. Sodium carbonate and caustic soda is diluted into a
regeneration solution in U03-MU02 using demineralised water. The regenerated solvent is
recycled back to the organic storage tank (U03-ST03). The regeneration solution is re-used
until it is exhausted. When the regeneration solution is spent, a valve in stream U03-S12 is
opened into a storage tank (U03-ST08) from where it is sent to the waste treatment facility.

3.9.4. Unit 4 (U04): Precipitation

The final unit in the uranium extraction process is the precipitation process unit which is
necessary to form a solid product with the desired specifications. The precipitation of ADU is
achieved by adding ammonia gas to the OK liquor containing uranyl sulphate ions. The final
ADU product is sent to NUFCOR for further processing. This unit can be divided into three
sections which are discussed below.

Precipitation
Settling
Washing and drying

Precipitation (U04-PT01 & U04-PT02)

The precipitation of ADU is done in two vessels, the first in which nucleation occurs and the
second in which crystal growth occurs. Crystal growth is important because no floculant is
used for the settling of the solids therefore the ADU particles should be substantially large to
settle naturally. To ensure sufficient formation of particles two vessels are used.

The OK liquor received from solvent extraction is heated to 30C in a heat exchanger (U04-
HX01) using steam at 175C and 12 bar. The heated OK liquor flows into the first
precipitation tank (U04-PT01) where ammonia is mixed with compressed air and added to
U04-PT01 from below. From U04-PT01 the entire slurry flows to the second precipitation
tank (U04-PT02) where ammonia and compressed air are added again. The ammonia is
added in combination with the compressed air to ensure even distribution of the ammonia in
the vessel and helps with the agitation of the solution. It is important to provide sufficient
agitation in the precipitation vessels to ensure that settlement do not occur in these vessels.
Therefore the vessels are also mechanically agitated. The vessels that are available on the
School of Chemical and Minerals Engineering
104
Chapter 3: Process development
existing plant, is large enough to process the increased load and therefore the existing 3.5
m
3
tanks will be used.

The reactor temperature should be controlled at 30C and the pH of the slurry should be
controlled at 7.5 to ensure that the optimal precipitation occurs. The OK liquor feed flow rate
and liquid levels in the precipitation tanks should be controlled to ensure adequate residence
time for the ADU particles to form. It is also necessary to control the ammonia gas and
compressed air entering the reactors to ensure the correct ratio of ammonia to air is fed to
the reactors.

Settling (U04-TH01)

The slurry containing the precipitated ADU particles is sent to the ADU thickener (U04-TH01)
to allow the solids to settle in order to separate the ADU cake from the liquid containing
impurities. As mentioned, no floculant is added to the thickener and only one thickener is
sufficient for the necessary separation. The necessary thickener volume is calculated to
ensure sufficient residence time. The calculated value showed that the existing thickener
with a volume of 710 m
3
can handle the increased feed.

The overflow of U04-TH01, containing a high concentration ammonium sulphate, flows into a
storage tank (U04-ST01) with a capacity of 35 m
3
. From U04-ST01 the liquid is pumped
back to solvent extraction where it is used as stripping agent to strip the uranium from the
organic phase. It is necessary to control liquid level in U04-TH01 to ensure enough
residence time for the solids to settle and to produce an effective overflow. The underflow of
U04-TH01 flows to the first stage mixing tank (U04-MT01) and from U04-MT01 the slurry is
pumped to two centrifuges in parallel. The centrifuges are used to wash and dry the ADU
product.

Washing and drying (U04-CG01 to U04-CG03)

The centrifuges are used to wash off the impurities such as SO
4
2-
and to continuously
dewater the ADU slurry. The current capacity of the existing centrifuges is 100 150 kg
ADU/hr which is sufficient to process the increased capacity.

The slurry from U04-MT01 is pumped to the first stage centrifuge which consists of two
centrifuges in parallel (U04-CG01 and U04-CG02) and demineralized water is added to both
centrifuges in the form of a spray to wash the solids in the slurry. The wash solution from
School of Chemical and Minerals Engineering
105
Chapter 3: Process development
the centrifuges flows to a storage tank (U04-ST02) and is eventually pumped back to U04-
TH01 to ensure that no uranium is lost in the wash solution. A part of this recycle stream is
bled to the waste water treatment facility to prevent built-up in the system. The solids from
U04-CG01 and U04-CG02 are pumped to the second stage mixing tank (U04-MT02) and
from there it is pumped to the second stage centrifuge U04-CG03. In the second stage
centrifuge no water is added and this centrifuge is only used for the dewatering of the
product. The liquid from U04-CG03 continues to a storage tank (U04-ST03) and is pumped
to the waste water treatment facility. The solids coming from U04-CG03 is pumped to the
final storage tank (U04-ST04) where the final product is continuously stirred and air agitated
to ensure that settlement does not occur in the storage tank.

The flow of the ADU slurry to all the centrifuges should be controlled to ensure optimal
operating of the equipment. It is also necessary to regulate the wash water added to U04-
CG01 and U04-CG02 to achieve effective washing of the solids in the slurry. The final
product, stored in U04-ST04 with a capacity of 100 m
3
, has a solid content of 35% and this
product is sent to NUFCOR which process the ADU to produce U
3
O
8
.

3.10. Innovations

The design procedure proposed by Douglas (1988) creates a platform for the development
of creative solutions for specific design problems. All the proposed solutions are
economically evaluated to ensure that profitability is not compromised and the impact on the
environment and working conditions are also studied. Table 3.16 shows the innovative
steps taken in the process development.


School of Chemical and Minerals Engineering
106
Chapter 3: Process development
Table 3.16: Innovations
Innovations Advantages
Use of nitric acid instead of MnO
2
for
leaching oxidant.
- Nitric acid provides faster kinetics
- Less nitric acid is used due to the
auto-catalytic effect which reduces
the raw material costs.
Use of existing leaching pachucas. Due to
the faster kinetics, the increased feed can
still be processed in the existing columns.
- No additional pachucas needs to be
constructed which means less capital
is needed for the expansion project.
Use of existing leaching pachuca for
neutralization because not all the pachucas
are used for the new leaching process.
- No additional construction is needed
for the neutralization plant.
- All the existing puchacas are used
therefore decommissioning is not
necessary.
Construction of additional CCD train instead
of decommissioning of existing train. The
existing CCD equipment will not be able to
handle the additional feed, therefore an
additional CCD train is built and the feed
stream is divided between the two trains.
- The existing CCD equipment is still
used therefore no decommissioning
is necessary.
- This setup will be able to process
more feed than required for this
project which allows further increase
in ore processing if necessary.
Use of existing adsorption ion exchange
columns for fixed-bed ion exchange.
- No additional ion exchange columns
needs to be built, therefore less
capital is needed.
Use of after-settlers in the solvent extraction
process.
- The after-settlers provide additional
residence time to allow phase
separation and this reduces solvent
loss. The solvent used is extremely
expensive thus it is critical to
minimize solvent loss.

School of Chemical and Minerals Engineering
107
Chapter 3: Process development
Use of four stripping mixer-settlers in the
solvent extraction process, instead of three.
- The pH control is simplified by adding
an additional mixer-settler. The pH
control is difficult because the recycle
stream from the precipitation is too
high. Therefore the first settler is
used to allow the solution pH to reach
a steady state which simplifies the pH
control in the remaining settlers.
Use of bio-organisms to remove the nitrate
ions in the waste water.
- Removes the excess nitrate ions in
the waste water which is dangerous
to the environment.
Construction is done over a three-year
period of which the first year the existing
plant will still operate at normal capacity
while the additional equipment is build. The
second year the plant will shut-down for
construction and the final year the plant will
be operated at 50% of the new capacity.
- Several of the new equipment can be
constructed while the old plant is still
operating, therefore production is not
lost.
- Because various existing equipment
is used and only a few facilities needs
to be decommissioned, the
construction period is short, therefore
production loss is minimized.


School of Chemical and Minerals Engineering
108
Chapter 4: Detail design

Chapter 4: Detail design

The purpose of the solvent extraction (liquid-liquid separation) process is to purify and
concentrate the uranyl sulphate ions. The principle of the solvent extraction process is
based upon immiscible properties of aqueous and organic liquids. The existing solvent
extraction section on the South Uranium Plant (SUP) is out-dated and should be upgraded.

The detail design for the solvent extraction process includes the following considerations:

Choice or type of system
The kinetics and thermodynamics of the system.
A detail chemical design.
The start-up and shut-down procedures.
Optimization of the process which includes sensitivity analysis.
Mechanical aspects that should be considered for the process.

The solvent extraction process consists of four separate stages which include: extraction,
scrubbing, stripping, and regeneration. In the extraction stage the uranyl sulphate ions are
removed from the aqueous phase into the organic phase. In the scrubbing stage the organic
phase is cleansed from impurities and in the stripping stage the uranyl sulphate ions are
transferred from the organic phase to aqueous phase. During the stripping stage the
concentration of the uranyl sulphate ions is increased. The regeneration stage removes any
impurities entrained in the organic solvent from where the solvent is recycled to the
extraction stage.

4.1. Choice of system type

The different process alternatives for the solvent extraction section are discussed in Section
2.5.3.1. The first choice is the type of extractant that should be used for the system,
followed by the composition of the solvent. Alamine

336 is used as extractant for this

system which has been used for the past 20 years, and is a tried and tested extractant for
the SUP process. The current solvent composition (volume %) is also used which contains:
93% kerosene, 5% Alamine

336, and 2% isodecanol.

School of Chemical and Minerals Engineering
109
Chapter 4: Detail design

There is a choice which should be made for the equipment between mixer-settlers and
pulsed columns. Mixer-settler equipment, shown in Figure 4.1, was chosen above pulsed
columns to ensure easier pH control for the stripping stages of the process (Law & Tod: 4).
The current operators are experienced in the operation of mixer-settler equipment therefore
operating optimization will need less time and fewer resources will have to be spent on
training. If pulsed columns is used the operating personnel will have to be engineers,
making the labour cost extremely high. In the case of unforeseen plant disturbances all the
solvent in the pulsed columns will be lost which is economically unfavourable as the solvent
is very expensive (Watson, 2009).


Figure 4.1: Representation of mixer-settler equipment

The operating choices for the flow of the mixer-settler equipment are cross-current and
counter-current flow configuration of the organic and aqueous phases. Counter-current flow
is used for this system due to higher overall efficiency. Counter-current flow allows for the
best compromise between high recovery and good separation for the process.


School of Chemical and Minerals Engineering
110
Chapter 4: Detail design
4.2. Kinetics and thermodynamics

The mixer-settler equipment, as seen in Figure 4.1, consists of two separate compartments,
i.e. mixing and settling. In both these compartments the kinetic and thermodynamic
parameters are vital to the design of the mixer-settler equipment. The kinetics for the
extraction of uranium from the aqueous phase to the solvent phase containing Alamine

336
is extremely fast, therefore equilibrium parameters are more important for the mixing
compartments (Mackenzie, 1997: 11).

To describe the equilibrium parameters for the extraction of uranium, loading isotherms are
needed. Loading isotherms are described by the equilibrium distribution of uranium in the
aqueous and organic solvent phases. It is reported by Stnner & Wiesner (1982) that the
loading isotherm for the extraction stage of solvent extraction is modelled by Equation 4-1.

| |
| |
| | | |

=
( + + +

aq
1
org
2 3 2 4 4
aq aq
aq
U
U k
U k Cl k H SO k
(4-1)


All the concentrations are in g/L and the concentration of uranium is in U
3
O
8
. The constants
k
1
to k
4
are fitted parameters with k
1
being the maximum uranium loading capacity of the
organic phase. Modelling using Equation 4-1 agrees with data supplied by the Metallurgy
Division of the South African Atomic Energy Board (Stnner & Wiesner, 1982: 97).

The loading isotherm for the stripping stages can be found in Morais & Gomiero (2005) and
is for chloride free uranium ores. This loading isotherm is displayed in Figure 4.2.



School of Chemical and Minerals Engineering
111
Chapter 4: Detail design














Figure 4.2: Stripping loading isotherm with (NH
4
)
2
SO
4


Using the above mentioned equilibrium data, it is possible to design the required number of
stages for effective separation. Due to the fast kinetics for the uranium transfer between the
phases a short residence time in the mixing compartments is required.

The main objective of the settlers is to achieve efficient separation between the aqueous and
organic phases. Stnner & Wiesner (1982) reported that the settling kinetics can be
modelled by two consecutive steps. The first step is a volume-controlled reaction which
describes the coalescence of small droplets to large droplets. In the second step which is
area-controlled, the large droplets enter its original phase. The kinetics for these steps is
given in Equations 4-2 and 4-3 (Stnner & Wiesner, 1982: 98).

=
1 1
dV V
k.F.H
dt V
(4-2)

=
2 2
dV V
c.F
dt V
(4-3)




School of Chemical and Minerals Engineering
112
Chapter 4: Detail design
Equation 4-2 represents the kinetics of the first step where V
1
is the volume of dispersion
contains small droplets, F is the settling area, H is thickness of the dispersion band and V is
the total volume of dispersion. Equation 4-3 represents the kinetics of the second step
where V
2
is the volume of dispersion containing large droplets and c is the velocity of these
droplets (Stnner & Wiesner, 1982: 98).

4.3. Detail chemical design

The following procedure is followed to do a detail chemical design on the solvent extraction
unit. This unit consists of four sections as mentioned before. Each of these stages consists
of a number of mixer-settler units connected using counter-current flow. The first step in the
design procedure is to determine the amount of mixer-settler units in each section. The next
step is to size the equipment, starting with the mixers. After the mixers are sized, the
settlers, pumps and tanks are sized. The assumptions made for the chemical design are the
following:

Uranium recovery of the extraction section is 98%.
Extraction and stripping stage efficiency of 98%.
Mixer agitators work in turbulent flow regime.
Height of the dispersion layer is directly proportional to the volumetric flow of the
droplets.
Initial volumetric fraction of small droplets in the dispersion is 100%.

4.3.1 Number of stages

Using the extraction and stripping loading isotherms together with a basic mass balance, the
McCabe-Thiele method is applied to determine the number of stages. Accurate isotherms
are required for this step, which is stated in Section 4.2 (see Appendix C).

The isotherm for the extraction section is obtained from Equation 4-1 and the constants were
manipulated to obtain a similar trend. The given constants, reported by Stnner & Wiesner
(1982), and manipulated constants for Equation 4-1 is shown in Table 4.1.


School of Chemical and Minerals Engineering
113
Chapter 4: Detail design
Table 4.1: Extraction loading isotherm constants
Constant Value reported Manipulated value
k1 6.772 6.772
k2 0.77218 0.772
k3 0.02864 0.002864
k4 2.51 0.5

The validation of the manipulated variables is shown in Appendix C. The design method
using McCabe-Thiele is discussed in detail in Appendix C. From Figure C.4 and C.5 the
exiting uranium loading is obtained and shown in Table 4.2.

Table 4.2: Results from McCabe-Thiele method
Stage Organic loading g U
3
O
8
/L Aqueous loading g U
3
O
8
/L
Extraction
1 3.5918 0.95
2 1.05 0.15
3 0.3 0.07
Stripping
1 11.95 1.4
2 4.15 0.7
3 1.7 0.28

The organic phase fed to the extraction section has uranium loading of 0.25 g U
3
O
8
/L and
exits at 3.5918 g U
3
O
8
/L, while the oraganic phase fed to the stripping section has uranium
loading of 3.5918 g U
3
O
8
/L and exits at 0.28 g U
3
O
8
/L. Next the mixers are designed to
obtain the desired efficiency.

4.3.2 Mixer

The design considerations for the mixer are residence time, impeller rotary speed, flow
regime, impeller tip speed, produced head and mixer box dimensions. The residence time
for all mixers was assumed from literature to be 2 minutes (Mackenzie, 1997: 11). The
results obtain using the design philosophy in Appendix C is shown in Table 4.3.


School of Chemical and Minerals Engineering
114
Chapter 4: Detail design
Table 4.3: Mixer box design results
Specifications Value
Hydraulic efficiency 0.23
Impeller diameter 1.1 m
Impeller rotary speed 70 rpm
Impeller tip speed 4 m/s
Mixer box diameter 1.4 m
Mixer box height 1.7 m
Mixer box liquid volume 2 m
3

N
h
2.46
N
p
0.5493
N
q
0.01048
Produced head 2 m
Residence time 2 min
Reynold number 1.7 x 10
6

Total volumetric flow (10% over design) 59.6

From Table 4.3 it is seen that turbulent flow is achieved which allows for adequate mixing
between the different phases. The produced head is greater than the height of the mixer
box, which will ensure continuous flow into the settler vessel. The residence time is 2
minutes which ensures sufficient mixing, and will therefore allow for high stage efficiency.

4.3.3 Settler

The settling kinetics used in this design is derived from Equation 4.2 and 4.3 and shown in
Appendix C. The kinetics were solved using Polymath

and sensitivity analyses were done

on the dispersion layer velocity, width of the setter vessels and construction cost. The
design constraints are identified as the following:

Height of the vessel should not exceed 1.5 m, to reduce the mixer head required.
The width of the vessel should not exceed 2 m, to ensure adequate dispersion
distribution into the vessel.
Final dispersion layer height should not exceed 1 mm, to reduce the solvent loss.
A realistic residence time should be achieved, to ease operability.

School of Chemical and Minerals Engineering
115
Chapter 4: Detail design
Using the settling kinetics and above mentioned constraints, the settler vessel dimensions
are sized for the extraction and stripping sections. The designed dimensions are shown in
Table 4.4.

Table 4.4: Designed dimensions for the settler vessels
Dimension Value
Extraction
Height 1.5 m
Width 2 m
Length 5.5 m
Residence time 12.5 min
Initial dispersion height 1.1 m

The sizes of the vessels for the extraction and stripping sections differ due to the difference
in aqueous to organic feed ratio. The results in a lower flow for the stripping section,
therefore, smaller settler vessels are required for the stripping section. However, since the
material of construction is vacuum infused fibreglass, a mould is used to construct the
settling vessels. The construction cost for this manufacturing method consist of the price for
each mould and construction materials used, therefore, using one standard vessel size will
optimize construction cost. Due to the smaller vessels required for the stripping section, it is
decided that the size for both the stripping and scrubbing settler vessels are exactly the
same as for the extraction settling vessels.

4.3.3 Pipe sizing

The pipes are sized to achieve a velocity flow less than 1 m/s to reduce the probability of a
fire hazard caused by statistic discharge. The design methodology used to sizes the pipes
are shown in Appendix C and the results given in Table 4.5.



School of Chemical and Minerals Engineering
116
Chapter 4: Detail design
Table 4.5: Solvent extraction pipe sizes

Over-designed flow Pipe inner diameter Flow velocity

(m
3
/hr) (m) (m/s)
Eluate flow 28.4 0.115 0.76
Organic flow 31.2 0.115 0.84
Stripping flow 8.7 0.065 0.73
Scrubbing flow 3.1 0.04 0.69

All the chosen pipe inner diameters are standard pipe sizes (converted from inches) to avoid
the cost for custom made pipes.

4.3.4. Pump sizing

The design of the solvent extraction section requires 11 pumps. The pumps are required to
transport the liquid from the respective storage tanks to the mixers. The design
methodology used for the pumps are discussed in Appendix C and the results obtained for
the pump sizing is shown in Table 4.6.

Table 4.6: Pump sizing results
Pump equipment name Head required (m) Liquid flow (L/s)
U03-PM01 4 7.2
U03-PM02 17 6.8
U03-PM03 4 7.2
U03-PM04 3.5 7.9
U03-PM05 4.5 1.6
U03-PM06 3.5 1.6
U03-PM07 4.5 2.2
U03-PM08 2.5 2.2
U03-PM09 5 1.6
U03-PM10 4.5 1.6
U03-PM11 6.5 1.6

Once a pump supplier is contracted, the pump curves are used to determine which pumps to
use. When choosing a pump, using the pump curves, it is important to use a pump with high
efficiency at the desired head and liquid flow. It is also important to calculate the NPHS (net
positive head suction) required for the pump and compare it with the achieved PHS (positive
head suction) for the system.
School of Chemical and Minerals Engineering
117
Chapter 4: Detail design
In case of a pump malfunction or maintenance on the pump, it is important to design a back-
up pump system. The system should always be on standby to ensure continuous operation
in case of pump failure. The configuration of the back-up pump (used for all the pumps) is
shown in Figure 4.3.


Figure 4.3: Back-up pump configuration

A control valve is placed before and after each pump configuration to open and close liquid
supply as needed. Routine maintenance needs to be done weekly, which includes oil
change and re-greasing of the pump motors. Weekly inspections are done to check the
gaskets for any leaks, while vibration and temperature tests must be done semi-annually.
Due to maintenance the pumps are alternated each week to ensure continuous operation of
the plant.

4.3.5. Control valve sizing

For the preliminary feasibility study it is important to determine the pressure drop (P) across
the valve, volumetric flow (Q) and the valve coefficient (CV). It is also important to determine
the inherent valve characteristics required to ease upgrade and controllability of the plant.
The inherent valve characteristics chosen for all the valves are equal percentage valve
characteristics. The equal percentage characteristics is chosen due to the forgive nature of
the valve. If the pumps are incorrectly design an equal percentage valve will show better
valve characteristics which will not be seen for linear or quick opening valves. Due to this
fact, the equal percentage valve is used when upgrading a system (Svrek et al, 2006:34).


School of Chemical and Minerals Engineering
118
Chapter 4: Detail design
The volumetric flow is calculated in the mass balance and is therefore a known variable.
The pressure drop is calculated using Equation C-22 and that leaves the valve coefficient
which is calculated using Equation C-23. The resulting pressure drop and valve coefficients
are shown in Table 4.7.

Table 4.7: The calculated pressure drop and valve coefficients

P
CV
(kPa) CV (m
3
/hr.bar
0.5
) CV (gpm/psi
0.5
)
U03-FCE01 17.522 7.16687 8.37251
U03-FCE02 19.4698 5.85162 6.836
U03-FCE03 19.4698 5.85162 6.836
U03-FCE04 17.522 7.16687 8.37251
U03-FCE05 19.222 5.85504 6.84
U03-FCE06 19.6822 1.2803 1.49568
U03-FCE07 19.672 2.25463 2.63392
U03-FCE08 18.772 2.30805 2.69632
U03-FCE09 18.772 1.15402 1.34816
U03-FCE10 19.6757 1.77874 2.07797
U03-FCE11 17.522 7.16687 8.37251
U03-FCE12 19.672 0.45093 0.52678
U03-FCE13 19.672 0.45093 0.52678
U03-FCE14 19.672 0.45093 0.52678
U03-FCE15 18.7257 1.8233 2.13002
U03-FCE16 17.522 7.16687 8.37251
U03-FCE17 17.522 1.19448 1.39542
U03-FCE18 17.522 2.38896 2.79084
U03-FCE19 19.672 6.40316 7.48032
U03-FCE20 17.522 6.1325 7.16414

To chose a valve once a supplier is contracted, the calculated valve coefficient shown in
Table 4.7 is used. It is also very important to size the valve according to the pipe diameter to
ensure a small additional pressure drop which will lead to superior valve characteristics.


School of Chemical and Minerals Engineering
119
Chapter 4: Detail design
4.4. Start-up and shut-down procedures

Once a solvent extraction section is designed and built, it is important to have a proper
start-up procedure. During the start-up procedure for the commissioning period, all process
equipment and instrumentation are tested to ensure the desired performance. It is also
important to have a shut-down procedure in case of an emergency or upstream or
downstream process units. The start-up procedure after such a shut-down is less intense
and time consuming, since most of the equipment testing was done during the
commissioning phase.

4.4.1. Commissioning of process unit

The commissioning of the solvent extraction section can be categorised into several steps
during which specific procedures are followed to achieve successful operation of the process
unit. The different commissioning steps are given chronologically below.

Dry inspection

When the solvent extraction section is constructed, all process instrumentation is manually
inspected for defects and other problems. A list of the more important inspections is given
below.

Valves
Connections at all flanges.
Clear all obstructive objects from mixers and settlers.
Ensure all pH probes are installed.
Ensure all conductivity meters are installed.
Ensure all flow meters are installed.
Electrical power to all equipment and instrumentation.
Fire extinguisher equipment.

The dry inspection is a tedious step in commissioning of a process unit but is of utmost
importance to prevent equipment damage at start-up. When these inspections are done the
process unit is ready to receive liquid feed.


School of Chemical and Minerals Engineering
120
Chapter 4: Detail design
Wet inspection

In this step the solvent extraction section is flooded with demineralised water. It is important
to make sure that all the valves are in the open position before the wet inspection is started,
to avoid high differential damage to the equipment due to water hammering. It is necessary
to make sure that the control system is not active during this step. When all the equipment is
at maximum capacity several inspections are carried out.

Flow direction of the pumps.
Rotational direction of the mixer impellers.
Pipe and vessel leaks.
Operation of the control equipment.
Accuracy of the pH probes.
Accuracy of flow meters.
Operation of the conductivity meter.
Effective working of weirs in the mixers and settlers.

It is also important to test the control system for the process unit. Once these inspections
are complete, the solvent extraction section is ready to receive the organic feed.

Preparation for operation

When the solvent extraction section is flooded with water and all required inspection are
completed to satisfaction, the organic solvent is allowed onto the section. The organic
solvent is introduced into the system through its designed flow. Since the density of the
organic solvent is lower than that of the water still being fed to the process unit, the top
sections of the settler vessels will be filled with organic solvent. Approximately 66 m
3
of
solvent is required before the settler vessels are operated at the designed levels.

From here the control system is activated to achieve the correct flow rates and to start the
pH control on the stripping section. When everything is in place, the water in the solvent
extraction section is replaced with the specific streams, first the stripping agent for the
stripping section and finally the eluate in the extraction section to introduce uranium into the
system. As soon as the first uranium is produced in the OK-liquor the solvent extraction
process unit is signed over to the operating personnel who will further optimise it.

School of Chemical and Minerals Engineering
121
Chapter 4: Detail design
4.4.2. Shutdown procedures

There are two possible shut down procedures for the solvent extraction section, one for
when the settlers are required to be emptied and one where only the flows in and out of this
process unit is stopped. The latter procedure is used more often since it holds a lower
safety hazard while effectively halting operation by requiring no feed from the previous
section and producing none for the next. Since counter-current flow is used through the
solvent extraction section, if one mixer-settler needs to be shut down, then all mixer-settlers
are shut down.

During the shut down procedure mentioned first, the pumps to the section is stopped while
the drain valves on the settler vessels are opened to drain the contents of these vessels.
This procedure is dangerous since the highly flammable organic solvent runs freely through
a drain. The organic solvent then ends up in a sump where it is preserved and the aqueous
phase is separated from it. If this shut down procedure is used it is important to follow the
start up procedure of a wet inspection followed by the preparation for operation procedure.

The simpler shutdown only halts the feed from the ion exchange and to the ADU
precipitation. This is achieved by stopping all the pumps while keeping the impellers in the
mixers rotating. This causes the internal recycle streams to be continuously recycled while
no organic or aqueous phases transfers between the different mixer-settler stages. The
pump around through the internal recycle may be carried out for long periods supplying
enough time to repair problems and complete inspections on the plant. If the shut-down is
expected to last longer than 2 days, the pumps and mixers are turned off. The start-up
procedure after the above mentioned shut-down procedures are to firstly start the mixers to
allow for internal recycle flow. When adequate internal flow is achieved, the pumps are
started, which will result in normal operation.

4.4.3. Emergency shutdown procedures

Emergency shutdown procedures are set in place to ensure plant and personal safety in
case of unforeseen circumstances. Three possible disturbances are evaluated and a
shutdown procedure is suggested for each.



School of Chemical and Minerals Engineering
122
Chapter 4: Detail design
Low flow or high flow

Disturbances upstream can lead to unwanted high or low flow velocities. These unwanted
flows can cause phase inversion which results in solvent lost which is economically
unfavourable as solvent is very expensive. If high or low flow occurs, the following
procedure should be followed:

The pumps should be turned off, while the mixers are still running. Switching of the
all pumps will stop any feed to enter the system which will reduce the probability of
phase inversion. Internal recycle, due to the mixers, will ensure that the liquids inside
the settler vessel are continuously moving.
The situation should be evaluated to determine the correct action that should be
done to restore the operability of the process.
If the proposed corrective action is time consuming, the mixers can be turned off to
save electricity and therefore decreasing operating costs.
The corrective action should then be implemented and the situation should be
monitored closely to ensure that the problem is solved and normal operating
conditions are resorted.

No flow or low level

No flow in the system or low levels in the equipment will cause a solvent extraction section
trip, which can lead to a plant-wide shut down. These low levels and no flow rate will also
cause phase inversion which will lead to solvent loss. In case of a plant trip due to no flow or
low level the following emergency shutdown procedure is followed:

All the pumps and mixers should be shut down to ensure equipment safety and avoid
loss of solvent due to phase inversion.
The situation should be evaluated to determine the corrective action.
The corrective action should then be implemented and the situation should be
monitored closely to ensure that the problem is solved and normal operating
conditions are resorted.


School of Chemical and Minerals Engineering
123
Chapter 4: Detail design
Maintenance

On any plant, maintenance is always required to repair equipment and instrumentation.
Maintenance usually involves grinding or welding which is a major fire risk on solvent
extraction section. It is therefore crucial to have a detailed shut down procedure for when
maintenance is required on the solvent extraction section. The following emergency
shutdown procedure should be followed:

The entire solvent extraction system should be drained to the sump except for the
storage tanks. The tanks are isolated from the system using the different control
valves used for flow control.
All the streams and equipment should be flushed with potable water to wash away
the flammable organic solvent.
If possible, the damaged equipment pieces should be removed from the demarcated
area and repaired outside the dead zone (discussed in Chapter 6).
Finally, when the repair work is completed, the wet inspection start-up procedure
should be followed and the preparation for operating should be conducted.

4.5. Mechanical aspects

The mechanical design of the mixer-settler equipment is a direct result of the chemical
design with some extra mechanical considerations. In this section the mechanical drawings
created in SolidWorks

2009 SP2.1 are given and discussed. As mentioned before, the

material of construction for the mixer and settler vessels is vacuum infused fibre glass, while
the extra components such as the impeller, inlet manifold and picket fence are constructed
from stainless steel 316L. The mechanical drawings consist of three pages:

Page 1 of 3: Overview and assembly of equipment.
Page 2 of 3: Mixer equipment with important dimensions.
Page 3 of 3: Settler equipment with important dimensions.


School of Chemical and Minerals Engineering
124
Chapter 4: Detail design
In the assembly drawing of the mixer-settler equipment (Page 1 of 3) the mixer and settler
vessel is shown together with the basic equipment. These are shown in an assembled and
exploded view. A list of this equipment follows.

Motor
Mixer vessel
Mixer lid
Impeller
Inlet manifold
Settler vessel
Picket fence

The assembly of these parts are indicated on the mechanical drawing. The mixer lid is also
constructed from vacuum infused fibre glass. More detailed drawings of the mixer and
settler equipment are given in pages 2 and 3 of the mechanical drawings.

The mixer equipment shown on Page 2 of 3 includes the mixer vessel, mixer lid, impeller,
inlet manifold and the representation of the electric motor. The mixer vessel has an inner
diameter of 1.4 m with a height of 1.7 m. The bottom part of the vessel slopes down at an
angle of 30to meet the inlet manifold. The liquid outlet of the vessel starts 0.3 m from the
top with a height of 0.2 m and width equal to the inner diameter of the mixer vessel (1.4 m).
A representation of the impeller is given in Figure 4.4.


Figure 4.4: Pumper mixer Impeller

School of Chemical and Minerals Engineering
125
Chapter 4: Detail design
The impeller is designed with a diameter of 1.2 m while the mixer vessel is equipped with
baffles extending 0.05 m as indicated on the mechanical drawing. This leaves a space of
0.05 m between the impeller and baffles. The impeller used for this design is a product of
MC Process, and the detail mechanical specifications are confidential. The mixer vessel is
fitted to the inlet manifold with a diameter of 0.152 m at the bottom. The mixer and settler
vessels are attached with a flange plate at the liquid outlet.

In order to extent the life-time of the pH probes used in the settlers, it is important to ensure
that it does not come in contact with the organic solvent. To reduce lag time in the control of
pH the measurement should be done in the mixer. To overcome the problem a small
external pH box is attached to the mixer vessel which causes separation. This allows for pH
measurement without damage to the expensive pH probes. Figure 4.5 is a schematic
representation of the pH box.



Figure 4.5: pH box for measurement

The settler equipment is shown on Page 3 of 3 of the mechanical drawings. The settler
vessel has a total length of 8.3 m which consists of the connection to the mixer vessel, the
settling compartment and the aqueous outlet compartment. The connection to the mixer
vessel has a directional length of 0.6 m. The settler compartment is defined as the length
between the liquid inlet and the aqueous weir and has a length of 5.5 m, as designed. The
aqueous outlet compartment is separated from the settling compartment by the bottom weir
and is 1.8 m in length as indicated in the mechanical drawing. It is also important to note
that the settler has a breathable lid to reduce solvent loss due to evaporation.
School of Chemical and Minerals Engineering
126
Chapter 4: Detail design
The mechanical drawing on Page 3 of 3 also shows the organic phase weir, aqueous phase
weir and the overflow. Each of these weirs is equipped with pipes to enable the different
streams to exit the settler vessel. It should be noted that the aqueous outlet weir is 0.14 m
lower than the organic outlet weir to account for the difference in liquid levels due to the
density difference between the organic and aqueous phase. Picket fences are used in the
settling compartment to enhance phase separation. As indicated on the mechanical
drawings, these picket fences fit into the picket fence slots on the settler vessel. Three
picket fence slots are situated 1 m apart in the settler vessel which allows for adjustment of
the picket fence position to obtain optimum performance.

