Red Mud Minimisation and Management for the …...Bauxite residue (red mud), a waste from the Bayer...
Transcript of Red Mud Minimisation and Management for the …...Bauxite residue (red mud), a waste from the Bayer...
Red Mud Minimisation and Management for the
Alumina Industry by the Carbonation Method
A thesis submitted in fulfilment of
the requirement for the degree of
Doctor of Philosophy Environmental Engineering
By CUONG PHUOC TRAN
February 2016
School of Chemical Engineering
Faculty of Engineering, Computer and Mathematical Sciences
The University of Adelaide
i
DECLARATION
I certify that this work contains no material which has been accepted for the
award of any other degree or diploma in my name, in any university or other tertiary
institution and, to the best of my knowledge and belief, contains no material
previously published or written by another person, except where due reference has
been made in the text. In addition, I certify that no part of this work will, in the
future, be used in a submission in my name, for any other degree or diploma in any
university or other tertiary institution without the prior approval of the University of
Adelaide and where applicable, any partner institution responsible for the joint-award
of this degree.
I give consent to this copy of my thesis, when deposited in the University
Library, being made available for loan and photocopying, subject to the provisions of
the Copyright Act 1968.
I also give permission for the digital version of my thesis to be made
available on the web, via the University’s digital research repository, the Library
Search and also through web search engines, unless permission has been granted by
the University to restrict access for a period of time.
CUONG PHUOC TRAN
Adelaide, 2016
ii
ACKNOWLEDGEMENTS
This thesis has been a truly exciting and extremely enriching experience for
me, both academically and personally. The outcomes of the research had the
assistance and cooperation of many individuals and organisations. I would like to
offer my grateful thanks to all of them. Particularly, I am thankful to the University
of Adelaide and the Vietnamese Government for offering me a prestigious
scholarship with which to conduct this study.
My special thanks must first go to my supervisor Associate Professor Dzuy
Nguyen, for his continuous assistance, guidance, active supervision, and kindness to
support me during my research candidature. He guided me initially on
comprehensive research of materials and encouraged me since the early stages of
research. I am greatly indebted to Associate Professor Dzuy Nguyen.
I would also like to express my appreciation to Associate Professor Brian
O’Neill, who was my previous co-supervisor and initially helped me in critical
thinking, supported and corrected my English at the beginning of this project. I am
also thankful to Associate Professor Yung Ngothai, who replaced my previous co-
supervisor in 2014. She always encouraged and shared with me difficulties occurring
during the research. She provided me with the best facilities from her lab for my
experiments.
My sincere thanks also go to Professor Allan Pring from Museum of South
Australia, who provided me with permission to use facilities in SA Museum. My
heartfelt thanks go to all academic members, office staff and other PhD students in
the School of Chemical Engineering, School of Physical Sciences, and Adelaide
Microscopy for their helps, friendship, encouragement, and understandings on many
occasions. My grateful thanks go to my colleagues, who shared with me their
research experience and made me feel confident through their talks and
iii
companionship. It has been a memorable time of my life and one that I will never
forget. I also thank Rio Tinto Alcan for kindly donating the red mud sample used in
this study.
To my family members, who gave support from the beginning, I have greatly
appreciated it and will try to do all those things that I promised. I would like to thank
my lovely wife, Loan Thi Thuy Nguyen, for her love, inspiration, and endless
encouragement. She has gracious and much patience in looking after our two lovely
and gentle children, Vy Thuy Tran and Trong Phuoc Tran (Ken Tran). Their love and
affection kept me sane during my research in Adelaide. Finally, I am thankful to my
parents and my siblings for their unconditional love and constant prayers, as well as
my parents-in-law, for their great understanding and support.
Adelaide, 2016
iv
ABSTRACT
Bauxite residue (red mud), a waste from the Bayer process for refining bauxite
to alumina, is highly alkaline (pH~13) and its treatment and management have posed
environmental challenges to the alumina industry. Carbonation of red mud using
carbon dioxide (CO2) has previously been demonstrated to be feasible in both
permanently capturing the CO2 and neutralising this solid waste. A systematic study
of the neutralisation of red mud by CO2 over a range of different operating
conditions is essential in order to optimise the carbonation process and maximise the
volume of CO2 captured by red mud.
The objectives of this study were to determine the acid neutralisation capacity
of red mud and its solid and aqueous phase contribution to the acid neutralisation
capacity via the analyses of red mud compositions. A red mud sample, provided by
Rio Tinto Alcan, was carbonated in a range of different operating conditions with the
intent of establishing the optimal condition for the carbonation process. The
carbonation was carried out at room temperature and atmospheric pressure using a
stirred tank reactor operating at different conditions such as total gas flow rate, CO2
concentrations, stirring speeds, and solids concentrations in red mud. A range of
analytical techniques such as X-ray diffraction (XRD), scanning electron microscopy
(SEM) coupled with Energy Dispersive X-ray (EDX), Inductively Coupled Plasma
Mass Spectrometry (ICP-MS) and Carbon-Hydrogen-Nitrogen Elemental Analyser
(CHN) were used to ascertain the different mineral phases, change of chemical
composition before and after carbonation, and carbonation capacity of the mud.
Finally, based on the information of red mud composition, an equilibrium chemical
model using MINEQL+ version 5.0 was developed for the carbonation process.
v
The acid neutralisation capacity of red mud was measured by both rapid and
long-term titration of red mud slurries to pH endpoint of 4.5, 6, 8, and 10. At the
endpoint of pH 4.5 corresponding to the bicarbonate endpoint, the acid neutralisation
capacity of the red mud was found to be 0.79 and 1.91meq/g red mud for rapid and
long-term titration, respectively. Furthermore, it is estimated that the solid phase
contributed approximately 81% to the acid neutralisation capacity, while contribution
from the liquid phase was only 19% in the final long-term acid neutralisation
capacity determination.
The carbonation process was observed to be significantly dependent on
concentration of CO2, total gas flow rate and stirring speeds, whereas the
concentration of solids in red mud seemed to have a little effect based on only three
concentrations studied. For the carbonation of red mud slurry, it took from 30-75
minutes to establish the equilibrium pH of 7.5-6.6 in the range of CO2 concentrations
of 10%-100%. In contrast, when the carbonation of red mud liquor only was
performed at the same range of CO2 values, the stable pH of 7.0-6.3 (0.3-0.5 pH unit
lower) was reached within 15-30 minutes. After carbonation, the pH from carbonated
red mud slurries, exposed to atmosphere CO2, rebound quickly and took about 20-25
days to reach pH of 9.7. The carbonated liquor, however, showed a lower rate of pH
recovery, and took a month to equilibrate to pH of 9.7.
The XRD patterns of carbonated red mud revealed the appearance of calcite
and the increase of gibbsite due to the dissolution of sodalite and the breakdown of
cancrinite minerals in the carbonation of red mud. The quantifications confirmed the
precipitation of calcite from 0% to 1.51%, and the increase of gibbsite from 1.04% to
5.15% in raw red mud and carbonated red mud, respectively. XRD patterns and the
quantifications associated with other results such as EDX and CHN analyses
vi
indicated that the most optimal conditions for carbonation process were 30% CO2
concentration and total gas flow rate of 200mL/min. At this condition, the amount of
CO2 captured for the whole red mud (both solid and liquid phases) was highest at
65g CO2/kg of red mud, and the alkalinity decreased from 11,610mg/L to 2,104mg/L
as CaCO3. Stirring speeds were found to be effective in boosting the extent of red
mud carbonation and the amount of CO2 sequestration. The results showed that when
stirring speeds rose from 250rpm to 700rpm, the amount of CO2 sequestration
increased by 3.4g/kg of red mud, from 65 to 68.4g CO2/kg of red mud.
The simulation for heavy metals dissolved in long-term titration of red mud at
different pH levels of 4.5, 6, 8, 10, and 12.5 was performed using chemical
equilibrium modelling system MINEQL+ 5.0. The modelling suggested that four key
dominant metals Al, Na, Ca, and Fe were found to govern the aqueous chemistry of
the red mud carbonation process due to their presence in both soluble and solid forms
in red mud. Measured metal concentrations from long-term titration at various pH
values indicated that boehmite (AlO(OH)) and hematite (Fe2O3) did not dissolve in
the system, therefore, both Al and Fe were not responsible for the control of
carbonation process as their concentrations remained unchanged. However, Na and
Ca were considered the major solids controlling the process. The dissolution of
sodalite (Na8(AlSiO4)6(OH)2.4H2O) and cancrinite (Na6(AlSiO4)6(CaCO3)(H2O)2)
were attributable to Na and Ca concentrations in the system. The key reactions are as
below:
Na8(AlSiO4)6(OH)2.4H2O + 18H+ = 8Na+ + 6Al3+ + 6Si(OH)4 + 2H2O
Na6(Al6Si6O24)(CaCO3)(H2O)2+24H+ = 6Na++6Al3+ + 6Si(OH)4 +Ca2++CO32-+2H2O
vii
For carbonation process, a chemical model was formulated in MINEQL+ 5.0
to calculate the final equilibrium pH values for both carbonation of RM slurry and
RM liquor at different concentration of CO2. The results revealed that the simulated
pH values for the carbonation process at different PCO2 were 0.3-0.45 pH units higher
than the experimental pH values. In other words, the difference in final pH
equilibrium values between experimental and simulated carbonation of red mud
varies from 4.0-6.0%. This difference is about 2 times lower than that of previous
work done by Khaitan (2009b). The key reactions of carbonation of red mud are as
follows:
Liquid phase reactions:
2OH-(aq) + CO2(aq) ↔ CO3
2- + H2O
H2O + CO32- + CO2(aq) ↔ HCO3
-(aq) + H+
(aq)
[Al(OH4)-](aq) + CO2(aq) + Na+
(aq) ↔ Al(OH)3(s) + Na+(aq) + HCO3
-(aq)
Solid phase reactions:
Na8(AlSiO4)6(OH)2.4H2O + 18H+ ↔ 8Na+ + 6Al3+ + 6Si(OH)4 + 2H2O
Na6(AlSiO4)6CaCO3(H2O)2 ↔ 6NaAlSiO4 + CaCO3(s) + H2O
viii
TABLE OF CONTENTS
DECLARATION .......................................................................................................... i
ACKNOWLEDGEMENTS ......................................................................................... ii
ABSTRACT ................................................................................................................ iv
TABLE OF CONTENTS .......................................................................................... viii
LIST OF FIGURES .................................................................................................... xi
LIST OF TABLES ................................................................................................... xvii
CHAPTER 1 INTRODUCTION .............................................................................. 1
1.1. Background ....................................................................................................... 1
1.2. Objectives .......................................................................................................... 6
1.3. Organisation of Thesis ....................................................................................... 7
CHAPTER 2 LITERATURE REVIEW .................................................................. 8
2.1. Overview of Bauxite Residue Management ...................................................... 8
2.1.1. Bauxite Residue Generation Process .......................................................... 8
2.1.2. Physicochemical and Mineralogical Properties of Red Mud .................... 11
2.1.3. Methods Utilised for Disposal of Red Mud .............................................. 14
2.2. The Red Mud Utilisation Options ................................................................... 20
2.2.1. The Utilisations of Red Mud in Construction ........................................... 21
2.2.2. The Utilisations of Red Mud in Chemical Applications........................... 22
2.2.3. The Utilisations of Red Mud in Metallurgy.............................................. 23
2.2.4. The Utilisations of Red Mud in Agriculture ............................................. 24
2.2.5. The Utilisations of Red Mud in Environmental Treatment ...................... 26
2.3. Red Mud Neutralisation Methods ................................................................... 28
2.3.1. Neutralisation of Red Mud with Seawater ................................................ 29
2.3.2. Neutralisation with Gypsum (CaSO4.2H2O) ............................................ 30
2.3.3. Neutralisation of Red Mud by Acid Mine Wastes .................................... 31
2.3.4. Neutralisation of Red Mud Using Mineral Acid ...................................... 32
2.3.5. Neutralisation of Red Mud by Fly Ash ..................................................... 33
2.3.6. Neutralisation of Red Mud by Carbon Dioxide (CO2) Gas ...................... 33
2.3.7. Perspectives of Red Mud Carbonation ..................................................... 35
2.3.8. Mechanism of the Carbonation of Red Mud ............................................ 42
2.4. Summary ......................................................................................................... 45
ix
CHAPTER 3 MATERIALS AND METHODS .................................................... 49
3.1. Materials .......................................................................................................... 49
3.2. Materials Preparation ...................................................................................... 50
3.3. Methods ........................................................................................................... 50
3.3.1. Acid Titration Procedures ......................................................................... 50
3.3.2. Determination of Total Alkalinity of Raw RM and Carbonated RM ....... 52
3.3.3. X-ray Diffraction (XRD) .......................................................................... 53
3.3.4. Scanning Electronic Microscopy and Energy Dispersive X-ray (SEM-
EDX) ................................................................................................................... 53
3.3.5. Carbon-Hydrogen-Nitrogen Elemental Analyser ..................................... 54
3.3.6. Thermal Analysis (TGA-DSC) ................................................................. 55
3.3.7. Fourier Transform Infrared Spectroscopy (FT-IR) ................................... 55
3.4. Carbonation Experiments ................................................................................ 56
3.4.1. Construction of Reaction Chamber........................................................... 56
3.4.2. Carbonation of RM ................................................................................... 58
3.4.3. pH Rebound of the Carbonated Red Mud ................................................ 59
3.5. Chemical Equilibrium Modelling .................................................................... 59
CHAPTER 4 RESULTS AND DISCUSSIONS ................................................... 62
4.1. Acid Neutralising Capacity (ANC) of the raw RM ......................................... 62
4.1.1. Rapid Titration of RM slurry and RM liquor ........................................... 62
4.1.2. Long Term Titration of Red Mud ............................................................. 64