These mechanical aspects are a preliminary guideline for the mechanical design of the
mixer-settler equipment. These mixer-settlers are used in the extraction, scrubbing,
stripping and regeneration sections of solvent extraction. The given specifications result
from the chemical design and ensure adequate sizes for an over-designed solvent extraction
process.




















School of Chemical and Minerals Engineering
127
Chapter 4: Detail design




Mechanical drawing: Page 1

School of Chemical and Minerals Engineering
128
Chapter 4: Detail design










Mechanical drawing: Page 2

School of Chemical and Minerals Engineering
129
Chapter 4: Detail design









Mechanical drawing: Page 3
School of Chemical and Minerals Engineering
130
Chapter 5: Techno-economical evaluation

Chapter 5: Techno-economical evaluation


The purpose of any chemical plant is to make a profit and to increase the value of the initial
investment into the venture. The techno-economical evaluation is important to ensure that a
profit is obtained after the proposed lifetime of the project; this is done by certain economical
analyses. The considerations for the techno-economical evaluation are:

Equipment cost.
Fixed and working capital.
Fixed and variable cost.
Revenue from product and by-product sales.

Due to the fact that this project is an upgrade for an existing plant, some of the existing
equipment and facilities will remain in the process which will decrease the equipment and
overhead plant cost involved in this evaluation. This techno-economical evaluation consists
of:

Assumptions and definitions.
Estimation of capital, operating cost and income.
A cash flow analysis.
Economic sensitivity analysis.
Recommendations for profitability.


5.1. Definitions and assumptions

The economic evaluations done in the following sections are better understood if the
definitions and assumptions are known. The different definitions and assumptions are
described in this section.

School of Chemical and Minerals Engineering
131
Chapter 5: Techno-economical evaluation
5.1.1. Definitions

Fixed capital investment (FCI) The investment needed to purchase or manufacture and
install all the process equipment with all components and necessary plant facilities (Peters et
al, 2004: 233).

Working capital The capital investment necessary for the operation of the plant which
include the money needed for raw materials, taxes, monthly expenses such as salaries and
wages and accounts that need to be paid (Peters et al, 2004: 233).

Fixed costs The fixed costs include all the expenses which are independent of the
production rate of the product as well as the plant overhead costs and general expenses
associated with the management of the company (Peters et al, 2004: 262).

Variable costs The variable costs is all the expenses associated with the manufacturing of
the product and is depended on the operability of the plant (Peters et al, 2004: 262).

Revenue from sales The sale of the products produced by the plant is known as the
revenue and is calculated as the sum of the unit price of each product multiplied by the rate
of sales (Peters et al, 2004: 258).

Income taxes Income taxes are paid on a corporate-wide basis with the total gross profit
being the taxable income of a corporation. The tax rate is subject to change as it is
depended on the taxable income (Peters et al, 2004: 304).

Discounting The calculation used to determine the present worth of a future amount using
a discount rate which can be calculated with Equation 5-1. In Equation 5-1, i is the interest
rate and N is the number of years (Peters et al, 2004: 298).

( )

= +
N
Discount factor 1 i
(5-1)


A countrys central bank also has a discount rate at which it charges commercial banks for
loans to meet the temporary shortage of funds.

School of Chemical and Minerals Engineering
132
Chapter 5: Techno-economical evaluation
Inflation The percentage change in consumer prices of goods and services compared with
the previous years prices, and this is regardless of the time value of money (Peters et al,
2004: 290).

The financial resources required for a project is limited and should be used in an appropriate
and efficient manner. Different alternatives are available to determine the most efficient use
of these resources. To evaluate these alternatives, methods are used to calculate the
profitability of the project and these methods will be discussed.

Return on investment (ROI) This method does not consider the time value of
money. The method calculates the annual return on an investment as percentage
per year, by dividing the annual net profit by the total capital investment. It is
recommended that the average ROI over the entire project life is calculated using
Equation 5-2, to obtain a better representation of the profitability (Peters et al, 2004:
323).

( ) ( )
=
=
=

N
p,j
j 1
N
j
j b
1
N
N
ROI
F
(5-2)


In Equation 5-2 N is the evaluation period, N
p,j
is the net profit in year j, -b the year in
which the first investment is made in the project with respect to zero as the startup
time, and F
j
is the total capital investment in year j.

Payback period (PBP) This method does not consider the time value of money
and is a method used to calculate the time needed for the cash flow to equal the
original fixed-capital investment. This payback period is calculated using Equation 5-
3 in which V is the manufacturing fixed-capital investment, A
x
the nonmanufacturing
fixed-capital investment and (A
j
)
ave
the average cash flow (Peters et al, 2004: 324).

( ) ( )
( )
=
+ +
= =

x x
N
j
ave
j
j 1
V A V A
PBP
A
1
A
N
(5-3)


A projects PBP should be less than or equal to the reference value calculated using
Equation 5-4 (Peters et al, 2004: 324).
School of Chemical and Minerals Engineering
133
Chapter 5: Techno-economical evaluation

=
+
ar
0.85
PBP
0.85
m
N
(5-4)


Equation 5-4 is the PBP obtained from the minimum acceptable rate of return, m
ar

expressed as a percentage per year (Peters et al, 2004: 321).

Net present value (NPV) This method consider the time value of money. Using the
present worth of all cash flow and all capital investments in Equation 5-5, the net
present worth can be calculated (Peters et al, 2004: 327).


( )( )
= =
(
= + +


N N
cf,j j oj j j j v,j j
j 1 j b
NPV PWF s c d 1 rec d PWF F
(5-5)


In Equation 5-5, PWF
cf,j
is the selected present worth factor for the cash flow, s
j
is the
value of sales, c
oj
is the total product cost not including depreciation, rec
j
is the
dollars recovered from the working capital and sale of physical assets, PWF
v,j
the
appropriate present worth factor for investments occurring, and F
j
the total
investment. All these variables are values for year j.

Internal rate of return (IRR) This method is also known as the discounted cash
flow rate of return (DCFR) and consider the time value of money. This method
calculates the return using all investments and cash flows which are all discounted.
Equation 5-5, which calculates the net present value, is set equal to zero and the
discount rate is solved (Peters et al, 2004: 328).

Depreciation Physical facilities decreases in value with time in terms of physical
depreciation and functional depreciation. The causes of physical depreciation include wear
and tear, corrosion, and accidents while functional depreciation is caused by technological
advances, making an existing property obsolete (Peters et al, 2004: 307).





School of Chemical and Minerals Engineering
134
Chapter 5: Techno-economical evaluation
5.1.2. Assumptions

It is necessary to use certain assumptions to simplify the calculations. It is a common use to
assume fixed amounts for the different rates; although rates are subject to change, these
changes cannot be accounted for in these analyses. The following assumptions were used
in this techno-economic evaluation.

Tax rate of 28% (The Low Tax Network, 2009);
Discount rate of 11% (Photius, 2009);
Inflation rate of 6.4% (Viljoen, 2009);
Rand/dollar exchange rate of 7.42 R/dollar (X-rates, 2009);
The construction of the upgraded equipment for this plant will be done over a period
of two years. For this economic evaluation the assumption was made that the plant
will be in operation for 20 years therefore the evaluation was done for a 22-year
period.
During the first year of construction the plant will be operated at the usual capacity of
producing 624 ton of ADU per year while the additional CCD equipment is being
constructed. The second year of construction the plant will be shut down to conduct
necessary piping changes for the new process and construction of the new solvent
extraction section.
The first year of operation after the construction period the plant will run at 50% of the
new capacity to allow for operational troubleshooting.

5.2. Estimation of capital, operating cost and revenue

Whenever a new project is launched, sufficient capital is required to construct the equipment
and facilities needed to produce a product which will sell at a profit. This capital investment
should be calculated accurately to ensure the maximum rate of return on the initial
investment. It is also important to estimate certain operating costs that will be required to
operate the plant throughout the project lifetime. The revenue produced from sales should
surpass the initial capital investment and operating costs to ensure economic feasibility of
the project.



School of Chemical and Minerals Engineering
135
Chapter 5: Techno-economical evaluation
5.2.1. Capital investment

Different methods can be used to calculate the total capital investment that is needed for a
project. The percentage of delivered-equipment cost method is based on the determination
of delivered-equipment cost and other direct and indirect costs. This method uses
Equation 5-6 to calculate the total capital investment (C
n
) (Peters et a.l, 2004:250).

( ) = + + + +
n 1 2 n
C E 1 f f ... f
(5-6)

In Equation 5-6 f
1
through f
n
are the multiplying factors for the specific direct and indirect
costs and E is the total purchased cost of the equipment. The values of these multiplying
factors are given in Table 5.1 on the following page. The working capital is 15% of the total
capital investment. Because this is an expansion project for the South Uranium Plant,
buildings and services already exist which will lower the total capital investment needed by a
factor of 0.8.

For the extraction of uranium the solid-fluid processing plant factors are used to calculate the
total capital investment with the delivered-equipment cost method. These calculations and
assumptions are given in Appendix D, with the following results:

Fixed capital investment (FIC) = R 720 million.
Working capital = R 126 million.
Total capital investment = R 847 million.

After calculating the total capital investment, it is important to estimate the operating cost
required to run the plant.


School of Chemical and Minerals Engineering
136
Chapter 5: Techno-economical evaluation
Table 5.1: Multiplying factors used in the delivered-equipment cost method
Fraction of delivered equipment cost for
Solid
processing
plant
Solid-fluid
processing
plant
Fluid
processing
plant
Direct cost
Purchased equipment (E) 1 1 1
Delivery 0.10 0.10 0.10
Purchased equipment installation 0.45 0.39 0.47
Instrumentation and controls (installed) 0.18 0.26 0.36
Piping (installed) 0.16 0.31 0.68
Buildings (including services) 0.25 0.29 0.18
Electrical system (installed) 0.10 0.10 0.11
Yard improvements 0.15 0.12 0.10
Service facilities (installed) 0.40 0.55 0.70
Total direct plant cost 2.69 3.02 3.60
Indirect plant cost
Engineering and supervision 0.33 0.32 0.33
Construction expenses 0.39 0.34 0.41
Legal expenses 0.04 0.04 0.04
Contractors fee 0.17 0.19 0.21
Contingency 0.35 0.37 0.44
Total indirect plant cost 1.28 1.26 1.44
Fixed-capital investment 3.97 4.28 5.04
Working capital 0.70 0.75 0.89
Total capital investment 4.67 5.03 5.93


School of Chemical and Minerals Engineering
137
Chapter 5: Techno-economical evaluation
5.2.2. Operating cost

The operating costs considered for this project is for the entire plant, and not just for the
upgraded sections. The operating costs consist of variable production cost and fixed
charges which include plant overhead costs and general expenses. The factors used to
calculate the operating costs and the amounts calculated is showed in Table 5.2 with
detailed calculations given in Appendix D.

Table 5.2: Operating costs
Fraction Fraction basis Calculated amount
(R million per
annum)
Raw materials 116.39
Operating labor 4.96
Operating supervision 0.15 Operating labor 0.74
Utilities 25.87
Maintenance and repairs 0.07 Fixed-capital investment 50.45
Operating supplies 0.15 Maintenance and repairs 7.57
Laboratory charges 0.15 Operating labor 0.74
Royalties 0.04 Revenue from sales -
Catalyst and solvents 4.86
Total variable cost 211.59
Taxes (property) 0.02 Fixed-capital investment 14.42
Financing (interest) 0.105 Fixed-capital investment 75.68
Insurance 0.01 Fixed-capital investment 7.21
Rent 0 Fixed-capital investment -
Plant overhead costs 0.5 Sum of operating labor,
supervision and
maintenance

28.08
General expenses 0.2 Operating labor 0.99
Total fixed cost 126.37
Total product cost 337.96

From Table 5.2 it is seen that the fixed cost is R 126 million per annum while the total
variable cost is manipulated to obtain the variable cost per ton product. At a production rate
of 1068 ton ADU per year, the variable cost amounts to R 198 000 per ton ADU.
School of Chemical and Minerals Engineering
138
Chapter 5: Techno-economical evaluation
5.2.3. Revenue

Up to now the necessary capital investment has been calculated, which is only a part of the
complete cost estimation. The revenue generated from the plant operation also plays an
important role in the complete cost estimation. The product of the South Uranium Plant is
ADU with the by-product of 0.5 g gold per ton ore. The commercial and proposed sale price
of the product and by-product are given in Table 5.3 with the annual revenue produced.
These calculations are also supplied in Appendix D.

Table 5.3: Revenue from product and by-product sales
Product Commercial selling
price (R/ton)
Proposed selling price
(R/ton)
Annual revenue
(R million per annum)
ADU 743 786.41 198 079.21 211.55
Gold 254 552 770.00 127 276 385.00 135 931.18

The assumptions made for the proposed selling price of the ADU and gold are:

AngloGold Ashanti only receives half of the commercial selling price of ADU.
South Uranium Plant receives 50% of the gold extracted.

The selling price per unit is calculated from data received on 26 October 2009 from Mr.
William Manana at the South Uranium Plant with a spot price of 45.50 $/lb for uranium and
254 552.77 R/kg for gold.

5.3. Cash flow analysis

The total capital investment, operating cost and revenue from sales calculated are used to
perform a cash flow analysis for the South Uranium Plant. The proposed lifetime of the plant
is 20 years with a construction period of two years to construct the new counter-current
decantation and solvent extraction sections and to upgrade the required piping. After the
two years of construction, one year is given to achieve designed through-put.

The calculations for the cash flow analysis are given in Appendix D and the results are
discussed in this section. Figure 5.1 is the graph of cumulative cash position.
School of Chemical and Minerals Engineering
139
Chapter 5: Techno-economical evaluation


Figure 5.1: Cash flow diagram

From this analysis the different representations of the plant profitability is calculated, these
are ROI, PBP, NPV, and IRR with their different definitions in Section 5.1.1. These results
are given in Table 5.4.

Table 5.4: Profitability results from cash flow analysis
Profitability representation Value
ROI (return on Investment) 60.5%
PBP (payback period) 3.98
NPV (net present value) R 2.36 billion
IRR (internal rate of return) 352%

-2
0
2
4
6
8
10
12
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
C
a
s
h

f
l
o
w

(
R

b
i
l
l
i
o
n
s
)
Years (yr)
Cumulative cash flow Cumulative discounted cash flow
School of Chemical and Minerals Engineering
140
Chapter 5: Techno-economical evaluation
According to Peters et al. (2004) the minimum rate of return (m
ar
) for projects that increases
the capacity of an existing plant and has an established corporate market position is
between 8 and 16% per year. For this project the ROI is calculated as 60.5% which is
significantly higher than the minimum rate of return which provides a favorable investment to
the investor.

For the payback period to be acceptable it should be less than or equal to the value
calculated by Equation 5.4. The value calculated from this equation is 3.95 years which is
slightly lower than the calculated PBP from the cash flow analysis. A positive NPV is
obtained from the cash flow analysis which is an indication of the money earned from the
project regardless of the investment earnings. Finally, as seen in Table 5.4, the NPV is
greater than zero which means the IRR should be calculated to give an indication of the
discounted cash flow rate of return.

5.4. Economic sensitivity analysis

The economic sensitivity analysis is done to investigate the effect of various factors on the
profitability indicators. The factors under discussion are:

Revenue from a unit product.
Inflation rate.
Fixed cost.
Variable cost.
Tax rate.

These factors are varied by increasing and decreasing the factors with increments of 5% up
to 20% to obtain the effect on the NPV value. The sensitivity analysis is given in the graph in
Figure 5.2 with the calculations in Appendix D.
School of Chemical and Minerals Engineering
141
Chapter 5: Techno-economical evaluation


Figure 5.2: Economic sensitivity analysis

As seen from Figure 5.2, the factor that positively affects the NPV the most is the revenue
received from a unit ADU with an increase of 30.9% in the NPV as the revenue increase with
20%. The other factor positively affecting the NPV is the inflation rate. The factors that
negatively affects the NPV the most is the variable cost per ton product. This means that as
the raw material costs, utility costs, operating labour costs, ect. increase or decrease the
NPV will be affected the most. The other factors negatively affecting the NPV are the tax
rate and fixed cost per annum.

-100%
-80%
-60%
-40%
-20%
0%
20%
40%
60%
80%
100%
-20% -15% -10% -5% 0% 5% 10% 15% 20%
P
e
r
s
e
n
t
a
g
e

c
h
a
n
g
e

i
n

N
V
P
Percentage change in variable
Revenue from a unit product Inflation Fixed Cost Variable Cost Tax Rate
School of Chemical and Minerals Engineering
142
Chapter 5: Techno-economical evaluation
5.5. Recommendation for profitability

Although this project already shows a desirable profit, there is always room for improvement
without jeopardizing the quality of the product and safety on the plant. In this section some
recommendation are given to improve profitability. These recommendations are:

Optimization of process:
After the construction of the additional equipment and the desired through-put is
achieved, it is necessary to optimise the process. This optimization will be done to
better understand the process conditions which can be improved to achieve optimum
efficiency of the process.

Optimization of reagents:
At the completion of the process optimization, the reagent use should be optimised to
achieve a maximum amount of product with the least amount of raw materials.
Recycle streams should be considered to achieve this optimization.

Minimum loss of solvent:
One of the most expensive reagents is the solvent used in the solvent extraction
section, thus measures should be considered to minimize the loss of this reagent.

Optimal use of existing structures:
Since this project is an upgrade of an existing plant, it should be considered to use
existing processing equipment and administrative buildings on the plant instead of
replacing them. The existing equipment that is not used should be sold, if possible.

Commission a processing plant for ADU:
At present the ADU is sent to NUFCOR for processing. It is recommended that an
economic feasibility study is conducted to compare the commissioning of a new ADU
processing plant to the outsourcing of this processing to NUFCOR.

School of Chemical and Minerals Engineering
143
Chapter 6: Safety and environment

Chapter 6: Safety and environment

From an economical point of view it is better to prevent accidents rather than to repair the
damage after the accident. This is why, when considering safety and environment, the focus
should be on preventing the hazards associated with the chemicals used, process
equipment and overall safety on the plant. In this chapter the importance of ensuring the
safety of the employees, the general public and surrounding environment are emphasized.

The sections under discussion in this chapter are:

The overall safety specifications that should be considered for the plant and the
corrective measures taken in case of hazard occurrence.
Hazard and Operatibility (HAZOP) level 1 study which ensures that the hazards
associated with the chemicals involved in the plant is understood properly.
Environmental impact and management. This is done to ensure that the wastes
generated from the plant are treated in the correct manner to minimize the plants
impact on the surrounding environment.
The plant location, which is important to determine the environment surrounding the
plant. A preliminary plant layout is discussed to ensure safety of the employees,
general public surrounding the plant, and the environment.

In all the sections the necessary occupational health and safety laws and standards should
be implemented.

6.1. Overall safety specifications

This section considers the overall safety specifications of the plant, which entails a
discussion about the possible hazards of the chemicals in each processing section as well
as some of the effects of these chemicals on the equipment used. The complete hazard
assessment of each chemical is done in Section 6.2 and only the most important hazards
are discussed in this section. There are preventative measures that can help reduce the
possibility of a hazard becoming an accident.

School of Chemical and Minerals Engineering
144
Chapter 6: Safety and environment
6.1.1. Leaching

In the acid leaching section the raw materials used are sulphuric acid, nitric acid, and
potable water. Both the sulphuric and nitric acid is extremely hazardous and corrosive
materials and human contact should be avoided at all cost. The correct signage should be
provided where these materials are used and stored as well as the correct personal
protective equipment (PPE) should be worn at all times when near the chemicals. The
sulphuric acid is diluted with the potable water, thus the correct procedure should be
followed when mixing these two chemicals together. The acid is added to the water, and not
the other way around, otherwise an extremely exothermic reaction will occur and could
cause an explosion.

The leaching equipment used is steam-heated, air-agitated, open pachucas tanks. The
hazards present for this equipment type is the heating steam and the fact that these tanks
are open with a large amount of hazardous chemicals inside. To prevent personnel falling
into the pachuca tanks the correct protective railings should be provided at the top of these
tanks. To ensure safe working conditions at this plant section the tanks should never be
operated at full capacity to prevent overflow of the tanks. The steam lines to the tanks
should be provided with the correct signage to indicate that the lines are hot and the lines
should be inspected regularly to prevent leakages.

The counter-current decantation process is also part of the leaching section. The magnafloc
used in this process is extremely slippery and spillage of this compound should be avoided.
In case of spillage, safety shoes with adequate grip should be worn and workers should
avoid the contaminated area. The processing equipment used is open thickeners. The
preventative measures for the thickeners are the same as that for the pachuca tanks, the
correct railings should be provided to prevent personnel from falling in and the correct
signage should be provided to indicate that the thickeners contain hazardous chemicals.


School of Chemical and Minerals Engineering
145
Chapter 6: Safety and environment
6.1.2. Ion exchange

The second unit in the process is the ion exchange unit where the uranium is loaded onto
the resin. The resin used is extremely slippery and spillage of this resin should be avoided.
In case of spillage, safety shoes with the adequate grip should be worn and workers should
avoid the contaminated area. The uranium complexes are a heavy metal and spillage of
these complexes will be dangerous to workers and the environment. Sulphuric acid is used
as eluant and is harmful to the workers. The acid and sulphate complexes that are present
in the system are corrosive which will damage the equipment..

The ion exchange unit consists of vertical pressure vessels with dished ends and safety
railings should be provided to the walkway at the top of the vessels to prevent workers from
falling down. To protect the environment and workers against spillage, containment walls
should be built around the units. The appropriate signage should be used to warn workers
against the corrosive nature of the chemicals in case of spillage. Special PPE should be
worn at all times in case of spillages or leaks.

6.1.3. Solvent extraction

The next processing step is the solvent extraction process which contains a number of
hazardous chemicals. The chemicals present in this section are the organic solvent and the
aqueous feed from the ion exchange section. The solvent is a composition of kerosene,
alamine

336 (alkyl amine), and isodecanol. All these organic chemicals pose a fire hazard
that should be prevented at all cost. The solvent extraction section of the plant should be
isolated from the rest of the plant to prevent fire damage to the rest of the plant in case a fire
start in the solvent extraction section. These chemicals should be stored away from ignition
sources and the correct fire fighting equipment should be present in case of fire. The fire
fighting procedures are discussed in section 6.1.5. Exposure of these chemicals to
personnel and the environment should be kept to a minimum.

The processing equipment used in the solvent extraction section is mixer-settlers. The
potential hazards associated with this type of equipment are static and operational sparks
which act as ignition sources and should be strictly controlled. Sparks are prevented by
thoroughly grounding the equipment and to isolate the mechanical and electrical equipment
to enclose the sparks generated. Additional precautions are discussed in section 6.1.5.

School of Chemical and Minerals Engineering
146
Chapter 6: Safety and environment
6.1.4. Precipitation

Precipitation is the final unit of the process where ammonia and compressed air are used to
precipitate the product, ADU, out from the OK liquor. The OK liquor, coming from the
solvent extraction unit, and the ADU, is both radioactive and the ADU is corrosive. The
ammonia is toxic when inhaled, having a toxic affect on the respiratory system and other
internal organs. When working at the precipitation unit it is necessary to wear respiratory
equipment and safety goggles to protect workers against ammonia inhalation and ADU dust.

To minimise the environmental impact of these substances, all the equipment of the
precipitation unit is inside a building and access to the building is limited. The centrifuges
that is used to wash the ADU product causes hazardous noise which means ear protection
must be used and the noise should be monitored to make sure it does not exceed the noise
limit given by the occupational health and safety act. The necessary protective hand railings
should be provided at the top of the thickener. As mentioned, the ADU product is radioactive
and extra caution must be taken to ensure that workers are not exposed to the product be
means of leakage. Radiation control is discussed in detail in section 6.1.9. The equipment
must be checked regularly to prevent spills and leakage.

6.1.5. Fire fighting

One of the most dangerous hazards on the metallurgical plant is the possibility of fire. The
largest fire hazard is located at the solvent extraction section of the plant. Thus fire
prevention and control is crucial to the design of the plant. Fire prevention and control
includes the detection, suppression, and mitigation of existing fires.

The standard fire prevention and control systems include the following (BHP billiton, 2009:
673):








School of Chemical and Minerals Engineering
147
Chapter 6: Safety and environment
Fire protection requirements in the design of plant and equipment.
Training and induction for all personnel.
Specialist training for high-risk areas.
Dedicated response personnel.
Centralised fire detection.
Fixed fire suppression systems.
Planned workplace inspections.
Hazardous materials management.

Since there may already be existing fire prevention and control measures implemented at
the South Uranium Plant, it is necessary to access these measures and equipment with the
criteria listed above. If the requirements are not met the procedure and equipment should
be upgraded and new measures should be implemented for the upgraded section of the
plant.

Special attention is dedicated to the fire prevention and control on the solvent extraction
section of the plant, since the largest fire hazard is located in this section. The main cause
of fire that should be considered is static discharge. Static discharge can be avoided by
applying certain rules including (Watson, 2009):

No flow velocities above 1 m/sec.
Adequate grounding of all equipment.
Use of PTFE diaphragm valves.
Fibre flooring and walking grid.
Pipes should be operated at full capacity to avoid friction between liquid and vapour.

Additional precautions that should be implemented for safety in the solvent extraction section
are a controlled entrance, specialized automated fire protection system, and a 15 m dead
zone around the section. The purpose of the controlled entrance is to prohibit personnel and
guests from entrance the section with equipment or accessories which may cause ignition.
All equipment and accessories are placed in a safety box at the entrance of the section and
this is controlled by a security guard.



School of Chemical and Minerals Engineering
148
Chapter 6: Safety and environment
The specialized automated fire protection system consists of both detection and
extinguishing equipment. Each vessel containing solvent is equipped with three fire
detection probes. A fire will only be registered if two of these probes detects a fire to prevent
accidental release of extinguishing foam which causes total loss of solvent. Each tank is
equipped with three extinguishing foam release points to ensure adequate dispersion of
extinguishing foam.

The 15 m dead zone is to minimize the damage to surrounding process equipment in case of
fire. Additional purpose of this dead zone is to isolate all the electrical transmitters from the
hazardous environment. All the electrical signals are transferred to Zener barriers, which is
located outside the dead zone, to prevent electrical sparks inside the hazardous
environment. A Zener barrier is an intrinsically safe explosion-proof system for electronic
instrumentation equipment (Fuji electric systems Co., Ltd., 1989).

It must be kept in mind that the necessary fire fighting equipment and procedures should be
in place to comply with insurance guidelines to ensure the safety of the investment. All
these prevention and protection measures should be implemented to ensure a safe working
environment for personnel and protection of equipment in the solvent extraction section and
the entire plant.

6.1.6. Training and personal safety

AngloGold Ashanti management and personnel are driven by the company values. The first
value concerns safety and is given as (Cutifani, 2009):

We place people first and correspondingly put the highest priority on safe and healthy
practices and systems of work. We are responsible for seeking out new and innovative ways
to ensure that our workplaces are free of occupational injury and illness. We live each day
for each other and use our collective commitment, talents, resources and systems to deliver
on our most important commitment ... to care.

To achieve this important value the company should provide training for all personnel and
visitors before allowing them to enter the working environment. The training should include
the following categories:


School of Chemical and Minerals Engineering
149
Chapter 6: Safety and environment
General process information training.
Risk assessment training.
General plant safety training which include PPE, radiation, fire, and gas leak training.
Plant specific training which includes chemical hazard training, and handling of tools
and equipment.

The general process information training is important to inform all personnel of the process
and layout of the entire plant. The overview of the process will provide the personnel with a
better understanding of potential hazards associated with the plant environment. The
personnel must know the layout of the plant to ensure that safety equipment and assembly
points are easily found. To further equip the personnel with the necessary skills to ensure
safety on the plant, the risk assessment training is provided.

The general plant safety training is important to protect the personnel and equipment on the
plant. No visitors are allowed on the plant without the proper plant safety training to ensure
public safety. This training entails PPE, radiation, fire, and gas leak training. The PPE that
should be worn at all times is; hardhat, safety goggles, safety boots, ear protection, gloves, a
safety overall, and respiratory protection at the ADU section. The radiation, fire and gas leak
training is required to ensure that safety equipment is used in the correct manner to protect
the personnel and equipment.

Each process unit should have their own plant specific training to inform the personnel
working in that section of the specific hazards involved and how to handle these situations.
In this training, a more detailed assessment of the chemical hazards and chemistry for each
section is given. Tools and equipment training is important to avoid operation errors and to
protect the specialised equipment used in the section.

6.1.7. Danger zones and signs

The most efficient safety measure is to use safety signs to indicate the potential hazards at
the relevant location. If there are dangerous zones present on the plant that is a high risk to
personnel, it should be clearly indicated using the appropriate signage. This signage should
provide information regarding the nature of the hazard as well as the safe operating
procedures when working in this dangerous zone. It is also very useful to provide safety
signs which indicate the necessary safety measures such as the PPE that have to be worn
and the location of emergency assembly points.
School of Chemical and Minerals Engineering
150
Chapter 6: Safety and environment
All safety equipment, such as fire extinguishers and breathing apparatus, must be indicated
with the correct signage. It is essential to make sure all personnel understand, are aware of
and obey the signs used on the plant, by means of general and plant specific training and
continuous inspection. Some examples of the necessary signage are shown in Figure 6.1.









Figure 6.1: Examples of necessary signage.

6.1.8. Emergency response plan

Although more resources are dedicated to the prevention of incidents, it is also important to
be prepared when accidents do happen. The emergency response plan is provided to
minimise the damage to equipment and injuries to personnel that might be caused by
accidents. The emergency response plan should include a response plan to ensure the
safety of personnel, a response plan to minimise damage to equipment and to control the
accident, and a response plan to notify the surrounding residential and business areas that
might be affected.

The safety of personnel is ensured by using a variety of alarms to notify all personnel of the
accident or potential hazard. Different alarms should be used for different accidents. When
the personnel are notified, everyone must follow the evacuation procedure provided and
gather at the assigned emergency assembly points. At these assembly points roll call must
be taken to ensure all personnel are safe. Personal safety equipment, such as breathing
apparatus, is provided on the plant for additional protection if the assembly points can not be
safely reached.



School of Chemical and Minerals Engineering
151
Chapter 6: Safety and environment
To protect the equipment on the plant in case of an accident and to contain the hazardous
chemical spills, emergency shutdown procedures should be implemented. Each section on
the plant should have a specific shutdown procedure which should be integrated into the
entire process. The hazards of chemical spills can further be contained by building
containment walls around the equipment with a high potential of chemical spill.

If the accident can not be controlled, the necessary safety services should be notified to
respond as quickly as possible. To ensure quick communication to the safety services,
dedicated phone lines should be provided on site.

6.1.9 Radiation control

Radiation is an unseen hazard and little information is available to the public, therefore this
hazard is often misunderstood which means more resources should be dedicated to the
control of radiation exposure. The International Commission on Radiology Protection (ICRP)
provides a standard dosage limitation measured in sieverts. The ICRP states that the
radiation limit for workers is 20 mSv per annum while the limitation for the public is 1 mSv
per annum (BHP Billion, 2009).

The radiation limits of each employee should be monitored each month by means of a
continuous personal dosimeter which monitors radon gas. At the end of each month a report
should be compiled, listing all the radiation exposure results, and sent to the National
Nuclear Regulator (NNR) (AngloGold Ashanti, 2007:41). Another way of determining the
dosage that the workers have been exposed to, is defined by the ICRP and involves the
determination of the different ways in which workers might be exposed to radiation. The
radiation levels must be measured in the identified areas and together with the time spent in
the exposed areas, the dosage can be determined using an internationally accepted
conversion factor (BHP Billion, 2009). In Table 6.1 some examples of radiation dosages are
given (NECSA, 2009).







School of Chemical and Minerals Engineering
152
Chapter 6: Safety and environment
Table 6.1: Examples of radiation dosages
Dosage Effect
0.01 mSv Radiation dosage received by patient having his/her teeth X-rayed.
0.01 mSv Radiation dosage received by patient having his/her lungs X-rayed.
2 mSv Annual dose of cosmic radiation received by a person working in an
aeroplane.
4 mSv Average annual radiation dose for South Africans caused by indoor radon, X-
ray examinations, etc.
100 mSv Highest permitted dose for a radiation worker over a period of five years.
1 000 mSv The dose which may cause symptoms of a radiation sickness if received
within 24 hours.
6 000 mSv The dose which may lead to death when received all at once.

To ensure that the workers do not reach the exposure limit, efficient ventilation systems
should be installed, and old buildings which might cause leaks should be monitored and
repaired. Internal radiation limits should be established, which is lower that the standard
limit, and if this limit is reached the workers should be moved to another working area where
the exposure risk is lower (AngloGold Ashanti, 2007:41).

6.2. HAZOP level 1

Hazard and Operability Studies (HAZOPS) is an integral part of the design process of a new
plant or piece of equipment. These studies analyse the potential hazards and operational
problems which is divided into six sub-sections. The most important sub-sections for the
preliminary design of a new plant are:

HAZOP level 1: Examination of the entire input-output structure to access the health
and safety considerations of the raw materials used and products produced.
HAZOP level 2: A study which investigates the possible dangers associated with
specific plant equipment. This will ensure adequate protective measures and
environmental safety.
HAZOP level 3: Any risks associated with the general operation of the process are
examined and necessary control measures are applied.

The three HAZOP levels above should be carried out in a certain sequence; this sequence is
displayed in Figure 6.2.
School of Chemical and Minerals Engineering
153
Chapter 6: Safety and environment



Figure 6.2: HAZOP structure for plant

For the scope of this conceptual design, only a HAZOP level 1 study is completed in detail.
A HAZOP level 1 study is important to understand the overall operation of the process and
the hazards associated with the raw materials and products. It is important to complete this
study at an early stage of the design process, since it is more economic favourable to
prevent hazards than it is to repair the damage after an accident. The HAZOP level 1 study
is divided into separate sections, these sections are discussed below.

6.2.1. Project definition

This project entails the upgrade of the current South Uranium Plant (SUP) of AngloGold
Ashanti. The ore processing rate needs to be increased from 240 000 to 360 000 ton/month.

To accommodate the additional feed the equipment should be upgraded or redesigned. The
process units under discussion are leaching, ion exchange, solvent extraction and
precipitation. The operating time is 8150 hours per year, which allows for maintenance time
and unscheduled shut-downs. Table 6.2 gives the projected production rate of the ADU and
gold for the upgraded plant.

Table 6.2: Projected production rate for the upgraded plant
Product Production rate (ton/annum)
ADU 1068
Gold 2.16

This increase in production, although expensive, is expected to provide long-term
economical benefits.
School of Chemical and Minerals Engineering
154
Chapter 6: Safety and environment
6.2.2. Process description

A detailed description of the process is provided in Chapter 3 hence a very short description
of the process is given in this section.

Unit 1 (U01): Leaching and CCD

The uranium extraction process starts with the leaching of uranium containing ores. The
uranium is liberated using nitric acid as oxidant and sulphuric acid as lixiviant. The leach
product stream contains both solids and liquids which are separated using counter-current
decantation. Magnafloc 90L is used as a flocculant to enhance the efficiency of
sedimentation.

Unit 2 (U02): Ion exchange

The liquid product from the counter-current decantation unit contains the uranium product
and other impurities. The uranyl sulphate ions are selectively adsorbed onto the resin,
Ambersep TM400. Using eluant, the uranyl sulphate ions are desorbed and sent to the
solvent extraction unit. The desorbtion process results in an increased uranyl sulphate ion
concentration and the removal of some impurities.

Unit 3 (U03): Solvent extraction

The solvent extraction process consists of four sections which includes extraction,
scrubbing, stripping and regeneration. The extraction section removes the uranyl sulphate
complexes from the aqueous phase into the organic phase. Most of the iron sulphate
complexes are entrained into the organic phase during this section. The scrubbing section
removes the entrained iron sulphate complexes from the organic phase. Stripping, using
ammonium sulphate, of the uranyl sulphate complexes from the organic phase into the
aqueous phase takes place in the next section. The organic solvent is regenerated in the
final section from where it is recycled to the first section.





School of Chemical and Minerals Engineering
155
Chapter 6: Safety and environment
Unit 4 (U04): Precipitation

The final unit of the uranium extraction process is the precipitation of the product, ammonium
diuranate (ADU). The OK-liquor is sent from the solvent extraction unit to the precipitation
tanks where ammonia gas is combined with compressed air, and added to precipitate the
ADU. The pH has to be controlled at 7.5 to maintain optimum precipitation. The precipitated
ADU is washed in centrifuges and the final project is stored in ADU storage.

6.2.3. Assessment of chemical hazards

The assessment of the chemical hazards only focus on individual hazards associated with
the raw materials and the products. The services, instrumentation, and plant equipment is
not considered within this assessment. The Chemical Hazard Proforma HS1A serves as a
guideline in this assessment. Table 6.3 is the chemical guide used for the HS1A performa,
Table 6.4.

Table 6.3: List of chemicals

Chemical compound Physical state
Quantity (Inventory or
through-put) (ton/annum)
A ADU Solid 624
B Alamine

336 (alkyl amine) Liquid 9.26

C Ammonia Gas 174.67
D Calcium oxide Solid 83 016.98
E Ferrous sulphate Liquid 3 548.11
F Isodecanol Liquid 9.26
G Kerosene Liquid 283.28 m
3

H Magnafloc 90L Liquid 215.98
I Nitric acid Liquid 13 844.41
J Nitric oxide Gas 2 149.38
K Potable water Liquid 3586
L Resin (Ambersep 400) Solid 123.25
M Silicon dioxide Solid 501
N Sodium carbonate Solid 84.16
O Sodium hydroxide Solid 168.32
P Sulphuric acid Liquid 143 005.73
Q Uraninite Solid 0.25

School of Chemical and Minerals Engineering
156
Chapter 6: Safety and environment
Table 6.4: Chemical hazard data sheet (HS1A)
Hazard potential -
Key: K
No.
Insignificant hazard
Hazards known and understood
See numbered notes

A B C D E F G H I J K L M N O P Q
Explosion and flammability hazards
Fire K K - - - K K - - - - - - - - - -
Deflagration/Detonation - K K - - - K - K - - - - - K K -
Electrical static - K - - - K K - - - - - - - - - -
Reactivity/stability hazards - - K - - K - K - - - - - - - - -
Immediate health hazards
Inhalation toxicity K K K K K K K - K K - - K K K K K
Skin absorption K K - K K K K - K K - - K K K K K
Corrosive - - K K - - - K K K - - - - K K -
Chronic health hazards
Digestive K K - K K K K - K - - K K K K K K
Sensitiser K K - - - K K - K K - - - - K K -
Continual K - - - K - K - K K - - K K K K K
Other health hazards
Odor - K K - - K - - K K - - - - - - -
Radiation K - K - - - - - - - - - - - - - K
Environmental hazards
Aqueous K K K - - K - - K K - - - - K K -
Gaseous - - K K - - - - - K - - - - - - -
Ground K K - K - K - - K K - K - - K K -
Hazard breakdown products K K - - K K - - K - - - - - K K K

6.2.4. Assessment of chemical interactions

After the individual hazards have been assessed, the chemical interactions between the
chemicals and construction materials should be considered to prevent any unknown and
undesirable reactions within the process. This assessment must be taken into consideration
when designing the entire plant. The Chemical Interaction Proforma HS1B, Table 6.5, is
used with the chemical guide, Table 6.3, to evaluate these interactions.
School of Chemical and Minerals Engineering
157
Chapter 6: Safety and environment
Table 6.5: Chemical interactions data sheet (HS1B)
Hazard potential -
Key: K
No.
Insignificant hazard
Hazards known and understood
See numbered notes

Chemical or
group of chemicals
B C D E F G H I J K L M N O P Q
A ADU A - - - - - K - K K - K - - - K -
B Alamine

336 B - K - - - - K K - - - - - - -
C Ammonia C - - - K - K - - - - K K - -
D Calcium oxide D - K K K K - K - - - - K -
E Ferrous sulphate E - - - - - - K - - K - -
F Isodecanol F - - K K - - - - - - -
G Kerosene G - K K - - - - - K -
H Magnafloc 90L H - - - - - - - - -
I Nitric acid I K - K K K K K K
J Nitric oxide J - K - - K - K
K Potable water K - - - - - -
L Resin L K - K K -
M Silicon dioxide M - - - -
N Sodium carbonate N - K -
O Sodium hydroxide O K K
P Sulphuric acid P K
Q Uraninite Q
Chemical
compound
A B C D E F G H I J K L M N O P Q
Material of
construction
Carbon steel - - K K - - - K K K - - - - K K -
Concrete - - K - - - - - K K - - - - K K -
Glass reinforced
plastics
- - - - - - - - - - - - - - - - -
Rubber - - - - - - - - - - - - - - - - -
Stainless steel316L - - - - - - - - - - - - - - - - -
Vacuumed infused
fibreglass
- - - - - - - - - - - - - - - - -
School of Chemical and Minerals Engineering
158
Chapter 6: Safety and environment
6.2.5. Means of handling

When handling hazardous chemicals the possibility of an accident is inevitable if the hazards
are not carefully evaluated. The means of handling Proforma HS1C is used to evaluate the
handling hazards of the chemicals on the plant. The HS1C for is given in Table 6.6.

Table 6.6: Means of handling (HS1C)
Chemical or
chemical compound
A B C D E F G H I J K L M N O P Q
Storage K - K K K K K K K K - - K K K K K
Transport K K K K - - K K K K - - - - K K K
Materials handling K - K K K K K K K K - - K K K K K
Extreme process conditions K - - K K - K - K K K - K K K K K
Other construction materials - - K K - - - - K K - - - - K K -
Decontamination K - - K - K K K - K - - K K K - K
Gaseous emissions - - K K - - K - K K - - - - - - -
Liquid release K K - - K K K K K - - - - - K K -
Disposal wastes K K K K K K K - K K - K K K K K K
Destruction of material K - - - - - - - - - - - - - - - K
Quality control K K K - - K - - K K - - - - - K -
Emergency procedures K K K K K K K K K K K K K K K K K
Plant layout, spacing, access K K K K - K K - K K - - - - K K -
Area classification K K K K K K K K K K K - K K K K K
Provision of services K - K - - - K - K - - - - - - K -
Codes of practice K K K K K K K K K K K K K K K K K

As seen in Table 6.6 there is an abundance of hazards associated with the handling of the
chemicals on the South Uranium Plant. This should be taken into careful consideration
when designing and upgrading the plant.

School of Chemical and Minerals Engineering
159
Chapter 6: Safety and environment
6.3. Environmental impact and management

According to the Environmental Conservation Act of 1989, the protection and rehabilitation of
the natural environment should be enforced by all. To accomplish this, the environmental
statement of the HAZOP level 1 study is done. This environmental statement will include a
waste block diagram analysis, handling and disposal of wastes, accidental releases of
hazardous materials and other occupational hazards to the environment.

6.3.1. Waste block diagram analysis

A waste block diagram for each processing unit is given in this section to make sure all raw
materials, wastes and products are accounted for. A typical waste block diagram is given in
Figure 6.3.



Figure 6.3: General waste block diagram structure

The first processing step is leaching where the valuable minerals are dissolved using lixiviant
and oxidizing agent. Figure 6.4 is an illustration of the waste block diagram for the leaching
process.






School of Chemical and Minerals Engineering
160
Chapter 6: Safety and environment



Figure 6.4: Leaching waste block diagram

The solid waste is transported to the gold processing plant next to the SUP, while the
pregnant leach liquor is sent to the ion exchange unit. In the ion exchange unit the
concentration of U
3
O
8
is upgraded using a resin. Figure 6.5 shows the waste block diagram
for the ion exchange unit.



Figure 6.5: Ion exchange waste block diagram

The barren liquor is recycled to the CCD unit in the leaching section to provide wash solution
for the CCD process. The eluate is sent to the solvent extraction section where the
concentration of U
3
O
8
will be further upgraded. The waste block diagram of the solvent
extraction section is shown in Figure 6.6.



School of Chemical and Minerals Engineering
161
Chapter 6: Safety and environment




Figure 6.6: Solvent extraction waste block diagram

The unwanted ADU gunk is added back into the process at the leaching section to be
processed again. The FeSO
4
waste should be disposed of in the correct manner to ensure
environmental safety. The OK liquor proceeds to the precipitation section. Figure 6.7 is a
representation of the precipitation waste block diagram.




Figure 6.7: Precipitation waste block diagram

The SO
4
2-
rich waste is recycled to the stripping unit in the solvent extraction section to strip
the U
3
O
8
from the solvent. The water waste contains harmful chemical and should be
disposed of in the correct manner.



School of Chemical and Minerals Engineering
162
Chapter 6: Safety and environment
6.3.2. Handling and disposal of wastes

The greatest concern when handling the waste of a uranium plant is the radioactive nature of
the materials that are released, in solid form, into the surrounding air and into the water
environment. It is important to consider all forms of waste which include solid, liquid and air
borne waste. When planning how to handle and dispose all waste it is essential to consult
the various safety standards and principles that should be met. International atomic energy
agency (IAEA, 1995) provides the following principles for the management of radioactive
waste:

Protection of human health.
Protection of the environment.
Protection beyond national borders.
Protection of future generations.
Burdens of future generations.
National legal framework.
Control of radioactive waste generation.
Radioactive waste generation and interdependencies.
Safety of facilities.

To further assist with the design of the waste disposal units, the different wastes can be
classified in to four levels, shown in Table 6.7 (NECSA, 2009).

Table 6.7: Waste classification
Classification Description
Very low-level waste (VLLW) Contains very low concentrations of radioactivity,
originating from the operation and decommissioning of
nuclear facilities.
Low- and intermediate level
waste (LILW)
Contains concentrations or quantities of radionuclides
above the clearance levels established by the regulator.
High-level waste (HLW) Contains heat-generating radionecluides with long- and
short-lived radionuclide concentrations.
NORM Contains low concentrations of naturally occurring
radioactive materials.

School of Chemical and Minerals Engineering
163
Chapter 6: Safety and environment
The waste generated on this plant is classified as NORM waste. The South African
department of minerals and energy (DME) provides certain regulations for the waste
management of radioactive waste and there are an additional three policies that should be
considered. These policies are:

The Radioactive waste management policy and strategy for the Republic of South
Africa (2005).
The Principles of Radioactive Waste Management (IAEA, 1995).
Management of Radioactive Waste from the Mining and Milling of Ores (IAEA, 2002).
Storage of Radioactive Waste, Safety Guide No. WS-G-6.1 (IAEA, 2006).

A few of the considerations that should be addressed are to ensure an adequate ventilation
system and that equipment is enclosed to prevent wind dispersion of contaminated
materials. The unit where the product is formed should be located in a separate building
with limited access to minimize the exposure of workers to radiation.

On the existing plant a waste management plan is already in practice. This system will be
used for the already processed feed and the upgraded feed. An upgrade might be
necessary when more information is available. The gold slurry produced at the CCD section
should be neutralization to increase the pH from 1.8 to 10.5, lime slaking is used for this
neutralization and this is done in one of the existing pachucas.

The upgrade of the plant involves a new process using nitric acid as oxidizing agent, causing
the release of nitrate ions. Although this waste is less hazardous than the waste that is
already produced, it can not be ignored. A process is already developed which uses
biochemical operations to remove nitrogen from the wastewater. This process is known as
denitrification and involves the reduction of nitrate-N to N
2
by heterotrophic bacteria that
grow in the absence of oxygen and the presence of nitrates. The nitrates act as the terminal
electron acceptor and the micro-organisms used grow in suspension in the liquid that is
treated. Because the wastewater treatment bioreactors is depended on the way in which the
micro-organisms grow, for denitrification a continuous stirred tank reactor (CSTR) is needed
(Grady, 1999:216). In the USA 100 wastewater treatment plants use methanol in the
denitrification process which serves as a carbon source for bacteria (Methanol Institute,
2009). These wastewater treatment facilities serve as an indication that this method of
wastewater treatment is efficient and can be used.

School of Chemical and Minerals Engineering
164
Chapter 6: Safety and environment
6.3.3. Accidental releases of hazardous materials

It is not possible to completely avoid accidental releases of hazardous chemicals. Therefore
it is necessary to have adequate preventive measures and emergency response in case of a
release. Table 6.8 lists the hazards materials that may lead to potential spillage.

Table 6.8: Accidental releases
Materials Impact Avoidance Measures or Actions
ADU Exposure to radiation Signage and confined work area.
Acids Chemical burn
Signage and containment dams around
equipment.
Magnafloc 90L Slippery surfaces Signage and correct PPE.
Kerosene Fire hazard
Signage and security measures must
be taken.
Ore
Dust dispersion and small dose
radiation
Signage, correct PPE and enclosed
structures.

Strict implementation of the avoidance measurements should be enforced to ensure a safe
working environment for personnel and the natural environment.

6.3.4. Rehabilitation and decommissioning of plant

The decommissioning of a plant is required by the government in order to protect the
environment and reduce the exposure to radioactive materials. The decommissioning of a
plant starts after plant operations seized and involves the decontamination and dismantling
of equipment and the disposable of all radioactive materials. Decommissioning can be
described as the administrative and technical actions that should be taken to remove the
facilities which have been exposed to radioactive material on such a scale that it can be
considered a safety hazard. Decommissioning of facilities is becoming a major issue and
such be considered in every plant design (IAEA safety standards, 2006: 1).






School of Chemical and Minerals Engineering
165
Chapter 6: Safety and environment
The decommissioning of a plant can be divided into three phases. The first phase involves
the plant shutdown, followed by the disposal of process materials. In this first phase the
plant site is made safe and prepared for care and maintenance. During the second phase
some of the process equipment is dismantled, component sizes are reduced and the
equipment is decontaminated. In the final phase the buildings on site is decontaminated for
de-regulations (NECSA, 2009).

It is recommended that a preliminary decommissioning plan should be submitted to the
government five years before the projected end of operations. This plan should include the
cost estimation, discussion of the major technical actions that will need to be taken and
lifetime of the decommissioning project (Fentiman et al, 2009:1).

Some guidelines are provided by NECSA (2009) that can be followed when the preliminary
decommissioning plan is constructed.

A complete inventory of all facilities, equipment and materials that is exposed to
nuclear radiation should be complied.
The materials and processes that need to be decontaminated and dismantled should
be categorised. These categories are defined by the extent to which the materials
and equipment should be processed.
The capacities and cost of each of the processes are determined.
The nuclear exposure of all the equipment and materials should be assessed with
specially developed software packages.

6.4. Plant layout and positioning

The plant layout and positioning is done to ensure that enough land is available for the new
upgraded equipment and to get an overall representation of the positioning of all the different
units. It is important to note that the equipment sizes are not on scale but the plant area is a
more accurate representation. The layout was constructed using the air photo obtained from
the program Google Earth and the old plant layout received from William Manana which is
included in Appendix E. The new plant layout is given in Figure 6.8 and the legend included
assists in the interpretation of the figure.


School of Chemical and Minerals Engineering
166
Chapter 6: Safety and environment






















Figure 6.8: Plant layout

School of Chemical and Minerals Engineering
167
Chapter 7: Process control

Chapter 7: Process control

In the previous sections the process is design assuming steady-state operating conditions using
a specified feed. It should however be kept in mind that there are multiple variations in the
process conditions due to external disturbances on the process. These variations influence the
process which results in dynamic operating conditions. Since the process is designed from
steady-state assumptions it is important to manipulate the controllable variations to ensure that
the process conditions are within the design bounds.

When the process conditions are not within the design bounds, the process will operate under
either hazardous conditions or production quality loss conditions or both. These conditions
should be avoided since it will endanger the environment and plant personnel, damage process
equipment and result in economical loss. For these reasons it is important to have an effective
integrated control system to keep the process conditions within the design bounds.

The following aspects are addressed in this chapter:

An integrated control strategy for the entire plant.
A detailed process control for the solvent extraction section.
Specific safety considerations in the control of the solvent extraction section.

When all these aspects are addressed, the basic control for the uranium extraction plant is
described. These control strategies will be used to optimise the plant operations and ultimately
achieve optimum plant operating conditions.

7.1. Plant wide control

The plant wide control section will discuss the control objectives and strategies for each unit in
the process. A control schematic is shown for each of the important equipment used in each of
the units


School of Chemical and Minerals Engineering
168
Chapter 7: Process control
7.1.1. Unit 1: Leaching, CCD and neutralization

In this section the control strategies for the leaching, counter-current decantation and
neutralization processes are given and discussed.

Leaching

Firstly the control for the leaching section is discussed. In the leaching section the uranium is
liberated through several chemical reactions. To ensure the effective operation of this section
all parameters influencing reaction kinetics must be controlled. The following control objectives
for the leaching section are identified:

Reactant flow rates
Pachuca pH
Pachuca temperature
Level of agitation

In order to control these process conditions, control strategies are devised. These control
strategies are shown in the control schematic in Figure 7.1.

School of Chemical and Minerals Engineering
169
Chapter 7: Process control



Figure 7.1: Control schematic of the leaching section

In Figure 7.1, four control loops are seen, each one controlling one of the identified control
objectives. Figure 7.1 shows three reactant streams fed into the first pachuca; the ore feed,
H
2
SO
4
feed and the HNO
3
feed. Of these three reactant feeds only the flows of the H
2
SO
4
and
HNO
3
feed can be controlled since a constant ore feed in the slurry form is received from the
mills of the gold extraction plant. The H
2
SO
4
feed provide the H
3
O
+
ions required in the process
and is therefore used to control the pH of the reactor system which is discussed later.





School of Chemical and Minerals Engineering
170
Chapter 7: Process control
It is decided to control the flow for the HNO
3
feed stream as a constant ratio in regards to the
received ore feed flow. This is achieved by implementing a ratio control loop between the ore
feed and HNO
3
feed streams. The volumetric flow of the ore feed is measured (FI 01) and
transmitted to control unit (FRC 01) where the flow is multiplied with the required ratio to obtain
a set point for the HNO3 feed flow. The volumetric flow rate of the HNO
3
feed is also
continuously measured and transmitted to the control unit (FRC 01), which will then control the
valve opening according to the actual flow and calculated set point. This control loop operates
as a direct controller which implies that the valve opening should increase when the volumetric
flow rate of the ore feed increases. Since the disturbance (ore feed volumetric flow) is
measured and accounted for before it enters the process (pachuca), the control loop is
feedforward control.

As mentioned the pachuca pH is controlled using the H
2
SO
4
feed flow. The pH is measured at
the exit stream from the pachuca. This measured pH represents the pH of the whole pachuca
contents due the intense agitation and mixing. The pH value is transmitted to the pH control
unit (PHC 01) from which a set point for the H
2
SO
4
volumetric flow rate is controlled. This
controlled set point and the measured actual volumetric flow of the H
2
SO
4
feed (FI 03) is used to
control valve opening. This concept where the flow set point rather than the valve opening is
controlled is known as cascade control and is widely implemented for liquid systems (Svrcek et
al., 2006: 131-135). It should be noted that the pH control is implemented on the first two
pachucas only, since the most H
3
O
+
is used in these initial stages. The overall control loop
operates as a direct controller, since the H
2
SO
4
flow rate should increase when the pH
increases. This control loop is seen as a feedback control loop, since the disturbance is
measured downstream of the process.

The next important parameter to be controlled is the temperature of the pachuca content which
has a significant effect on the reaction kinetics. It is assumed in the design that a temperature
loss of 0.5 C is experienced over each pachuca, from which it is derived that the temperature
should be controlled at each fourth pachuca. It is imperative to keep the temperature in the
pachuca above 30 C to achieve the reaction rate the process is design for. Therefore the
lowest temperature in the pachuca (at the top) is measrured to control the temperature above
30 C. This temperature measurement is transmitted to the temperature control unit (TIC 01)
which controls the steam feed to the pachuca by implementing a cascade control loop. This
control loop operates as a reverse controller since the steam flow should be reduced if the
temperature increases. This control loop is seen as a feedback control loop.
School of Chemical and Minerals Engineering
171
Chapter 7: Process control
The pachuca contents is continuously agitated using air agitation which ensures adequate
mixing. The air entering at the bottom of the pachuca is controlled at or above a pressure of
6 bar to ensure adequate mixing. The pressure control is achieved by measuring the pressure
of the air where it enters the pachuca and transmitting it to the pressure control unit (PIC 01).
The pressure control unit (PIC 01) controls the valve opening to achieve the desired pressure.
The compressed air is supplied at 6 bar by the gold extraction plant and introduced to each of
the pachucas. This control loop operates as a reverse controller since the air flow should be
reduced if the pressure increases. This control loop is seen as a feedback control loop.

Counter-current decantation

In the following section the control of the counter-current decantation section is discussed. In
this section the leach product is washed to effectively remove the dissolved uranium in the liquid
phase from the solids. This is achieved using a series of thickeners which results in solid-liquid
separation. This process requires long residence times in thickeners with large capacity. The
most important parameters in controlling the thickeners are:

Residence time
Wash ratio
Solid content of pregnant leach liquor
Rake resistance

Each of these control objectives are controlled using a practical control strategy. These control
strategies are shown in a control schematic in Figure 7.2.


School of Chemical and Minerals Engineering
172
Chapter 7: Process control



Figure 7.2: Control schematic of the counter current decantation section

The control schematic of Figure 7.2 shows three feed streams, the leach product, wash solution
and floculant. The residence time is an important parameter and therefore the flow rates into
the thickeners must be kept as stable as possible. Further it is desired to achieve a low solid
content in the pregnant leach liquor to prevent fouling of the resin in the ion exchange column.
To ensure effective washing, the wash ratio is controlled at an optimum. The rakes gather the
slurry at the bottom of the thickener to the underflow, which at times can experience high
resistance due to the thickness of the settled slurry.

The residence time of the thickeners are fixed and determined by the feed flow rate of the leach
product and wash solution. Since a fixed wash solution to leach product ratio is used, only the
leach product feed is controlled to achieve the desired residence time. The leach product feed
flow is measured (FI 01) and transmitted to the flow controller (FIC 01) which manipulate the
valve opening to reach the determined flow. This control loop operates as a reverse controller
since the leach product flow should be reduced if the flow increases. This control loop is seen
as a feedforward control loop.
School of Chemical and Minerals Engineering
173
Chapter 7: Process control

The wash ratio is controlled by manipulating the set point of the wash solution feed flow rate
according to the set point of the leach product feed rate. The wash ratio is designed at one,
which will give an efficiency of 99.99%. The set point of the wash solution flow rate is not
directly dependent on the leach product flow rate, therefore a process upset introduced by the
leach product flow rate is not duplicated in the wash solution. The flow rate for the wash
solution is measured (FI 02) and transmitted to the flow controller (FIC 02) which controls the
valve opening to manipulate the wash solution feed rate. This control loop operates as a
reverse controller since the wash solution flow should be reduced if the flow increases. This
control loop is seen as a feedforward control loop.

It is important to keep the solid content of the pregnant leach liquor below the prescribed
50 ppm. This is achieved by manipulating the floculant addition. The solid content of the
pregnant leach liquor is determined by sampling this stream once a shift for analysis. The
results of the analysis will determine the set point for the floculant addition. The set point of the
floculant addition will vary greatly in the initial commissioning stage, however this variable will be
optimised for the specific process and will eventually require less sampling. This control loop
operates as a direct controller since the floculant addition should be increased if the solid
content of the pregnant leach liquor increases. This control loop is seen as a feedback control
loop.

The thickness of the settled slurry may cause equipment damage since the rake is driven from
the centre of the thickener. This resistance is observed in the amps which the electric motor
requires. This effect is used to control the rake and reduce the probability of equipment
damage. The rake is controlled by hydraulically lifting the rake and allowing it to pass over most
of the settled slurry whenever high amps are measured. This control loop operates as a direct
controller since the rake height should be increased if the amps increase. This control loop is
seen as a feedback control loop.






School of Chemical and Minerals Engineering
174
Chapter 7: Process control
Neutralization

The last process in this unit is the neutralization where the slurry product from the CCD section
is neutralised before it is sent to the gold extraction plant. This process is carried out in an
existing pachuca where the pH of the contents is increased to 10.5 by the addition of slaked
lime. The following control objectives are identified.

Pachuca pH
Level of agitation

The neutralised slurry is treated with cyanide at the gold extraction plant, which releases toxic
gases if the received slurry is acidic. Therefore it is important to control the neutralization
process effectively to ensure the safety of the environment. In Figure 7.3 a control schematic
for the neutralization process is shown.



Figure 7.3: Control schematic of the neutralization process



School of Chemical and Minerals Engineering
175
Chapter 7: Process control
In Figure 7.3 the feed streams to the neutralization section is shown; the slurry feed, slaked lime
and air feed. The slurry feed is received from the CCD section and cannot be controlled due to
the immense influence it will have on the CCD steady-state operation. Adequate agitation is an
important factor to ensure efficient neutralization. The pH of the product must be 10.5 which is
acceptable for the gold extraction plant.

The pH of the pachuca content is controlled using the slaked lime feed flow rate. The pH of the
pachuca is measured (PH 01) and transmitted to the pH controller (PHC 01) which then controls
the flow rate of the slaked lime feed. The flow rate is controlled using a cascade control method
as discussed for Figure 7.1. This control loop operates as a reverse controller since the slaked
lime feed flow rate should be increased if the pH decreases. This control loop is seen as a
feedback control loop.

To ensure adequate agitation the air pressure of the air feed at the bottom of the pachuca is
controlled at 6 bar. This is achieved by measuring the air pressure (PI 01) and transmitting it to
the pressure controller (PIC 01) which controls the valve opening to control the pressure. This
control loop operates as a reverse controller since the valve opening should be increased if the
air pressure decreases. This control loop is seen as a feedforward control loop.

7.1.2. Unit 2: Ion exchange

The pregnant leach liquor from the CCD section is fed to the ion exchange system where the
concentration of uranyl sulphate complexes is increased. This is achieved by the selective
adsorption of these complexes onto resin, and after saturation of the resin the complexes are
desorbed during the elution stage. The liquid eluant (diluted H
2
SO
4
), used for the elution of the
uranyl sulphate complexes, has a lower volumetric flow rate to ensure a higher concentration.

This process is operated in a semi-continuous procedure in which each column of resin goes
through four separate stages; adsorption, back wash, elution and resin regeneration. The
control for each of these stages is discussed separately in the following sections. These
processes are commonly controlled according to time cycles which is calculated and determined
during the commissioning period.

School of Chemical and Minerals Engineering
176
Chapter 7: Process control
When a column is in the adsorption stage, the uranyl sulphate complexes are adsorbed onto the
resin. The adsorption kinetics is fast and therefore an adsorption front is formed where the
adsorption takes place as described in Section 2.5.2. When this front reaches the bottom of the
column, break through is observed. The control objectives identified for this stage is given as:

Pregnant leach liquor flow rate
Column saturation

When the resin is saturated, the adsorption column is taken of the adsorption train to be back-
washed and eluted. The pregnant leach liquor flow rate influences the thickness of the
adsorption front. The control schematic for the adsorption stage is shown in Figure 7.4.




Figure 7.4: Control schematic of the adsorption stage of ion exchange

Figure 7.4 shows that the pregnant leach liquor is pumped into the resin-containing column and
exits to the next column in the train. The volumetric flow rate of pregnant leach liquor is
controlled to ensure adequate residence time and an acceptable adsorption front thickness.
The volumetric flow rate is measured (FI 01) and transmitted to the flow rate controller (FIC 01)
which controls the valve opening. This control loop operates as a reverse controller since the
valve opening should be decreased if the flow increases. This control loop is seen as a
feedforward control loop.
School of Chemical and Minerals Engineering
177
Chapter 7: Process control
In many cases the required time to saturate the resin in a column is used as control mechanism
to indicate when the resin is saturated. Due to resin poisoning the required time should be
analysed frequently to reduce uranium loss. The saturation of the resin can however be
measured using its conductivity. In this method the solution temperature (TI 01), solution
conductivity (CI 01) and resin conductivity (CI 02) are measured to determine the level of
saturation of the resin (Serra & Sol, 1986:34). These measurements are transmitted to an
on/off controller (CIC 01) which determines the saturation of the resin and decides accordingly
to switch the pump on or off. This control loop operates as a multi-input-single output controller,
which is seen as a feedback control loop. When the resin is saturated the pump is switched off
and back-wash may start. The back-wash stage control schematic is shown in Figure 7.5.




Figure 7.5: Control schematic of the back-wash stage of ion exchange

During the back-wash stage the resin bed is fluidised to ensure the effective removal of small
solid impurities from the resin and remove pregnant leach liquor remaining in the resin bed.
Therefore the back-wash water flow rate is controlled to achieve fluidization. The flow rate is
measured (FI 01) and transmitted to the flow rate controller (FIC 01) which controls the valve
opening. This control loop operates as a reverse controller since the valve opening should be
decreased if the flow increases. This control loop is seen as a feedforward control loop. When
the back-wash stage is completed, the resin is eluted as shown in the control schematic in
Figure 7.6.
School of Chemical and Minerals Engineering
178
Chapter 7: Process control



Figure 7.6: Control schematic of the elution stage of ion exchange

The eluant volume used to elute the resin is kept constant relative to the resin loading to
achieve the desired uranyl sulphate concentration. For this process, 11 bed volumes of eluant
is used with a residence time 15 minutes. To ensure this flow, the eluant flow rate is measured
(FI 01) and transmitted to the flow controller (FIC 01) which controls the valve opening. This
control loop operates as a reverse controller since the valve opening should be decreased if the
flow increases. This control loop is seen as a feedforward control loop.

The regeneration step of the resin is an important step to ensure a longer resin lifetime, and
better resin capacity. This stage is carried out in specific steps in a separate regeneration
column to prevent resin shock. Resin is pumped from the adsorption columns to this separate
adsorption column. Therefore the resin is first contacted with H
2
SO
4
, then water, then caustic
and backwards again. Figure 7.8 shows the regeneration control schematic.

School of Chemical and Minerals Engineering
179
Chapter 7: Process control


Figure 7.7: Control schematic of the regeneration stage of ion exchange

Each of the flows through this regeneration column takes place in the different steps and is
controlled using flow controllers similar that seen in the elution and back-wash stages (Figure
7.6 and 7.7).

7.1.3. Unit 3: Solvent extraction

The eluate from the ion exchange unit contains an increased uanyl sulphate concentration and
is sent to an eluate feed storage tank with a large capacity. From the eluate feed storage tank,
the eluate is pumped to the solvent extraction unit where the concentration of the uranyl
sulphate complexes is increased further and impurities are removed.