4.2. Carbonation of Red Mud ................................................................................. 68
4.2.1. Effect of CO2 Concentration on Carbonation of RM ............................... 68
4.2.2. Effect of Total Gas Flow Rate on Carbonation of RM ............................. 72
4.2.3. Effect of Stirring Speed on Carbonation of RM ....................................... 74
4.2.4. Effect of Solids Concentrations in RM on Carbonation of RM ............... 76
4.2.5. pH Rebound in Carbonated RM ............................................................... 78
4.2.6. Longer Carbonation of RM....................................................................... 80
4.3. Mineralogical Characterisation of Red Mud and Carbonated Red Mud ......... 84
4.3.1. X-ray Diffraction Analysis ....................................................................... 84
4.3.2. Micro-morphological Characterisation of raw RM and Carbonated RM by
SEM .................................................................................................................... 96
4.3.3. Chemical Composition Changes by EDX ................................................ 99
4.3.4. Determination of Alkalinity of RM and Carbonated RM ....................... 111
x
4.3.5. Thermal Analysis using TGA-DSC ........................................................ 113
4.3.6. FT-IR Spectroscopy ................................................................................ 115
4.4. Determination of CO2 Sequestration ............................................................. 116
4.4.1. Determination of CO2 sequestered in 2-hour carbonation of RM .......... 116
4.4.2. Determination of CO2 sequestered in 5-day carbonation of RM ............ 118
4.5. Modelling of Carbonation Process ................................................................ 122
4.5.1. Modelling of potentially dissolved metals .............................................. 123
4.5.2. Modelling of RM carbonation ................................................................ 130
4.6. Summary ....................................................................................................... 136
CHAPTER 5 FINDING OUTCOMES AND CONCLUSIONS .......................... 138
5.1. Major Findings of This Research .................................................................. 138
5.1.1. Acid Neutralisation Capacity (ANC) of Red Mud ................................. 138
5.1.2. Carbonation of Bauxite Residue ............................................................. 139
5.1.3. Modelling of the Carbonation Process.................................................... 142
5.2. Conclusions ................................................................................................... 144
CHAPTER 6 RECOMMENDATIONS FOR THE FUTURE WORK ................ 146
REFERENCES ......................................................................................................... 147
APPENDIX .............................................................................................................. 165
xi
LIST OF FIGURES
Figure 1.1. Global production rate and cumulative inventory ..................................... 2
Figure 2.1. Schematic of a general Bayer process ....................................................... 9
Figure 2.2. Lagooning red mud disposal .................................................................... 16
Figure 2.3. Schematic of dry stacking system............................................................ 19
Figure 2.4. A possible flowsheet for recovery of Fe, Al, and Ti from bauxite residue
.................................................................................................................................... 24
Figure 3.1. Carbonation reaction chamber ................................................................. 57
Figure 3.2. The experimental apparatus system for carbonation of RM .................... 58
Figure 4.1. Rapid RM liquor titration compared with that of RM slurry (44%wt).... 63
Figure 4.2. Long-term titration of RM ....................................................................... 64
Figure 4.3. XRD pattern of raw RM overlapped with titrated RM at pH 6 ............... 67
Figure 4.4. XRD pattern of raw RM overlapped with titrated RM at pH 4.5 ............ 67
Figure 4.5. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
200mL/min and stirring speed of 250rpm .................................................................. 69
Figure 4.6. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
200mL/min and stirring speed of 250rpm .................................................................. 69
Figure 4.7. Comparison of carbonation between RM slurry and RM liquor at some
different CO2 concentrations, fixed TF of 200mL/min and stirring speed of 250rpm
.................................................................................................................................... 70
Figure 4.8. Carbonation rate constant (k) for both RM slurry and RM liquor at
different CO2 concentration, total gas flow rate 200mL/min and speed 250rpm ...... 72
Figure 4.9. Carbonation of red mud by 30% of CO2, 250rpm and different TF of gas
.................................................................................................................................... 73
Figure 4.10. Carbonation rate constant (k) for RM slurry at 30% CO2 concentration,
stirring speed 250rpm, and different total gas flow rate ............................................ 74
xii
Figure 4.11. Carbonation of red mud by 30% of CO2, TF of 200mL/min and different
stirring speeds ............................................................................................................ 75
Figure 4.12. Rate constant (k) for carbonation of RM slurry by 30% CO2
concentration, TF of 200mL/min and different stirring speeds ................................. 76
Figure 4.13. Carbonation of red mud by 30% of CO2, TF of 200mL/min and stirring
speed of 250rpm, and different solids concentrations in RM .................................... 77
Figure 4.14. Rate constant (k) for carbonation of RM slurry by 30% CO2
concentration, TF of 200mL/min, and different solids concentrations in RM........... 78
Figure 4.15. pH rebound for both RM slurry and liquor at three CO2 concentrations,
TF of 200mL/min, stirring speed of 250rpm ............................................................. 79
Figure 4.16. pH rebound of carbonated RM slurries at different solids concentrations
.................................................................................................................................... 80
Figure 4.17. Longer carbonation of RM slurry at different CO2 concentrations ....... 81
Figure 4.18. Longer carbonation of RM slurry at fixed 30% CO2, stirring speed of
250rpm and at different total gas flow rate ................................................................ 82
Figure 4.19. Longer carbonation of RM slurry at fixed 30% CO2, total gas flow rate
of 200mL/min and different stirring speeds ............................................................... 83
Figure 4.20. Longer carbonation of RM slurry at fixed 30% CO2, total gas flow rate
of 200mL/min, stirring speeds of 250rpm and different solids concentrations of RM
.................................................................................................................................... 83
Figure 4.21. Variation of powder XRD pattern of raw RM ....................................... 84
Figure 4.22. Phase composition quantification of raw RM ....................................... 85
Figure 4.23. XRD pattern of carbonated RM compared with raw RM...................... 86
Figure 4.24. Phase composition quantification of carbonated RM at 15% CO2
concentration and total gas flow rate of 200mL/min ................................................. 88
Figure 4.25. Phase composition quantification of carbonated RM at 30% CO2
concentration and total gas flow rate of 200mL/min ................................................. 89
Figure 4.26. Phase composition quantification of carbonated RM at 40% CO2
concentration and total gas flow rate of 200mL/min ................................................. 90
xiii
Figure 4.27. Phase composition quantification of carbonated RM at 60% CO2
concentration and total gas flow rate of 200mL/min ................................................. 90
Figure 4.28. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 100mL/min ................................................. 91
Figure 4.29. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 300mL/min ................................................. 92
Figure 4.30. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 400mL/min ................................................. 93
Figure 4.31. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 350rpm ............................................................... 94
Figure 4.32. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 500rpm ............................................................... 95
Figure 4.33. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 700rpm ............................................................... 95
Figure 4.34. SEM imaging of raw RM: (a) Sodalite in “cotton ball” form, and (b)
Structure of crystalline sodalite .................................................................................. 96
Figure 4.35. SEM imaging of carbonated RM at different CO2 concentration, TF of
200mL/min and stirring speed of 250rpm .................................................................. 98
Figure 4.36. The amounts of C and CO2 absorbed by RM after 2-hour carbonation at
different CO2 concentration, TF of 200mL/min and stirring speed of 250rpm ....... 101
Figure 4.37. Amounts of C and CO2 absorbed by RM after 5-day carbonation ...... 103
Figure 4.38. Amounts of C and CO2 absorbed by RM at a given 30% CO2
concentration, 250rpm and different TF of gas ........................................................ 105
Figure 4.39. Amounts of C and CO2 absorbed by RM at given 30% CO2, TF of
200mL/min and different stirring speeds ................................................................. 106
Figure 4.40. Amounts of C and CO2 captured by RM in different solids
concentrations .......................................................................................................... 109
xiv
Figure 4.41. Comparison of amounts of C and CO2 captured between 2-hour and 5-
day carbonation at fixed TF of 200mL/min, 250rpm and different CO2 concentrations
.................................................................................................................................. 111
Figure 4.42. Changes in HCO3-, CO3
2-, and OH- alkalinity in raw RM and carbonated
RM at different concentrations of CO2, TF of 200mL/min, 250rpm ....................... 112
Figure 4.43. Acid titration curves for a) Raw RM and b) Carbonated RM at 30%
CO2, TF of 200mL/min and stirring speed of 250rpm ............................................. 113
Figure 4.44. TGA-DSC plots indicating weight loss of RM ................................... 114
Figure 4.45. TGA-DSC plots indicating weight loss of carbonated RM ................. 115
Figure 4.46. Fourier Transform Infrared (FT-IR) spectra of RM and carbonated RM
.................................................................................................................................. 116
Figure 4.47. Amounts of CO2 sequestered by RM after 2-hour carbonation at
different CO2 concentrations, stirring speed of 250rpm .......................................... 117
Figure 4.48. Amounts of CO2 captured by RM (A): solid, (B): liquor, after 5-day
carbonation at different CO2 concentrations, TF of 200mL/min and speed of 250rpm
.................................................................................................................................. 119
Figure 4.49. Comparison of CO2 amounts captured between 2-hour and 5-day
carbonations ............................................................................................................. 120
Figure 4.50. Amounts of CO2 captured by RM carbonated 30% CO2 concentration,
TF of 200mL/min and at different stirring speeds ................................................... 120
Figure 4.51. Amounts of CO2 captured by RM with different solids concentrations
carbonated at 30% CO2 concentration, TF of 250mL/min and 250rpm .................. 121
Figure 4.52. Metal concentrations in RM liquor as a function of pH ...................... 124
Figure 4.53. Comparison of simulated and experimental carbonation of RM liquor at
different CO2 concentration and TF of (A): 100mL/min, (B): 200mL/min ............. 133
Figure 4.54. Comparison of simulated and experimental carbonation of RM liquor at
different CO2 concentration and TF of (C): 300mL/min, (D): 400mL/min ............. 133
Figure 4.55. Comparison of simulated and experimental carbonation of RM slurry at
different CO2 concentration and TF of (A): 100mL/min, (B): 200mL/min ............. 134
xv
Figure 4.56. Comparison of simulated and experimental carbonation of RM slurry at
different CO2 concentration and TF of (C): 300mL/min, (D): 400mL/min ............. 134
Figure B-1. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
100mL/min and stirring speed of 250rpm ................................................................ 172
Figure B-2. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
300mL/min and stirring speed of 250rpm ................................................................ 172
Figure B-3. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
400mL/min and stirring speed of 250rpm ................................................................ 172
Figure B-4. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
100mL/min and stirring speed of 250rpm ................................................................ 173
Figure B-5. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
300mL/min and stirring speed of 250rpm ................................................................ 173
Figure B-6. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
400mL/min and stirring speed of 250rpm ................................................................ 173
Figure B-7. Carbonation of RM by 25% CO2 at different TF of gas ....................... 174
Figure B-8. Carbonation of RM by 40% CO2 at different TF of gas ....................... 174
Figure B-9. Carbonation of RM by 50% CO2 at different TF of gas ....................... 174
Figure B-10. Carbonation of red mud by 30% of CO2, TF of 100mL/min at different
stirring speeds .......................................................................................................... 175
Figure B-11. Carbonation of red mud by 30% of CO2, TF of 300mL/min at different
stirring speeds .......................................................................................................... 175
Figure B-12. Carbonation of red mud by 30% of CO2, TF of 400mL/min at different
stirring speeds .......................................................................................................... 175
Figure B-13. Carbonation of red mud by 30% of CO2, TF of 100mL/min, stirring
speed 250rpm at different solid concentrations of RM ............................................ 176
Figure B-14. Carbonation of red mud by 30% of CO2, TF of 300mL/min, stirring
speed 250rpm at different solid concentrations of RM ............................................ 176
Figure B-15. Carbonation of red mud by 30% of CO2, TF of 400mL/min, stirring
speed 250rpm at different solid concentrations of RM ............................................ 176
xvi
Figure B-16. Carbonation of red mud by 40% of CO2, TF of 200mL/min, stirring
speed 250rpm at different solid concentrations of RM ............................................ 177
Figure B-17. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 350rpm at different solid concentrations of RM ............................................ 177
Figure B-18. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 500rpm at different solid concentrations of RM ............................................ 177