School of Chemical and Minerals Engineering
180
Chapter 7: Process control
This process is similar to ion exchange in that the uranyl sulphate complexes are transferred to
an organic phase and back to an aqueous phase in the different sections. The process consists
of the extraction, scrubbing and stripping stages with an extra regeneration section where the
organic solvent is regenerated. The main control objectives and strategies for the solvent
extraction unit are discussed below, while a more detailed discussion on the control of this unit
is given in Section 7.3. The following main control objectives are identified for the solvent
extraction unit.

Volumetric flow rates of the various streams
pH on the stripping stages

This process is designed to process a certain eluate feed flow rate with set ratios between the
flow rates of the aqueous and organic phases through the mixer-settler units. The stage
efficiency and phase separation in the stripping section is greatly dependent on the pH which
varies because of extraction reactions. A control schematic of the flow rates for the solvent
extraction unit is given in Figure 7.9.



Figure 7.8: Basic control schematic of the solvent extraction unit


School of Chemical and Minerals Engineering
181
Chapter 7: Process control
Figure 7.8 shows that the eluate feed flow rate from the storage tank is controlled to ensure a
flow rate close to that for which the unit is designed for. From this the flow rate ratios in all the
sections are controlled to maintain the design conditions. The pH of each stage in the stripping
section is controlled using the caustic feed stream which is fed to each stage separately.

The eluate feed volumetric flow rate is controlled at the designed flow. This is achieved by
measuring the volumetric flow rate (FI 01) and transmitting it to the flow rate controller (FIC 01)
which controls the valve opening. This control loop operates as a reverse controller since the
valve opening should be decreased if the flow increases and is seen as a feedforward control
loop.

The ratios of the flow rates for the aqueous and organic phases through the stages are
controlled using a series of control loops as seen in Figure 7.8. All of these control loops
implement the ratio control method as described in Section 7.1.1. The organic to aqueous ratio
control for the scrubbing is described as an example. The set point for the volumetric flow rate
of the uranium loaded stream (organic) is transmitted to the flow rate controller (FIC 03) which
then controls the flow rate of the aqueous stream by using cascade control. These control loops
operate as direct controllers since the controlled flow rate should increase if the set point of the
wild flow rate increases and is seen as feedforward control loops.

The pH control in the respective stripping section stages is important and is difficult to control
due to the logarithmic dependence of the pH. The control for each of the stages is represented
by the single control loop in Figure 7.8. The pH is measured using an external pH probe
(PH 01), as discussed in Section 4.5, which does not come into contact with the organic phase
to prolong the lifetime of these probes. The measured pH is transmitted to a pH controller
(PHC 01) which then controls the flow of caustic to the respective stripping stage by
implementing cascade control. This control loop operates as a reverse controller since the
caustic flow rate should decrease if the pH increases and is seen as a feedback control loop.


School of Chemical and Minerals Engineering
182
Chapter 7: Process control
7.1.4. Unit 4: Precipitation

The precipitation unit recovers the product from the uranium containing OK-liquor leaving the
solvent extraction unit. The product specifications for the ADU product are also achieved in this
unit to obtain optimum economical benefit. The volumetric flow rates in this unit are relatively
low since the concentration of uranyl sulphate complexes in the OK-liquor is high. To achieve
the desired product specifications and ensure adequate recovery, the relevant process
conditions are controlled within the design bounds. The following control objectives are
identified:

Reactor temperature
OK-liquor feed flow rate
Reactor tank liquid levels
Reactor liquid pH
Ammonia gas make-up

Each of these process conditions are controlled to ensure the optimum operation of this unit.
The control schematic which shows the control loops for the two precipitation reactors is given
in Figure 7.9.















School of Chemical and Minerals Engineering
183
Chapter 7: Process control



Figure 7.9: Control schematic of the precipitation reactors in the precipitation unit

In Figure 7.9 it is seen that the ADU is formed and precipitated in two precipitation reactors by
increasing the pH and ammonia gas addition. The OK-liquor is heated to the desired
temperature with steam in a heat exchanger before it is fed to the first precipitation reactor. The
uranyl sulphate complexes react with the ammonium fed to the reactor to form ADU which
precipitates at a higher pH. From here the solid ADU containing stream is sent to the solid-
liquid separation section.

The temperature of the precipitation reactors are controlled at a certain level to ensure the
reaction rate achieves the desired conversions of the uranyl sulphate complexes. The
temperature of the OK-liquor is controlled to manipulate the temperature in the reactors. To
achieve this the reactor temperature is measured (TI 01) and transmitted to the temperature
controller (TIC 01) which controls the flow of the steam to the heat exchanger by using a
cascade control loop. This control loop operates as a reverse controller since the steam flow
rate should be increased if the reactor temperature decreases. This control loop is seen as a
feedback control loop.
School of Chemical and Minerals Engineering
184
Chapter 7: Process control
The OK-liquor feed flow rate together with the precipitation reactor liquid level determines the
residence time in the reactors. These parameters are therefore controlled to ensure sufficient
residence time which will achieve the desired conversions. The OK-liquor feed flow rate is
measured (FI 02) and transmitted to the flow rate controller (FIC 02) which controls the valve
opening. This control loop operates as a reverse controller since the valve opening should be
increased if the flow rate decreases. This control loop is seen as a feedforward control loop.

The reactor tank liquid levels are also controlled to ensure the desired residence time. This is
achieved by measuring the reactor liquid level (HI 01) in the first reactor and transmitting it to
the liquid level controller (HIC 01). The liquid level controller then controls the flow rate of the
exit stream from the first reactor by implementing a cascade control loop. The liquid level for the
second reactor is controlled similarly by measuring the liquid level (HI 02), transmitting the
height to the height controller (HIC 02) and controlling the flow rate of the exit stream from the
second reactor. These control loops operate as direct controllers since the exit flow rate should
be increased if the liquid level increases. These control loops are seen as feedback control
loops.

The pH levels of the reactors are controlled to achieve the desired reaction kinetics and ensure
precipitation of the ADU. The pH control in the first reactor is achieved by measuring the pH in
the reactor (PH 01) and transmitting it to the pH controller (PHC 01) which controls the ammonia
gas flow rate to the reactor. The ammonia flow rate is controlled by using a cascade control
loop. The pH control for the second reactor is done similarly. These control loops operate as
reverse controllers since the ammonia gas feed flow rate should be increased if the pH level
decreases. These control loops are seen as feedback control loops.

The ammonia gas is fed to the reactors as a mixture with air. To control the ammonia
concentration in the gas stream fed to the reactors, a ratio controller is implemented between
the air and ammonia gas feeds. This ratio control loop measures the volumetric flow rate of the
ammonia gas (FI 04) stream and transmits it to the ratio controller (FIC 01). The ratio controller
calculates a set point for the air volumetric flow rate by multiplying the flow rate of the ammonia
with the fixed ratio. The air flow rate is measured (FI 03) and transmitted to the ratio controller
(FIC 01) which controls the valve opening in the air feed stream according to the calculated set
point. This control loop operates as a direct controller since the valve opening in the air feed
School of Chemical and Minerals Engineering
185
Chapter 7: Process control
stream should be increased if the flow rate of the ammonia gas feed stream increases. This
control loop is seen as a feedforward control loop.

The slurry stream containing the precipitated solid ADU is sent the solid-liquid separation
section where the final ADU product is produced to specification. The ADU solids are separated
from the liquids and washed to achieve the desired ADU purity and density. This section
consists of a thickener and three centrifuges. In this section it is important to control the flow
rates to ensure adequate separation and purification. The following control objectives are
identified for this section:

Thickener liquid level
ADU feed to centrifuges
Wash water to the centrifuges

By controlling these process conditions the final ADU product is produced to specification. The
control schematic for the solid-liquid separation section in the precipitation unit is given in Figure
7.10.

















School of Chemical and Minerals Engineering
186
Chapter 7: Process control



Figure 7.10: Control schematic of the solid-liquid separation in the precipitation unit

In Figure 7.10 it is seen that the solids mass fraction of the ADU product is increased in a
thickener from where it is sent to two parallel centrifuges. In the centrifuges the ADU is washed
with water to remove remaining liquid impurities.





School of Chemical and Minerals Engineering
187
Chapter 7: Process control
The liquid level in the thickener is controlled to ensure adequate residence time and effective
overflow. This is achieved by measuring the liquid level in the thickener (HI 01) and transmitting
it to the liquid level controller (HIC 01) which controls the flow rate of the thickener underflow
flow rate by using a cascade control loop. This control loop operates as a direct controller since
the thickener underflow rate should be increased if the liquid level in the thickener increases.
This control loop is seen as a feedback control loop.

The ADU streams to the two parallel centrifuges are controlled to ensure efficient operation of
the equipment. To achieve this, the flow rate of ADU is measured (FI 02 & FI 03) and
transmitted to the flow rate controllers (FIC 02 & FIC 04) which then control the valve position.
These control loops operate as reverse controllers since the valve opening should be increased
if the flow rate decreases. These control loops are seen as feedforward control loops.

The wash water flow rates used to wash the ADU product in the centrifuges are controlled at a
fixed ratio according to the ADU stream. Ratio control loops as discussed for the ammonia gas
feed make-up for the precipitation reactors are implemented to control the flow rate of the wash
water streams. These control loops operate as direct controllers since the wash water flow rate
should be increased if the ADU flow rate increases. These control loops are seen as
feedforward control loops.

The control strategies discussed above are used to control the many important process
conditions on the uranium extraction plant. These strategies are control loops and procedures
that each controls a certain process parameter. An integrated control system, that incorporate
and monitor all above control strategies, must be designed and implemented to achieve
effective and practical plant wide control.






School of Chemical and Minerals Engineering
188
Chapter 7: Process control
7.2. Specific safety considerations for solvent extraction

To ensure the safety of the environment, equipment and employees, it is important to consider
several control aspects in the solvent extraction unit. These control aspects influence both the
economical and production performance. The influence of several process disturbances in the
solvent extraction unit are studied using a HAZOP level 3 analysis. The process disturbances
include no flow, low flow, high flow and several other disturbances. The HAZOP level 3 analysis
indicates shortcomings in the control strategies, which is rectified to reduce the safety risk of a
plant.

In this section a HAZOP level 3 analysis is done on the solvent extraction section. Before
commencing the HAZOP level 3 analysis, an fire hazard due to the extremely flammable
organic solvent is identified. The process control must also prevent solvent loss. The process
flow diagram used for which illustrates the different flow loops considered for the HAZOP level 3
analysis.
School of Chemical and Minerals Engineering
189
Chapter 7: Process control

























School of Chemical and Minerals Engineering
190
Chapter 7: Process control
The five different liquid flow loops, seen in the process flow diagram, used for the HAZOP level
3 analysis are colour coded. The first liquid flow used is the aqueous phase flow in the
extraction section and is represented with the colour brown. The eluate stream is the uranium
containing feed stream to the solvent extraction section. The eluate flows into a tank which
supply capacitance to the system to reduce the effects of process disturbances. The aqueous
phase is contacted repeatedly with the organic solvent which extracts the uranium, before it is
recycled back to the leaching section.

The streams indicated as green, as seen in the process flow diagram, represent the organic
solvent flow loop. This loop flows counter-current in respect to the aqueous phase flows. The
organic flow enters the extraction section where the organic is loaded with uranium and
continues to the scrubbing section. At the scrubbing section impurities are removed from the
solvent before the uranium is stripped in the stripping section. Finally the organic solvent is
regenerated and recycled back to U03-ST03 (organic storage tank).

The demineralised water flow loop is indicated with the blue stream. The demineralised water is
stored in U03-ST04 from which it is continuously recycled through the scrubbing section to
remove impurities. When the demineralised water is saturated with impurities, it is drained from
the system to allow for fresh demineralised water.

The next flow loop is indicated with the orange stream flow and symbolises the aqueous phase
through the stripping section. This loop contains two feed streams i.e. the stripping solution and
caustic feed. The stripping solution flows counter-current in respect to the organic solvent and
removes the uranium from the organic solvent. The caustic is fed into each tank to control the
pH which has a significant effect on stripping efficiency.

The organic phase is regenerated by contacting it with caustic solution, which is represented
with the colour black. To achieve the desired concentrations the caustic and demineralised
water are mixed in a make-up tank (U03-MU02). The caustic solution is continuously recycled
through the regeneration section until the solution is saturated with impurities.

These five streams are used as an approach to the HAZOP level 3 analysis to simplify the
reasoning. The HAZOP level 3 analysis is show in Table 7.1 and Table 7.2.

School of Chemical and Minerals Engineering
191
Chapter 7: Process control
Table 7.1: HAZOP 3 Proforma HS3A form
Path/line/node
Identification
Eluate loop Organic loop Demin water
loop
OK liquor loop Regeneration
loop
Problem? No Yes Ref
no.
No Yes Ref
no.
No Yes Ref
no.
No Yes Ref
no.
No Yes Ref
no.
High flow * 1 * 17 * 22 * 27 * 34
Low flow * 2 * 18 * 23 * 28 * 35
No flow * 3 * 19 * 24 * 29 * 36
Reverse flow * 4 * 4 * 4 * 4 * 4
High pressure * * * * *
Low pressure/vac * * * * *
High temperature * * * * *
Low temperature * * * * *
High level * 5 * 20 * 25 * 30 * 37
Low level * 6 * 21 * 26 * 31 * 38
High composition * * * * *
Low composition * * * * *
High pH * * * * 32 * 39
Low pH * * * * 33 * 40
Fast reaction, mix * * * * *
Slow reaction, mix * * * * *
High differential * 7 * 7 * 7 * 7 * 7
High stress * 8 * 8 * 8 * 8 * 8
Poor integrity * 9 * 9 * 9 * 9 * 9
Malfunction * 10 * 10 * 10 * 10 * 10
Impurities * 11 * 11 * 11 * 11 * 11
Lost containment * 12 * 12 * 12 * 12 * 12
Radiation * 13 * 13 * 13 * 13 * 13
Generation * 14 * 14 * 14 * 14 * 14
Maintenance * 15 * 15 * 15 * 15 * 15
Start/stop * * * * *
Emergency/ test * 16 * 16 * 16 * 16 * 16
Inoperability * * * * *
School of Chemical and Minerals Engineering
192
Chapter 7: Process control

Table 7.2: HAZOP 3 record for Proforma HS3A
Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
1 High flow Upstream disturbances
Pump failure
Overflow of tanks U03-ST01
and U03-ST02
Flooding of settler
Loss of settling efficiency
Increase of organic flow
Overflow weirs in settlers
Overflow valves on tanks
U03-ST01 and U03-ST02
Back-up pumps
High capacitance in tank
U03-ST01
Built in drain on U03-
ST01 with level control
Level control on tank
U03-ST02
Flow control on stream
U03-AQ02
Adequate alarm system

2 Low flow Blockages in pipes
Pump failure
Solvent loss
Loss of production
Phase inversion in settlers
Capacitance due to tank
U03-ST01
Back-up pumps
Filters
Emergency shut-down
procedure to ensure
safety of solvent
Adequate alarm system

3 No flow Upstream shut-downs
Pump failure
Power failure
Blockages in pipes
Loss of production
Pump cavitations
Up-stream down time
Capacitance due to tank
U03-ST01
Back-up pumps
Emergency shut-down
procedure to ensure
safety of solvent
High capacitance in tank
U03-ST01
Adequate alarm system

4 Reverse flow Height difference Pump damage None Non-return valves
Insure pumps are able to
handle reverse flow

5 High level High flow up stream
Low flow downstream
No flow downstream
Flooding of tanks and
settlers
Loss of product
Overflow weirs in settlers
Overflow valves on tanks
U03-ST01 and U03-ST02
Adequate alarm system
Level control on tank
U03-ST02

6 Low level Low flow up stream
High flow downstream
No flow up stream
Cavitation of pumps
Loss of production
Capacitance due to tank
U03-ST01
Emergency shut-down
procedure
Adequate alarm system


School of Chemical and Minerals Engineering
193
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
7 High
differential
Valve position at start-up Equipment damage None Adequate start-up
procedure
Pipe layout to reduce
water hammer effect

8 High stress Vacuum inside tanks Internal collapse Fire proof vent openings
on tanks
None
9 Poor
integrity
Corrosion Spillage
Loss of raw materials
Down time
Material of construction Frequent inspections
10 Malfunction Failure of instruments
Power failure
Instrument air
Pump failure
Mixer failure
Loss of production
Unsafe working conditions
High, low or no flow
Back-up pumps Instrument redundancy
Alarm system on mixers

11 Impurities Solids from upstream
processes
Impeller damage
Lower stage efficiency
None Filter
12 Lost
containment
Leaks
Evaporation
Fire
Damage to environment
Unsafe working conditions
Drainage systems
Enclosed tanks and
settlers
Containment walls
Frequent inspections

13 Radiation Nuclear radiation Unsafe working conditions
Damage to environment
Regular radiation testing
personnel
Correct PPE
None
14 Generation High flow rates
Half full pipes
Instrumentation short
circuits
Static
Personal accessories
(phones, etc.)
Fire Ensure flows lower than 1
m/s
Air bleeds in pipe system
Fire proof instrumentation
Allow no personal
accessories on the SX
plant
Ground all equipment
Fibre grating cat walks

School of Chemical and Minerals Engineering
194
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
15 Maintenance Malfunction of
instrumentation and
equipment
Corrosion
Removal of precipitated
solids from settlers
Down time
Unsafe working conditions
Frequent inspections Operating procedure
16 Emergency/
test
Fire Equipment damage
Unsafe working conditions
Emergency shut-down
procedure
Frequent inspection
Adequate alarm system

17 High flow Pump failure Flooding of settlers
Solvent loss
Production loss
Overflow weirs in settlers
Sump to recycle solvent
overflows
Back-up pumps
Overflow valve on tank
U03-ST03
After settler to reduce
solvent loss
Level control on tank
U03-ST03
Flow control on streams
U03-OG03 and U03-
OG16
Adequate alarm system
Built in drain on U03-
ST03 with level control

18 Low flow Pump failure
Blockages in pipes
Loss of production
Phase inversion
Capacitance due to tan k
U03-ST03
Back-up pumps
Filters
Emergency shut-down
procedure to ensure
safety of solvent
Adequate alarm system

19 No flow Pump failure
Blockages in pipes
Power failure
Up- and downstream
shut-downs
Loss of production
Pump cavitations
Up-stream down time
Back-up pumps
Capacitance due to tank
U03-ST03
Filters
Adequate alarm system

School of Chemical and Minerals Engineering
195
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
20 High level High flow of recycle
stream U04-OG16
Low flow downstream
No flow downstream
Flooding of tanks and
settlers
Loss of product
Overflow weirs in settlers
Overflow valves on tanks
U03-ST03
Adequate alarm system
Level control on tank
U03-ST03

21 Low level Low flow of recycle
stream U04-OG16
High flow downstream
No flow up stream
Pump cavitations
Loss of production
Capacitance due to tank
U03-ST03
Emergency shut-down
procedure
Adequate alarm system

22 High flow Upstream disturbances at
demin. water plant
Pump failure
Flooding of tank U03-ST04
and settlers
Overflow weirs in settlers
Overflow valve on tank
U03-ST04
Back-up pumps
Level control on tank
U03-ST04
Adequate alarm system

23 Low flow Upstream disturbances
Pump failure
Blockages in pipes
Loss of production
Phase inversion
Capacitance due to tan k
U03-ST04
Back-up pumps
Filters
Emergency shut-down
procedure to ensure
safety of solvent
Adequate alarm system

24 No flow Pump failure
Blockages in pipes
Power failure
Up- and downstream
shut-downs
Loss of production
Pump cavitations
Up-stream down time
Back-up pumps
Capacitance due to tank
U03-ST04
Filters
Adequate alarm system

25 High level High flow of upstream
demin. water plant
Low flow downstream
No flow downstream
Flooding of tanks and
settlers
Loss of product
Overflow weirs in settlers
Overflow valves on tanks
U03-ST04
Adequate alarm system
Level control on tank
U03-ST04 using the flow
of U03-DW01

School of Chemical and Minerals Engineering
196
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
26 Low level Low flow up stream
demin. water plant
High flow downstream
No flow up stream
Pump cavitations
Loss of production
Capacitance due to tank
U03-ST04
Emergency shut-down
procedure
Adequate alarm system

27 High flow Upstream disturbances
Pump failure
Flooding of tank U03-ST05,
U03-ST06 and settlers
Overflow weirs in settlers
Overflow valve on tanks
U03-ST05 and U03-ST06
Back-up pumps
Level control on tank
U03-ST05
Adequate alarm system
High capacitance in tank
U03-ST06
Flow control on stream
U03-AQ09

28 Low flow Upstream disturbances
Pump failure
Blockages in pipes
Loss of production
Phase inversion
Solvent loss
Capacitance due to tan k
U03-ST05
Back-up pumps
Filters
Emergency shut-down
procedure to ensure
safety of solvent
Adequate alarm system

29 No flow Pump failure
Blockages in pipes
Power failure
Up- and downstream
shut-downs
Loss of production
Pump cavitations
Up-stream down time
Back-up pumps
Capacitance due to tank
U03-ST05
Filters
Adequate alarm system

30 High level High flow of upstream
processes
Low flow downstream
No flow downstream
Flooding of tanks and
settlers
Loss of product
Overflow weirs in settlers
Overflow valves on tanks
U03-ST05
Adequate alarm system
Level control on tank
U03-ST05 and U03-
ST06
Height control on tank
U03-ST07 using stream
U03-DW08

School of Chemical and Minerals Engineering
197
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
31 Low level Low flow up stream
processes
High flow downstream
No flow up stream
Pump cavitations
Loss of production
Capacitance due to tank
U03-ST05
Emergency shut-down
procedure
Adequate alarm system
Height control on tank
U03-ST07 using stream
U03-DW08

32 High pH pH of precipitation recycle
Caustic soda addition
Precipitation of solids
Inadequate phase
separation
4 stripping settlers instead
of 3
Addition of caustic soda to
stripping settler 2
pH measurement at
mixer instead of
aqueous stream out
pH control on tank U03-
ST07
Flow control on caustic
soda addition to mixers

33 Low pH pH of precipitation recycle
No flow of caustic soda
Lower stage efficiency 4 stripping settlers instead
of 3
Addition of caustic soda to
stripping settler 2
pH measurement at
mixer instead of
aqueous stream out
Flow control on caustic
soda addition to mixers
pH control on tank U03-
ST07

34 High flow Upstream disturbances
Pump failure
Flooding of tank U03-MU02,
U03-ST08 and settlers
Overflow weirs in settlers
Overflow valve on tanks
U03-MU02 and U03-ST08
Back-up pumps
Level control on tank
U03-MU02 using stream
U03-DW07
Adequate alarm system

35 Low flow Upstream disturbances
Pump failure
Blockages in pipes
Loss of production
Phase inversion
Solvent loss
Capacitance due to tank
U03-MU02
Back-up pumps
Filters
Emergency shut-down
procedure to ensure
safety of solvent
Adequate alarm system

School of Chemical and Minerals Engineering
198
Chapter 7: Process control

Ref.
no:
Deviation Causes Consequences or Hazards Safeguards already
provided
Recommendations or
Actions
By
36 No flow Pump failure
Blockages in pipes
Power failure
Up- and downstream
shut-downs
Loss of production
Pump cavitations
Up-stream down time
Back-up pumps
Capacitance due to tank
U03-MU02
Filters
Adequate alarm system

37 High level High flow up stream
processes
Low flow downstream
No flow downstream
Flooding of tanks and settler
Loss of product
Overflow weirs in settler
Overflow valves on tanks
U03-MU02
Adequate alarm system
Level control on tank
U03-MU02 using stream
U03-DW07

38 Low level Low flow up stream
processes
High flow downstream
No flow up stream
Pump cavitations
Loss of production
Capacitance due to tank
U03-MU02
Emergency shut-down
procedure
Adequate alarm system
Level control on tank
U03-MU02 using stream
U03-DW07

39 High pH Caustic soda and Na
2
CO
3

addition
Inadequate phase
separation
None pH measurement at
mixer instead of
aqueous stream out
pH control on tank U03-
MU02 using U03-ST09

40 Low pH No flow of caustic soda
and Na
2
CO
3

Lower stage efficiency
Inadequate solvent
regeneration
None pH measurement at
mixer instead of
aqueous stream out
pH control in mixer using
stream U03-S10



School of Chemical and Minerals Engineering
199
Chapter 7: Process control
The following recommended actions are identified and rectified. From the HAZOP level 3
analysis the emergency shutdown and start-up procedures are revised to account for the
identified risks and to ensure adequate operating response. Filters are installed to remove alien
objects from the system which cause blockages. Non-return valves are used to protect
equipment against reverse flow. The pipe layout on the plant is specifically designed to reduce
the risk of damage if they experience a water hammer. Each tank on the solvent extraction site
is surrounded by a containment wall to prevent the fires from spreading and reduce spill
contaminated area. Due to the high acidic process conditions it is important to conduct frequent
inspections on equipment. It is necessary to remove all possible sources of ignition from the
solvent extraction plant and therefore fibre grated catwalks are used rather than metal catwalks
to prevent charge build-up.

After the HAZOP level 3 analysis is done, the recommended corrective actions are
implemented. In the following section the implementation procedure and discussion for these
corrective actions are given. Each of the protective measures is imperative for the continuous
efficient and safe operation of the plant.

7.3. Detailed process control for solvent extraction

The most important process parameters to control are the feed ratios of each section, feed flow
of the eluate to the extraction section and the pH in the stripping section. As identified in the
HAZOP level 3 analysis it is also important to control the level of the storage tanks to reduce the
probability of tank flooding. These levels are controlled by draining the tank to a sump when a
critical level is reached and therefore a draining system and sump is required. To prevent
unnecessary emergency shutdowns due to instrument failure, redundant instrumentation is
installed. A RAT (range, alarms and trip) list is composed for the implemented control loops to
ensure notification. All these identified control strategies are illustrated in the following process
flow diagram.





School of Chemical and Minerals Engineering
200
Chapter 7: Process control

























School of Chemical and Minerals Engineering
201
Chapter 7: Process control
The process flow diagram shows twenty one control loops and extra specific indicators that
are installed in order to enable desired control of the solvent extraction section. According to
Svrcek et al. (2008) it is helpful to create a relative gain array to indicate the best control
scheme for the pairing of manipulated and controlled variables. The simplicity of the control
objectives and existing solvent extraction control strategies makes a relative gain array
unnecessary for this specific case. The control loops are discussed in five groups with
similar control objectives which are listed below:

Basic flow control.
Level control on tanks using the drain system.
Level control on tanks using feed to the tank.
pH control on make-up tanks.
pH control on the mixer-settlers.

However, as mentioned certain instrumentation is installed to indicate malfunctioning of
equipment. For this reason electric current indicators are installed on all pumps and mixers,
and enables operators to switch any of this equipment on or off. These indicators allow
operators to observe the working status of these pumps and mixers from the control room.
Malfunctioning of the pump is shown by these indicators, for instance, if the pump is running
and a zero reading is shown, the pump is malfunctioning. Only selected electric current
indicators are shown in the process flow diagram to simplify the sketch.

7.3.1. Flow control loops

The basic flow control loops on selected streams are indicated with pink on the process flow
diagram. These flows are controlled at points where a certain flow rate is desired for
effective operation of the solvent extraction unit and other units. The flow control loops on
the streams entering the mixer-settlers are implemented to control the flow at a certain ratio
to the uranium containing phase flow through the mixer-settlers.






School of Chemical and Minerals Engineering
202
Chapter 7: Process control
Basic flow control loops are fast responding with little dead time and no capacitance. It is
important to place the flow sensor upstream of the control valve to ensure more accurate
flow readings. Due to the no capacitance in this system, flow measurements are noisy and
therefore derivative action control is not implemented. PI controllers are widely used in the
industry for flow control and are simpler to tune. The quarter decay ratio is commonly not
observed for these control loops, due to the fast response of the flow controllers (Svrcek et
al., 2008: 147).

The simpler flow control loops control the flows of the raffinate recycle to the leaching and
the OK-liquor to the precipitation unit. These control loops control the flows to reduce
disturbances to the leaching and precipitation units. The tanks from which these streams
flow provide efficient capacitance to dampen disturbances.

The eluate feed to the extraction section is controlled using FIC 02 and is kept constant at a
set point. The organic phase flow fed to the extraction section is controlled with FIC 06 with
a set point calculated from the set point for FIC 02 multiplied with the specified ratio. The
demineralised water fed to the scrubbing section is similarly controlled with FIC 07 at a set
point which is calculated from FIC 06. The stripping solution flow entering the stripping
section is controlled with FIC 11 at a certain ratio to the set point of FIC 06. The caustic feed
to the regeneration section is controlled with FIC 20 which also has a set point at a certain
ratio to that of FIC 06. Finally the organic solvent recycle flow is controlled with FIC 21
which also has a set point at a certain ratio to that of FIC 06. Table 7.3 shows the
specifications for control loops discussed above.


School of Chemical and Minerals Engineering
203
Chapter 7: Process control
Table 7.3: Specifications for flow controllers
Controller Description Process variable Set point
FIC 02
Controls eluate flow to
U03-MSE01.
Volumetric flow of
U03-AQ02.
25.816 m
3
/hr
FIC 04
Controls raffinate flow
to leaching.
Volumetric flow of
U03-AQ07.
1 x Set point of
FIC 02
FIC 06
Controls organic flow
to U03-MSE03.
Volumetric flow of
U03-OG03.
1.1 x Set point of
FIC 02
FIC 07
Controls demineralised
water flow to U03-
MSS03.
Volumetric flow of
U03-DW02.
0.2 x Set point of
FIC 06
FIC 11
Controls stripping
solution flow to U03-
MST01.
Volumetric flow of
U03-AQ09.
0.28 x Set point of
FIC 06
FIC 16
Controls OK-liquor flow
to precipitation.
Volumetric flow of
U03-AQ15.
1 x Set point of
FIC 11
FIC 20
Controls regeneration
solution flow to U03-
MSR01.
Volumetric flow of
U03-S10.
1 x Set point of
FIC 06
FIC 21
Controls organic
recycle flow to U03-
ST03.
Volumetric flow of
U03-OG16.
1 x Set point of
FIC 06

Controlling the flow and flow ratios are a simple and effective method to keep the mixer-
settler levels at the designed conditions. Therefore these flow control loops are important to
ensure effective operation of the mixer-settlers and prevent solvent loss.

7.3.2. Tank level control loops

There are two types of tank level control loops implemented in the solvent extraction unit.
One type is where the level is controlled by draining the tank, indicated with orange, while
the other controls the feed flow to the tank and is indicated with blue. The tanks supply
capacitance to the system to simplify control and dampen process disturbances. The
disturbances on the liquid surface may cause noise on the measured process variable and
therefore PID control is used with caution (Svrcek et al., 2008: 151).

School of Chemical and Minerals Engineering
204
Chapter 7: Process control
The level on tanks U03-ST01, U03-ST02, U03-ST05 and U03-ST06 are controlled by
draining the tank to a sump. This control is implemented here, since it is important that the
tanks do not overflow. For this reason the liquid height itself is not controlled, but the tanks
is only drained at a dangerous liquid level and therefore on/off controllers are used in these
control loops. The draining system rather than the exit or feed streams are controlled to
reduce the influences of the level control on the greater system. The on/off control loops
make use of a dead band due to incapability of the controller to throttle the actuator. The
dead band used here has a maximum of 95 % and minimum of 90 % of the liquid level.
When the liquid level rises above the maximum point, the draining flow is opened and only
closed once the liquid level is below the minimum point of the dead band (Svrcek et al.,
2008: 94).

These controllers are cascade control loops, which is used seldom and malfunctions may
occur. Therefore it is important to have a flow indicator on the draining stream to ensure that
flow occurs when the valve is open. The specifications for these control loops are given in
Table 7.4.

Table 7.4: Specifications for drain system level control
Controller Description Process variable Set point
HIC 01
Level control on
U03-ST01.
Liquid level of
U03-ST01.
On 95 %
Off 90 %
HIC 05
Level control on
U03-ST02.
Liquid level of
U03-ST02.
On 95 %
Off 90 %
HIC 12
Level control on
U03-ST05.
Liquid level of
U03-ST05.
On 95 %
Off 90 %
HIC 17
Level control on
U03-ST06.
Liquid level of
U03-ST06.
On 95 %
Off 90 %

The level on tanks U03-ST04, U03-ST07 and U03-MU02 are controlled by the demineralised
water feed to the tanks. This is done to ensure adequate liquid levels and to simplify the pH
control on these tanks. The control loops implemented here are also cascade control loops
for more accurate control of the flow. The specifications for these control loops are shown in
Table 7.5.



School of Chemical and Minerals Engineering
205
Chapter 7: Process control
Table 7.5: Specifications for feed flow level control
Controller Description Process variable Set point
HIC 08
Level control on
U03-ST04.
Liquid level of
U03-ST04.
75 %
HIC 09
Level control on
U03-ST07.
Liquid level of
U03-ST07.
75 %
HIC 19
Level control on
U03-MU02.
Liquid level of
U03-MU02.
75 %

The control loops of the liquid levels of these tanks create no disturbances that will influence
the solvent extraction process or downstream processes. Since the liquid levels of these
tanks are not of utmost importance to the system, their influences on the system are
minimised.

7.3.3. pH control loops

The pH is controlled in two equipment types in the solvent extraction unit, in the caustic
make-up tanks and in the mixers. The pH measurement and configuration of these two
control loops are different and are therefore discussed separately. To control the pH of any
system accurately, it is important to have a constant pH in the feed stream used to control it.
For this reason the pH on the make-up tanks are controlled to provide feed streams with a
stable pH.

The pH on tank U03-ST07 and U03-MU02 are controlled using similar control loops
represented by the green control loops. The caustic feed flows to these tanks are controlled
to control the pH in the tanks. The control loops implemented for this function is cascade
control loops to ensure more accurate control. It is important to note that the height of these
tanks are controlled using the demineralised water addition. These two control loops on
each tank are interdependent and therefore influence one another in the dynamic state.
However, this control strategy results in adequate control response to achieve the desired
control conditions. A simple case study shows that this control strategy is capable of
reaching the control objectives.



School of Chemical and Minerals Engineering
206
Chapter 7: Process control
For the case study a situation where the liquid level is high and the pH is also high is
assumed and is a worst case scenario. In this case, the caustic feed to the tank is closed
and the demineralised water feed is also closed. The liquid level will soon drop due to a
constant exit stream, causing the demineralised water feed to open slowly, which will
decrease the pH. Quicker response time for any other situation is expected.

The pH of the mixers is simply controlled using a cascade control loops on the caustic flows
indicated by the yellow control loops. The measurement of the pH in the mixers is
complicated since the mixers contain a dispersion of aqueous and organic phases. Direct
pH measurement of the dispersion is possible, from which the aqueous pH is derived.
However, the pH probes used in the direct measurement comes in contact with the organic
phase which significantly reduces the lifetime of the costly pH probes. For this reason, a
small settling box is used (as described in Section 4.5), where settling takes place, and the
pH of the aqueous phase is measured. This vessel must have a small volume to reduce the
residence time and ultimately reduce the dead time on the control loop. The specifications
for these control loops are given in Table 7.6.

Table 7.6: Specifications for pH control
Controller Description Process variable Set point
PHC 10
pH control on
U03-ST07.
pH of U03-ST07. 9.5
PHC 13
pH control on
U03-MST02.
pH of U03-MST02. 5.1
PHC 14
pH control on
U03-MST03.
pH of U03-MST03. 4.8
PHC 15
pH control on
U03-MST04.
pH of U03-MST04. 3.9
PHC 18
pH control on
U03-MU02.
pH of U03-MU02. 9.5

The pH control on the mixers in the stripping section is important to ensure adequate stage
efficiency and phase separation. The pH in the in the first mixer of the stripping section is
high due to the stripping solution recycle from the precipitation unit. Thereafter, the pH is
controlled in a decreasing order in the second, third and fourth mixers, as indicated in Table
7.6.

School of Chemical and Minerals Engineering
207
Chapter 7: Process control
7.3.4. Range, alarms and trips

The HAZOP level 3 analysis done on the solvent extraction section shows that alarm
systems and trips are required to ensure plant safety. For this reason a RAT list is compiled
and shown in Table 7.7.

Table 7.7: RAT list for the solvent extraction section
Loop
number
Tag
number
Instrument
type
Service
RAT information
Range Alarm Trip
HIC 01
HI 01
Ultra sonic
level
Indicate level on U03-
ST01
0-100%
H: 95%
L: 20%
HH:99%
LL: 5%
FI 01 Flow meter
Indicates flow to U03-
MSE01
0-35
m
3
/hr

FIC 02 FI 02 Flow meter
Indicates flow to U03-
MSE01
0-35
m
3
/hr
H: 26.5
L:24.5
HH:28.5
LL: 22

AI 03-
25
Amp meter
Indicates amp of the
respective pump or
mixer
0-20A
H:12
L: 3
HH: 15
LL: 0
FIC 04 FI 04 Flow meter
Indicates flow to
leaching
0-35
m
3
/hr
H: 26.5
L:24.5
HH:28.5
LL: 22
HIC 05
HI 05
Ultra sonic
level
Indicate level on U03-
ST01
0-100%
H: 95%
L: 20%
HH:99%
LL: 5%
FI 05 Flow meter
Indicates flow to
Drain
0-35
m
3
/hr

FIC 06 FI 06 Flow meter
Indicates flow to U03-
MSE03
0-35
m
3
/hr
H: 29.5
L:27.5
HH:31
LL: 25
FIC 07 FI 07 Flow meter
Indicates flow to U03-
MSS03
0-35
m
3
/hr
H: 8.5
L:7
HH:9.5
LL: 6

School of Chemical and Minerals Engineering
208
Chapter 7: Process control
HIC 08
HI 08
Ultra sonic
level
Indicate level on U03-
ST04
0-100%
H: 80%
L: 20%
HH:95%
LL: 5%
FI 08 Flow meter
Indicates flow to U03-
ST04
0-35
m
3
/hr

HIC 09
HI 09
Ultra sonic
level
Indicate level on U03-
ST07
0-100%
H: 80%
L: 20%
HH:95%
LL: 5%
FI 09 Flow meter
Indicates flow to U03-
ST07
0-35
m
3
/hr

PHC 10
PH 10 pH meter
Indicate pH on U03-
ST07
1-14pH
H: 9
L: 6
HH:10.5
LL: 4.5
FI 10 Flow meter
Indicates flow to U03-
ST07
0-35
m
3
/hr

FIC 11 FI 11 Flow meter
Indicates flow to U03-
MST01
0-35
m
3
/hr
H:8.5
L:7
HH: 9.5
LL: 6
HIC 12
HI 12
Ultra sonic
level
Indicate level on U03-
ST07
0-100%
H: 95%
L: 20%
HH:99%
LL: 5%
FI 12 Flow meter
Indicates flow to U03-
ST07
0-35
m
3
/hr

PHC
13-15
PH pH meter
Indicates pH of U03-
ST07
1-14pH
H: 8
L: 3
HH:9.5
LL: 2
FI Flow meter
Indicates flow to U03-
ST07
0-35
m
3
/hr

FIC 16 FI 16 Flow meter
Indicates flow to U03-
ST07
0-35
m
3
/hr
H:8.5
L:7
HH:9.5
LL:6
HIC 17
HI 17
Ultra sonic
level
Indicate level on U03-
ST06
0-100%
H: 95%
L: 20%
HH:99%
LL: 5%
FI 17 Flow meter
Indicates flow to U03-
ST06
0-35
m
3
/hr

PHC 18
PH 18 pH meter
Indicates pH of U03-
MU02
1-14pH
H: 8
L: 3
HH:9.5
LL: 2
FI 18 Flow meter
Indicates flow to U03-
MU02
0-35
m
3
/hr

HIC 19
HI 17
Ultra sonic
level
Indicate level on U03-
MU02
0-100%
H: 80%
L: 20%
HH:95%
LL: 5%
FI 17 Flow meter
Indicates flow from
U03-MU02
0-35
m
3
/hr

School of Chemical and Minerals Engineering
209
Chapter 7: Process control
PHC 20
PH pH meter
Indicates pH of U03-
MSR01
1-14pH
H: 8
L: 3
HH:9.5
LL: 2
FI Flow meter
Indicates flow to U03-
MSR01
0-35
m
3
/hr

FIC 21 FI 21 Flow meter
Indicates flow to U03-
MSE03
0-35
m
3
/hr
H: 29.5
L:27.5
HH:31
LL: 25

Installing these alarms and trips will safeguard the equipment and effectively notify plant
personnel of the process disturbances. The overall plant safety and efficiency is greatly
enhanced by the implementation of these guidelines.

7.4. Dynamic control analysis

In this section the dynamic control response is simulated and analysed. This gives a greater
understanding of the dynamic behaviour of the implemented control system. The solvent
extraction unit is simulated using Aspen HYSYS

. It is decided to simulate the different

sections separately to simplify the flowsheet which contains recycle streams. The flow and
pH control loops are analysed and optimised in this section.

HYSYS does not simulate open tanks which produced several obstacles in the simulation of
the effectively open tank system. For this reason air is included in the component list and
streams to simulate the open tanks. Each mixer-settler unit is represented with a three
phase separator with the same volume and footprint area as that designed for the settlers.
In the design, the organic phase is 93 % kerosene, and therefore the organic phase is
represented by 100 % C
12
H
24
in the simulation. The aqueous phase is always represented
by water.

For the desired separation to take place the thermodynamic model data in the simulation is
required. The NRTL thermodynamic package is used to simulate the system. The binary
activity coefficients for the liquid-liquid equilibrium are acquired using Aspen data bank. The
binary coefficients between the air and both liquids are set to zero to ensure no
thermodynamic interactions.



School of Chemical and Minerals Engineering
210
Chapter 7: Process control
The HYSYS simulation is specifically used to simulate the liquid flow and the storage tank
liquid levels. Since this is a complex system and HYSYS does not have specialised process
blocks to simulate all the different equipment, only the above mentioned process parameters
are simulated.

7.4.1. Tuning control loops

To achieve effective control with the control loops, it is tuned to give satisfactory response
time with limited overshoot. By tuning the control, the control efficiency is optimised and the
control equipment, which includes control valves, is protected. An example of this is the
controllers in the flow control loops, which are PI-controllers instead of PID-controllers to
reduce the noise on control valves.

There are three parameters in a PID-controller that determines the control aggressiveness of
the controller. The gain (K
c
), the integral time (T
i
) and the differential time (T
d
) are tuned or
varied to obtain the desired control action on the specific system. HYSYS does include an
auto-tune function which automatically derive specific values for the controller parameters
(K
c
, T
i
and T
d
) to achieve satisfactory control (Svrcek et al., 2008: 107-110).

There are also methods, such as the Ziegler-Nichols open-loop method, available to obtain
these parameters from the simulated process (Svrcek et al., 2008: 125). Because of the
complex system and all the control equipment such as pumps, steady state is not easily
reached in the solvent extraction simulation. Therefore the Ziegler-Nichols method is not
implemented on the simulation, since it requires the simulation to reach steady state. In this
section the auto-tune function in HYSYS is used to optimise the control action of each
control loop.

In Section 7.3.1 it is mentioned that several of the flow control loops are ratio control loops.
The tuning of these ratio control loops are described by taking the flow ratio control in the
extraction section as an example. The simulation flowsheet in HYSYS for the extraction
section is shown in Figure 7.11.





School of Chemical and Minerals Engineering
211
Chapter 7: Process control


Figure 7.11: Extraction simulation flowsheet

In Figure 7.11, the flow controllers that are using ratio control are indicated with the red
circles, where the eluate flow on the left is the independent stream and the organic flow on
the right is the dependent stream. The set point of the organic feed controller (FIC 06) is
directly calculated from the set point of the eluate flow controller (FIC 02). This is different
from the ratio control methods discussed by Svrcek et al. (2008) in that set point of the
independent stream, rather than the process variable, is measured to calculate the set point
of the dependent stream. This is done to ensure that process disturbances in the eluate flow
do not cause disturbances in the organic feed to the extraction section. Since the flow
control action of the flow control loops are fast and effective, the desired ratio is achieved
quickly. Figure 7.12 shows the eluate and organic flow into the extraction section when a
step change in the set point of the eluate feed is made.








School of Chemical and Minerals Engineering
212
Chapter 7: Process control


Figure 7.12: Ratio control results in extraction section

The intended ratio is 1.1:1 for the organic to eluate flow rate into the system. From Figure
7.14 it is calculated that the ratio remains 1.1:1 before and after the change in the set point
of the eluate feed. Figure 7.14 shows that the implemented ratio control system is effective.

As an example of the tuning procedure followed to auto-tune the liquid level control loops in
the simulation, the liquid level controller (HIC 08) on U03-ST04 is tuned to optimise the
control action. This is a cascade control loop with the liquid level controller as the primary
controller and the flow controller as the secondary controller. The simulation flowsheet for
the scrubbing section in HYSYS is shown in Figure 7.13.



School of Chemical and Minerals Engineering
213
Chapter 7: Process control



Figure 7.13: Scrubbing simulation flowsheet

Figure 7.13 shows the cascade control loop in the red circle which is used to control the
liquid level in the tank. In a cascade control loop it is important to ensure that the secondary
control loop has a faster (approximately four times faster) response time than that of the
primary control loop. If this is achieved the primary controller is not influenced by the control
of the secondary controller and therefore cascade control loop will function properly. Since
this control loop is a cascade control loop it is tuned by following the procedure below
(Svrcek et al., 2008: 134).

Place the primary control loop in the manual control mode.
Tune the secondary control loop as an independent loop.
Place the primary control loop back in the automatic control mode.
Tune the primary control loop normally.

Using this procedure, the cascade control loop is tuned and the effect is simulated. Figure
7.14 shows the liquid level control loop results.


School of Chemical and Minerals Engineering
214
Chapter 7: Process control


Figure 7.14: The primary control loop results

Figure 7.14 shows the liquid level in the tank as the black line, while the set point for the
liquid level is indicated by the red line. When the set point is increased with 5 %, the control
response is fast, with an overshoot of approximately 39 % and a decay ratio of 0.08. The
high overshoot is not desired, but the small decay ratio shows adequate control. The rise
time is approximately 4 minutes and the response time 64 minutes. When the set point of
the liquid level is decreased with 5 %, a slower control action is observed. The overshoot for
this step is 27 % with a decay ratio of 0.18. The rise time for the downward step is 28
minutes with a response time of 64 minutes. The less aggressive control action of the
downward step is observed due to the limit on the exit stream flow caused by the pump. In
School of Chemical and Minerals Engineering
215
Chapter 7: Process control
Figure 7.15 the set point and process variable (feed flow rate) of the secondary control loop
is shown for the same action as in Figure 7.14.



Figure 7.15: The secondary control loop results.

Figure 7.15 shows that the process variable in the secondary control loop follows the varying
set point closely and this implies a fast control action. This results in the effective control
action of the primary loop seen in Figure 7.14. All the control loops seen in Figures 7.11 and
7.15 are tuned using the auto-tuner function in HYSYS to simplify and aid the simulation. A
list of the controller parameters obtained for each of these control loops are given in Table
7.8.
School of Chemical and Minerals Engineering
216
Chapter 7: Process control

Table 7.8: Control loop tuning values
Control loop Description K
C
T
i
T
D

Extraction
FIC-100 Level control on tank U03-ST02 0.10000 0.01000 -
FIC-101 OK liquor feed control to U03-MSE01 0.04860 0.00896 -
FIC-102 Organic feed control to U03-MSE03 0.05890 0.01120 -
LIC-100 Level control on tank U03-ST01 62.10000 1.31000 -
LIC-102 Level control on tank U03-ST03 1.00000 0.01000 -
Scrubbing
FIC-100 Control demin water feed to U03-MSS03 0.14400 0.00852 -
FIC-101 Control loaded organic feed to U01-MSS01 0.03940 0.00900 -
FIC-102 Cascade level control on tank U03-ST04 0.11800 0.01090 -
LIC-102 Cascade level control on tank U03-ST04 14.80000 1.47000 0.32600

After all these control loops are tuned, the simulation will run smoothly and possible
modifications are analysed. With all the short-comings of the HYSYS simulation it still
serves as a good indication of the validity of the control strategies.

7.4.2. Variable pairing

In the control of the solvent extraction unit there are several systems with multi-variable
inputs to control the process conditions. These single-loop control schemes may interact
and therefore influence each other. The combination of variables that results in the best
overall control is the combination of control loops that have the smallest influence on each
other. In many cases this pairing is obtained through logic reasoning, however there are
methods to indicate the best combination of variable pairing (Svrcek et al., 2008: 215).

The control system on the caustic storage tank (U03-ST07), used to control the pH in the
mixers, has a liquid level and pH control loop on the tank. It is decided to control these
parameters using the caustic and demineralised water feed streams to the tank. As an
example of variable pairing, these control loops are analysed to obtain the optimum variable
pairing. Before the variable pairing is done, the system is simulated with an appropriate
mathematical model in Excel. The HYSYS simulations are not used since the available
software package is incapable of simulating electrolyte systems.

School of Chemical and Minerals Engineering
217
Chapter 7: Process control
In this simulation the height of the liquid level and the pH of the tank contents are calculated
as functions of the caustic and demineralised water feed. In this simulation the following
assumptions are made:

Steady state conditions
1.5 mole/L NaOH in caustic feed
K
w
= 10
-14

Demineralised water pH = 7

The system is then solved to satisfy mass balances and equilibrium conditions. The variable
pairing is obtained by determining the relative gain array (RGA) for the system. This is done
by increasing the demineralised water volumetric feed to the tank with 20 % and recording
the pH and liquid level change. This is also done for a 20 % change in the caustic feed.
From these results a RGA is determined and given in Table 7.9 (Svrcek et al., 2008: 218-
219).

Table 7.9: Relative gain array for control on U03-ST07

Caustic feed Demineralised water feed
pH 1 0
Liquid level 0 1

From Table 7.9 it is clear that the pH of the tank must be controlled with the caustic feed and
the liquid level in the tank must be controlled with the demineralised water feed. In this
control scheme the two control loops has a minimal influence on each other. This is
expected from logical reasoning, since the demineralised water feed stream to the tank has
a little influence on the pH compared to the caustic feed. The Niederlinski index described
by Svrcek et al., (2008) estimates the stability of a control scheme in the dynamic state. The
Niederlinski index for this system is calculated as one (positive), which concludes that this
variable pairing control scheme is stable in dynamic state.


School of Chemical and Minerals Engineering
218
Chapter 7: Process control
7.5. Conclusion

The control strategies for the plant wide control for the uranium extraction plant are able to
control all important process parameters. These strategies maintain the desired process
conditions and ensure the safety of the plant personnel, equipment and the environment.
The detail discussion of the control strategies for the solvent extraction unit shows that the
intended control is sufficient and also enables the optimisation of the process.

The simulation of the control system on the solvent extraction unit in HYSYS gave a better
insight into specific obstacles in the control. From this simulation it is for instance decided to
increase the volume of specific storage tanks in the solvent extraction unit to provide greater
capacitance to the system. This control system for the uranium extraction plant is a
conceptual system which serves as a starting point for further work to create a proper
functioning control system.






School of Chemical and Minerals Engineering
219
References

References

ACAT. 2006. Technical information: Magnafloc 90L.
www.acat.com
Date of use: 5 August 2009.

AIR PRODUCTS. 1998. Material safety data sheet: Nitric oxide.
www.avogadro.chem.iastate.edu/MSDS/nitric_oxide.pdf.
Date of use: 5 August 2009.

ANGLOGOLD ASHANTI. 2007. Radiation monitoring. AngloGold Ashanti report to
society Occupational safety and health. 25 53p.

AVVARU, B., ROY, S.B., LADOLA, Y., CHOWDHURY, S., HAREENDRAN, K.N. & PANDIT,
A.B. 2007. Sono-chemical leaching of uranium. Chemical Engineering and Processing,
47(2008):2107-2113, November.

BASF. 2006. Safety data sheet: Isodecanol.
www.worldaccount.basf.com
Date of use: 5 August 2009.

BATEMAN. 2004. Aflease uranium plant to utilise pressure leacing. p.3

BHP BILLITON. 2009. Olympic dam expansion draft environmental impact statement 2009.
665 - 692p.

BOC GASES. 1996. Material safety data sheet: Ammonia.
www.brown.edu/Departments/Visual_Art/documents/Ammonia.pdf
Date of use: 5 August 2009.

BRITISH GEOLOGICAL SURVEY. 2007. Uranium: mineral profile. Natural environment
research counsel. 33p.



School of Chemical and Minerals Engineering
220
References
BROWN, R.A. 1982. Solvay mining: Uranium.
http://www.solvayh2o2-mining.com
Date of use: 04 November 2009

BRUGESS, L. 2008. A copper mining company to watch.
http://www.wealthdaily.com
Date of use: 19 August 2009

CARTWRIGHT, P.F.S., NEWAN, E.J. & WILSON, D.W. 1967. Precipitation from
homogeneous solution: A review. The analyst, 92(1100): 664 679, Nov.

CEMI, L. 2005. Thermodynamics in minerals sciences: an introduction. Berlin: Springer.
367p.

CLEVELAND, C.J. 2008. Uranium. Encyclopedia of Earth.
http://www.eoearth.org/article/Uranium.
Date of use: 04 November 2009.

COGNIS CORPORATION. 2007. Material safety data sheet: Alamine 336.
www.apps.cognis.com
Date of use: 5 August 2009.

CUTIFANI, M. 2009. Values booklet. AngloGold Ashanti.
http://www.anglogold.co.za/Values/Values.htm.
Date of use: 13 October 2009.

DAMARUPURSHAD, A. 2007. Uranium: future sources (South Africa). Department:
Minerals and Energy. Republic of South Africa. 16p.

DEPARTMENT: Minerals and Energy see SOUTH AFRICA. Department: Minerals and
Energy

DOUGLAS, J.M. 1988. Conceptual design of chemical processes. Singapore: McGraw-Hill
Book Co. 601p.


School of Chemical and Minerals Engineering
221
References
ENERGY INFORMATION ADMINISTRATION. 2008. International energy annual 2006.
www.eia.doe.gov/iea/wecbtu.html
Date of use: 4 August 2009

ENGHAG, P. 2004. Encyclopedia of the elements. Wiley-VCH Weinheim. 1243p.

FACILITY FOR THE ANALYSIS OF CHEMICAL THERMODYNAMICS. 2009. Reaction-
web properties of a species or chemical reaction.
http://www.crct.polymtl.ca
Date of use: 03 November 2009

FLEMING, C.A. & NICOL, M.J. 1980.A comparative study of kinetic models for the extraction
of uranium by strong-base anion-exchange resins.

FENTIMAN, A.W., KARAM, P.A. & MEYERS, R.B. 2009. Low-level radioactive waste from
decontamination and decommissioning of facilities that use radioactive materials. Factsheet
Low-level radioactive wate, 14:5.

FOGLER, H.S. 2006. Elements of chemical reaction engineering. 4
th
ed. Westford,
Massachusetts. 1080p.

FUJI ELECTRIC SYSTEMS CO., LTD. 1989. Zener barrier data sheet. Japan. 6p.

GOLDPRICE. 2009. Gold price history.
http://goldprice.org/gold-price-history.html#5_year_gold_price
Date of use: 4 August 2009

GRADY, C.P.L., DAIGGER, G.T. & LIM, H.C. 1999. Biological wastewater treatment. 2
nd

edition. New York: Marcel Dekker, Inc. 1076p.

GUPTA, R., PANDEY, V.M., PRANESH, S.R. & CHAKRAVARTY, A.B. 2004. Study of an
improved technique for precipitation of uranium from eluted solution. Hydrometallurgy, 71(3-
4): 429-434, Jan.

GUPTA, C.K. 2003. Chemical metallurgy: Principles and practice. Weinheim: Wiley-VCH
Verlag GmbH & Co. 811p.

School of Chemical and Minerals Engineering
222
References
GUPTA, C.K. & SINGH, H. 2003. Uranium resource processing: Secondary resources.
Springer. 519p.

HAUSEN, D.M. 2009. Yellow cake.
www.wikipedia.com
Date used: 11/08/2009
HEMINGWAY, B.S. 1987. Quartz: heat capacities from 340 to 1000 K and revised values
for the thermodynamic properties. American Mineralogist, 72:273-279, 1987.

IAEA. 2009. Standards and guides
www.iaea.org
Date of use: 14 October 2009

IAEA SAFETY STANDARDS. 2006. Decommissioning of facilities using radioactive
material. Vienna. 25p.

IKEDA, Y., YASUIKE, Y., NISHIMURA, K., HASEGAWA, S. & TAKASHIMA, Y. 1995.
Kinetic study on dissolution of UO
2
powders in nitric acid. Journal of Nuclear Materials, 224
(1995): 266-272, March.

INTERNATIONAL BIO-ANALYTICAL INDUSTRIES, INC. 2006. Material safety data sheet:
Ammonium diuranate.
www.ibilabs.com
Date of use: 5 August 2009.

JANOV, J., ALFREDSON, P.G. & VILKAITIS, V.K. 1971. The influence of precipitation
conditions on the properties of ammonium diuranate and uranium dioxide powders.
Australian atomic energy commission: Research establishment. 42p.

JOHNSON, T. 2007. Global uranium supply and demand.
http://www.cfr.org/publication/14705/
Date of use: 4 August 2009

KARACA, S., KADIOGLU, Y.Y., & BAYRAKCEKEN, S. 1994. Kinetics of pyrite oxidation in
aqueous suspension by nitric acid. Fuel Processing Technology, 41(1995): 273-287,
August.

School of Chemical and Minerals Engineering
223
References
KEMPSTER, P.L. 1999. Report on the radioactivity monitoring programme in the Mooi
River (Wonderfonteinspruit) catchment. Institute for water quality studies. Department of
water affairs and forestry.
http://www.dwaf.gov.za/iwqs/radioact/mooi/summary.htm.
Date of use: 8 August 2009.

KORDOSKY, G., VIRNIG, M., & MACKENZIE, M. Solvent extraction reagents and
selectivity control. Henkel kgaA. 20p.

KORETSKY, M.D. 2004. Engineering and chemical thermodynamics. USA: John Wiley &
Sons, Inc. 553p.

LAW, J.D. & TODD. A. Liquid-liquid extraction equipment. Idaho National Laboratory. 14p.

LENNTECH. 2008. Lenntech water treatment & air purification holding B.V. Chemical
properties of uranium Health effects of uranium Environmental effects of uranium.
http://www.lenntech.com/Periodic-chart-elements/U-en.htm.
Date of use: 8 August 2009.

LEVENSPIEL, O. 1999. Chemical reaction engineering. 3
rd
ed. New York: John Wiley and
Sons. 668p.

LINDEMANN, I. 2008. Hazards of uranium. 17p.

LOTTERING, M.J., LORENZEN, L., PALA, N.S., SMIT, J.T. & SCHALKWYK, G.A.C. 2007.
Mineralogy and uranium leaching response of low grade South African ores. Minerals
Engineering. 21 (2008) 16-22

LOZOWSKI, D. 2009. Economic indicators. Chemical Engineering, 116(8): 56, Aug.

LUNT, D. & HOLDEN, A. 2006. Uranium extraction: the key issues for processing.
GRDMinproc limeted. p.20

LURIE, W., PINKNEY, E.T. & VAN ZYL, P.C.N. 1962. Chemical processing of uranium ores.
Vienna: International Atomic Energy Agency. p.148

School of Chemical and Minerals Engineering
224
References
MACKENZIE, J.M.W. 1997. Uranium solvent extraction using tertiary amines. Uranium Ore
Yellow Cake Seminar, Melbourne, Australia. February 1997.

MADHAVAN, R. 2008. Optimize liquid-liquid extraction.
http://www.cheresources.com/extraction.shtml
Date used: 12/08/2009

MANANA, W. ([email protected]). 26 October 2009. Spot prices for ADU
and gold. E-mail to Nel, A. ([email protected]).

MERRITT, R.C. 1971. The extractive metallurgy or Uranium. Colorado school of mines
research institute. 576p.

METHANOL INSTITUTE. 2009. Wastewater treatment with methanol denitrification.
http://www.methanol.org.
Date of use: 27 October 2009.

MINERALS COUNCIL OF AUSTRALIA. 2006. Hydrometallurgy: Theory and practice.
Murdoch: Murdoch University. 87 p.

MORAIS, C.A. & GOMIERO, L.A. 2005. Uranium stripping from tertiary amine loaded
solution by ammonium sulfate. Minerals Engineering, 18(2005): 1277-1281.

MUDD, G.M. 2000. Acid in situ leach uranium mining: 2 Soviet block and Asia. Tailings &
Mine Waste 00. 23(26): 527-536, January

MURTY, B.N., BALAKRISHNA, P., YADAV, R.B. & GANGULY, C. 2002. New approaches
in the characterization of uranium dioxide. Materials Characterization, 49(2003): 103-111,
June.

NECSA. 2009. Nuclear waste.
www.necsa.co.za
Date of use: 14 October 2009

NECSA. 2009. South African fundamental atomic research installation. Brochure handed
out at SAFARI-1 reactor. Date obtained: 28 July 2009

School of Chemical and Minerals Engineering
225
References
NEOMAGUS, H.W.J.P. 2008. CEMI311: Fluid dynamics I dictate. South Africa: North-West
University. 87p.

NUNNGLOW. 2008. In-situ leaching (ISL) impacts.
http://www.nunnglow.com
Date of use: 20 August 2009

OECD NUCLEAR ENERGY AGENCY & INTERNATIONAL ATOMIC ENERGY AGENCY.
2007. Uranium 2007: resources, production and demand. USA: Cameco. p.420

PERRY, R.H., GREEN, D.W. & MALONEY, J.O. 1997. Perrys chemical engineers
handbook. 7th ed. McGraw Hill.

PETERS, M.S., TIMMERHAUS, K.D. & WEST, R.E. 2004. Plant design and economics for
chemical engineers. 5
th
ed. New York: McGraw Hill. 988p.

PHOTIUS. 2009. Central bank discount rate (%) 2009 Country Ranks.
http://www.photius.com
Date of use: 26 October 2009.

RYDBERG, J., COX, M., MUSIKAS, C., & CHOPPIN, G.R. 1992. Solvent extraction
principles and practices. 2
nd
ed. New York: Marcel Dekker, Inc.

SCIENCE LAB. 2008. Material safety data sheets.
http://www.sciencelab.com
Date of use: 5 August 2009.

SEADER, J.D. & HENLEY, E.J. 2006. Separation process principles. 2
nd
ed. New York:
John Wiley and Sons. 756p.

SENQU MUNICIPALITY. 2009. 2008/2009 Tariff structure.
www.senqumunicipality.co.za
Date of use: 26 October 2009.

SERRA, A. & SOL, C. 1986. Control of water softening process by ion exchange using
electrical conductivity measurements. Industrial and engineering chemistry process design
and development, 25(1):31-35.
School of Chemical and Minerals Engineering
226
References
SKIPTON, S.O., DVORAK, B.I., BAUMERT, F. & WOLDT, W. 2008. Neb guide. Drinking
water: uranium. University of Nebraska-Lincoln Extension. 4p.

SNLL, S. & LILJEFORS, T. 2000. Leachability of major element from minerals in strong
acids. Journal of Geochemical Exploration, 71(2000): 1-12, March.

SNOOPER. 2008. US removes 550 tons of yellow cake uranium from Iraq.
http://patdollard.com/2008/07/ap-exclusive-us-removes-550-tons-of-yellow-cake-uranium-
from-iraq/
Date used: 11/08/2009

SOUTH AFRICA. Department: Minerals and Energy. 2008. South Africas mineral industry
2007/2008. Pretoria: Mineral Economics. 182 p.

STNNER, H.M. & WIESNER, P. 1982. The design of equipment for a solvent-extraction
plant for the recovery of metal salts. Journal of the South African Institute of Mining and
Metallurgy. April 1982, 97-105.

SVRCEK, W.Y., MAHONEY, D.P., & YOUNG, B.R. 2006. A real-time approach to process
control. 2
nd
ed. England: John Wiley & Sons Ltd. 325p.

THE LOW TAX NETWORK. 2009. South Africa: Summary of domestic taxation regime.
http://www.lowtax.net
Date of use: 26 October 2009.

TUTOR VISTA. 2008. Determination of enthalpy change.
www.turorvista.com
Date of use: 03 November 2009

ULRICH, G.D., & VASUDEVAN, P.T. 2009. Capital costs quickly calculated. Chemical
Engineering, 116(4): 46-52

VENTER, R. & BOYLETT, M. The evaluation of various oxidants used in acid leaching of
uranium. Hydrometallurgy Conference 2009, The South African Institute of Mining and
Metallurgy, 2009.

VILJOEN, F. 2009. CPI inflation in South Africa.
School of Chemical and Minerals Engineering
227
References
http://liberta.co.za
Date of use: 26 October 2009.

WATSON, P. 2009. Solvent extraction equipment design. MC Process in Glenvista,
Johannesburg.

WEBER, A.S. 1996. Hazardous and industrial wastes. USA: Technomic Publishing
Company, Inc. 809p.

WEISS, N.L. 1985. SME Mineral processing handbook. American Institute of Mining,
Metallurgical, and Petroleum Engineers Inc. Volume 1 & 2.

WENDEL, G. 1998. The Journal of the South African institute of mining and metallurgy.
Radioactivity in mines and mine water sources and mechanisms. 92p.

WILSON, P.D. 1996. The nuclear fuel cycle: from ore to wastes. Oxford science
publications. 323p.

WORLD NUCLEAR ASSOCIATION. 2009. Nuclear power in France.
www.world-nuclear.org/info/inf40.html
Date of use: 4 August 2009

WORLD NUCLEAR ASSOCIATION. 2009. World uranium mining.
http://www.world-nuclear.org/info/inf23.html
Date of use: 4 August 2009

WORLD NUCLEAR ASSOCIATION. 2009. Nuclear power in South Africa.
http://www.world-nuclear.org/info/inf88.html
Date of use: 4 August 2009

WORLD NUCLEAR ASSOCIATION. 2008. Environmental aspects of uranium mining.
http://www.world-nuclear.org.
Date of use: 8 August 2009.

X-RATES.COM. 2009. South African rand.
http://www.x-rates.com
Date of use: 26 October 2009.
School of Chemical and Minerals Engineering
228
References

ZHAO, Y. & CHEN, J. 2007. Studies on the dissolution kinetics of ceramic uranium dioxide
particles in nitric acid by microwave heating. Journal of Nuclear Materials, 373(2008): 53-58,
March.