Figure B-19. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 700rpm at different solid concentrations of RM ............................................ 178
Figure B-20. pH rebound for both RM slurry and liquor at some CO2 concentrations,
TF of 200mL/min, stirring speed of 250rpm ........................................................... 178
Figure C-1. Phase composition quantification of carbonated RM at 20% CO2
concentration, total gas flow rate 200mL/min ......................................................... 200
Figure C-2. Phase composition quantification of carbonated RM at 50% CO2
concentration, total gas flow rate 200mL/min ......................................................... 201
xvii
LIST OF TABLES
Table 2.1. Chemical constituents of red mud in different locations (%wt) ............... 12
Table 2.2. Buffering common reactions in aqueous solution of bauxite residue ....... 13
Table 2.3. Dissolution reactions of common buffering solids present in bauxite
residues ....................................................................................................................... 14
Table 2.4. Summary of CO2 amount captured in previous studies on RM carbonation
.................................................................................................................................... 41
Table 2.5. Reactions taking place in the carbonation of red mud .............................. 43
Table 3.1. Major mineral composition of raw RM .................................................... 49
Table 3.2. Concentration of raw RM and liquor ........................................................ 61
Table 4.1. Comparison between rapid and long term ANC for RM .......................... 65
Table 4.2. Metal concentrations in RM liquor at different pH values ....................... 66
Table 4.3. Effect of CO2 concentrations on the composition of solid phase in
carbonated RM as quantified by XRD ....................................................................... 88
Table 4.4. Effect of total gas flow rate on the composition of solid phase in
carbonated RM as quantified by XRD ....................................................................... 92
Table 4.5. Effect of stirring speed on the composition of solid phase in carbonated
RM as quantified by XRD.......................................................................................... 94
Table 4.6. Major elemental composition (%w/w in average) of RM and carbonated
RM at different concentrations of CO2, TF of gas 200mL/min, stirring speed 250rpm
.................................................................................................................................... 99
Table 4.7. Major compound composition (%w/w in average) of RM and carbonated
RM at different concentrations of CO2, TF of gas 200mL/min, stirring speed 250rpm
.................................................................................................................................. 101
Table 4.8. Major elemental composition (%w/w in average) of RM and carbonated
RM at 15%-60% CO2, TF of 200mL/min, 250rpm in 5 days of carbonation .......... 102
Table 4.9. Major compound composition (%w/w in average) of RM and carbonated
RM at 15%-60% CO2, TF of 200mL/min, 250rpm in 5 days of carbonation .......... 103
Table 4.10. Major elemental composition (%w/w in average) of RM and carbonated
RM at 30% CO2 concentration, 250rpm and different total gas flow rate ............... 104
xviii
Table 4.11. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2 concentration, 250rpm and different total gas flow rate ............... 105
Table 4.12. Major element composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min and different stirring speeds ........................... 107
Table 4.13. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min and different stirring speeds ........................... 107
Table 4.14. Major element composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min, 250rpm and different solids concentrations .. 108
Table 4.15. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min, 250rpm and different solids concentrations .. 109
Table 4.16. Concentration of raw RM and liquor .................................................... 123
Table 4.17. Solid precipitation/dissolution reactions in red mud model.................. 125
Table 4.18. Solid dissolution/precipitation and liquid reactions in RM simulation 131
Table A-1. Rapid titration of RM by 0.1N HCl ....................................................... 165
Table A-2. Rapid titration of RM liquor to pH 4.5 by 0.1N HCl ............................ 166
Table A-3. Long-term titration of RM to pH 4.5 by 0.1N HCl................................ 167
Table A-4. Long-term titration of RM to pH 6.0 by 0.1N HCl................................ 168
Table A-5. Long-term titration of RM to pH 8.0 by 0.1N HCl................................ 169
Table A-6. Long-term titration of RM to pH 10 by 0.1N HCl................................. 170
Table A-7. Metal concentrations in RM liquor measured at different pH values .... 171
Table A-8. Simulated metal concentrations in RM liquor at different pH values ... 171
Table B-1. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 100mL/min, stirring speed of 250rpm ................................................................. 179
Table B-2. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 100mL/min, stirring speed of 250rpm ................................................................. 180
Table B-3. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 200mL/min, stirring speed of 250rpm ................................................................. 181
Table B-4. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 200mL/min, stirring speed of 250rpm ................................................................. 182
Table B-5. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 300mL/min, stirring speed of 250rpm ................................................................. 183
xix
Table B-6. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 300mL/min, stirring speed of 250rpm ................................................................. 184
Table B-7. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 400mL/min, stirring speed of 250rpm ................................................................. 185
Table B-8. Carbonation of RM at different CO2 concentrations and total gas flow rate
of 400mL/min, stirring speed of 250rpm ................................................................. 186
Table B-9. Carbonation of RM by 30% CO2 concentrations, stirring speed of 250rpm
and different total gas flow rate ............................................................................... 187
Table B-10. Carbonation of RM by 30% CO2 concentrations, total gas flow rate of
200mL/min and different stirring speeds ................................................................. 188
Table B-11. Carbonation of RM by 30% CO2 concentrations, TF of 200mL/min,
speeds of 250rpm and different solids concentrations in RM .................................. 189
Table B-12. Longer carbonation of RM at 15% - 30% CO2 concentrations, TF of
200mL/min and stirring speed of 250rpm ................................................................ 190
Table B-13. Longer carbonation of RM at 40% - 60% CO2 concentrations, TF of
200mL/min and stirring speed of 250rpm ................................................................ 190
Table B-14. Longer carbonation of RM at by 30% CO2 concentrations, stirring speed
of 250rpm and different total gas flow rate .............................................................. 191
Table B-15. Longer carbonation of RM at by 30% CO2 concentrations, stirring speed
of 250rpm, TF of 200mL/min and different solids concentrations in RM............... 191
Table B-16. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 100mL/min and stirring speed of 250rpm .................................................... 192
Table B-17. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 100mL/min and stirring speed of 250rpm .................................................... 193
Table B-18. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 200mL/min and stirring speed of 250rpm .................................................... 194
Table B-19. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 200mL/min and stirring speed of 250rpm .................................................... 195
Table B-20. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 300mL/min and stirring speed of 250rpm .................................................... 196
Table B-21. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 300mL/min and stirring speed of 250rpm .................................................... 197
xx
Table B-22. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 400mL/min and stirring speed of 250rpm .................................................... 198
Table B-23. Carbonation of RM liquor at different CO2 concentrations, total gas flow
rate of 400mL/min and stirring speed of 250rpm .................................................... 199
Table D-1. Simulated carbonation of RM at different CO2 concentrations and total
gas flow rate of 100mL/min ..................................................................................... 202
Table D-2. Simulated carbonation of RM at different CO2 concentrations and total
gas flow rate of 200mL/min ..................................................................................... 202
Table D-3. Simulated carbonation of RM at different CO2 concentrations and total
gas flow rate of 300mL/min ..................................................................................... 203
Table D-4. Simulated carbonation of RM at different CO2 concentrations and total
gas flow rate of 400mL/min ..................................................................................... 203
1
CHAPTER 1 INTRODUCTION
1.1. Background
Bauxite residue, commonly called red mud (RM), is a principal waste products
created from the Bayer process in the alumina production. It is composed of hydrous
muddy silt combined with very high alkaline solid waste with a pH in the range of
10.5-13 (Borges et al. 2011). This cocktail is generated by physical and chemical
treatments of bauxite in the alumina production (György & Tran 2008; Zhang et al.
2001). The worldwide alumina industry has experienced significant growth over past
decades and many producing countries are now coping with severe problems in
managing bauxite residues. The quantity of RM has been increasing alarmingly due
to the rapid increases in the global demand of aluminium metal for the development
of construction, packaging, and transportation.
Normally, the production of 1 tonne of alumina results in the generation of 1-
1.5 tonnes of RM (Brunori et al. 2005; Kumar et al. 2006). However, depending on
the efficiency of alumina extraction and quality of bauxite source, the quantity of
waste generated may vary from 0.3 to 2.5 tonnes for the high-grade or for very low
grade bauxite, respectively (Borges et al. 2011; Kalkan 2006; Nguyen & Boger
1998; Paramguru et al. 2005; Sushil & Batra 2008). Consequently, the volume of
bauxite residue waste has grown exponentially. In 1940, the inventory of bauxite
waste was 22 million tonnes associated with the annual production rate of 1 million
tonnes of aluminium metal. By 1985, the generation rate of the waste grew to
roughly 48.5 million tonnes per annum, and the global inventory of red mud reached
2
1 billion tonnes (Klauber et al. 2011; Power et al. 2009, 2011). In 2007, it was
predicted that approximately 120 million tonnes of red mud were produced per year
and the inventory of the waste had grown to 2.7 billion tonnes (Fig. 1.1) (Clark et al.
2009; Klauber et al. 2011; Power et al. 2011). Of greater significance was the rapid
increase in the rate of waste generation. It can be seen that the first billion tonnes of
RM were produced within nearly a century, but the second billion tonnes were
accumulated approximately 15 years.
Figure 1.1. Global production rate and cumulative inventory (Power et al. 2011)
Future predictions suggest that approximately 3 billion tonnes of RM will be
generated in the period from 2010 to 2015 (Power et al. 2009, 2011). The bauxite
residue tonnage was often estimated by applying an overall ratio of 1.5 to alumina
production (Power et al. 2011; Sutar et al. 2014). In 2012, the global alumina
production as reported by Dentoni et al. (2014) reached 90.17 million tonnes. It is
projected that the annual amount of bauxite residue will be generated about 135
million tonnes worldwide, thus, the global production of red mud could possibly
reach over 5 billion tonnes in 2030. This mass of bauxite residue will pose a
significant environmental problem as the RM is highly alkaline (pH>13), fine-
3
grained in nature (more than 90% is under 10µm), and contains elevated
concentrations of sodium (over 50g/kg of RM) (Johnston et al. 2010). Clearly,
appropriate management methods and utilisation practices must be devised to
improve this problem. Unfortunately, there are no acceptable solutions for dealing
with the RM in terms of its management and its potential utilisation. Treating the
waste would lead to a significant penalty to production cost. As a result, it seems that
the RM disposal problem may be ignored by the industry, the public and
governments (Power et al. 2009, 2011).
The most important constraints to the management, utilisation, and remediation
of RM are its very high level of sodicity and alkalinity, and the complex physical and
chemical properties. These properties result in the high impedance of reusing red
mud. One of the efforts amongst the several pH-reduction processes to be
incorporated is neutralisation by acid. To determine the amount of acid needed for
neutralisation, the acid neutralisation capacity (ANC) of red mud must be measured.
In addition, the contributions of both solid and liquid phases to the ANC must be
identified. This process is not only challenging due to the complicated chemical
constitutes in red mud, but because of slow reactions taking place between the acid
and the solids. The ANC of red mud was measured by rapid and long-term acid
titrations. Data from the titration of red mud slurries was fitted by using chemical
equilibrium modelling to identify the aqueous and solid reactions in the system.
Furthermore, metal concentrations from long-term titrations at pH endpoints of 4.5,
6, 8 and 10 were determined to confirm the dissolution/precipitation of Na, Al, Ca
and Fe corresponding to their solid phases (Khaitan, et al. 2009a).
4
Another promising effort is the use of industrial wastes such as CO2 emission
for solving the problems of other wastes. The capture and storage of CO2 by RM
could play a significant role in mitigating the release of greenhouse gases to the
environment. It is estimated that one third of all CO2 emissions due to human’s
activities derive from fossil fuels used for power plants, which annually emit several
million tonnes of CO2 (Rai 2013). The aluminium industry is of special interest to the
global warming mitigation strategy as it is one of the largest energy consumers with
approximately 14MWh/ton of aluminium required for the aluminium refining
process.
The use of CO2 from the atmosphere or from industrial emissions is a
potentially significant source of acid for neutralizing red mud. Up to date, a variety
of solutions have been conducted for neutralising RM by aqueous CO2, (Enick et al.
2001; Shi et al. 2000), SO2 from flue gases (Fois et al. 2007), seawater treatment
(Cooling 2007; Hanahan et al. 2004), Mg and Ca rich salts, acid treatment
(Bonenfant et al. 2008; Khaitan et al. 2009b). Also, CO2-containing emission gas or
CO2 gas phase can be bubbled through red mud slurries to form the carbonic acid in
the aqueous phase that would react with basic components of red mud lowering the
pH of slurries (Szirmai et al. 1991). The extent of neutralisation of red mud by CO2
gas as a function of CO2 partial pressure has been measured to identify the
geochemical reactions responsible for carbon sequestration (Khaitan et al. 2009b).
Carbonation of red mud was observed to be affected by operating parameters.