School of Chemical and Minerals Engineering
229
Appendix A: Mass and energy balances

Appendix A: Mass and energy balances

A.1. Mass balance

The individual mass balances for the separate units are given in Appendix A. In these mass balances all the materials present in the
entire system is tabulated and the mass flows of the materials that is present in the specific unit is given in Table A.2 to A.7. The
stream names given correspond to the stream names used in the Aspen Tech

simulation and a short description of each stream is

given after the table. This simulation is shown in Figure A.1.

A summary of all the process units is given in Table A.1 with the mass flow in and out of the unit. As seen in Table A.1 the errors are
very small which is an indication that the inflow and outflow of the processes match and the law of conservation is confirmed.

Table A.1: Summary of mass balances for individual units
Process unit Mass flow in (kg/hr) Mass flow out (kg/hr) % error
Leaching 925 163 925 163.1 3.68E-08
CCD 1 526 960 1 526 960 6.044E-08
Neutralization 992 488 992 487.8 1.01E-08
Ion exchange 629 626 629 612.1 2.14E-03
Solvent extraction 41 815.8 41 836.78 5.01E-02
Precipitation 10 355.1 10 355.08 8.94E-06
School of Chemical and Minerals Engineering
230
Appendix A: Mass and energy balances

Figure A.1: Aspen Tech

simulation
School of Chemical and Minerals Engineering
231
Appendix A: Mass and energy balances
Table A.2: Mass balance for leaching process
IN

OUT
Mass flow kg/hr

Mass flow kg/hr

ORE HNO3 H2SO4 LOAD

LEACHPRO

880 000 2 498.15 15 323.52 27 341.5

925 163.1

Liquids Solids Liquids Solids Liquids Solids Liquids Solids

Liquids Solids
H2O 350000 0 648.148 0 0.055 0 24064 0

380110 0
SIO2 0 379149 0 0 0 0 0 0

0 383824
MUSCV 0 54550 0 0 0 0 0 0

0 51167.9
CLINOCL 0 10802 0 0 0 0 0 0

0 4320.79
PYROPHLT 0 52389.6 0 0 0 0 0 0

0 49770.1
PYRITE 0 7021.28 0 0 0 0 0 0

0 6810.65
ALBITE 0 25924.7 0 0 0 0 0 0

0 25665.5
UO2 0 75.614 0 0 0 0 0 0

0 0.03025
UTI2O6 0 70.213 0 0 0 0 0 0

0 30.8936
UCLPO6 0 5.401 0 0 0 0 0 0

0 0.10802
UO2SIO2 0 10.802 0 0 0 0 0 0

0 0.21604
UO2-2W 0 1.08 0 0 0 0 0 0

0 0.0216
U3O8 0 0 0 0 0 0 0 0

0 0
TIO2 0 0 0 0 0 0 0 0

0 14.6155
K+ 0 0 0 0 0 0 0 0

331.985 0
AL+++ 0 0 0 0 0 0 0 0

2233.55 0
MG++ 0 0 0 0 0 0 0 0

507.751 0
School of Chemical and Minerals Engineering
232
Appendix A: Mass and energy balances
FE++ 0 0 0 0 0 0 0 0

1166.72 0
FE+++ 0 0 0 0 0 0 0 0

0 0
FESO4- 0 0 0 0 0 0 74.527 0

509.879 0
NA+ 0 0 0 0 0 0 0 0

22.7283 0
UO2++ 0 0 0 0 0 0 0 0

10.3293 0
US2O10-- 0 0 0 0 0 0 0 0

176.491 0
US3O14-- 0 0 0 0 0 0 4.785 0

4.78546 0
H3O+ 0.001 0 159.719 0 323.55 0 982.086 0

165.462 0
OH- 0.001 0 0 0 0 0 0 0

1.84E-08 0
CL- 0 0 0 0 0 0 0 0

0.4686 0
PO4--- 0 0 0 0 0 0 0 0

1.25531 0
H2SO4 0 0 0 0 13348.8 0 0 0

2.30E-08 0
HSO4- 0 0 0 0 1651.07 0 0 0

2526.22 0
SO4-- 0 0 0 0 0 0 2216.1 0

14351.3 0
HNO3 0 0 1169.67 0 0 0 0 0

0.03813 0
HNO2 0 0 0 0 0 0 0 0

0 0
NO3- 0 0 520.612 0 0 0 0 0

1076.51 0
NO2 0 0 0 0 0 0 0 0

37.1426 0
NO 0 0 0 0 0 0 0 0

263.728 0
R-NO3 0 0 0 0 0 0 0 0

0 0
R2-UC 0 0 0 0 0 0 0 0

0 0
R2-SO4 0 0 0 0 0 0 0 0

0 0
R-FEC 0 0 0 0 0 0 0 0

0 0
School of Chemical and Minerals Engineering
233
Appendix A: Mass and energy balances
L4US3O14 0 0 0 0 0 0 0 0

0 0
L2-SO4 0 0 0 0 0 0 0 0

0 0
KEROSENE 0 0 0 0 0 0 0 0

0 0
ISODEC 0 0 0 0 0 0 0 0

0 0
NH3 0 0 0 0 0 0 0 0

0 0
AMMON-01 0 0 0 0 0 0 0 0

0 0
NH4+ 0 0 0 0 0 0 0 0

0 0
ADU 0 0 0 0 0 0 0 0

0 0
H4SIO 0 0 0 0 0 0 0 0

61.4052 0
NH4NO3 0 0 0 0 0 0 0 0

0 0
LIME 0 0 0 0 0 0 0 0

0 0
CA++ 0 0 0 0 0 0 0 0

0 0
CAOH+ 0 0 0 0 0 0 0 0

0 0
CA(OH)2 0 0 0 0 0 0 0 0

0 0
CALCI(S) 0 0 0 0 0 0 0 0

0 0
AL(OH)3 0 0 0 0 0 0 0 0

0 0
FE(OH)2 0 0 0 0 0 0 0 0

0 0
MG(OH)2 0 0 0 0 0 0 0 0

0 0
Total 350000 530000 2498.15 0 15323.5 0 27341.5 0

403558 521605

Ore = The ore feed received from the mine, already mixed with water Leachpro = Leach product
HNO
3
= 68% Nitric acid feed
H
2
SO
4
= 98% Sulphuric acid feed
Load = Solvent extraction recycle
School of Chemical and Minerals Engineering
234
Appendix A: Mass and energy balances
Table A.3: Mass balance for CCD process

IN

OUT

Mass flow kg/hr

Mass flow kg/hr

LEACHPRO 6 21

SOLIDS PREG-OF

925 163.14 601 796 0.338

924 903.6 602 056.5

Liquids Solids Liquids Solids Liquids Solids

Liquids Solids Liquids Solids
H2O 380110.45 0 567211.1 0 0.998 0

380121.2 0 567186.9 0
SIO2 0 383824.16 0 0.002 0 0

0 383824.1 0 0.045
MUSCV 0 51167.877 0 0 0 0

0 51167.87 0 0.006
CLINOCL 0 4320.79 0 0 0 0

0 4320.79 0 0.001
PYROPHLT 0 49770.101 0 0 0 0

0 49770.1 0 0.006
PYRITE 0 6810.645 0 0 0 0

0 6810.645 0 0.001
ALBITE 0 25665.493 0 0 0 0

0 25665.49 0 0.003
UO2 0 0.03 0 0 0 0

0 0.03 0 0
UTI2O6 0 30.894 0 0 0 0

0 30.894 0 0
UCLPO6 0 0.108 0 0 0 0

0 0.108 0 0
UO2SIO2 0 0.216 0 0 0 0

0 0.216 0 0
UO2-2W 0 0.022 0 0 0 0

0 0.022 0 0
U3O8 0 0 0 0 0 0

0 0 0 0
TIO2 0 14.616 0 0 0 0

0 14.616 0 0
K+ 331.985 0 495.384 0 0 0

331.985 0 495.384 0
AL+++ 2233.547 0 3332.871 0 0 0

2233.547 0 3332.871 0
MG++ 507.751 0 757.66 0 0 0

507.751 0 757.66 0
School of Chemical and Minerals Engineering
235
Appendix A: Mass and energy balances
FE++ 1166.719 0 1740.963 0 0 0

1166.719 0 1740.963 0
FE+++ 0 0 0 0 0 0

0 0 0 0
FESO4- 509.879 0 637.268 0 0 0

427.07 0 720.077 0
NA+ 22.728 0 33.915 0 0 0

22.728 0 33.915 0
UO2++ 10.329 0 0.001 0 0 0

0.001 0 9.828 0
US2O10-- 176.491 0 0.018 0 0 0

0.012 0 177.357 0
US3O14-- 4.785 0 0 0 0 0

0 0 4.785 0
H3O+ 165.462 0 269.336 0 0 0

180.496 0 269.605 0
OH- 0 0 0 0 0 0

0 0 0 0
CL- 0.469 0 0.699 0 0 0

0.469 0 0.699 0
PO4--- 1.255 0 1.873 0 0 0

1.255 0 1.873 0
H2SO4 0 0 0 0 0 0

0 0 0 0
HSO4- 2526.22 0 3655.107 0 0 0

2449.504 0 3653.731 0
SO4-- 14351.285 0 21603.75 0 0 0

14477.9 0 21554.06 0
HNO3 0.038 0 0.055 0 0 0

0.037 0 0.055 0
HNO2 0 0 0 0 0 0

0 0 0 0
NO3- 1076.514 0 1606.363 0 0 0

1076.516 0 1606.363 0
NO2 37.143 0 55.424 0 0 0

37.143 0 55.424 0
NO 263.728 0 393.531 0 0 0

263.728 0 393.531 0
R-NO3 0 0 0 0.004 0 0

0 0.004 0 0
R2-UC 0 0 0 0.062 0 0

0 0.062 0 0
R2-SO4 0 0 0 0.547 0 0

0 0.546 0 0
R-FEC 0 0 0 0.024 0 0

0 0.024 0 0
School of Chemical and Minerals Engineering
236
Appendix A: Mass and energy balances
L4US3O14 0 0 0 0 0 0

0 0 0 0
L2-SO4 0 0 0 0 0 0

0 0 0 0
KEROSENE 0 0 0 0 0 0

0 0 0 0
ISODEC 0 0 0 0 0 0

0 0 0 0
NH3 0 0 0 0 0 0

0 0 0 0
AMMON-01 0 0 0 0 0 0

0 0 0 0
NH4+ 0 0 0 0 0 0

0 0 0 0
ADU 0 0 0 0 0 0

0 0 0 0
H4SIO 61.405 0 0 0 0 0

0 0 61.405 0
NH4NO3 0 0 0 0 0 0

0 0 0 0
LIME 0 0 0 0 0 0

0 0 0 0
CA++ 0 0 0 0 0 0

0 0 0 0
CAOH+ 0 0 0 0 0 0

0 0 0 0
CA(OH)2 0 0 0 0 0 0

0 0 0 0
CALCI(S) 0 0 0 0 0 0

0 0 0 0
AL(OH)3 0 0 0 0 0 0

0 0 0 0
FE(OH)2 0 0 0 0 0 0

0 0 0 0
MG(OH)2 0 0 0 0 0 0

0 0 0 0
Total 403558.19 521604.95 601795.4 0.639 0.998 0

403298.1 521605.5 602056.4 0.061

Leachpro = Leach product Solids = Solids to neutralization
6 = Barren liquor Preg-of = Pregnant leach liquor
21 = Wash water make-up

School of Chemical and Minerals Engineering
237
Appendix A: Mass and energy balances
Table A.4: Mass balance for neutralization process

IN

OUT

Mass flow kg/hr

Mass flow kg/hr

SOLIDS LIME

GOLD

924 903.6 67 584.13

992 487.8

Liquids Solids Liquids Solids

Liquids Solids
H2O 380121.2 0 57398 0

435043.3 0
SIO2 0 383824.1 0 0

0 383824.1
MUSCV 0 51167.87 0 0

0 51167.87
CLINOCL 0 4320.79 0 0

0 4320.79
PYROPHLT 0 49770.1 0 0

0 49770.1
PYRITE 0 6810.645 0 0

0 6810.645
ALBITE 0 25665.49 0 0

0 25665.49
UO2 0 0.03 0 0

0 0.03
UTI2O6 0 30.894 0 0

0 30.894
UCLPO6 0 0.108 0 0

0 0.108
UO2SIO2 0 0.216 0 0

0 0.216
UO2-2W 0 0.022 0 0

0 0.022
U3O8 0 0 0 0

0 0
TIO2 0 14.616 0 0

0 14.616
K+ 331.985 0 0 0

331.985 0
AL+++ 2233.547 0 0 0

0 0
MG++ 507.751 0 0 0

46.205 0
School of Chemical and Minerals Engineering
238
Appendix A: Mass and energy balances
FE++ 1166.719 0 0 0

0 0
FE+++ 0 0 0 0

0 0
FESO4- 427.07 0 0 0

427.07 0
NA+ 22.728 0 0 0

22.728 0
UO2++ 0.001 0 0 0

0.001 0
US2O10-- 0.012 0 0 0

0.012 0
US3O14-- 0 0 0 0

0 0
H3O+ 180.496 0 0 0

0 0
OH- 0 0 0 0

4.411 0
CL- 0.469 0 0 0

0.469 0
PO4--- 1.255 0 0 0

1.255 0
H2SO4 0 0 0 0

0 0
HSO4- 2449.504 0 0 0

0 0
SO4-- 14477.9 0 0 0

16901.98 0
HNO3 0.037 0 0 0

0 0
HNO2 0 0 0 0

0 0
NO3- 1076.516 0 0 0

1076.552 0
NO2 37.143 0 0 0

37.143 0
NO 263.728 0 0 0

263.728 0
R-NO3 0 0.004 0 0

0 0.004
R2-UC 0 0.062 0 0

0 0.062
R2-SO4 0 0.546 0 0

0 0.546
R-FEC 0 0.024 0 0

0 0.024
School of Chemical and Minerals Engineering
239
Appendix A: Mass and energy balances
L4US3O14 0 0 0 0

0 0
L2-SO4 0 0 0 0

0 0
KEROSENE 0 0 0 0

0 0
ISODEC 0 0 0 0

0 0
NH3 0 0 0 0

0 0
AMMON-01 0 0 0 0

0 0
NH4+ 0 0 0 0

0 0
ADU 0 0 0 0

0 0
H4SIO 0 0 0 0

0 0
NH4NO3 0 0 0 0

0 0
LIME 0 0 0 10186.13

0 0
CA++ 0 0 0 0

7272.343 0
CAOH+ 0 0 0 0

10.529 0
CA(OH)2 0 0 0 0

0 0
CALCI(S) 0 0 0 0

0 0
AL(OH)3 0 0 0 0

0 6457.574
FE(OH)2 0 0 0 0

0 1877.368
MG(OH)2 0 0 0 0

0 1107.526
Total 403298.1 521605.5 57398 10186.13

461439.8 531048

Solids = Solids to neutralization Gold = Solids to gold plant
Lime = Lime used for neutralization
School of Chemical and Minerals Engineering
240
Appendix A: Mass and energy balances
Table A.5: Mass balance for ion exchange process

IN

OUT

Mass flow kg/hr

Mass flow kg/hr

PREG-OF RES-MU ELUANT WATERSIO 6 ELUATE SIO2-OUT

602 056.5 15.12 27 528.58 25.35 601 796 27 752.28 63.78

Liquids Solids Liquids Solids Liquids Liquids

Liquids Solids Liquids Liquids
H2O 567186.9 0 1 0 24451.18 25.352

567211.1 0 24451.99 2.307
SIO2 0 0.045 0 0 0 0

0 0.002 0 0
MUSCV 0 0.006 0 0 0 0

0 0 0 0
CLINOCL 0 0.001 0 0 0 0

0 0 0 0
PYROPHLT 0 0.006 0 0 0 0

0 0 0 0
PYRITE 0 0.001 0 0 0 0

0 0 0 0
ALBITE 0 0.003 0 0 0 0

0 0 0 0
UO2 0 0 0 0 0 0

0 0 0 0
UTI2O6 0 0 0 0 0 0

0 0 0 0
UCLPO6 0 0 0 0 0 0

0 0 0 0
UO2SIO2 0 0 0 0 0 0

0 0 0 0
UO2-2W 0 0 0 0 0 0

0 0 0 0
U3O8 0 0 0 0 0 0

0 0 0 0
TIO2 0 0 0 0 0 0

0 0 0 0
K+ 495.384 0 0 0 0 0

495.384 0 0 0
AL+++ 3332.871 0 0 0 0 0

3332.871 0 0 0
MG++ 757.66 0 0 0 0 0

757.66 0 0 0
School of Chemical and Minerals Engineering
241
Appendix A: Mass and energy balances
FE++ 1740.963 0 0 0 0 0

1740.963 0 0 0
FE+++ 0 0 0 0 0 0

0 0 0 0
FESO4- 720.077 0 0 0 0 0

637.268 0 82.808 0
NA+ 33.915 0 0 0 0 0

33.915 0 0 0
UO2++ 9.828 0 0 0 0 0

0.001 0 75.942 0
US2O10-- 177.357 0 0 0 0 0

0.018 0 68.141 0
US3O14-- 4.785 0 0 0 0 0

0 0 0 0
H3O+ 269.605 0 0 0 575.859 0

269.336 0 575.006 0
OH- 0 0 0 0 0 0

0 0 0 0
CL- 0.699 0 0 0 0 0

0.699 0 0 0
PO4--- 1.873 0 0 0 0 0

1.873 0 0 0
H2SO4 0 0 0 0 0 0

0 0 0 0
HSO4- 3653.731 0 0 0 2073.462 0

3655.107 0 2077.817 0
SO4-- 21554.06 0 0 0 428.076 0

21603.75 0 420.579 0
HNO3 0.055 0 0 0 0 0

0.055 0 0 0
HNO2 0 0 0 0 0 0

0 0 0 0
NO3- 1606.363 0 0 0 0 0

1606.363 0 0 0
NO2 55.424 0 0 0 0 0

55.424 0 0 0
NO 393.531 0 0 0 0 0

393.531 0 0 0
R-NO3 0 0 0 0 0 0

0 0.004 0 0
R2-UC 0 0 0 0 0 0

0 0.062 0 0
R2-SO4 0 0 0 14.123 0 0

0 0.547 0 0
R-FEC 0 0 0 0 0 0

0 0.024 0 0
School of Chemical and Minerals Engineering
242
Appendix A: Mass and energy balances
L4US3O14 0 0 0 0 0 0

0 0 0 0
L2-SO4 0 0 0 0 0 0

0 0 0 0
KEROSENE 0 0 0 0 0 0

0 0 0 0
ISODEC 0 0 0 0 0 0

0 0 0 0
NH3 0 0 0 0 0 0

0 0 0 0
AMMON-01 0 0 0 0 0 0

0 0 0 0
NH4+ 0 0 0 0 0 0

0 0 0 0
ADU 0 0 0 0 0 0

0 0 0 0
H4SIO 61.405 0 0 0 0 0

0 0 0 61.474
NH4NO3 0 0 0 0 0 0

0 0 0 0
LIME 0 0 0 0 0 0

0 0 0 0
CA++ 0 0 0 0 0 0

0 0 0 0
CAOH+ 0 0 0 0 0 0

0 0 0 0
CA(OH)2 0 0 0 0 0 0

0 0 0 0
CALCI(S) 0 0 0 0 0 0

0 0 0 0
AL(OH)3 0 0 0 0 0 0

0 0 0 0
FE(OH)2 0 0 0 0 0 0

0 0 0 0
MG(OH)2 0 0 0 0 0 0

0 0 0 0
Total 602056.4 0.061 1 14.123 27528.58 25.352

601795.4 0.639 27752.28 63.781

Preg-of = Pregnant leach liquor 6 = Barren liquor
Res-mu = Resin make-up Eluate = Eluate
Eluant = Eluant SiO2-out = Backwash water out
WaterSiO = Backwash water in
School of Chemical and Minerals Engineering
243
Appendix A: Mass and energy balances
Table A.6: Mass balance for solvent extraction process

IN

OUT

Mass flow kg/hr

Mass flow kg/hr

ELUATE 7 A-SCRUB

LOAD 16 19

27 752.28 8 459.47 5 604.08

27 341.47 5 635.98 8 859.33

Liquids Liquids Liquids

Liquids Liquids Liquids
CLINOCL 0 0 0

0 0 0
PYROPHLT 0 0 0

0 0 0
PYRITE 0 0 0

0 0 0
ALBITE 0 0 0

0 0 0
UO2 0 0 0

0 0 0
UTI2O6 0 0 0

0 0 0
UCLPO6 0 0 0

0 0 0
UO2SIO2 0 0 0

0 0 0
UO2-2W 0 0 0

0 0 0
U3O8 0 0 0

0 0 0
TIO2 0 0 0

0 0 0
K+ 0 0 0

0 0 0
AL+++ 0 0 0

0 0 0
MG++ 0 0 0

0 0 0
FE++ 0 0 0

0 0 0
FE+++ 0 0 0

0 0 0
FESO4- 82.808 0 0

74.527 8.281 0
School of Chemical and Minerals Engineering
244
Appendix A: Mass and energy balances
NA+ 0 0 0

0 0 0
UO2++ 75.942 0 0

0 0 0
US2O10-- 68.141 0 0

0 0 0
US3O14-- 0 0 0

4.785 0 234.511
H3O+ 575.006 0.002 0

982.086 0.098 0.002
OH- 0 0 0

0 0 0
CL- 0 0 0

0 0 0
PO4--- 0 0 0

0 0 0
H2SO4 0 0 0

0 0 0
HSO4- 2077.817 0.04 0

0 0 0.041
SO4-- 420.579 860.464 0

2216.1 0 1025.982
HNO3 0 0 0

0 0 0
HNO2 0 0 0

0 0 0
NO3- 0 0 0

0 0 0
NO2 0 0 0

0 0 0
NO 0 0 0

0 0 0
R-NO3 0 0 0

0 0 0
R2-UC 0 0 0

0 0 0
R2-SO4 0 0 0

0 0 0
R-FEC 0 0 0

0 0 0
L4US3O14 0 0 0

0 0 0
L2-SO4 0 0 0

0 0.047 0
KEROSENE 0 0 0

0 21.581 0
School of Chemical and Minerals Engineering
245
Appendix A: Mass and energy balances
ISODEC 0 0 0

0 0.048 0
NH3 0 0.009 0

0 0 0.009
AMMON-01 0 0 0

0 0 0
NH4+ 0 322.679 0

0 0 322.679
ADU 0 0 0

0 0 0
H4SIO 0 0 0

0 0 0
NH4NO3 0 0 0

0 0 0
LIME 0 0 0

0 0 0
CA++ 0 0 0

0 0 0
CAOH+ 0 0 0

0 0 0
CA(OH)2 0 0 0

0 0 0
CALCI(S) 0 0 0

0 0 0
AL(OH)3 0 0 0

0 0 0
FE(OH)2 0 0 0

0 0 0
MG(OH)2 0 0 0

0 0 0
Total 27752.28 8459.468 5604.08

27341.47 5635.981 8859.33

Eluate = Eluate Load = Solvent extraction recycle
7 = ADU recycle 16 = Spend demineralised water
A-scrub = Aqueous scrub water 19 = OK liquor
School of Chemical and Minerals Engineering
246
Appendix A: Mass and energy balances
Table A.7: Mass balance for precipitation process
IN

OUT
Mass flow kg/hr

Mass flow kg/hr

19 NH3 20

8 12 18 22

8 859.33 85.73 1 410.02

143.98 8 539.27 374.01 1 297.82

Liquids Liquids Liquids

Liquids Liquids Liquids Solids Liquids
H2O 7276.107 64.295 1410.022

119.723 7100.497 243.085

1275.785
SIO2 0 0 0

0 0 0 0 0
MUSCV 0 0 0

0 0 0 0 0
CLINOCL 0 0 0

0 0 0 0 0
PYROPHLT 0 0 0

0 0 0 0 0
PYRITE 0 0 0

0 0 0 0 0
ALBITE 0 0 0

0 0 0 0 0
UO2 0 0 0

0 0 0 0 0
UTI2O6 0 0 0

0 0 0 0 0
UCLPO6 0 0 0

0 0 0 0 0
UO2SIO2 0 0 0

0 0 0 0 0
UO2-2W 0 0 0

0 0 0 0 0
U3O8 0 0 0

0 0 0 0 0
TIO2 0 0 0

0 0 0 0 0
K+ 0 0 0

0 0 0 0 0
AL+++ 0 0 0

0 0 0 0 0
MG++ 0 0 0

0 0 0 0 0
School of Chemical and Minerals Engineering
247
Appendix A: Mass and energy balances
FE++ 0 0 0

0 0 0 0 0
FE+++ 0 0 0

0 0 0 0 0
FESO4- 0 0 0

0 0 0 0 0
NA+ 0 0 0

0 0 0 0 0
UO2++ 0 0 0

0.003 0.165 0 0 0
US2O10-- 0 0 0

0 0 0 0 0.004
US3O14-- 234.511 0 0

0 0 0 0 0
H3O+ 0.002 0 0

0 0 0 0 0
OH- 0 0 0

0 0 0 0 0
CL- 0 0 0

0 0 0 0 0
PO4--- 0 0 0

0 0 0 0 0
H2SO4 0 0 0

0 0 0 0 0
HSO4- 0.041 0 0

0 0 0 0 0.002
SO4-- 1025.982 0 0

18.738 1111.32 0.017 0 17.016
HNO3 0 0 0

0 0 0 0 0
HNO2 0 0 0

0 0 0 0 0
NO3- 0 0 0

0 0 0 0 0
NO2 0 0 0

0 0 0 0 0
NO 0 0 0

0 0 0 0 0
R-NO3 0 0 0

0 0 0 0 0
R2-UC 0 0 0

0 0 0 0 0
R2-SO4 0 0 0

0 0 0 0 0
R-FEC 0 0 0

0 0 0 0 0
School of Chemical and Minerals Engineering
248
Appendix A: Mass and energy balances
L4US3O14 0 0 0

0 0 0 0 0
L2-SO4 0 0 0

0 0 0 0 0
KEROSENE 0 0 0

0 0 0 0 0
ISODEC 0 0 0

0 0 0 0 0
NH3 0.009 21.432 0

0 0 0 0 0
AMMON-01 0 0 0

0 0 0 0 0
NH4+ 322.679 0 0

5.518 327.286 0.005 0 5.012
ADU 0 0 0

0 0 0 0 0
H4SIO 0 0 0

0 0 0 130.904 0
NH4NO3 0 0 0

0 0 0 0 0
LIME 0 0 0

0 0 0 0 0
CA++ 0 0 0

0 0 0 0 0
CAOH+ 0 0 0

0 0 0 0 0
CA(OH)2 0 0 0

0 0 0 0 0
CALCI(S) 0 0 0

0 0 0 0 0
AL(OH)3 0 0 0

0 0 0 0 0
FE(OH)2 0 0 0

0 0 0 0 0
MG(OH)2 0 0 0

0 0 0 0 0
Total 8859.33 85.727 1410.022

143.983 8539.268 243.107 130.904 1297.819

19 = OK liquor 8 = ADU bleed
NH
3
= Ammonia 12 = ADU recycle
20 = Wash water for centrifuge 18 = ADU product
22 = Spent liquid
School of Chemical and Minerals Engineering
249
Appendix A: Mass and energy balances
A.2. Energy balance

The approach for calculating the energy balance is to consider each section individually.
Additional energy is required at the leaching and precipitation units to increase the
temperature as required. Each process where significant energy transfer or generation
occurs is seen as a black box to validate the energy balance. Finally an energy balance
over each unit in the uranium extraction plant is given. In this section it is assumed that the
ambient temperature is 20 C.

The energy balance for the leaching pachucas is done by calculating the amount of steam
required to heat the pachuca contents to the required temperature. The feeds into the
pachuca is simplified to ore, water, air, and steam. The energy balance over the leaching
pachucas is given in Table A.8.

Table A.8: Leaching pachucas energy balance
IN
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 395175.0 4384.0 0.0 0.0
Ore 530000.0 800.0 0.0 0.0
Steam 59197.9 4246.0 155.0 10822313.8
Air 10000.0 1005.0 0.0 0.0
Total 10822313.84
OUT
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 454372.8589 4384 16 8853202.725
Ore 530000 800 16 1884444.444
Air 10000 1005 16 44666.66667
Total 10782313.84




School of Chemical and Minerals Engineering
250
Appendix A: Mass and energy balances
The heat capacities for the water, steam (sub-cooled water, 175 C, 12 bar) and air is
derived from literature (Koretsky, 2004: 507). The heat capacity for the ore is assumed to be
820 J/kg.K which is derived by taking the average of the heat capacities of several
components of the ore. The heat capacity of quartz which contributes approximately 70% of
the ore weight is 742 J/kg/K (Hemingway,1987: 275) and for muscovite which contributes
approximately 10% of the ore weight is 818 J/kg/K (Cemi,2005: 91). From this average
heat capacity is estimated at 800 J/kg.K. Mass flows for the streams are approximated from
the mass balance. The reference temperature used for the calculations in Table A.8 is
20 C.

Further it is assumed that the air for agitation is fed at 10 000 kg/hr at a feed pressure of 6
bar. The energy loss to the environment through evaporation and conduction is assumed as
40 kW which is less than 1 % of the system energy. The pachuca content is first heated to
30 C and thereafter the temperature of each fourth pachuca is increased with 2 C to
maintain the desired temperature. For this reason the outlet temperature is taken as 36 C,
since there are three pachucas that require additional heating. Using solver in Microsoft

Office Excel, the mass flow of the steam into the system is solved to satisfy the energy
balance. The steam required for the leaching pachucas is 60 000 kg/hr.

The leach product flows to the counter-current decantation section to wash the uranium
containing liquids from the solids, recovering the product. This process takes place in large
open tanks with a high residence time. The wash water entering the counter-current
decantation from the ion exchange system is at 20 C and therefore does not introduce
thermal energy into the system. Here the liquids and solids lose all their thermal energy to
the environment through evaporation and conduction. In the energy balance shown in Table
A.9 it is assumed that the liquids and slurry leaving the counter-current decantation section
are at 20 C, the same as the surrounding environment and therefore all potential thermal
energy leaves the system.


School of Chemical and Minerals Engineering
251
Appendix A: Mass and energy balances
Table A.9: Counter-current decantation energy balance
IN
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Ore 530000 800 10 1177778
Water 395175 4384 10 4812353
Wash water 610000 4384 0 0
Total 5990131
OUT
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Ore 530000 800 0 0
Water 403298.1 4384 0 0
Leach liquor 600000 4384 0 0
Total 0

Table A.9 shows that an energy loss of approximately 6 MW to the environment. This is
possible since a large open area will enhance evaporation which causes great energy
losses. It is also important to note that liquid surface is in continuous circular movement,
enhancing evaporation. The wash water entering the system is at 20 C which will create an
energy gradient for energy transfer. It is calculated that the temperature of the exit streams
are 24 C if no energy loss is assumed, but this is highly unlikely.

The slaked lime and slurry stream entering the neutralisation pachuca are at the reference
temperature (20 C). Since neutralisation reactions occur in this process it is important to
take the energy generation caused by these reactions into account. The heat of reaction for
several of the more important reactions, such as the precipitation of metal hydroxides due to
the high pH and neutralisation reactions are calculated. From this it is calculated that
approximately 1.6 MW is produced by the reactions in the system. The energy balance for
the neutralisation process is given in Table A.10.


School of Chemical and Minerals Engineering
252
Appendix A: Mass and energy balances
Table A.10: Neutralisation energy balance
IN
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 403298.0 4384.0 0.0 0.0
Ore 530000.0 800.0 0.0 0.0
Slaked lime 57398.0 4384.0 0.0 0.0
Solid lime 10186.0 2845.2 0.0 0.0
Reactions 1645318.0
Total 1645318.0
OUT
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 461439.0 4384.0 0.0 1359756.7
Ore 531047.0 800.0 0.0 285561.3
Total 1645318.0

The heat of reaction for the reactions is obtained from the Facility for the analysis of
chemical thermodynamics (2009). The reactions are the only source of energy in the system
as seen in Table A.10. If it is assumed that no energy loss occurs, and therefore leaves the
system in the exit stream. The temperature of the exit stream is solved using Microsoft

Office Excel. Table A.10 shows the exit temperature under these conditions increases with
7 x 10
-4
C. If energy loss to the environment occurs the temperature increase is even
lower, therefore it is assumed that the temperature increase over the neutralisation process
is negligible.

No energy increases are present in the ion exchange or solvent extraction. The small
amounts of reactions occurring in these sections are highly diluted and are therefore
negligible. From this it is assumed that the temperature increase over these sections is also
negligible.


School of Chemical and Minerals Engineering
253
Appendix A: Mass and energy balances
At the precipitation the reactor temperature is controlled at 30 C to enhance reaction
kinetics. This temperature is achieved by heating the OK-liquor in a heat exchanger using
steam. An ammonia gas and air mixture is also introduced to the precipitation reactors. It is
found in the calculations, similar to that of neutralisation, that the thermal energy generated
by the reactions has no significant temperature effect. The energy balance for the
precipitation reactors is given in Table A.11.

Table A.11: Precipitation energy balance
IN
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 8.1 4384.0 0.0 0.0
NH3 21.4 2058.8 0.0 0.0
air 50.0 1006.0 0.0 0.0
Q

725759.9
Total 725759.9
OUT
Stream
Mass flow
(kg/hr)
Heat capacity
(J/kg.K)
T-T
ref
(C)
Thermal power
(Watt)
Water 8.1 4384.0 10.0 356945.3
ADU 130.9 281.8 10.0 368814.6
Air 50.0 1006.0 10.0 503080.5
Total 725759.9

In Table A.11 it is seen that 0.75 MW is required to heat the reactor contents to the desired
30 C. All the energy transferred to the precipitation reactors leaves with the product to the
thickener, where all the energy is lost to the environment through evaporation and
conduction.

It should be noted that these calculations are only an approximation for the complex
hydrometallurgical system. From these calculations it is concluded that the energy
distribution over the uranium extraction plant follows the law of energy conservation and that
no heat integration is practical.
School of Chemical and Minerals Engineering
254
Appendix B: Equipment sizing

Appendix B: Equipment sizing

Appendix B consists of the description of the sizing methods followed for Section 3.7.

B.1. Leaching

The kinetics for the nitric acid leaching of uranium is found in Ikeda et al. (1995) as Equation
B-1

| |

| | | | | |
( = +
| | |

\ . \ . \ .
2.3
1 2
a 1 2 2 3
H H
r A exp A exp HNO NO
RT RT
(B-1)


For Equation B-1 the constants are given in Table B.1.

Table B.1: Leaching kinetics constants
Constant Description Value
A1 Frequency factor 2.2 x 10
4

A2 Frequency factor 0.46
H1 Activation energy -79 500
H2 Activation energy -36 800

The design equation used is for a batch reactor and is manipulated to give Equation B-2
(Fogler, 2006: 70):

=
a
a
dN
r S N
dt
(B-2)

It was assumed that the shrinking core model is applicable to the leaching of uranium ore.
Ikeda et al. (1995) derived the shrinking core model for the reaction kinetics. Since other
side reactions occur during the leaching process, the reduction in particle radius is taken into
account with an additional linear constant. This derivation is given in Equation B-3.

School of Chemical and Minerals Engineering
255
Appendix B: Equipment sizing

| |
= +
|

\ .
8 a
r MW dr
2.95 10
dt 100
(B-3)

The extra term (2.95x10
-8
) in Equation B-3 is manipulated to reduce the radius of the particle
to10% of the initial radius. Further it is assumed that the uranium content of the particles is
0.034%. The amount of particles (N) is calculated using Equation B-4.

=
Total solid feed volume
N
Volume of single spherical particle
(B-4)

The feed nitrate concentration ([NO
3
-
]
0
) is taken from the mass balance in Appendix A. It
was derived from the reaction stoichiometry that the rate at which the nitrates decrease in
the system is given by Equation B-5.


( ( =

a
3 3
0
0
N 2
NO NO
3 v
(B-5)

In Equation B-5 v
0
is the volumetric flow of the liquid feed. The above equations were
implemented in Polymath

to determine the influence of several design parameters. The

Polymath

program is shown next.

A1 = 2.2e4
A2 = 0.46
R = 8.314
T = 273 + 30
MW = 270
rho = 2.650
GramU0 = 530 * 0.34 * 1000 / 60
MolU0 = GramU0 / MW

CHNO2 = 0
C0NO3 = 0.0618

ra = (A1 * exp(-79500 / R / T) + A2 * exp(-36800 / R / T) * CHNO2) * (CNO3) ^ 2.3
w = 1 / ra
d(r)/d(t) = -(ra * MW / rho / 100 + 0.0000000295)
r(0) = 0.0000375

School of Chemical and Minerals Engineering
256
Appendix B: Equipment sizing
S = 4 * (22 / 7) * (r ^ 2) * (100 ^ 2)

d(MolU)/d(t) = ra * S * N
MolU(0) = 0

GramU = MolU * MW
dXdt = ra * S * N / MolU0

N = 6.15e17 * (0.34 / 1000)
CNO3 = C0NO3 - (2 / 3) * (MolU / 381645)
t(0) = 0
t(f) = 1137

X = GramU / GramU0

Using the above code the influence of temperature and nitrate concentration was determined
in order to optimize the system. Firstly the influence of the nitrate concentration was
determined, while a constant temperature was used. It is important to note that the
concentration of HNO
2
was assumed to be zero, since it is not fed as raw material and it has
no effect on the overall chemical reaction (given in Equation B-6).

(B-6)

The concentration of the nitrates in the feed was varied between 0.06 and 0.1 mole/L, at a
temperature of 25 C. The influence of the nitrate concentration is shown in Figure B.1.

+ +
+ + + +
2
2 3 3 2 2
3UO 2NO 8H O 3UO 12H O 2NO
School of Chemical and Minerals Engineering
257
Appendix B: Equipment sizing

Figure B.1: Influence of nitrate concentration on reaction kinetics

In Figure B.1 it is seen that an increase in nitrate concentration results in a significant
increased amount of product. It is also noticed, from Figure B.1, that the amount of uranium
leached increases exponentially as the nitrate concentration increases. This effect is caused
by the exponential function of the reaction kinetics in respect to the nitrate concentration.

The influence of temperature was found to have a significant effect on the reaction kinetics.
The temperature was varied between 25 and 55 C while keeping the nitrate concentration
constant at 0.0182 mole/L. Figure B.2 illustrates the results of this variation of temperature
on the reaction kinetics.

0
2
4
6
8
10
12
14
16
18
20
0 200 400 600 800 1000 1200
M
o
l
e

u
r
a
n
i
u
m

l
e
a
c
h
e
d
Time (min)
0.06 mole/L 0.07 mole/L 0.08 mole/L 0.09 mole/L 0.1 mole/L
School of Chemical and Minerals Engineering
258
Appendix B: Equipment sizing

Figure B.2: Influence of temperature on reaction kinetics

As seen in Figure B.2, the temperature also has a significant effect on the reaction kinetics.
The influence of the temperature forces the use of steam to ensure a proper conversion in
the leaching section, especially during the winter. It is therefore important to optimise both
the temperature and the nitrate concentration to ensure economical conversion.

Taking the cost considerations for temperature increase and nitrate addition into account, it
was optimised and found that the nitrate recycle from solvent extraction is sufficient while the
temperature is controlled at 30 C. It was also approximated that the heat loss will cause a
temperature drop of about 0.55 C per tank and therefore it was decided to heat each fourth
leaching tank.

The sizing of the leaching process is done according to the graphical method described by
the Minerals Council of Australia (2006). The kinetic data is obtained at a nitrate
concentration of 0.0618 mole/L and a temperature of 30 C. Figure B.3 illustrates the
conversion rate against the conversion of uranium, and from it the existing stages are
evaluated.




0
1
2
3
4
5
6
7
8
0 200 400 600 800 1000 1200
M
o
l
e

u
r
a
n
i
u
m

l
e
a
c
h
e
d
Time (min)
55 C 45 C 25 C 35 C
School of Chemical and Minerals Engineering
259
Appendix B: Equipment sizing



Figure B.3: Leach tank size evaluation

From Figure B.3 it was found that 11 of the existing leaching stages will give a uranium
conversion of 96%. From this it is concluded that 11 of the existing leaching tanks will be
used while maintaining a temperature of 30 C and a nitrate feed concentration of 0.0618
mole/L.


School of Chemical and Minerals Engineering
260
Appendix B: Equipment sizing
B.2. Counter-current decantation

The existing counter-current decantation equipment consists of six thickeners in series. The
first thickener has a larger area to ensure minimal solids in the feed to the ion exchange
section. For the expansion project a larger feed must be processed in this section. It was
decided to split the feed into two trains, Train 1 consisting of the existing thickeners while
Train 2 will consist of six new thickeners in series. The smaller capacity of Train 2 is sized
according to the existing equipment capacity and some guidelines from the literature.

The capacity of the existing thickeners is calculated from available plant data. Thickener 1
of the existing train has a cross-sectional area of 2 800 m
2
and that of Thickener 2 to 6 is 2
400 m
2
. Currently Train 1 is processing a leach liquor from an ore feed of 8 000 ton/day
from which a capacity coefficient is calculated by dividing the cross-sectional area with the
ore feed. This capacity coefficient for Thickener 1 is 0.35 and for Thickener 2 to 6 it is 0.3
m
2
per ton per day. This correlates with figures from Merritt (1971) which suggest this
capacity coefficient should be between 0.23 and 0.6 m
2
per ton per day. The cross-sectional
area of the thickeners of Train 2, which should process an ore feed of 5 000 ton/day, is sized
directly from this and was found to be 1 750 m
2
for Thickener 1 and 1 500 m
2
for Thickener 2
to 6. From this the diameter of Thickener 1 of Train 2 is approximated as 50 m and for
Thickener 2 to 6 as 45 m. The total capacity of the two trains will process an ore feed of 13
000 ton/day.

The wash water that enter Thickener 1 in both Train 1 and 2 is calculated with Equation B-7.

=
n
100
%Uranium loss
(wash ratio)
(B-7)

In Equation B-7, n is the number of stages in the counter current decantation section, which
is 6 in this case. A uranium loss of 0.01% is assumed resulting in a wash ratio of 1.
Therefore the volumetric flow of the wash solution is equal to that of the leach liquor.


School of Chemical and Minerals Engineering
261
Appendix B: Equipment sizing
B.3. Ion exchange

For the ion exchange section fixed-bed column ion exchange process is used. The sizing of
the columns was done using resin specifications and basic mass balance calculations. The
Ambersep 400 resin is used which has a capacity of 1.4 eq/L and according to Merritt (1971)
40% of the resin capacity is occupied by uranyl complexes. The bed voidage is assumed to
be 30%.

From the kinetics and resin capacity either the column dimensions or the adsorption time
can be calculated. The column dimensions of the existing adsorption columns are used. It
was calculated that the resin capacity is 78 g UO
2
(SO
4
)
3
4-
per litre resin and from the mass
balance the uranium content is 0.339 g UO
2
(SO
4
)
3
4-
per litre solution. A mass balance over
this system shows that the adsorption time for one column is 20 hours. Therefore during this
period there is sufficient time for elution and washing of the resin.

The elution of saturated resin is done with seven bed volumes of eluant (acidified NH
4
NO
3
).
This amount of eluant is chosen to ensure the desired concentration of uranyl complexes
(4 g U
3
O
8
/L) in the feed to the solvent extraction. According to Merritt (1971) the residence
time for eluant in the column should be between 12 and 20 minutes. A residence time for 20
minutes was assumed for this design which resulted in an elution time of 8 hours. From this
it is seen that the elution is easily done in the available time with ample time left for washing.
A total volume of 500 m
3
of eluate is sent to the solvent extraction feed tank every 20 hours,
resulting in a volumetric flow of 25 m
3
/hr with a uranium content of 3.9 g U
3
O
8
/L.

B.4. Precipitation

The precipitation of ADU is achieved by bubbling ammonia and air mixture through the OK
liquor. The existing reactor has a diameter of 1.5 m and height of 1.5 m with a volume of
2.65 m
3
. The tank is operated at 60% of its capacity to ensure adequate space for gasses
and emissions (Motsau, 2008:64). Therefore the effective volume of the reactor is 1.6 m
3
.

The reaction kinetics as described by Motsau (2008) is given in Equation B-8.

=
2
A A
r kC
(B-8)

School of Chemical and Minerals Engineering
262
Appendix B: Equipment sizing
In Equation A-7 r
A
is the rate at which ammonia reacts (mole/m
3
.min), k is the reaction rate
constant with a value of 857.8 m
3
/mole.min and C
A
is the concentration of UO
2
(SO
4
)
3
4-
in the
solution. The design equation for a continuous stirred tank reactor (CSTR) is taken from
Fogler (2006) and shown in Equation B-9

=

A0
2
A0
vC X
V
k(C (1 X))
(B-9)

In Equation B-9 X is the conversion of UO
2
(SO
4
)
3
4-
to ADU, V is the effective volume of the
reactor, C
A0
is the feed concentration and v is the volumetric flow of the OK liquor.

From the available plant data for the existing precipitation reactor and the OK liquor
concentrations and flow, the conversion was calculated. Currently the concentration of
UO
2
(SO
4
)
3
4-
is 39 mole/m
3
and the volumetric flow rate is 6 m
3
/hr for the OK liquor. This
gives a conversion of 99.86%. The same reactor is evaluated to determine whether its
capacity is sufficient for the increased OK liquor flow. From the mass balance the increased
OK liquor has a volumetric flow rate of 7.8 m
3
/hr and UO
2
(SO
4
)
3
4-
concentration of 44
mole/m
3
. The conversion obtain from the increased OK liquor is 99.85% which is a very
small decrease in conversion. Therefore the existing precipitation reactor will be used to
process the increased OK liquor and will result in 173 mole ADU/hr which is sent to a
thickener.

The existing thickener has a diameter of 15 m and a height of 4 m, giving a cross-sectional
area of 180 m
2
. Currently this thickener processes 1.8 ton ADU/day and from this the
capacity coefficient is calculated as 98.17 m
2
per ton per day. According to Weiss (1985)
the capacity coefficient should be between 4.7 and 11.6 m
2
per ton per day but the
calculated coefficient is much higher which shows the thickener is over-designed. For the
increased feed of 2.6 ton ADU/day the capacity coefficient is 68 m
2
per ton per day which is
still higher than the prescribed range. Therefore the existing ADU thickener is used to
process the increased feed.


School of Chemical and Minerals Engineering
263
Appendix B: Equipment sizing
B.5. Neutralisation

The pH of the slurry from the counter current decantation section must be 10.5 before it is
sent to the gold extraction plant. Therefore a neutralisation reactor is necessary to
neutralise this slurry by adding lime. Since only 11 of the 14 existing leaching pachucas are
used, one of the remaining pachucas is used to neutralise the slurry. This pachuca are
already equipped with air agitation to ensure good mixing of the slurry and lime before it is
sent to the gold extraction plant.

From basic calculations involving the solubility constants of several metal hydroxide species,
the mass of lime required to reach the desire pH was found as 10 700 kg/hr. This lime is
slaked with 46 700 kg/hr water to achieve a slaked lime feed density of 1.15 kg/L. The total
volumetric flow to the neutralisation pachuca , including the slaked lime and slurry feed, is 60
m
3
/hr. For the effective volume of 750 m
3
for the neutralisation pachuca, a residence time of
1 hour and 15 minutes is achieved, which is sufficient to neutralise the slurry.



School of Chemical and Minerals Engineering
264
Appendix C: Detail design calculations

Appendix C: Detail design calculations

Appendix C consists of a detailed explanation for the calculations and logical reasoning
followed during the detail design. For this feasibility study, the detail design is done for the
solvent extraction section.

C.1. Loading isotherms

The loading isotherm for the extraction section is obtained using Equation 4-1 and the given
constants. The loading isotherm using the reported constant does not fit the experimental
data; therefore the constants are manipulated to obtain a satisfactory modelled isotherm. In
Figure C.1 the experimental data is modelled using Equation 4-1 with the respective
constants.


Figure C.1: Modelling of the extraction loading isotherm

0
1
2
3
4
5
6
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
y

(
U
o
r
g
)

(
g
/
L
)
x (U
aq
) (g/L)
Model - repoted constants Model - manipulated constants Experimental data
School of Chemical and Minerals Engineering
265
Appendix C: Detail design calculations
In Figure C.1 it is noticed that the reported constants for Equation 4-1 does not fit the
experimental data well. The manipulated constants however fit the experimental data well
and are used to obtain the loading isotherm for the specific operating conditions.

The experimental data obtained from Figure 4.2 is shown in Figure C.2. The data is plotted
in Excel to obtain a smooth line between the data points.


Figure C.2: Loading isotherm data for the stripping section

The operating line is calculated from a basic mass balance of the uranium over the process,
given in Figure C.3.


Figure C.3: Solvent extraction process flow schematic
0
5
10
15
20
25
30
35
40
0 1 2 3 4 5 6 7 8 9
y

(
U
3
O
a
q
)

(
g
/
L
)
x (U
3
O
8org
) (g/L)
School of Chemical and Minerals Engineering
266
Appendix C: Detail design calculations
In Figure C.3, x represents the uranium content (g U
3
O
8
/L) of the phase the uranium is
transferred from while y represents the uranium content of the phase the uranium is
transferred to. The operating lines for the extraction and stripping sections are derived from
the mass balance and given in Equation C-1 and C-2 respectively.


= +
A A
E,in E,out E,in E,out
O O
V V
y y x x
V V
(C-1)

= +
O O
S,in S,out S,in S,out
A A
V V
y y x x
V V
(C-2)

As seen in Figure C.3, the extraction and stripping section are interdependent due to the
recycling of the organic solvent. The uranium content of the aqueous stream, x
E,in
, into the
extraction section and the aqueous stream, y
S,out
, out of the stripping section is specified for
the design. A recovery of 0.98 is assumed in the extraction section, which sets the value for
y
E,out
. The uranium content of the loaded solvent from the extraction section, y
E,out
, is equal
to that of the solvent fed to the stripping section, x
S,in
. The process is iteratively solved to
obtain equal value for y
E,in
and x
S,out
. This is achieved by varying the ratio of the volumetric
flows of the aqueous, V
A
, and the organic, V
O
, phases for the stripping section. The V
A
to V
O

ratio for the extraction section is specified as 1.1:1.

The resulting McCabe-Thiele graphs from the procedure discussed above are shown in
Figures C.4 and C.5. These graphs are the final result in the iterative solving, involving four
iterations, to obtain equal values for y
E,in
and x
S,out
.
School of Chemical and Minerals Engineering
267
Appendix C: Detail design calculations
= +
E,in
E,out E,in
O
A
0.98x
y y
V
V

Figure C.4: McCabe-Thiele for the Extraction section

The calculated operating line for the extraction section, seen in Figure C.4, is given in
Equation C-3.

= +
E,in E,out
1 1
y 3.5918 3.751 x
1.1 1.1
(C-3)

The operating line for the extraction section is fixed due to the specified V
A
to V
O
ratio. For
the first iteration of McCabe-Thiele the value for y
E,in
was assumed to be 0 g U
3
O
8
/L from
which a new value of 0.25 g U
3
O
8
/L for y
E,in
is found. The value for y
E,out
in Equation B-1 is
calculated from the assumed recovery of 0.98 in the extraction section (see Equation C-4).
This gave the final operating line given in Equation C-3.



(C-4)
School of Chemical and Minerals Engineering
268
Appendix C: Detail design calculations
= +
O O
S,in S,out
A A
V V
y 12 3.5918 x
V V
The efficiency used in the McCabe-Thiele method is taken as 0.98, which depends on the
level of mixing and residence time in the mixers. The value for y
E,out
is very close to the
assumed value of 0.25 g U
3
O
8
/L, which is satisfactory due to the human factor in the
McCabe-Thiele method. Once the McCabe-Thiele method gives satisfactory results, the
method is applied to the stripping section to iteratively solve the mass balance as mentioned
above.


Figure C.5: McCabe-Thiele for the Stripping section

The calculated operating line for the stripping section, seen in Figure C.5, is given in
Equation C-5.


(C-5)

School of Chemical and Minerals Engineering
269
Appendix C: Detail design calculations
The value for y
S,out
is specified as 12 g U
3
O
8
/L in Equation B-5, while the value for x
S,in
is
equal to y
E,out
(3.5918 g U
3
O
8
/L). The value for the V
O
to V
A
ratio is varied to obtain a value
of 0.25 g U
3
O
8
/L for x
S,out
. The final ratio for V
O
to V
A
is obtained as 3.6 which then solves
the mass balance as seen in Figure C.4. From the McCabe-Thiele method the uranium
loading (g U
3
O
8
/L) exiting each stage in the extraction and stripping sections are obtained.

C.2 Mixer design

The residence time is used to calculate the volume of the liquids in the mixer box. An over-
design of 10% was used throughout the calculations for the mixer design. In Equation C-6
the volume of the liquids (V in m
3
) was calculated by multiplying the residence time (t in hr)
with the aqueous and organic volumetric flow rate (Q
A
and Q
O
) and the over-design. The
calculation is shown in Equation C-6.

= +
A O
V t (Q Q ) 1.1 (C-6)

The residence time is specified as 2 minutes to ensure height stage efficiency and the total
design volumetric flow (Q
A
+ Q
O
) is 54.2 m
3
/hr. The liquid volume is calculated as 2 m
3
. A
rule of thumb is used to determine the dimensions of the mixer box which states that the
height is equal to the diameter of the mixer. Using the above information the dimensions of
the mixer box is as follows:

A height of 1.7 m, which includes an extra 0.3 m for instrumentation.
A diameter of 1.4 m.


The diameter of the impeller blade is approximated as 80% of the mixer box diameter and
results in a 1.1 m impeller diameter. The used impeller acts as a mixer and a pump and is
supplied by MC Process revered to as the MC Process SX impeller (Watson, 2009). The
three mixer design equations and their constant, for the specific impeller, as well as the
equation for the hydraulic efficiency follows.




School of Chemical and Minerals Engineering
270
Appendix C: Detail design calculations
( ) +
= =
=
= =

=
A O
q
p
p
h 2 2 2
p
2
h q
p
Q Q 1.1
N 0.01048
ND
N 0.5493
2gH
N 2.46
N D
N N
2N
(C-7)

(C-8)

(C-9)

(C-10)

The most important parameters in the above equations are N and D
p
, which is the impeller
rotary speed (revolutions per second) and the impeller diameter (m) respectively. Due to the
fact that the impeller acts as both a mixer and pump, it is also important to calculate the
head produced H in meters, and the hydraulic efficiency, .

Since the over-designed volumetric flow and the impeller diameter is known, it is possible to
calculate the impeller rotary speed using Equation C-7. The tip speed of the impeller is a
crucial design parameter, which determines the level of crud formation. Crud is an over-
emulsified mixture of organic and aqueous phases which does not settle. An approved
design prescribed by MC Process has a tip speed of 3.9 m/s, compared to the 4 m/s
resulting from the above design. It is therefore accepted that the designed impeller rotary
speed is within the design bounds. Once the impeller rotary speed is known, the produced
head and hydraulic efficiency is calculated using Equation C-9 and C-10.

The produced head should be able to supply sufficient pressure increase to ensure
continuous flow into the settler vessel. According to Eckhart (2004) stirring devices always
work in the turbulent flow regime, thus Re
R
(Equation C-7) should be larger than 5x10
5
.
Using Equation C-11 together with the calculated impeller rotary speed the Reynolds
number, Re
R
, is calculated to indicate the flow regime (Eckhart, 2004:15):

=

2
m
R
c
ND
Re (C-11)

In Equation C-11 the subscript m is an indication of the agitator and c is for the continuous
phase. The symbols n
R
and d
R
is the rotor speed and mixer diameter, respectively. The
kinematic viscosity of the continuous phase is given by the symbol
c
. For systems with an
aqueous-organic ratio of close to one, the continuous phase is seen as the denser phase
which in this system is seen as the aqueous phase for this system (Minerals Council of
School of Chemical and Minerals Engineering
271
Appendix C: Detail design calculations
Australia, 2006). The Reynolds number is calculated as 1.7 x 10
6
and is therefore in the
turbulent flow regime.

C.3. Settler design

The settler are design according to the settling kinetics, Equation 4-2 and 4-3, and constants
reported by Stnner and Wiesner (1982). The above mentioned equations are for a batch
system, which are manipulated in order to represent a plug flow reactor system. Equation C-
12 and C-13 are used in this design.

=
1 1
T
dQ Q
k.H.w.
dL Q
(C-12)

| |
=
|
\ .
2 1 2
T T
dQ Q Q
w k.H. c.
dL Q Q
(C-13)

In Equations C-12 and C-13, Q
1
is the volumetric flow (m
3
/hr) of the small droplets, Q
2
is the
volumetric flow of the large droplets, w is the width of the settler vessel (m) and L is the
length of the vessel (m). In the above equations Q
T
is the total volumetric flow of the
dispersion layer, which is the sum of the volumetric flow of the aqueous (Q
A
) and organic
phases (Q
O
). The base of these differential equations is the length, L (m), of the vessel.
The height of the dispersion layer, H, is assumed to be directly proportional to the total
volumetric flow of the droplets. The initial height of the dispersion layer is dependent on the
w, Q
T
and the velocity of the dispersion through the vessel (v in m/hr), see Equation C-14.

=
T
0
Q
H
w.v
(C-14)

In Equation C-14, H
0
is the initial height of the dispersion layer and Q
T
is the total volumetric
flow (aqueous and organic) into the settler vessel. The values of the constants k and c
(settling velocity of large droplets) in equation C-12 and C-13 is reported as 17.6 hr
-1
and
13.2 m/hr respectively. In the design of the extraction settlers, the droplets contain the
organic solvent, while the aqueous phase is the continuous phase. It is further assumed that
the total volumetric flow of droplets is initially small droplets. This assumption is valid if the
School of Chemical and Minerals Engineering
272
Appendix C: Detail design calculations
mixing efficiency is high and will result in the slowest possible settling rates. The slowest
settling rates are chosen to act as an over-design for the settler vessel.

Combining the above equations, a Polymath

program is created to simultaneously solve the

two differential equations, which follows. This program solves the differential equations (C-
12 and C-13) in terms of the length. The vessel is designed with an adequate settling length
to allow the height of the dispersion layer, H, to reach a desired height.

d(Q1)/d(s) = -rate1 * dikte
d(Q2)/d(s) = (rate1 - rate2) * dikte
rate1 = k * H * Q1 / QT
rate2 = c * Q2 / QT
Q1(0) = 28.317
Q2(0) = 0
s(0) = 0
s(f) = 5.5

speed = 25
dikte = 2

k = 17.6
c = 13.2
Q0 = 28.317

QT = (Q1 + Q2) + (Q1 + Q2) / 1.1
H0 = (Q0 + Q0 / 1.1) / (dikte * speed)
H = H0 * QT / (Q0 + Q0 / 1.1)

In the above Polymath

program there are two variable design parameters (w and v). These
parameters are varied to determine their influence on the settling kinetics. The extraction
and stripping sections differs in regard to the V
A
to V
O
ratio, therefore it is sized separately.
The extraction section is sized first where a V
O
to V
A
ratio of 1.1:1 is used with the aqueous
feed of 25.8 m
3
/hr. The width parameter, w, is varied between 2 and 4 m, while keeping v
constant at 30m/hr and is shown in Figure C.6.

School of Chemical and Minerals Engineering
273
Appendix C: Detail design calculations

FigureC.6: Vessel width sensitivity analysis

In Figure C.6 it is seen that the vessel width has a significant influence on the initial height of
the dispersion layer, while having an insignificant influence on the required length of the
vessel. Figure C.6 shows that the initial height of the dispersion layer decreases when the
vessel width increase. In Figure C.7 the velocity (v) is varied between 30 and 55 m/hr, while
keeping the width at 1.5 m.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 3 6 9 12 15
H
e
i
g
h
t

-
H

(
m
)
Length -L (m)
2 m
3 m
4 m
School of Chemical and Minerals Engineering
274
Appendix C: Detail design calculations

Figure C.7: Dispersion layer velocity sensitivity analysis

From Figure C.7 it is noticed that the dispersion layer velocity has a significant effect on its
initial height, while influencing the required length to some extent. Both the vessel width and
the dispersion layer velocity effects the dimensions of the required settler vessels, which
directly influence the equipment costs. Therefore it is important to do a sensitivity analysis
combining the vessel width and dispersion layer velocity to investigate the economical
influence. This economical sensitivity analysis is based on the area of equipment material
and is shown in Figure C.8.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 3 6 9 12 15
H
e
i
g
h
t

-
H

(
m
)
Length - L (m)
30 m/hr
35 m/hr
40 m/hr
45 m/hr
50 m/hr
55 m/hr
School of Chemical and Minerals Engineering
275
Appendix C: Detail design calculations

Figure C.8: Economical sensitivity analysis on the extraction settlers

The cost of equipment is directly proportional to the area of materials needed for
construction. An increase in material area will increase the manufacturing and installation
cost. It is therefore important to determine the optimum configuration between dispersion
velocity of the layer and the width of the vessel. It is also important to ensure that the initial
height of the dispersion layer and the width of the vessel is within realistic bounds. A
realistic initial height of the dispersion layer is chosen as 1.5 m, which will include a 30%
over-design to ensure enough space for air. The width of the settler vessel should not be
greater than 2 m to ensure effective distribution of the dispersion at the top of the mixer. The
length calculated will ensure a final dispersion layer height of 2.5 cm. The calculated length
represents the distance from the distributor to the aqueous weir.

According to these limitations and Figure C.8 the best economical and efficient settler choice
has a width of 2 m and a dispersion layer velocity of 25 m/hr, a height of 1.5 m and a length
of 7.3 m which includes additional length of 1.8 m for the aqueous weir. The effective
settling length for the extraction settler is 5.5 m, which will ensure effective separation of the
two phases to reduce the height of the dispersion layer to less than 2.5 cm and results in a
residence time of 12.5 minutes. The final height of the dispersion layer will increase slightly
in practice and therefore the settlers are designed to reach a small final dispersion height.
0
10
20
30
40
50
60
70
0 10 20 30 40 50 60
A
r
e
a

(
m
2
)
Velocity flow (m/hr)
1.5 m
2 m
3 m
School of Chemical and Minerals Engineering
276
Appendix C: Detail design calculations
The dispersion height profile over the length of the settler vessel for the specified settler
dimensions is shown in Figure C.9.


Figure C.9: Dispersion height profile over the length of the extraction vessel

It is important to note, from Figure C.9, that in the last third of the settling vessel length, the
dispersion layer height does not show a drastic decrease. This effect is expected from the
kinetic equation for the large droplet settling (Equation C-9), where the settling rate is directly
proportional to the dispersion layer height. A larger final dispersion layer height will increase
the probability of solvent loss. Due to the great economical strain associated with solvent
loss, it is decided to increase the capital cost which will decrease the operating cost of the
solvent extraction section. The capital cost is increased by increasing the vessel length
which will reduce the solvent loss, therefore, decreasing the operating cost.

The settler vessels for the stripping section are sized next. The same influence of the
dispersion layer velocity and the vessel width, as seen in Figure C.6 and C.7 for the
extraction section, is observed for the settling efficiency for the stripping section. It is
imperative to compare the extraction and stripping kinetics to determine the compatibility. A
velocity of 25 m/hr is used for the dispersion layer and a vessel width of 2 m is used to
compare the settling kinetics. The height profile of the dispersion layer for the stripping and
extraction section is shown in Figure C.10.

0
0.3
0.6
0.9
1.2
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
H
e
i
g
h
t

-
H

(
m
)
Length - L (m)
School of Chemical and Minerals Engineering
277
Appendix C: Detail design calculations

Figure C.10: Economical sensitivity analysis on the stripping settlers

In Figure C.10 it is noticed that for the same dispersion layer residence time, the settling
kinetics of the stripping section is faster and will therefore require a smaller settling vessel.
However, since the material of construction is vacuum infused fibreglass, a mould is used to
construct the settling vessels. The construction cost for this manufacturing method consist
of the price for each mould and construction materials used, therefore, using one standard
vessel size will optimize construction cost. Due to the smaller vessels required for the
stripping section, it is decided that the size for both the stripping and scrubbing settler
vessels are exactly the same as for the extraction settling vessels.

According to Minerals Council of Australia (2006), it is important to consider an internal
recycle stream of the continuous phase to ensure the desired phase continuity if the feed
ratio is close to one. The internal recycle stream will result in a different internal V
O
to V
A

ratio and a larger total volumetric flow. Figure C.11 illustrates the different height profiles as
a result of internal recycling for the extraction section.

0
0.3
0.6
0.9
1.2
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
H
e
i
g
h
t

-
H

(
m
)
Length - L (m)
Extraction
Stripping
School of Chemical and Minerals Engineering
278
Appendix C: Detail design calculations

Figure C.11: Height profiles for internal recycle and without internal recycle

It is seen, in Figure C.11, that the initial dispersion height increases to approximately 1.4 m
with the addition of the internal recycling stream. The height of the settler vessels is chosen
as 1.5m and is therefore adequate to handle the internal recycle of 33% of the aqueous
phase. The final dispersion layer height resulting from the internal recycle is 3 cm which is
still acceptable. Due to the faster settling kinetics of the stripping section, it is derived that
an internal recycle in the stripping section is also viable.


C.4. Pipe sizing

Due to safety considerations with regards to fire hazards, it is important to design the inside
diameter of the pipe to ensure a velocity flow of less than 1 m/s. It is also imperative to use
normal pipe size standards to reduce the material cost of construction. Equation C-15 is
used to determine the velocity flow.

=
| |

|
\ .
2
Q*1.1
v
ID
2
(C-15)


The inside diameter (ID in meters) is varied using the standard pipe sizes to obtain a velocity
flow (v in m/s) of less than 1 m/s. The results are shown in Chapter 4.

0
0.3
0.6
0.9
1.2
1.5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
H
e
i
g
h
t

-
H

(
m
)
Length - L (m)
Without internal recycle With 33% internal recycle
School of Chemical and Minerals Engineering
279
Appendix C: Detail design calculations
C.5. Pump sizing

There are two important parameters to determine when sizing a pump, the first parameter is
the head required to pump a liquid to the required position and the second parameter is the
volumetric flow of the liquid. The first parameter is calculated using the law of conservation
of energy and the second parameter is calculated in the mass balance using the law of
conservation of mass.