Cardile et al. (1994) characterised the influence of temperature, gas flow rate, stirring
rate, total sodium content and slurry density by using factorial experiment design.
These operating factors can help to optimise the carbonation process.
5
Although the pH of the aqueous phase decreased quickly from a pH of 12-13 to
7-8 when red mud contacting with CO2 gas, it soon rose again to unacceptable levels
because of additional alkaline material leaching from red mud. In a study of red mud
carbonation by CO2 (Rai 2013), the pH was observed to rebound back to 9.5 each
time the red mud was neutralised in the multiple cycle experiments. In another study
of red mud neutralisation using CO2 sequestration cycle (Sahu et al. 2010), the pH of
carbonated slurry rebound slowly for few days after that it remained at 8.45. The pH
recovery back to 9.5 was also observed when red mud was carbonated by pure CO2
for 14 hours (Cardile et al. 1994). Additionally, the phenomenon of pH rebound was
found to be inversely proportional to the partial pressure of CO2 used for carbonation
of red mud (Khaitan et al. 2009b).
This research aims to investigate the RM neutralisation by carbon dioxide gas
over a range of different operating conditions such as CO2 concentrations, total gas
flow rate, stirring speeds and concentrations of RM. Furthermore, it was to identify
the equilibrium pH reached for different concentrations of CO2 and to understand the
potential for carbon sequestration by red mud. Liquid and solid phase reactions that
control the carbonation of red mud slurries were investigated with the assistance of
chemical equilibrium modelling. The final equilibrium pH values affected by
different concentrations of CO2 during carbonation were also simulated and
compared to the experimental values. Finally, the calculation of CO2 volume
permanently captured by RM should also be examined.
6
1.2. Objectives
Nowadays, global warming and climate change have been becoming the most
challenging environmental problems as the atmospheric CO2 concentration would be
reaching from 500-700ppm by the end of the 21st century (IPCC 2014). Global
warming is of special interest to alumina industry because it requires a high demand
of energy and disposes of large amounts of CO2 emissions and waste red mud.
Annually, direct atmospheric emissions of greenhouse gases such as CO2 from
bauxite refinery and aluminium smelting have been estimated worldwide at 110
million tonnes of carbon dioxide (Martcheck 2003). Additionally, approximately 135
million tonnes of red mud have been generated per year from the Bayer process in
over the world by applying the ratio of 1.5 to alumina production (Sutar et al. 2014).
The disposal of this waste is often expensive and may cause negative impacts on the
environment. Therefore, the reuse of red mud as an important material for the capture
of CO2 could potentially make a significant contribution to the global warming
mitigation. This research aims to focus on the investigation of neutralisation of red
mud using CO2 gas with the following objectives.
1. To investigate the acid neutralisation capacity (ANC) of red mud and
evaluate the contribution extent of liquid phase and solid phase to the ANC of
red mud.
2. To characterise the carbonation of red mud over a range of different operating
conditions such as concentrations of CO2, total flow rate of gas, stirring
speeds and concentrations of red mud in order to optimise the carbonation
process.
7
3. To characterise the materials using a range of techniques such as XRD, SEM,
EDX, FT-IR, TG-DSC, ICP-MS, CHN, and Total Organic Carbon (TOC)
Analyser.
4. To determine and evaluate the effect of operating parameters on the
carbonation process, then suggest the optimal condition for the carbonation
process.
5. To simulate the effects of different concentrations of CO2 on the equilibrium
pH values, then compared to the experimental values.
6. To calculate the potential amount of CO2 captured by the red mud over the
range of these conditions.
1.3. Organisation of Thesis
The thesis consists of 6 chapters. Chapter 1 describes background to the
problems studied. Chapter 2 provides a comprehensive literature review of the
process of bauxite residue generation and the properties of RM. In addition, this
chapter addresses current methods of RM management and treatment. Research
methodology and techniques deployed to conduct the research are detailed in Chapter
3. Chapter 4 presents and discusses the results and outcomes of the research. Chapter
5 summarises the findings and conclusion of the project and scope for future work is
detailed in Chapter 6.
8
CHAPTER 2 LITERATURE REVIEW
2.1. Overview of Bauxite Residue Management
2.1.1. Bauxite Residue Generation Process
Bauxite residue or red mud is produced as an undesirable by-product during the
extraction of alumina from bauxite ore in the Bayer process as shown in Figure 2.1.
This process was patented by Karl Josef Bayer in the period from 1887 to 1892
(Power et al. 2011; Sparks 2010). Now, the process becomes dominant for the
production of alumina because of its unexpected success. Though this process
contributes significantly to the alumina processing industry, it results in an
unavoidable consequence. The key issue is the generation of the very large amount
of difficult to treat waste. This waste could provide a source of valuable metals if an
economic treatment could be devised. The exponential increase in the RM produced
was due to the development of the Hall-Heroult process for smelting alumina to
aluminium in the 19th century (Klauber et al. 2009).
As can be seen from Figure 2.1, the Bayer process includes a variety of
processes that treat bauxite feed to produce RM waste and product liquor. Generally,
the Bayer process consists of bauxite milling, pre-desilication, digestion,
clarification, and of counter-current decantation (CCD) washing or thickener series.
However, before discharging the RM into a bauxite residue disposal area, a further
thickening or filtration step was employed. Each of these unit processes has an
impact on the chemical composition of the residue, as well as their physical and
mineralogical properties of red mud (Power et al. 2009).
9
Figure 2.1. Schematic of a general Bayer process (Power et al. 2011)
In the Bayer process, gibbsite (Al(OH)3) and/or boehmite (AlOOH) from
bauxite were extracted by dissolving these minerals in the concentrated NaOH
solution. The elemental composition, expressed in terms of oxide, of Al, Si, Ti, Fe,
Na and Ca, in the red mud varies from places to places as shown in Table 2.1. This
could be due to a difference in the chemical composition of the bauxite used as feed.
To minimise silica contamination of the product, lime (Ca(OH)2) may be added in
the pre-desilication stage, just before the digestion stage, to form insoluble
cancrinite. Sodalite is the most common desilication products formed during the pre-
desilication process, whereas cancrinite might be formed in the presence of Ca
during the digestion of boehmite-rich bauxite with high temperature. The
10
concentrations of sodalite and cancrinite were identified to have value of 16-24%
(Castaldi et al. 2008), and up to 50% (Garau et al. 2007), respectively in bauxite
residues from Eurallumina plant processing Weipa bauxite. However, these sodalite
and cancrinite were often disposed with the bauxite residues. These products
contributed significantly to the acid neutralisation capacity of the red mud (Gräfe &
Klauber 2011). Other additives such as iron oxides (Cook 1970), MgSO4, apatite
solids (Bernard et al. 2007), and other unwanted impurities were either separated
from the NaAl(OH)4 rich solution, to be used for specific purposes or be disposed of
with the bauxite residues.
The management of red mud starts from the separation process of green liquor
after digestion (Power et al. 2011). The residue will, then, be transferred to a series
of washing units called counter-current decantation (CCD), where the residue will be
washed and NaOH and Al(OH)4- will be recovered for recycling back to the process.
After washing in the CCD, the collected waste is often subjected to a further
thickening or filtration process to increase the solids content before being discharged
into the disposal area (Power et al. 2011).
Basically, bauxite residue comprises essentially insoluble compounds of Fe, Ti
and some un-digested soluble alumina minerals and others (György & Tran 2008).
Bauxite residue also consists of the quantity of coarse particles called “sand”. The
amount of sand can range from <1% to about 50% depending on the bauxite source,
but the average is about 5% (György & Tran 2008). The large quantity of RM
generated from the Bayer process is usually discharged directly into the environment
via a tailing dam without further treatment. This direct disposal method, being cheap
and convenient, will pose an adverse impact on the ecosystem because of high
alkalinity level of the RM.
11
2.1.2. Physicochemical and Mineralogical Properties of Red Mud
Red mud is a very fine material with a particle size distribution, typically
varying from 1 to 150µm (over 90% by volume is <10µm) (Clark et al. 2009;
Johnston et al. 2010). It has a very low settling rate (1.4x10-6-3.6x10-3m/s), a range
of density of 2.6x103-3.5x103kg/m3 (György & Tran 2008), and a high surface area
(Sushil & Batra 2008). The pH of RM varies from 10 to 13.5, therefore, it is a highly
caustic waste, which can harm the environment (György & Tran 2008; Johnston et
al. 2010; Sushil & Batra 2008).
Red mud consists of a mixture of metal oxides, which were originally present
in bauxite ore and may be created in the Bayer process (Agrawal et al. 2004; Singh et
al. 1997; Sushil & Batra 2008). Depending on the bauxite source and the parameters
of digestion process, the chemical and mineralogical properties of RM can vary
significantly (Altundogan et al. 2000; Bertocchi et al. 2006; Cengeloglu et al. 2003;
Collazo et al. 2005; Gong & Yang 2000; György & Tran 2008; Koumanova et al.
1997; Park & Jun 2005; Pontikes et al. 2007; Tsakiridis et al. 2004; Yalcin & Sevinc
2000). The main compositions of RM comprise a significant quantities of metal
oxides such as Fe2O3, Al2O3, SiO2, TiO2, Na2O, CaO, MgO and several other minor
elements like K, Cr, V, Ni, Cu, Zn, Mn, etc. (Singh et al. 1997; Sushil & Batra
2008). Table 2.1 shows the chemical compositions of RM produced in various plants.
The mineralogical composition of bauxite determined the preferred reaction
temperature, pressure, and concentration of NaOH to be employed in the extraction
process (Gräfe et al. 2011). These preferred reaction conditions in the digestion often
lead to the consequence of mineralisation in the Bayer process, which is related to
12
the common equilibrium reactions of liquid and solids in bauxite residues listed in
Tables 2.2 & 2.3.
Table 2.1. Chemical constituents of red mud in different locations (%wt) (adapted
from György & Tran (2008))
Weipa
(Australia)
Trombetas
(Brasil)
South
Manchester
(Jamaica)
Darling
Range
(Australia)
Iszka
(Hungary)
Parnasse
(Greece)
Digestion
temperature 2400C 1430C 2450C 1430C 2400C 2600C
Al2O3 17.2 13.0 10.7 14.9 14.4 13.0
SiO2 15.0 12.9 3.0 42.6 12.5 12.0
Fe2O3 36.0 52.1 61.9 28.0 38.0 41.0
TiO2 12.0 4.2 8.1 2.0 5.5 6.2
Na2O 9.0 9.0 2.3 1.2 7.5 7.5
CaO - 1.4 2.8 2.4 7.6 10.9
L.O.I 7.3 6.4 8.4 6.5 9.6 7.1
Others 3.5 1.0 2.8 2.4 4.9 2.3
Buffering reactions shown in Tables 2.2 & 2.3 represent the ability of the
solids in red mud to maintain the concentration of alkaline anions in the solution. The
solids that are likely soluble to some degree, and H+ acceptance by the alkaline anion
in the solution are necessary for the buffering reactions taking place. At pH>10.2,
Ca2+ species is absent as it is lost to the formation of insoluble calcite (CaCO3), and
therefore, Na2CO3 governs the concentration of HCO3-/CO3
2- in the solution. Apart
from that, there are some major alkaline anions existing in red mud such as HCO3-
/CO32-, Al(OH)4
- and OH- that are responsible for the buffering property.
Furthermore, some systems H2SiO42-/H3SiO4
-/H4SiO4 and PO43-/HPO4
2-/H2PO4-,
which are at lower concentrations, may also help to buffer the pH of solution as well.
As shown in Table 2.2, the pH region between two pKa values is the buffering
region, for instance the buffer region of HCO3- is around pH of 8.3 (the pH average
of 10.2 and 6.35), and the precipitation of gibbsite Al(OH)3 occurs rapidly at pH
below 10 (Gräfe et al. 2011).
13
Table 2.2. Buffering common reactions in aqueous solution of bauxite residue
(adapted from (Stumm & Morgan 1981))
Reaction Acidity constants
OH− + H3O + ⇆ 2H2O
Al(OH)4−·2H2O + H3O+ ⇆ Al(OH)3·3H2O(s) + 2H2O
CO32− + H2O ⇆ HCO3
− + OH−
HCO3− + H3O+ ⇆ H2CO3 + OH−
H2SiO42− + H2O ⇆ H3SiO4
–
H3SiO4− + H2O ⇆ H4SiO4
PO43− + H2O ⇆ HPO4
2− + OH−
HPO42− + H2O ⇆ H2PO4 + OH−
H2PO4− + H3O+ ⇆ H3PO4 + OH−
pKw = 14.0
pKa4 ~10.2
pKa2 = 10.2
pKa1 = 6.35
pKa2 = 12.95
pKa1 = 9.85
pKa3 = 12.35
pKa2 = 7.2
pKa1 = 2.25
The introduction of slaked lime (Ca(OH)2) at different stages before, during
and after digestion results in the formation of Bayer process characteristic solids
(BPCSs) (Whittington 1996). These solids impart a significant buffering capacity to
the red mud solutions via a number of reactions as listed in Table 2.3. The abundance
of these solids relies on the conditions of bauxite processing. Therefore, the addition
of Ca(OH)2 helps not only enhance the overall extraction of gibbsite and/or boehmite
from bauxite, but also reduce the formation of sodalite by instead transforming some
to cancrinite. Furthermore, the introduction of Ca(OH)2 also favours the formation of
TCA or hydrogrossular by the exchange of 4OH- for one SiO44-, although this
substitution reaction seems limited (Whittington 1996). As shown in Table 2.3, it is
impossible to evaluate the buffering capacity of most the solids from their reactions
because of the paucity of dissolution constants and/or solubility products.