The head is calculated using the momentum balance which is a mathematical representation
of the law of conservation of energy. The overall momentum balance is re-written to
represent the pressure change for each type of energy and is shown in Equation C-16
(Neomagus, 2008: 34):

(C-16)

The first term, P
EP
, in the momentum balance equation represent the change in energy due
to change in pressure. The change in pressure is equal to zero because the liquid level of
the tank and the liquid level of the settling vessel are both at atmospheric pressure. The
change in mechanical pressure due to the change in pressure is equal zero Pa. The
calculation for P
EP
is shown in Equation C-17 (Neomagus, 2008: 33):

(C-17)

The second term, P
EL
, in the momentum balance represents the change in energy due to
change in height, thus the change in potential energy. Since the liquid level in the tank and
the liquid level of the settling vessel are at different heights, a change in potential energy will
occur. The calculation for P
EL
is shown in Equation C-18 (Neomagus, 2008: 33):

(C-18)

In Equation B-18 the variables are as follows: = density of water (kg/m
3
)
g = gravitational force (m/s
2
)
z
1
= height at point 1 with datum as
reference (m/s)
z
2
= height at point 2 with datum as
reference (m/s)
+ + + =
EP EL KE f A
P P P P' P
=
EP 1 2
P P P
=
EL 2 1
P g(z z )
School of Chemical and Minerals Engineering
280
Appendix C: Detail design calculations
The third term, P
KE
, in the momentum balance represents the change in energy due to
change in velocity, thus the change in kinetic energy. The velocity, of the liquids at both the
tank and settler vessel liquid level, is close to 0 m/s and will not result in a pressure increase.
The change in mechanical pressure due to the change in kinetic energy is zero Pa. The
calculation for P
KE
is as follow (Neomagus, 2008: 33):

(C-19)

In Equation C-19 the variables are as follows: = density of water (kg/m
3
)

1
= velocity at point 1 (m/s)
v
2
= velocity at point 2 (m/s)

The fourth term, P
f
, in the momentum balance represents the change in energy due to
friction loss. There will always be friction loss due to transport of the liquid. The flow regime
is important in the calculation of the friction coefficient (f). The equation used to calculate
the friction loss is independent on the flow regime but dependent on the roughness of the
pipe (), diameter of the pipe (D) and the Reynolds number of the flow (Re). The calculation
for P
f
is as follow (Neomagus, 2008: 41-77):







(C-20)








In Equation C-20 the variables are as follows: K = resistance coefficient
L = length of the pipe (m)
D = diameter of the pipe (m)
=
KE 2 1
P ( )
( )
( )

=

(
| |
= + (
|
\ . + (

| | (
| | | |
| = +
(
| |
|
\ . \ .
(
\ .
| |
=
|
\ .
=
=
+
=
1
12 12
1.5
16
0.9
16
T
2
T
f
D
Re
8 1
f ' 8
Re
A B
7
A 2.457ln 0.27
Re D
37530
B
Re
f ' L
f
D
K K
f K
P
2
School of Chemical and Minerals Engineering
281
Appendix C: Detail design calculations
The term on the right hand side of Equation C-16, P
A
, represents the amount of pressure
required from an external source to pump the liquid to the required position. If the term has
a negative value, no pump is required and gravity flow can be used. The required pressure
is translated to head (H in meters) using Equation C-21:

(C-21)


It is also important to note that the mixer-pumper used for mixing in the solvent extraction
section gives an additional head of 2 m and should be subtracted from the head calculated
using Equation C-16 an C-21. The head and volumetric flow calculated is used to determine
the purchased cost in Chapter 5.

C.6. Control valve

The control valves are design according to the rules of thumb supplied by Svreck et al.
(2008). There are two important design parameters used to size a control valve. The first is
the pressure drop across the valve (P
CV
) and is calculated using Equation C-22 (Svreck et
al.,2008: 40).


(C-22)

The pressure across the valve is P
CV
(kPa), P
s
is the supply pressure, Q
m
is the maximum
feed pressure, Q
d
is the design feed pressure and P
f
is the pressure loss due to friction in
the pipes. The minimum pressure drop (P
b
) is assumed to be 10 kPa. P
f
is calculated
using the same methodology as discussed for the pump sizing. Once the pressure across
the control valve is calculated it is possible to calculate the valve coefficient.

The valve coefficient is defined in Svreck et al. (2008) as the numbers of US gallons that will
pass through a control valve in 1 minute, when the pressure drop across the valve is 1 psi.
From the definition it is seen that the units is US metric, however, there exists a conversion
factor for this problem. The equation used to calculate the valve coefficient is shown in
Equation C-23 and the conversion factor is shown in Equation C-24 (Svreck et al., 2008:41).
(C-23)

A
P
H
g
(
= + +
(

m
CV s f b
d
Q
P 0.05P 1.1 1 P P
Q
=

V
Q
C
P
SG
School of Chemical and Minerals Engineering
282
Appendix C: Detail design calculations

(C-24)

Equation C-23 is used using SI units and the CV is calculated and converted to determine
the CV as per definition. Table C.1 illustrates the calculated variables for Equation C-22 and
C-23.

Table C.1: Variables for Equation C-22 and C-23.

Qm (m
3
/hr) Qd (m
3
/hr) Ps (kPa) Pf (kPa) Pb (kPa) SG
U03-FCE01 33.00 30.00 95 12 10 1
U03-FCE02 28.40 25.82 134 12 10 1
U03-FCE03 28.40 25.82 134 12 10 1
U03-FCE04 33.00 30.00 95 12 10 1
U03-FCE05 31.24 28.40 129 12 10 0.817
U03-FCE06 6.25 5.68 138 12 10 1
U03-FCE07 11.00 10.00 138 12 10 1
U03-FCE08 11.00 10.00 120 12 10 1
U03-FCE09 5.50 5.00 120 12 10 1
U03-FCE10 8.68 7.89 138 12 10 1
U03-FCE11 33.00 30.00 95 12 10 1
U03-FCE12 2.20 2.00 138 12 10 1
U03-FCE13 2.20 2.00 138 12 10 1
U03-FCE14 2.20 2.00 138 12 10 1
U03-FCE15 8.68 7.89 119 12 10 1
U03-FCE16 33.00 30.00 95 12 10 1
U03-FCE17 5.50 5.00 95 12 10 1
U03-FCE18 11.00 10.00 95 12 10 1
U03-FCE19 31.24 28.40 138 12 10 1
U03-FCE20 31.24 28.40 95 12 10 0.817

Combining the variables in Table C.1 and Equation C-23 and C-24, the CV is calculated and
shown in Table 4.6. The calculated CV is used to determine which valve to use once the
control valve supplier is contracted.

=
3
0.5 0.5
gpm m
1 0.865
psi hr.bar
School of Chemical and Minerals Engineering
283
Appendix D: Techno-economic calculations

Appendix D: Techno-economic evaluation

D.1. Capital investment

The capital investment is calculated using the delivered-equipment cost method according to
Peters et al. (2004). For this method the equipment purchased cost needs to be calculated for
all the new equipment used. The Marshall & Swift (M&S) cost index is used to calculate most of
the equipment purchased cost. Cost indexes are used due to inflation in the value of money.
The M&S index values are published in each issue of the Chemical Engineering magazine, and
the starting value for the M&S index in 1926 equals to 100 and the present value equals 1504.8
(Lozowski, 2009: 56).

The time value of money can be corrected by adjusting for inflation/deflation. This adjustment
can be done by the M&S index values using Equation D-1 (Ulrich & Vasudevan, 2009: 49).

| |
=
|
\ .
s
P,v,s P,v,r
r
M&S
C C
M&S
(D-1)


In Equation D-1 the subscripts r and s is the different years in which the equipment is
purchased, with purchased cost C
P,v,i
. Equation D-1 should be used for all purchased cost
calculated from a cost graph instead of a mathematical calculation. The formulas that is used
for these calculations is given and in Table D.1 a summary of these costs are given

Mixer-settler

The mixer-settlers chosen for this process is specialized equipment which means that
conventional cost estimations with calculations involving M&S indexes will give unrealistic
purchased cost for the equipment. The company providing the equipment, MC Process, gave a
ball park figure for the process as R 14 000 000. This estimation includes the agitators,
impellers, mix boxes, settlers, instrumentation, fire protection, and design time (Watson, 2009).

School of Chemical and Minerals Engineering
284
Appendix D: Techno-economic calculations
Heat exchanger

The heat exchanger purchased cost is calculated using Equation D-2 (Douglas, 1988: 572).
( )
| |
=
|
\ .
0.65
c
M&S
Purchased cost $ 101.3A F
280
(D-2)
In Equation D-2, the symbols are: A = area (ft
2
)

F
c
= (F
d
+F
p
)F
m
F
m
= shell and tube material
F
d
& F
p
= correction factors for heat exchangers

The values for the correction factors are supplied in Douglas (1988).

Pressure vessels / storage vessels

The purchased cost for pressure vessels / storage vessels is calculated using Equation D-3
(Douglas, 1988; 574).
( ) ( )
| |
=
|
\ .
1.066 0.82
c
M&S
Purchased cost $ 101.9D H F
280
(D-3)

From Equation D-3 the symbols are describes as: D = diameter (ft)
H = height (ft)
F
c
= F
m
x F
p

F
m
& F
p
= correction factors for vessel
material and pressure

The values for the correction factors are supplied in Douglas (1988).

Pumps

The purchased cost of the pumps is calculated using Figures D.1 and D.2. Figure D.1 is the
purchased cost of the pump including pump, steel base, and coupling, but no motor. Figure D.1
also provides correction factors for material of construction and provides the pump power
requirements in kilowatts (Peters et al., 2004: 517). Figure D.2 is the purchased cost of the
electric motor needed for the pump (Peters et al., 2004: 520).

School of Chemical and Minerals Engineering
285
Appendix D: Techno-economic calculations


Figure D.1: Cost of general-purpose centrifugal pumps
















Figure D.2: Cost of electric motors
School of Chemical and Minerals Engineering
286
Appendix D: Techno-economic calculations
Thickeners

The installed cost for a single-compartment thickener is calculated using Figure D.3 (Perry et
al., 1997: 18-73). The installed cost include the raking mechanism (including drivehead and lift),
walkways and bridge of centerpier, cage, railings, and overflow launders.



Figure D.3: Installed cost for single-compartment thickeners

The calculations for the purchased equipment cost calculated from the equations and
correlations above are given in Table D.1.




School of Chemical and Minerals Engineering
287
Appendix D: Techno-economic calculations
Table D.1: Purchased equipment cost
Equipment Quantity Price per unit (R per unit)
Total price
(R)
Leaching
Storage tanks for nitric acid 4 3,641,746.76 14,566,987.06
Slurry pumps 2 156,478.09 * 312,956.19
Thickener capacitance tank 1 15,462,463.06 15,462,463.06
Wash solution capacitance tank 1 4,339,077.16 4,339,077.16
CCD
Clarifier for train 2 1 13,708,772.19 13,708,772.19
Thickeners train 2 5 10,344,921.71 51,724,608.53
Slurry pumps 30 152,209.38 * 4,566,281.35
Ion exchange
Wash water capacitance tank 1 1,789,331.97 1,789,331.97
Eluant make-up tank 1 1,173,448.55 1,173,448.55
Regeneration make-up tank 1 1,173,448.55 1,173,448.55
Liquid pumps 14 126,342.70 * 1,768,797.80
Slurry pumps 6 99,958.79 * 599,752.72
Solvent extraction
Mixer-settlers 13 1,076,923.08 * 14,000,000.00
Liquid pumps 22 11,832.86 * 283,988.64
Organic make-up tank 1 358,988.91 358,988.91
Raffinate tank 1 956,138.27 956,138.27
OK-liquor storage tanks 1 956,138.27 956,138.27
Regen make-up 1 459,475.00 459,475.00
Organic storage tank 1 956,138.27 956,138.27
(NH
4
)
2
SO
4
make-up tank 1 956,138.27 956,138.27
Caustic storage tank 1 956,138.27 956,138.27
Spent regen 1 956,138.27 956,138.27
Eluate capacitance tank 1 4,339,077.16 4,339,077.16
Demineralized water 1 956,138.27 956,138.27
School of Chemical and Minerals Engineering
288
Appendix D: Techno-economic calculations
Precipitation
Heat exchanger 1 12,736,721.48 12,736,721.48
Storage tank for solvent extraction recycle 1 769,550.32 769,550.32
Storage tank for stage 1 centrifuge recycle 1 263,056.52 263,056.52
Storage tank for stage 2 centrifuge recycle 1 88,392.18 88,392.18
ADU storage tank 1 1,457,967.02 1,457,967.02
Liquid pumps 6 24,649.08 * 147,894.46
Slurry pumps 14 21,907.19 * 306,700.62
Total purchased equipment cost (E) 153,090,705.31
* Average price per unit

The total capital investment is calculated using the total purchased equipment cost (E) and the
factors given in Table 5.1 for a solid-fluid processing plant. In Table D.2 the results of these
calculations are given.

Table D.2: Capital investment calculations

Solid-fluid
processing
plant factor
Calculated values (R million)
Direct costs
Purchased equipment cost (E) 153.09
Delivery, percent of purchased equipment 0.10 15.31
Subtotal: delivered equipment 168.40
Installation 0.39 65.68
Instrumental and controls 0.26 43.78
Piping 0.31 52.20
Electrical systems 0.10 16.84
Buildings 0.29 48.84
Yard improvements 0.12 20.21
Service facilities 0.55 92.62
Total direct cost 508.57
School of Chemical and Minerals Engineering
289
Appendix D: Techno-economic calculations
Indirect costs
Engineering and supervision 0.32 53.89
Legal expenses 0.04 6.74
Construction expenses 0.34 57.26
Contractors fee 0.19 32.00
Contingency 0.37 62.31
Total indirect cost 212.18
Total fixed capital 720.75
Working capital 0.75 126.30
Total capital invesment 847.05


D.2. Operating cost

The detailed operating cost calculations are given in Table D.3 with the following assumptions:

6 skilled operators per shift.
3 semi-skilled operators per shift.
5 general assistants per shift.
The property is owned by AngloGold Ashanti, thus no rent is paid.
Already existing market and buyer for the product ADU and gold.
There is no research and development department on the plant.
No royalties costs are included as no equipment or processes needs to be patented.


School of Chemical and Minerals Engineering
290
Appendix D: Techno-economic calculations
Table D.3: Detailed operating cost calculation
Production cost
Suggested
factor
Rate or
quantity per
year
Cost per rate
or quantity
unit
Unit
Calculated
values (R
million per
annum)
Variable production costs
Raw materials
Nitric acid 13844 1,500.00 R/ton 20.77
Sulphuric acid 143006 500.00 R/ton 71.50
Resin (Ambersep 400) 123 57,000.00 R/ton 7.03
Caustic soda (NaOH) 168 4,200.00 R/ton 0.71
Kerosene 283 5,000.00 R/m
3
1.42
Isodecanol 9 22,410.00 R/ton 0.21
Alamine 9 57,260.00 R/ton 0.53
NH
3
175 15,000.00 R/ton 2.62
Lime (CaO) 83017 139.00 R/ton 11.54
Na
2
CO
3
84 900.00 R/ton 0.08
Subtotal: Raw materials 116.39
Operating labor
Skilled 48900 67.80 R/hr 3.32
Semi-skilled 24450 31.40 R/hr 0.77
General assistant 40750 20.20 R/hr 0.82
Call out feed 1630 33.60 R/hr 0.05
Subtotal: Operating labour 4.96
Operating supervision 0.15 of operating labour 0.74
Utilities
Water
Cooling 0 7.00 R/m
3
0.00
Process 3586 7.00 R/m
3
0.03
Electricity 24450000 0.40 c/kWh 9.78
Fuel 0 7.80 R/L 0.00
Refrigeration 0 148.32 R/GJ 0.00
Steam (175C and 12 bar) 312442 50.00 R/ton 15.62
School of Chemical and Minerals Engineering
291
Appendix D: Techno-economic calculations
Waste water
Disposal 0 3.93 R/ton 0.00
Treatment 16820 3.93 R/ton 0.07
Waste treatment and disposal
Hazardous 100 1,075.33 R/ton 0.11
Non-hazardous 1000 266.98 R/ton 0.27
Subtotal: Utilities 25.87
Maintenance and repairs 0.07 of Fixed capital investment 50.45
Operating supplies 0.15 of Maintenance and repairs 7.57
Laboratory charges 0.15 of Operating labour 0.74
Royalties 0.04 of Total product cost without depreciation
Catalysts and solvents
(Magnafloc 90L) 215.975 22,500.00 R/ton 4.86
Total: Variable costs 211.59

Fixed charges (without depreciation)
Taxes (property) 0.02 of Fixed capital investment 14.42
Financing (interest) 0.105 of Fixed capital investment 75.68
Insurance 0.01 of Fixed capital investment 7.21
Rent 0 of Fixed capital investment 0.00
Subtotal: Fixed charges 97.30
Plant overhead costs 0.5
of operating labour, supervision and
maintenance 28.08
Total manufacturing costs 336.97
Administrative costs 0.2 of operating labour 0.99
Distribution and marketing
expenses 0 of total product cost 0.00
Research and development
costs 0 of every sales Rand 0.00
Total: Fixed cost 126.37
Total product cost 337.96


School of Chemical and Minerals Engineering
292
Appendix D: Techno-economic calculations
The total variable cost calculated in Table D.3 is used in the overall cash flow analysis but is
used as rand per ton ADU produced. Using the variable cost of R 211.59 million per annum and
a production rate of 1068.20 ton ADU per annum the variable cost per ton ADU produced is
calculated to be R 198 079.21. The final product cost accumulates to approximately
R 337.96 million per annum.

D.3. Revenue

Revenue for this operating plant is generated from the uranium and gold product sales. In
Table D.4 the price at which ADU is sold internationally and to NUFCOR as well as the trading
gold price is given. These spot prices were received on 26 October 2009 from William Manana
at South Uranium Plant.

Table D.4: Revenue calculations
Product International trading price Selling price to NUFCOR
ADU 45.5 $/lb 22.75 $/lb
Gold 254 552.77 R/kg N/A

D.4. Cash flow analysis

Using the calculations above a cash flow analysis is done and given in Table D.5. The cash
flow diagram in Figure 5.1 is achieved from the cash flow analysis in Table D.5.

Table D.5: Cash flow analysis.
Capital

Fixed capital R 720,751,040.60

Results
Working capital R 126,299,831.88

ROI 60.49%
Costs

Fixed costs R 126,372,504.60 per annum

Payback (years) 3.98
Variable cost R 198,079.21 per ton

NPV R 2,363,686,680.54
Revenue from
sales

ADU R 371,893.21 per ton

IRR 352%
Gold R 127,276,385.00 per ton

Tax rate 28.0% per annum

Discount rate 11.0% per annum
Inflation rate 6.4% per annum

School of Chemical and Minerals Engineering
293
Appendix D: Techno-economic calculations




CCD construction (old
production)
SX construction (no
production)
Optimization (Half
production)
Operation (full
production)
Year 0 1 2 3 4
ADU production rate (ton/year) 624 534 1068
Gold production rate (ton/year) 1.44 1.08 2.16
Fixed capital R -720,751,040.60
Working capital R -126,299,831.88
Depreciation R -36,037,552.03 R -36,037,552.03
Inflation factor 1.00 1.06 1.13 1.20
Fixed costs R -84,248,336.40 R -143,065,806.97 R -152,222,018.62
Variable costs R -82,400,950.85 R -131,310,126.02 R -254,868,495.73
Revenue from sales for ADU R 232,061,360.12 R 224,866,045.87 R 478,514,945.60
Revenue from sales for gold R 183,277,994.40 R 155,616,213.26 R 331,151,301.82
Profit before tax R 248,690,067.27 R 106,106,326.14 R 402,575,733.07
Tax R -69,633,218.83 R -19,619,256.75 R -102,630,690.69
Profit after tax R 179,056,848.43 R 86,487,069.39 R 299,945,042.38
Cash flow R - R 179,056,848.43 R -847,050,872.48 R 86,487,069.39 R 299,945,042.38
Cumulative cash flow R - R 179,056,848.43 R -667,994,024.05 R -581,506,954.66 R -281,561,912.28
Discount factor 1.00 0.90 0.81 0.73
Discounted cash flow R - R 179,056,848.43 R -763,108,894.13 R 70,194,845.70 R 219,317,229.85
Cumulative discounted cash
flow R - R 179,056,848.43 R -584,052,045.70 R -513,857,199.99 R -294,539,970.14
IRR Discount Factor 1.00 0.22 0.05 0.01
IRR Discounted Cash Flow R - R 179,056,848.43 R -187,547,866.97 R 4,239,908.18 R 3,255,739.08
IRR Cumulative Discounted
Cash Flow R - R 179,056,848.43 R -8,491,018.53 R -4,251,110.36 R -995,371.27



School of Chemical and Minerals Engineering
294
Appendix D: Techno-economic calculations





Operation (full production)
Year 5 6 7 8 9
ADU production rate (ton/year) 1068 1068 1068 1068 1068
Gold production rate (ton/year) 2.16 2.16 2.16 2.16 2.16
Fixed capital
Working capital
Depreciation R -36,037,552.03 R -36,037,552.03 R -36,037,552.03 R -36,037,552.03 R -36,037,552.03
Inflation factor 1.28 1.36 1.45 1.54 1.64
Fixed costs R -161,964,227.81 R -172,329,938.39 R -183,359,054.44 R -195,094,033.93 R -207,580,052.10
Variable costs R -271,180,079.46 R -302,437,433.47 R -307,001,883.24 R -326,650,003.76 R -364,301,054.04
Revenue from sales for ADU R 509,139,902.12 R 541,724,855.85 R 576,395,246.63 R 613,284,542.41 R 652,534,753.13
Revenue from sales for gold R 352,344,985.14 R 374,895,064.19 R 398,888,348.29 R 424,417,202.58 R 451,579,903.55
Profit before tax R 428,340,579.99 R 441,852,548.18 R 484,922,657.24 R 515,957,707.31 R 532,233,550.54
Tax R -109,844,847.83 R -113,628,198.92 R -125,687,829.46 R -134,377,643.48 R -138,934,879.58
Profit after tax R 318,495,732.16 R 328,224,349.26 R 359,234,827.78 R 381,580,063.83 R 393,298,670.95
Cash flow R 318,495,732.16 R 328,224,349.26 R 359,234,827.78 R 381,580,063.83 R 393,298,670.95
Cumulative cash flow R 36,933,819.88 R 365,158,169.14 R 724,392,996.92 R 1,105,973,060.75 R 1,499,271,731.70
Discount factor 0.66 0.59 0.53 0.48 0.43
Discounted cash flow R 209,803,003.91 R 194,785,175.97 R 192,061,608.68 R 183,791,247.17 R 170,662,714.29
Cumulative discounted cash
flow R -84,736,966.24 R 110,048,209.73 R 302,109,818.41 R 485,901,065.58 R 656,563,779.87
IRR Discount Factor 0.00 0.00 0.00 0.00 0.00
IRR Discounted Cash Flow R 765,445.29 R 174,656.19 R 42,324.73 R 9,954.15 R 2,271.66
IRR Cumulative Discounted
Cash Flow R -229,925.98 R -55,269.79 R -12,945.06 R -2,990.91 R -719.25


School of Chemical and Minerals Engineering
295
Appendix D: Techno-economic calculations




Operation (full production)
Year 10 11 12 13 14
ADU production rate (ton/year) 1068 1068 1068 1068 1068
Gold production rate (ton/year) 2.16 2.16 2.16 2.16 2.16
Fixed capital
Working capital
Depreciation R -36,037,552.03 R -36,037,552.03 R -36,037,552.03 R -36,037,552.03 R -36,037,552.03
Inflation factor 1.75 1.86 1.98 2.11 2.24
Fixed costs R -220,865,175.43 R -235,000,546.66 R -250,040,581.65 R -266,043,178.87 R -283,069,942.32
Variable costs R -369,799,162.66 R -393,466,309.07 R -438,818,887.11 R -445,441,634.63 R -473,949,899.25
Revenue from sales for ADU R 694,296,977.33 R 738,731,983.88 R 786,010,830.85 R 836,315,524.02 R 889,839,717.56
Revenue from sales for gold R 480,481,017.38 R 511,231,802.49 R 543,950,637.85 R 578,763,478.67 R 615,804,341.30
Profit before tax R 584,113,656.61 R 621,496,930.63 R 641,101,999.94 R 703,594,189.18 R 748,624,217.29
Tax R -153,461,309.28 R -163,928,626.01 R -169,418,045.41 R -186,915,858.40 R -199,524,266.27
Profit after tax R 430,652,347.33 R 457,568,304.63 R 471,683,954.52 R 516,678,330.78 R 549,099,951.02
Cash flow R 430,652,347.33 R 457,568,304.63 R 471,683,954.52 R 516,678,330.78 R 549,099,951.02
Cumulative cash flow R 1,929,924,079.03 R 2,387,492,383.66 R 2,859,176,338.18 R 3,375,854,668.96 R 3,924,954,619.98
Discount factor 0.39 0.35 0.32 0.29 0.26
Discounted cash flow R 168,352,670.45 R 161,148,454.87 R 149,657,448.35 R 147,687,759.61 R 141,401,065.09
Cumulative discounted cash
flow R 824,916,450.32 R 986,064,905.19 R 1,135,722,353.54 R 1,283,410,113.15 R 1,424,811,178.24
IRR Discount Factor 0.00 0.00 0.00 0.00 0.00
IRR Discounted Cash Flow R 550.74 R 129.56 R 29.57 R 7.17 R 1.69
IRR Cumulative Discounted
Cash Flow R -168.50 R -38.94 R -9.37 R -2.19 R -0.51



School of Chemical and Minerals Engineering
296
Appendix D: Techno-economic calculations





Operation (full production)
Year 15 16 17 18
ADU production rate (ton/year) 1068 1068 1068 1068
Gold production rate (ton/year) 2.16 2.16 2.16 2.16
Fixed capital
Working capital
Depreciation R -36,037,552.03 R -36,037,552.03 R -36,037,552.03 R -36,037,552.03
Inflation factor 2.38 2.54 2.70 2.87
Fixed costs R -301,186,418.63 R -320,462,349.42 R -340,971,939.79 R -362,794,143.93
Variable costs R -528,579,353.65 R -536,556,785.14 R -570,896,419.39 R -636,700,336.56
Revenue from sales for ADU R 946,789,459.48 R 1,007,383,984.89 R 1,071,856,559.92 R 1,140,455,379.76
Revenue from sales for gold R 655,215,819.15 R 697,149,631.57 R 741,767,207.99 R 789,240,309.31
Profit before tax R 772,239,506.35 R 847,514,481.90 R 901,755,408.74 R 930,201,208.57
Tax R -206,136,547.21 R -227,213,540.36 R -242,400,999.88 R -250,365,823.83
Profit after tax R 566,102,959.14 R 620,300,941.54 R 659,354,408.86 R 679,835,384.74
Cash flow R 566,102,959.14 R 620,300,941.54 R 659,354,408.86 R 679,835,384.74
Cumulative cash flow R 4,491,057,579.12 R 5,111,358,520.65 R 5,770,712,929.51 R 6,450,548,314.25
Discount factor 0.23 0.21 0.19 0.17
Discounted cash flow R 131,332,956.81 R 129,645,593.01 R 124,151,294.99 R 115,322,255.03
Cumulative discounted cash flow R 1,556,144,135.05 R 1,685,789,728.06 R 1,809,941,023.05 R 1,925,263,278.08
IRR Discount Factor 0.00 0.00 0.00 0.00
IRR Discounted Cash Flow R 0.39 R 0.09 R 0.02 R 0.01
IRR Cumulative Discounted Cash
Flow R -0.12 R -0.03 R -0.01 R -0.00



School of Chemical and Minerals Engineering
297
Appendix D: Techno-economic calculations




Operation (full production)

Year 20 21 22

ADU production rate (ton/year) 1068 1068 1068

Gold production rate (ton/year) 2.16 2.16 2.16

Fixed capital

Working capital R 126,299,831.88

Depreciation R -36,037,552.03 R -36,037,552.03 R -36,037,552.03

Inflation factor 3.25 3.46 3.68

Fixed costs R -410,717,799.17 R -437,003,738.32 R -464,971,977.57

Variable costs R -687,673,364.18 R -766,937,482.09 R -778,512,264.90

Revenue from sales for ADU R 1,291,104,973.60 R 1,373,735,691.91 R 1,461,654,776.19

Revenue from sales for gold R 893,495,797.20 R 950,679,528.22 R 1,011,523,018.03

Profit before tax R 1,086,209,607.45 R 1,120,473,999.73 R 1,229,693,551.76

Tax R -294,048,175.52 R -303,642,205.36 R -334,223,679.92

Profit after tax R 792,161,431.93 R 816,831,794.38 R 895,469,871.83

Cash flow R 792,161,431.93 R 816,831,794.38 R 1,021,769,703.71

Cumulative cash flow R 7,987,829,318.43 R 8,804,661,112.80 R 9,826,430,816.52

Discount factor 0.14 0.12 0.11

Discounted cash flow R 109,062,913.97 R 101,314,838.89 R 114,174,854.50

Cumulative discounted cash flow R 2,148,196,987.14 R 2,249,511,826.03 R 2,363,686,680.54 NVP
IRR Discount Factor 0.00 0.00 0.00

IRR Discounted Cash Flow R 0.00 R 0.00 R 0.00
IRR Cumulative Discounted Cash
Flow R -0.00 R -0.00 R 0.00


School of Chemical and Minerals Engineering
298
Appendix D: Techno-economic calculations
D.5. Economic sensitivity analysis

The results of the economic sensitivity analysis done is given in Tables D.6 and D.7.

Table D.6: Chance in variables

Revenue from a
unit ADU (R/ton)
Inflation
Fixed Cost
(R/annum)
Variable Cost
(R/ton ADU)
Tax Rate
20% 446,271.85 7.7% 151,647,005.52 237,695.05 33.6%
15% 427,677.19 7.4% 145,328,380.29 227,791.09 32.2%
10% 409,082.53 7.0% 139,009,755.06 217,887.13 30.8%
5% 390,487.87 6.7% 132,691,129.83 207,983.17 29.4%
0% 371,893.21 6.4% 126,372,504.60 198,079.21 28.0%
-5% 353,298.55 6.1% 120,053,879.37 188,175.25 26.6%
-10% 334,703.88 5.8% 113,735,254.14 178,271.29 25.2%
-15% 316,109.22 5.4% 107,416,628.91 168,367.33 23.8%
-20% 297,514.56 5.1% 101,098,003.68 158,463.37 22.4%

Table D.7: Economic sensitivity analysis in terms of percentage rise/fall of NPV

Revenue from a unit ADU Inflation Fixed Cost Variable Cost Tax Rate
20% 30.9% 16.6% -10.3% -16.2% -9.4%
15% 23.2% 12.2% -7.7% -12.2% -7.0%
10% 15.5% 8.0% -5.1% -8.1% -4.7%
5% 7.7% 3.9% -2.6% -4.1% -2.3%
0% 0.0% 0.0% 0.0% 0.0% 0.0%
-5% -7.7% -3.8% 2.6% 4.1% 2.3%
-10% -15.5% -7.4% 5.1% 8.1% 4.7%
-15% -23.2% -10.9% 7.7% 12.2% 7.0%
-20% -30.9% -14.3% 10.3% 16.2% 9.4%


School of Chemical and Minerals Engineering
299
Appendix E: Plant layout and positioning

Appendix E: Plant layout and positioning

Since this project is an upgrade of the existing South Uranium Plant of AngloGold Ashanti
located near Orkney in the North-West Province, the old plant layout should be studied to
analyze available space to build new processing equipment. Figure E.1 is an air photo taken
from the program Google Earth

and display the old layout of the plant.

Figure E.1: Google Earth air photo

Figure E.2 and E.3 is the plant layout received from William Manana at AngloGold Ashanti for
the existing plant.
CCD
Leaching
Solvent extraction
IX and precipitation
Sulphuric acid storage
South Uranium Plant (SUP)
School of Chemical and Minerals Engineering
300
Appendix E: Plant layout and positioning

Figure E.2: Existing plant layout of South Uranium Plant with number legend
LEGEND
1. Water treatment salt
2. Water treatment salt
3. Floc N300 750kg 25kg
4. Soda Ash
Copper Sulphate
Floc N300 25kg
5. Lime
6. Manganese
7. Manganese
8. Coal
9. Diesel
10. Ammonia
11. Caustic
12. Pegasol, Armeen 380, Isodec
13. Steel Balls
14. Armeen drums
15. Resin
16. Resin
17. Manganese
School of Chemical and Minerals Engineering
301
Appendix E: Plant layout and positioning























Figure E.3: Existing plant layout of South Uranium Plant with colour legend
School of Chemical and Minerals Engineering
302
Appendix F: MSDS information

Appendix F: MSDS information

The MSDS forms for the following substances are provided on the compact disc (CD)
provided with the report:

Alamine

336 (Cognis corporation, 2007).

Ammonia gas (BOC Gases, 1996).
Ammonium diuranate (ADU) (International Bio-Analytical Industries, Inc, 2006).
Caustic soda (ScienceLab, 2009).
Ferrous sulphate (ScienceLab, 2009).
Isodecanol (BASF, 2006).
Kerosene (ScienceLab, 2009).
Magnafloc 90L (ACAT, 2006).
Nitric acid (ScienceLab, 2009).
Nitric oxide (Air products, 1998).
Potable water (ScienceLab, 2009).
Silicon dioxide (ScienceLab, 2009).
Sulphuric acid (ScienceLab, 2009).
Uranium (British Geological Survey, 2007).