Nevertheless, the BPCSs must be responsible for the buffering effect because the
oxide minerals of Fe, Al, Ti and Si are not the buffering solids (Gräfe et al. 2011).
14
Table 2.3. Dissolution reactions of common buffering solids present in bauxite
residues (adapted from (Greenberg & Chang 1965; Greenberg et al. 1960; Stumm &
Morgan 1981; Vieillard & Rassineux 1992))
Dissolution reaction Solubility constants
Natron-decahydrate
Na2CO3·10H2O(s) + H2O ⇆ 2Na+ + HCO3− + OH− + 10H2O
Calcite
CaCO3(s) ⇆ Ca2+ + CO32-
Hydrocalumite
Ca4Al2(OH)12·CO3·6H2O + 7H2O ⇆ 4Ca2+ + 2Al(OH)3(aq) +
HCO3− + 7OH− + 6H2O
Tri-calcium aluminate (TCA or hydrogrossular, n = 0)
Ca3Al2[(OH)12−4n](SiO4)n(s) + H2O ⇆ 3Ca2+ + 2Al(OH)3 + 6OH−
Hydroxysodalite
Na6[Al6Si6O24]∙2NaOH + 24H2O ⇆ 8Na+ + 8OH− + 6Al(OH)3 +
6H4SiO4
Cancrinite
Na6[Al6Si6O24]∙2CaCO3 + 26H2O ⇆ 6Na+ +
2Ca2++ 8OH− + 2HCO3− + 6Al(OH)3 + 6H4SiO4
pKsp = 1.31
pKsp = 8.42 (6.2)
pKsp = n/d
pKsp = n/d
pKsp = n/d
pKsp = n/d
2.1.3. Methods Utilised for Disposal of Red Mud
Prior to the 1970s, two major conventional wet disposal methods were utilised
in most alumina refineries, namely marine discharge and lagooning (Klauber et al.
2011). Generally, the marine disposal practice sounds simpler than the lagooning in
terms of economy and technology. However, these practices make a significant
contribution to the management of RM at that time.
2.1.3.1. Marine Disposal Practice
Marine discharge method was employed in France, Greece, the USA, Australia
and Japan for several years given the neutralisation capacity of seawater to the
causticity of RM (György & Tran 2008). During marine disposal, the RM slurry
15
produced from the Bayer process is directly discharged into the deep ocean via the
pipeline system. Although reports on environmental impact assessment revealed no
negative impacts on the coastal ecosystem at the sedimentation areas in Japan, this
disposal method has not been encouraged by the United Nations Industrial
Development Organisation (UNIDO 1985). Consequently, a transition from ocean or
river disposal to lagooning or alternative methods has been adopted, and the
available literature revealed that no refineries established after 1970 employed the
marine disposal method. Currently, 2-3% of alumina production worldwide use
marine discharge for the treatment of red mud (Power et al. 2011).
The Gramercy plant in Louisiana in the USA dumped its red mud into the
Mississippi river until 1974, when it adopted lagooning for waste storage and
subsequent treatment. The transition removing RM from the river and relocating it to
land-based impoundments was voluntarily initiated by the Kaiser Company
(Kirkpatrick 1996). Some Japanese refineries consider marine discharge a main
method due to the limitation of land area available for disposal. Nevertheless, these
refineries made a commitment to International Marine Organisation to stop dumping
their RM waste into the ocean by 2015, and will conduct other alternative methods
for disposal of red mud (International Marine Organisation 2005). Similarly,
refineries at Gardanne in France, which still utilise marine disposal method, are also
scheduled to stop discharging their waste into the sea by 2015 (Martinet-Catalot et
al. 2002).
2.1.3.2. Lagooning Method
The use of lagoon is now normally practised for dumping red mud (György &
Tran 2008). The bauxite residue slurry is directly discharged into the land-based
16
impoundments (Power et al. 2011) called a red mud pond (Fig. 2.2). The practice
requires increased engineering input considering topography, linings and material’s
issues, and construction complexity, etc. when compared to the marine option, but it
is now widely applied in many refineries in the world. These requirements aim to
reduce potential leakage of caustic and alkaline water into the soil and ground water.
Unfortunately, in many plants built before 1960, there was often no special sealing
applied at the bottom of the pond. This resulted in the contamination of the
surrounding soils and ground water, with consequent negative impacts on human’s
health (György & Tran 2008).
Figure 2.2. Lagooning red mud disposal (György & Tran 2008)
It is suggested that, the best lagooning method is to line the pond with a single
compacted clay bed or multiple ones of up to 300-400mm thick in order that the red
mud can be separated from the original soil or rock of the pond (György & Tran
2008; Liu et al. 2009). Moreover, another security solution can be employed by
utilising multiple layers of plastic or geo-membrane materials to create a seal
between the red mud and the supporting clay bed (Cooling 1989).
17
Nevertheless, previous studies have noted that the caustic soda content of the
bauxite residue reacts with the clays over several decades to form amorphous sodium
aluminium hydro silicates and ultimately zeolite through a series of complex
reactions (Gerrise & Thomas 2008). These reactions increase the hydraulic
conductivity of the clay bed, resulting in potential contamination of the aquatic
system after several decades. These problems could be reduced by neutralising the
RM slurry before discharging, for example, by using mineral acids such as sulphuric
acid, or by mixing the slurry with seawater (Power et al. 2009, 2011).
Generally, lagooning method is the simplest land-based disposal that is
globally utilised. However, this practice is still dependent on good engineering
practice for residue storage; hence, it may lead to the high cost of operation. In
addition, this method is often more problematic due to its high risk of leakage,
liquefaction, and instability. The statistics show that around 103 major tailings dam
failures have been recorded worldwide since 1960, leading to at least 1,838 human
deaths and untold environmental degradation (WISE 2015). The newest red mud
problem that tragically happened in Hungary in October 2010 (Enserink 2010), and
the other two tragedies of Hpakant jade plant in Myanmar and Germano iron mine in
Brasil demonstrated in November 2015 (WISE 2015) was the tailings dams failures
killing 140 people and causing long-term environmental damage.
2.1.3.3. Dry Stacking Method
An alternative disposal method for the storage of bauxite residue called dry
stacking has been introduced. This method relates to the progressive deposition of
dewatered red mud onto sloped and under-drained drying bed (Sofrá & Boger 2002)
to facilitate the consolidation process. This practice shows a paradigm shift of
18
engineering. In addition, the tailings should be engineered to meet the requirements
of residue discharge. This means it is very important to understand how the residue
transportation process and the characteristics of deposition can be influenced by
material properties and operational parameters (Sofrá & Boger 2002).
The dry stacking process was employed the first time at the Burntisland
Alumina plant in Scotland in 1941 (György & Tran 2008). Unlike lagooning, dry
mud stacking is termed thickened tailings disposal and implies that the washed
residue slurry is dewatered or consolidated to a paste with an initial solids content of
approximately 48-55% prior to disposal as shown in Figure 2.3 (Alcoa 2011).
Therefore, it was considered to be cost-effective and less environmental problems
(Purnell 1986).
The next pioneer in conducting dry stacking method in the early 1970s was the
Giulini GmbH refinery in Germany (Haerter & Shefer 1975). This practice considers
reducing the area of land needed for disposal and maximising the return of soda and
alumina to the process. This practice was introduced at the Alcan refinery followed
by the achievement of Alcoa in 1985 (Paradis 1992). By comparison with the
lagooning method, two sealing layers are simultaneously needed at the bottom of the
disposal site. The first is an at least 600mm thickness compacted clay liner, which is
placed at the bottom. Then, a plastic membrane, which is often high-density
polyethylene (HDPE) is used as the upper layer. This plastic layer offers a good
resistance to high soda and pH environment (György & Tran 2008).
19
Figure 2.3. Schematic of dry stacking system (Alcoa 2011)
2.1.3.4. Dry Cake Disposal Method
Dry cake disposal is also a dry disposal practice, where the slurry is
mechanically dewatered as much as possible in order to create a dry cake with more
than 65% of solid content before discharging (Power et al. 2011). The most
significant feature that differs from dry stacking practice is that there is no further
dewatering once the red mud waste has been dumped at the storage site. Similar to
the other practices, this method also offers an advantage feature that risk of alkalinity
and caustics can be further diminished by neutralising or washing in the filtration on
vacuum drum filters (Shah & Gararia 1995).
20
However, it is impossible to obtain a dry cake by using only means of
thickening; therefore, it does require a filtration phase (Power et al. 2011). This
combination is essential to overcome the capillary resistance inside the residue’s
pores in order to dewater the cake thoroughly. This practice has been successfully
deployed in a pilot scale, for example at the Hindalco plant at Renukoot (Shah &
Gararia 1995) and the Aluminiumoxid Stade (Germany) plant with the solids content
of over 75% achieved (Bott et al. 2002). However, this practice seems to be unlikely
specified for utilisation in alumina refineries as such in available information because
the dry stacking is often used as its advantages and preferred disposal strategy.
2.2. The Red Mud Utilisation Options
It is clear that the current rate of red mud production and its alkaline and
caustic properties will pose adverse impacts on the environment and human health if
there are no proper minimisation and management solutions. Currently, many
alumina plants in the world are still discharging the red mud into the ocean, or land-
based impoundments, or dry stacking ponds. These practices are just conventional
solutions for storing red mud in different ways. They will not actually help to
minimise and reuse this waste as a valuable material in terms of economic,
environmental, and social considerations. Confronted with the problem, many
different technologies have been developed in the efforts of long-term remediation
and utilisation applications of red mud in order to minimise to the environmentally
acceptable level.
21
2.2.1. The Utilisations of Red Mud in Construction
Building materials seem to be one of the most successful applications that
could consume a significant red mud quantity. The study conducted in early 1936
(Thakur & Sant 1983a) showed that the contents of iron and alumina in red mud can
contribute a crucial part to the setting and strength properties of the cement, but the
soda is detrimental. The use of calcium replacing soda in the study can foster the
ability of red mud as an additive; however, the process of residue calcinations has to
be conducted at a very high temperature of 10000C. Similarly, Liu et al. (2009) also
advocated that the production of cement from the mix of 50% red mud and other
solid wastes and modifiers has reached the standard of Portland 42.5 cement, but the
limited content of sodium in cement according to the newest standard could lead to
the prohibition of using red mud in cement production.
Additionally, research on using a mixture of gypsum, bauxite and red mud for
preparing special cements was performed by Singh et al. (1996, 1997). The authors
concluded that with the red mud content of 20-50% by dry weight added in the
composition of the material mix, it is possible to produce cements with superior
setting strengths when compared to normal Portland cement. A case study on using
red mud for manufacturing bricks and blocks have been conducted in Jamaica by
McCarthy et al. (1992). It indicated that the manufacturing of bricks from residue has
illustrated the technical feasibility, and the alumina producers were willing to support
the development of the process. However, the main brick properties such as long-
term stability, leaching and ionising radiation were not mentioned. Pinnock (1991;
1999) reported that the radon levels from bauxite residue bricks recorded were
22
approximately 2-3 times higher than that of the conventional concrete. This will pose
an impact on human health.
A recent investigation on developing unsintered bricks from red mud generated
by Shandong Aluminium Plant in China has been conducted (Yang & Xiao 2008). In
this study, researchers considered five key experimental plans in order to identify the
optimal ratio of red mud to other materials such as fly ash, the percentage of sand,
the effect of lime, gypsum, and Portland cement. It is noteworthy that the
manufacturing process of unfired bricks was implemented at the ambient
temperature. The results concluded that the optimal percentages (in weight) of red
mud bricks consist of red mud (25-40%), fly ash (18-28%), sand (30-35%), lime (8-
10%), gypsum (1-3%) and Portland cement (just 1%). The products have met the
first class brick standard of China with good durability in severe climatic conditions
and high strength. Furthermore, such products can compete with traditional bricks as
the production process consumed much lower energy and of low cost.
2.2.2. The Utilisations of Red Mud in Chemical Applications
In the context of using bauxite residue as catalysts, many potential applications
have been explored due to the presence of Fe2O3 and TiO2 and its high surface area
and the low cost of the source materials (Klauber et al. 2009, 2011). Recent studies
conducted by Sushil & Batra (2008, 2012) have demonstrated the use of red mud as a
catalyst in hydrogenation, hydro-dechlorination, exhausted gas clean-up and other
areas. These authors concluded that unmodified red mud had a poor performance as a
catalyst compared with pure iron oxide or commercial catalysts.
In the production of ceramics, red mud is considered a beneficial material that
creates a barrier for radiation shielding (Amritphale et al. 2007). This is because of
23
the formation of a dense ceramic matrix produced from liquid phase sintering. This
matrix demonstrated that the shielding thickness and compressive and impact
strength being better compared to normal Portland cement-based shielding materials.
Although the applications in this area are really broad, they consume a low volume
of bauxite residue and require some pre-treatments.
2.2.3. The Utilisations of Red Mud in Metallurgy
There were about 135 patents granted between 1964 and 2008 related to
metallurgical applications, 17% out of which are subject to bauxite residue (Klauber
et al. 2011). The major metals recovered from bauxite residue are iron, aluminium,
titanium, and sodium. The process of iron, aluminium and titanium recovery
separately and in combination has been reviewed in previous studies (Paramguru et
al. 2005; Thakur & Sant 1983b). However, the available literature showed that this
proposed recovery process is technically complex requiring large investment in an
energy intensive plant (Fig. 2.4) (Piga et al. 1993). As a result, there is no large-scale
extraction of metals from bauxite residue to date.
24
Figure 2.4. A possible flowsheet for recovery of Fe, Al, and Ti from bauxite residue
(Piga et al. 1993)
2.2.4. The Utilisations of Red Mud in Agriculture
A number of investigators have studied the potential use of untreated red mud
to improve soil condition for agriculture or safety given its ability for fixation of
heavy metals (Liu et al. 2011). The biological and chemical assessments study of in
situ fixation of metals in soil using red mud conducted by Lombi et al. (2002a;
2002b) is a typical example. This study reported a remarkable decrease in the
concentration of heavy metals in the soil pore water following treatment using 2%
red mud and 5% beringite (an alkaline alumino silicate). The research also confirmed
that red mud treatment could significantly accelerate the production of soil microbial
biomass.
25
Similarly, research has been conducted using batch, pot and field experiments
involving gravel sludge and red mud to deal with contaminated agricultural soils near
a former Pb/Zn smelter in Austria (Friesl et al. 2006). The field experiment results
showed that heavy metals such as Cd, Pb and Zn could be minimised by up to 96%,
99% and 99%, respectively in treated soils. As well, red mud can retain the content
of phosphorus significantly and reduce soil acidity effectively (Liu et al. 2011; Snars
et al. 2004; Summers et al. 2001). Furthermore, in a pot experiment, Friesl et al.
(2003) found that compared to unamended soil, red mud could considerably decrease
Cd, Zn, and Ni uptake in fescue and amaranthus (Amaranthus hybridus L.) by up to
87%, 81% and 87%, respectively.
It is evident that bauxite residue can improve the cycle of phosphorus in
agricultural areas with sandy soils having a low phosphate and other nutrient holding
capacities (Klauber et al. 2011). This occurs because the mud has two useful
properties with respect to phosphate cycling, namely reducing phosphate leaching
into ground and surface water, and creating a phosphate pool which is subsequently
available to plants for improved growth. A number of studies employed by the
Western Australian Department of Agriculture (Summers et al. 1996a; Summers et
al. 1996b; Summers et al. 1993; Summers & Pech 1997; Summers et al. 1999)
reported that bauxite residue is useful for improving P retention, reducing run-off
into the Peel Inlet and the Harvey Estuary by up to 75%. Concurrently, pasture yields
were increased by 25% on average and in well-controlled areas by up to 200%. This
retention of P in the soil reduces the damaging effects of eutrophication caused by
phosphate leaching into the Peel-Harvey ecosystems.
26
2.2.5. The Utilisations of Red Mud in Environmental Treatment
Bauxite residue can be used to mitigate environmental problems given its
capacity to remove metals and metalloids due to its high alkalinity and oxidising
potential. The mud’s alkalinity causes most metal ions to hydrolysed to form
hydrolysis products and hydroxide precipitates, and the presence of high
concentrations of iron, aluminium, and titanium oxides assists sorption reactions of
metals and metalloids to occur quickly. As well, red mud contains a considerable
amount of TiO2, which facilitates oxidisation reactions thereby lowering the toxicity
of metals for example by converting As+3 to As+5, Cr+6 to Cr+3 in their salted
compounds.
Wastewater treatment applications of red mud in the removal of contaminants
both metals and metalloids from water streams, have been widely touted (Genc-
Fuhrman et al. 2005; Genc-Fuhrman et al. 2007; Genc-Fuhrman et al. 2004a, 2004b;
Genç et al. 2003; Gupta et al. 2004; Gupta & Sharma 2002; Li et al. 2006; Liu et al.
2007; Liu et al. 2011; Palmer et al. 2010; Wang et al. 2009; Zhang et al. 2008).
These studies confirmed that red mud demonstrates a promising ability to remove
toxic heavy metals, inorganics and organics, metalloids, phenolic compounds and
bacteria alike in wastewater. Improving wastewater is a consequence of the basic
advantage of red mud, which is a versatile mixture of adsorbents and flocculants that
can sequester or adsorb pollutants from the wastewater (Liu et al. 2011).
The result from the study of removing phosphate from wastewater by using red
mud reported that up to 70% of phosphate in the pH range of 6.5-7.5 can be removed
(Couillard 1982). Similarly, at pH = 7, pH removal reached 99% if red mud was
treated with HCl (concentration = 0.25 mol/L) at 7000C for 2 hours (Li et al. 2006;
27
Liu et al. 2007), and its performance was better than treated fly ash. An experiment
on comparison of nitrate adsorption between activated red mud with 20% HCl and
raw red mud was conducted by Cengeloghu et al. (2006). The authors observed that
the adsorption capacity of the activated mud was fivefold higher than that of the raw
mud, and the rate of nitrate removal reduced at pH values exceeding 7. The effect of
pH on the rate of adsorption was explained by ligand exchange reactions occurring
between metal oxides and the nitrate ion in the red mud and the activated red mud.
Numerous studies using red mud as an adsorbent for removal of heavy metals
in wastewater have been conducted (Altundogan et al. 2002; Altundoğan & Tümen
2003; Genc-Fuhrman et al. 2004a; Gupta et al. 2001; Gupta & Sharma 2002; Li et al.
2010; Soner Altundogan et al. 2000; Zhu et al. 2007). These studies concluded that
red mud has an effective capacity for adsorbing and precipitating heavy metals due to
its high pH. Furthermore, after physical and chemical treatments such as heating, or
the addition of seawater, HCl, or H2O2, a very finely grained red mud with high
surface and charge ratios is produced. This mud strongly binds to heavy metal ions
(Altundogan et al. 2002; Brunori et al. 2005; Lin et al. 2004; Liu et al. 2011). As
well, the results confirmed that pH, contacting time, adsorbent dosage, initial
pollutant concentrations, and presence of other ions are the key factors for successful
adsorption. The efficiency of the adsorption process may increase if the solution pH
is kept above 5 (Gupta & Sharma 2002). Oxide components such as Fe2O3, Al2O3,
and TiO2 play an important role in the removal of heavy metals due to their high
adsorption affinity. Unfortunately, this study (Liu et al. 2011) did not identify which
oxide possessed the highest affinity for a given heavy metal ion.
28
Apart from the inorganic removal, substantial efforts have been also made in
eliminating organics such as phenol and its derivatives, dye and bacteria from
wastewater with red mud (Liu et al. 2011). Many methods of dye removal, such as
sedimentation, filtration, chemical treatment, oxidation, biological treatment, and
adsorption and ion exchange, have used low cost red mud as adsorbents (Gupta &
Suhas 2009; Liu et al. 2011). A study on using red mud as an adsorbent for removing
congo red and acid violet from aqueous solution has been conducted (Namasivayam
& Arasi 1997; Namasivayam et al. 2001). The result showed that the removal
capacity of red mud for the dye was recorded to be 4.05mg/g and the mechanism of
adsorption is mostly ion exchange.
An experiment on using red mud mixed with sand for removing bacteria and
virus has been reported (Wang et al. 2008). The study was implemented by column
technique and using red mud neutralised by 5% gypsum and incorporated to form
30% of amended sand. The result showed that the removal rate of Escherichia coli,
Salmonella adelaide and poliovirus-1 in secondary effluent from wastewater
treatment plant was significantly improved. The order of removal was Poliovirus>E.
coli>S. adelaide (Wang et al. 2008).
2.3. Red Mud Neutralisation Methods
Untreated red mud is alkaline with pH as high as 13. Therefore, this high
alkaline waste posed an environmental hazard, and needs to be neutralised before
discharging into the environment. Several red mud neutralisation methods such as
neutralisation with seawater (Hanahan et al. 2004; Menzies et al. 2004),
neutralisation with the addition of gypsum (CaSO4.2H2O) (Barrow 1982; Courtney &
Timpson 2005; Ho et al. 1989; Kopittke et al. 2004; Wong & Ho 1993),
29
neutralisation of red mud by acid mine wastes (Glenister & Thornber 1985; Ho et al.
1985; Wong & Ho 1994), neutralisation of red mud using mineral acid (Glenister &
Thornber 1985; Khaitan et al. 2009a; Thornber & Hughes 1986), neutralisation of
red mud by fly ash (Khaitan et al. 2009; Santini & Fey 2015), and carbonation of red
mud by CO2 gas (Cardile et al. 1994; Cooling et al. 2002; Shi et al. 2000) have been
reported. These neutralisation methods not only resolve the significant amount of
pollutants, but also enhance the importance of reusing the neutralised wastes as a
valuable resource in an economic aspect.
2.3.1. Neutralisation of Red Mud with Seawater
Neutralisation of red mud with seawater has been explored by several groups
(Glenister & Thornber 1985; Hanahan et al. 2004; Menzies et al. 2004). This
neutralisation method involves the addition of seawater to bauxite residue resulting
in the precipitation of insoluble hydroxides (Mg3(OH)6) and carbonates (CaCO3 and
MgCO3), and hydroxycarbonates (Mg6Al2(CO3)(OH)16.4H2O,
CaAl2(CO3)2(OH)4.3H2O) (Johnston et al. 2008). Seawater does not eliminate
hydroxide alkalinity from the system, but it uses the soluble Ca2+ and Mg2+ content
within the seawater solution to neutralise the red mud slurry. The neutralisation of
red mud by seawater not only helps to reduce the pH level of bauxite residue slurry
to approximately 8.5 (Hanahan et al. 2004; Menzies et al. 2004), but also to increase
the long-term acid neutralisation capacity of red mud without decreasing the short
term neutralisation capacity compared with untreated red mud (Paradis et al. 2007).
Furthermore, this neutralisation method could enhance the nutrient retention capacity
of the soil due to increasing removal of P, and improve physical and chemical
30
properties of red mud that promote plant growth (Akhurst et al. 2006; Hanahan et al.
2004; Johnston et al. 2008). Chemistry of seawater neutralisation shows that in the
neutralisation process Mg reacted effectively with hydroxide, while Ca was found
more effectively with carbonates reaction. Key chemistry reactions of the
neutralisation process are the precipitation of hydrotalcite and para-
aluminohydrocalcite as below, though other minor simple carbonates and hydroxides
reactions are also possible (Hanahan et al. 2004; Johnston et al. 2008):
Hydrotalcite:
6MgCl2+12OH-+2Al(OH)4- +CO3
2-+12Na++4H+↔Mg6Al2CO3(OH)16.4H2O+12NaCl
Para-aluminohydrocalcite:
CaCl2 + 2Al(OH)4- + 2CO3
2- + 2Na+ ↔ CaAl2(CO3)2(OH)4.3H2O + 4OH- + 2NaCl
2.3.2. Neutralisation with Gypsum (CaSO4.2H2O)
The use of gypsum as an ameliorant or soil amendment has been widely
applied at bauxite processing residue disposal sites (Courtney & Timpson 2005;
Renforth et al. 2012). The addition of gypsum as a soluble calcium source to the
bauxite residue provides Ca2+ displacing Na+ from exchange complexes resulting in
the reduction of pH of the solution (Wong & Ho 1993). The previous studies
(Barrow 1982; Courtney & Timpson 2005; Ho et al. 1989; Kopittke et al. 2004;
Wong & Ho 1993) noted that the addition of gypsum brought the pH of red mud to
around 8.3, and the alkalinity of red mud can be precipitated in similar reactions to
those of seawater neutralisation.
31
Ca2+ + CO32- ↔ CaCO3 (calcite/aragonite)
3Ca2+ + 4OH- + 2Al(OH)4- ↔ Ca3Al2(OH)2 (tricalcium aluminate)
4Ca2+ + 4OH- + 2Al(OH)4- + CO3
2- ↔ Ca3Al2(OH)12.CaCO3.5H2O (hydrocalumite)
6Ca2+ +4OH- + 2Al(OH)4- + 3SO4
2- ↔ Ca3Al2(OH)12.3CaSO4.26H2O (ettringite)
When adding gypsum to weathered red mud with pH typically around 11, the
pH of red mud was lower to an acceptable level of approximately 8.5-9.0. In contrast
to weathered red mud, Thornber and Hughes (1986) reported that there are carbonate
(CO32-), aluminate (Al(OH)4
-) and free hydroxide (OH-) ions existing in alkalinity of
fresh red mud, therefore, the pH reduction can be achieved at 10.5 with the addition
of gypsum. Moreover, the addition of gypsum to red mud resulted in the
improvement of chemical characteristics and concentrations of nutrients and
enhancement of plant growth because of enhanced hydraulic quality and drainage
that require for successful revegetation at red mud impoundment sites (Courtney &
Timpson 2005; Woodard et al. 2008).
2.3.3. Neutralisation of Red Mud by Acid Mine Wastes
Acidic industrial wastes such as copperas (FeSO4) or mixture of ferrous
sulphate and sulphuric acid produced from the extraction of rutile from ilmenite and
from metal cleaning using sulphuric acid were found effectiveness in the
neutralisation of bauxite residue (Glenister & Thornber 1985; Ho et al. 1985;
Thornber & Hughes 1986; Wong & Ho 1994). The studies showed that when
copperas was added to the red mud, it acted as a strong acid resulting in the reduction
of pH. The extent of the decrease in the pH level was found positively related to the
32
quantity of acidic ameliorants in the copperas. Because copperas was highly soluble,
it reduced quickly the pH of the mud to about 5.0 to 6.0. This pH level was ideal for
the disposal of red mud at impoundment sites (Glenister & Thornber 1985; Wong &
Ho 1994).
2.3.4. Neutralisation of Red Mud Using Mineral Acid
The use of strong acids (H2SO4, HCl, HNO3) to neutralise bauxite residue as
a rapid way of reducing the pH of the red mud for safer storage has been reported in
previous studies (Glenister & Thornber 1985; McConchie et al. 2002). However, this
method has not been carried out at a plant-scale due to its high cost. There were some
lab-scale experiments reported (Glenister & Thornber 1985; Khaitan et al. 2009a).
These studies found that after acid addition the alkalinity within the liquid phase
reacted relatively quickly compared to the solid phase and leading to the immediately
decreased pH level. As mineral acids such as H2SO4, HCl, HNO3 are often strong
acids, thus if neutralisation of red mud using mineral acids, in theory any equilibrium
pH levels can be achieved depending on the quantity of acid added to the red mud
solution (Kirwan et al. 2013). However, there are drawbacks with this neutralisation
method as it introduces large volumes of sulphate or chloride impurities to the
process water stream (Cooling 2007). Strong acids have also been used to create
activated red mud as cheap adsorbents for removal of toxic metals, fluoride, and
phosphate in water or wastewater treatment and soil reclamations (Liang et al. 2014;
Pradhan et al. 1998; Shannon & Verghese 1976; Shiao & Akashi 1977).
33
2.3.5. Neutralisation of Red Mud by Fly Ash
The neutralisation of red mud can also be achieved with fly ash solid waste
produced from coal plants. Fly ash varies significantly in composition and shows a
range of acid-base characteristics when exposure to water (Theis & Wirth 1977). A
study performed by Khaitan et al. (2009) investigated the potential of fly ash for
neutralisation of highly alkaline red mud slurry and evaluate the kinetics of the
neutralisation process. Different acidic fly ashes mixed with red mud at fly ash/red
mud weight ratios ranging from 0.05 to 1.0 were used in the study to identify the
dose of fly ash required to neutralise a fixed amount of red mud. Experimental results
revealed that a variety of fly ash at fly ash/red mud weight ratios of 0.6-1.0 could
lower the pH from 12.5 to 10.8 in approximately 150 days. In addition to the large
doses of fly ash needed, the kinetics of neutralisation was rather slow (Khaitan et al.
2009). Due to the large amounts of fly ash required, this approach of neutralization is
not economically feasible. Furthermore, the slow rate of neutralisation showed that
fly ash is not a promising agent for bauxite residue neutralisation.
2.3.6. Neutralisation of Red Mud by Carbon Dioxide (CO2) Gas
In the environmental context, the current volume of CO2 in the atmosphere is
rapidly increasing and reaching the level of approximately 398ppm on October, 2015
(Dlugokencky & Tans 2015), and it is projected that atmospheric CO2 concentration
will be at around 700ppm by the end of the 21st century (IPCC 2014). This will lead
to the climate change and severe global warming affecting human health, industries
and agricultural activities. Therefore, using industrial wastes, such as metal oxide
bearing materials, for capturing CO2 may potentially make a significant contribution
to global warming mitigation. Possible materials could be alkaline and alkaline earth
34
oxides (MgO, CaO) naturally present in silicate rocks or alkaline industrial residues
such as slags from aluminium industry (IPCC 2005).
Carbonation of red mud is defined as the process of adding gaseous or liquid
CO2 to the red mud slurry before discharging into the disposal areas (Cooling et al.
2002). The carbon dioxide reacts with alkaline compounds in the slurry to form
carbonate species leading to a reduction in pH of the red mud. The carbonation
process involves the following reactions (Cooling et al. 2002):
NaOH(aq) + CO2(aq) ↔ NaHCO3(aq) (2.1)
Na2CO3(aq) + CO2(aq) + H2O ↔ 2NaHCO3(aq) (2.2)
NaAl(OH)4(aq) + CO2(aq) ↔ NaAlCO3(OH)2(s) + H2O (2.3) (dawsonite)
Na6[AlSiO4]6.2NaOH(aq) + 2CO2(aq) ↔ Na6[AlSiO4]6(s) + 2NaHCO3(aq) (2.4) (Desilication product Na2O) (sodium alumino silicate)
3Ca(OH)2.2Al(OH)3(aq) + 3CO2(aq) ↔ 3CaCO3(s) + Al(OH)3(s) + 3H2O (2.5) (tricalcium aluminate-6) (calcite) (gibbsite)
In the carbonation process, CO2 neutralises the red mud by reacting with free
soda in the form of NaOH, Na2CO3, and Al(OH)4- in the liquid phase. Consequently,
the soluble carbonate and bicarbonate ions were the dominant products of the
carbonation process as shown in equations (2.1) & (2.2) (Johnston et al. 2008).
Nevertheless, in some previous studies (Guilfoyle et al. 2005; McConchie et al.
2002) on the carbonation of red mud by CO2, it is indicated that the CO2 reacted with
ionised sodium aluminate to form solid dawsonite as described by equation (2.3).
The equations (2.4) and (2.5) take place in the solid phase where the reactions of
desilication products (e.g. sodalite) (Glenister & Thornber 1985) and calcium
35
mineral (e.g. tricalcium aluminate) dissolution and the precipitation of calcite
(Khaitan et al. 2009b) occurred that would decrease the pH of red mud slurry in
long-term carbonation.
Carbonation of red mud is considered an inexpensive and safe treatment
process that results in the formation of thermodynamically stable products (Huijgen
et al. 2005). This process could prove to be an effective approach to make red mud
less hazardous prior to disposal. As well, carbonation of red mud may provide other
significant benefits (Cooling et al. 2002). It could decrease the risk of seal material
failure (clay or HDPE) in the storage pond, thereby reducing the risk of underground
water pollution. Further, this treated non-hazardous red mud could be used for useful
purposes such as soil amendment and construction materials. Finally, it removed
greenhouse gases and minimised its adverse impacts on the quality of drainage water.
2.3.7. Perspectives of Red Mud Carbonation
Carbonation of red mud offers opportunities for the reuse of red mud as a
valuable resource which to date has been limited due to its high pH. The carbonated
products could potentially be used in industrial and agricultural activities for the
removal of toxic metals and nutrients and for soil amendment. The capacity of this
waste in capturing CO2 has been reported in a number of studies.
The first pilot study on carbonation of red mud was implemented by Alcan at
their Saramenha Ouro Preto refinery in Brazil in 1983 (Power et al. 2011).
Subsequently, the process was employed by Alcoa with the aim of lowering the pH
of the slurry before disposing of as a dry stacking waste. Many trials were conducted
between 1991 and 1996 and these studies confirmed the potential of the red mud as
an effective CO2 adsorbent, for example it was employed at the Kwinana refinery in
36
2000 (Cooling et al. 2002). The carbonation plant established in the Kwinana has
operated at the range of CO2 dose rate varying from 16 to 25.5 kg of CO2 per m3 of
red mud slurry (48%wt solids). It was observed that the lowest pH was 8.5 at the
CO2 dose rate of 25.3 kg/kL or 17.8g CO2/kg RM (an estimated density of
2,600kg/m3) under the current operating conditions at the Kwinana (Cooling et al.
2002).
In one study (Shi et al. 2000), red mud was mixed with liquid CO2 at 297K and
10MPa for the period of 5-15 minutes. The experiment was performed under high
pressure liquid CO2 rather than vapour phase CO2 in an attempt to neutralise both the
free soda and bound soda in the red mud. The pH value of the water in contact with
liquid CO2 was less than 3.0 so that it can neutralise the basic compounds of the red
mud leading to the rapid reduction of pH. The experiment suggested that the pH of
the slurry reduced immediately to 7.0 following the exposure to liquid CO2. After
treatment, the pH rebound slowly back and levelled off at 9.5 due to the release of
bound soda via desilication reaction. The approximate quantity of CO2 sequestration
was found 23g of CO2/kg of dewatered red mud. The amount of CO2 sequestered
was higher than that reported by Alcoa.
From the literature on red mud carbonation, it can be seen that preliminary
experiments clearly demonstrated the potential of carbonation to achieve the desired
pH. However, the outcome of the carbonation process is also influenced by a range
of parameters, such as partial pressure of CO2 or CO2 concentrations, chemical and
physical properties of red mud, flow rate of gas, temperature of reaction, etc. A study
on carbon dioxide treatment of red mud (Cardile et al. 1994) was performed using
factorial experimental design to characterise the influence of five factors namely
temperature of reaction, CO2 gas flow rate, stirring rate, total sodium content in
37
solution and slurry density. The responses of five factors were investigated in a fully
structured analysis by using the Jass statistical computing package and “relative
significant” Pareto plots. The determination of fitted responses, residuals, and
Normal plots of the residuals were also included in the analysis procedure. The
outcomes of the study stated that the minimum pH of 7.0 can be achieved and it
remains firmly in the solution as long as the addition of CO2 gas continues. If the
CO2 addition is stopped the pH conversion starts increasing to 9.5 after 14 hours of
treatment because the new establishment of equilibrium between the solution and
atmospheric CO2 occurred. The rebound of pH is due to the most alkaline materials
such as tricalcium aluminate or sodalite becomes reactive to CO2.
Furthermore, the study (Cardile et al. 1994) suggested that the efficiency of the
CO2 gas/mud reaction depends largely on the stirring speed and flow rate of gas,
whereas the temperature and sodium content have a little effect. These factors also
have impacts on the final pH rebound of the slurry. Accordingly, total Na content in
the RM and slurry density were found the main factors showing a remarkable impact
on the final pH recovery of the equilibrated system, while other physical behaviours
do not seem to be effective on this aspect. However, this research did not work out
the specific operating parameters that would be necessary for achieving the optimal
conditions for the carbonation process in these experiments.
In a study conducted by Khaitan et al. (2009b), bauxite residue was carbonated
in different partial pressures of carbon dioxide in the air. The partial pressures of CO2
of 10-3.5, 0.01, 0.1, and 1atm were used in the study by mixing CO2 gas with air in
different volume proportions. The results showed that the rate and extent of
carbonation was directly proportional to the partial pressure of CO2. At the pressure
of 1atm CO2, an equilibrium pH level was achieved in one day of carbonation, while
38
it took about 9 days to establish a steady state pH of 9.8 at CO2 pressure of 10-3.5.
The pH conversion phenomenon for all treatments was found to reach the same value
of 9.9 within one day. The pH rebounded was attributable to the dissolution of
tricalcium aluminate in the red mud slurry (Smith et al. 2003). The CO2 sequestration
potential for red mud liquor was estimated to be 5.8mg CO2/kg RM, while the
amount of CO2 for the whole bauxite residue was found 11.9mg CO2/kg RM. It is
estimated that the amount of carbon sequestration potential of annual red mud
production (30 million metric tonnes/year) was estimated in the order of 0.029-0.057
million metric tonnes/year (Khaitan et al. 2009b). This quantity is small in
comparison with annual production of bauxite residues.
Based on the results from carbonation of red mud at different partial pressures
of CO2 of 10-3.5, 0.01, 0.1, and 1atm, the modelling program MINEQL+ was used to
calculate the final equilibrium pH values for the carbonation. The simulation results
were then plotted and compared with experimental data. The simulation showed that
the modelled curve was similar in shape to the experimental data. The modelled
results at PCO2 values of 0.01, 0.1, and 1atm yielded a higher pH value by about 0.5
pH units, a difference not well understood (Khaitan et al. 2009b). Furthermore, the
modelling suggested that the dominant solid-phase pH buffering was caused by the
dissolution of tri-calcium aluminate (Ca3Al2O6(s)), sodium-aluminate silicate
(NaAlSiO4) and calcite (CaCO3) (Khaitan et al. 2009b).
A preliminary study on carbonation of raw red mud to determine the capacity
of waste in capturing CO2 showed that the carbon removal process is rapid and the
added carbon dioxide produced a large increase in bicarbonate alkalinity (Jones et al.
2006). The raw red mud slurry contacted with gaseous CO2 at a constant pressure of
68.9kPa, and flow rate of 200mL/minute in a range of different times. Although this
39
is a preliminary study, the capacity of red mud to capture CO2 has been proven.
Based on the amount of CO2 (4 litres) captured by red mud and its density (1.87
kg/m3), it is estimated that there are about 748g of CO2 taken up by 1kg of wet red
mud (Jones et al. 2006). In other words, the volume of 748kg of CO2 (wet weight)
was captured by per tonne of red mud. However, this amount of CO2 captured was
not feasible because there was a discrepancy in data reporting and a mistake in CO2
sequestered calculation found by Johnston et al. (2010). Accordingly, the CO2
sequestration calculated in the study conducted by Johnston et al. (2010) was 17g
CO2/kg RM.
Yadav et al. (2009) studied the sequestration of carbon dioxide by using red
mud. Their work set out to determine the main mineral phases in red mud responsible
for the carbonation process, and to evaluate the capacity of CO2 uptake in different
size red mud fractions at the ambient temperature and a fixed CO2 pressure of 3.5bar.
The conclusion was that red mud has an effective potential for sequestration of CO2,
and the principal minerals responsible for the carbonation were chantalite
(CaAl2SiO4(OH)4) and cancrinite (Na6Ca2Al6Si6O24(CO3)2). In addition, the red mud
fraction of different sizes also has different effectiveness of CO2 capture, and red
mud fraction with the average size of 30µm, a pH of 7 and its relative density of 1.8
g/cm3 was found more effective for CO2 capture, whilst the carbonation capacity was
53g of CO2/kg RM.
In another study (Sahu et al. 2010), pure CO2 gas with the flow rate of
5mL/min was passed into the red mud slurry for 5, 10, 20, 24, 48 and 72 hours. The
neutralised red mud (NRM) collected after centrifugation for 20 minutes at 3000rpm
was treated as cycle-1 for each carbonation period. Cycle-2 and 3 were repeated by
using only 5h carbonation processes. The results showed that the amount of CO2
40
removed for cycle 1, 2, 3 and NRM were 3.54, 2.28, 0.63, and 0.57g CO2/100g of
RM, respectively. Totally, calculated CO2 removal was 7.02g/100g of RM or 70.2g
of CO2/kg RM. The phenomenon of pH rebound was also observed in this study. The
pH of NRM was observed to increase slowly in the first few days and then reached a
constant value of 8.45 (Sahu et al. 2010).
In another CO2 neutralisation of red mud, Rai (Rai 2013) investigated the pH
rebound phenomenon. Two different carbonation experiments (30%wt solids) were
performed. The first experiment was called multiple cycle carbonation, where red
mud slurry was carbonated at 0.1 atm pressure for 15 minutes. The initial and final
pH were recorded, and the slurry was then stored for a week. After a week, the pH of
the slurry was measured and the slurry was carbonated again. The pH was measured
after carbonation and the slurry was stored again for another week. This procedure
was repeated in the seven-week experimental program. The second one was named
as single cycle carbonation, where the red mud slurry (30%wt) was carbonated only
once for 15 minutes. The pH of the carbonated slurry was recorded and the slurry
was then stored for a week. At the end of each week, the pH of the slurry was
measured for a period of seven weeks.
The results of the study indicated that the pH rebound was observed in both the
multiple cycle and single cycle carbonation experiments. The pH rebound in both
experiments was found to be about 9.5-9.7 within a week suggesting that pH rebound
happened irrespective of the number of cycles the slurry was carbonated. This means
that only the liquid phase was getting carbonated and the pH rebound occurred due to
leaching of the alkaline by-product in the red mud solids driving up the pH (Rai
41
2013). Unfortunately, this study did not work out the amount of CO2 sequestered by
red mud.
Table 2.4 summarises the amount of CO2 sequestered by red mud from
previous carbonation studies.
Table 2.4. Summary of CO2 amount captured in previous studies on RM carbonation
Studies on carbonation of
RM by CO2
The amount of CO2 captured
References gCO2/kg
RM
gCO2/kg
RM solid
gCO2/kg
RM liquor
Carbonation of RM at
Kwinana refinery in 2000. 17.8 - -
(Cooling et al.
2002)
Carbonation of RM by CO2
liquid at 297K, 10MPa. 23 - - (Shi et al. 2000)
Carbonated of RM in
different CO2 partial
pressure.
11.9 - 5.8 (Khaitan et al.
2009b)
Carbonation of RM at CO2
pressure of 68.9kPa, flow
rate of 200mL/min.
17 - - (Johnston et al.
2010)
Carbonation of RM at
15vol% CO2 and 85vol%
N2 at flow rate of 5mL/min
41.5 - - (Bonenfant et al.
2008)
Carbonation of RM at fixed
CO2 pressure of 3.5bar. 53 - -
(Yadav et al.
2009)
Carbonation of RM by CO2
at flow rate of 5mL/min. 70.2 - - (Sahu et al. 2010)
42
2.3.8. Mechanism of the Carbonation of Red Mud
The mechanism of red mud carbonation has been identified in both liquid and
solid phase reactions as summarised in Table 2.5. In the liquid phase, it is the
reaction of CO2 with hydroxide to form bicarbonate and the reversibility of key
alkalinity reactions between hydroxide, carbonate, and bicarbonate (Johnston et al.
2010). When initial injection of CO2 gas into the red mud slurry (pH~13), it absorbs
into the red mud liquor, then diffuses through the liquor and reacts with free
hydroxide (OH-) present in the liquor to form carbonate (CO32-) alkalinity (Eqn.
(2.6)) in Table 2.5, and lower the pH of solution to ~10.3. With further CO2 addition,
the carbonate is converted to bicarbonate illustrated by equation (2.7), therefore
lowering the liquor pH to <8.5 (Cardile et al. 1994; Johnston et al. 2010; Khaitan et
al. 2009b; Kirwan et al. 2013; Sahu et al. 2010).
However, much of the hydroxide for reaction with CO2 also comes from the
solubilisation of residual aluminium such as aluminate (Al(OH)4-). Thus, the
reactions taking place in liquid phase also involves the consumption of free
hydroxide from aluminate anion resulting in the precipitation of alumina as shown in
equation (2.8) (Johnston et al. 2008; Jones et al. 2006; Kirwan et al. 2013).
43
Table 2.5. Reactions taking place in the carbonation of red mud
Reactions
Liquid phase reactions:
2OH-(aq) + CO2(aq) ↔ CO3
2- + H2O
H2O + CO32- + CO2(aq) ↔ HCO3
-(aq) + H+
(aq)
[Al(OH4)-](aq) + CO2(aq) + Na+
(aq) ↔ Al(OH)3(s) + Na+(aq) + HCO3
-(aq)
(2.6)
(2.7)
(2.8)
Solid phase reactions:
Calcium-bearing mineral dissolution (e.g. tricalcium aluminate):
Ca3Al2(OH)12 + 6H+ ↔ 3Ca2+ + 2Al(OH)3 + 6H2O
DSP dissolution (e.g. sodalite):
Na6Al6Si6O24.Na2CO3.yH2O + 18H2O + 7H+ ↔ 8Na+ + 6Al(OH)3 +
6Si(OH)4 + HCO3- + yH2O
The precipitation of carbonate (e.g. calcite):
Ca2+ + CO32- ↔ CaCO3(s)
(2. 9)
(2.10)
(2.11)
In the solid phase, the reactions involved hydrogen ions of carbonic acid
causing the dissolution of alkaline tricalcium aluminate (TCA) and the dissolution of
desilication products (DSPs) such as sodalite, cancrinite and the precipitation of
carbonates such as calcite (Glenister & Thornber 1985; Johnston et al. 2010; Smith
et al. 2003). Solids like TCA act as an alkaline store and its dissolution (Eqn. (2.9))
can serve to buffer any neutralisation agent to a pH value of about 11 (Gräfe et al.
2009). The DSPs solid such as sodalite, cancrinite can be an alkalinity contributor
and its anion could also offer some buffering capacity for pH in the range of 8.3 to 11
(Kirwan et al. 2013). However, the buffer associated with the dissolution of these
solids (Eqn. (2.10)) can lead to a stable pH of nearly 8.0 (Glenister & Thornber
44
1985). The dissolution of solids was observed to occur slowly as the red mud was
carbonated (Khaitan et al. 2009a). Apart from that, the precipitation of carbonates
such as calcite (CaCO3) also occurs following the dissolution of solids. The
precipitation of calcite (Eqn. (2.11)) was observed taking place after the dissolution
of TCA and lowering the equilibrium pH of carbonated red mud (Khaitan et al.
2009b). The final pH value remains constant at around 7.0 as long as the injection of
CO2 continues. If the CO2 addition is ceased, the equilibrium pH of the slurry
increases because of alkaline solids becoming reactive to CO2 (Eqns. (2.9)&(2.10))
causing the pH to rebound (Cardile et al. 1994; Glenister & Thornber 1985; Khaitan
et al. 2009b).
The rate of conversion of CO2 into H+ and HCO3- ions in an aqueous solution
plays an important role in many geological processes, especially the dissolution and
precipitation of carbonate minerals (Dreybrodt & Buhmann 1991; Dreybrodt et al.
1996). The dissolution of CO2 into water is slow and dependent on pH and
temperature, while the dissociation of carbonic acid into carbonate (CO32-) (Eqn.
(2.6)) and bicarbonate (HCO3-) (Eqn. (2.7)) as shown in Table 2.5 occurs rapidly in
an aqueous solution. The forward rate constant of CO2 dissolution in aqueous
solution is kCO2=3.10-2s-1, whereas the dissociation (Eqn. (2.7)) becomes dominant at
high pH (>9) with a forward rate constant k=8.5.10-3mol-1s-1 at 250C (Pan et al. 2012;
Stumm & Morgan 1981). The dissolution rate of calcium bearing (e.g. TCA) (Eqn.
(2.9)) and DSP (e.g. sodalite) (Eqn. (2.10)) minerals and the precipitation rate of
carbonates (e.g. calcite) (Eqn. (2.11)) in the carbonation process are related to the
dissociation of CO2 in the liquid phase and the pH of solution as well (Dreybrodt &
Buhmann 1991; Pan et al. 2012). As discussed earlier, the dissolution of these
minerals bearing alkaline anions occurs slowly at the mineral surface as the red mud
45
was being carbonated. Both TCA and DSP minerals begin dissolving slowly at pH
below 9, but the reaction of calcium ions combining with carbonate ions is very fast
at pH range of 4.5-6.0 (Khaitan et al. 2009a; Pan et al. 2012). The dissolution
kinetics of solids improved with increasing temperature; however, carbonation
precipitation was retarded at higher temperature because of reduced CO2 solubility
(Costa et al. 2007; Gerdemann et al. 2007; Park et al. 2003). Therefore, it is
important to keep the temperature stable at room condition during the experiments in
this study.
2.4. Summary
The mass of bauxite residue generated from the Bayer process has posed a
significant impact on the environment because of its high alkalinity. Many different
disposal methods have been employed in the efforts of the management of red mud
in the world. The first two conventional wet methods used in alumina refineries prior
to the 1970s were marine discharge and lagooning. Although the marine method
sounds simple, it had adverse impacts on the ocean ecosystem, so it has been
discouraged by the UNIDO. As a result, the lagooning practice has been widely
applied as a main method of disposal of red mud in refineries established after 1970.
This practice is the simplest land-based disposal and requires increased engineering
input for red mud storage. However, this method is also problematic due to high risk
of leakage, instability, and liquefaction that can lead to human deaths and long-term
environmental depletion caused by tailings dam failures.
Other disposal methods that have been introduced to alumina refineries for
red mud storage are dry stacking and dry cake disposal practices. Both methods are
related to the process of dewatering red mud as much as possible before disposing of.
46
While the dry stacking method is considered to be cost-effective and less
environmental problems, the dry cake practice offers an advance feature that the risk
of alkalinity and caustics can be further diminished by neutralising or washing by the
filtration process. However, the dry stacking is often employed in refineries because
of its advantages and preferred disposal strategy compared with the dry cake disposal
practice.
There are many different technologies deployed in the efforts of long-term
remediation and utilisation of red mud in the environmentally friendly applications.
Red mud has been used as a building material for manufacturing bricks and blocks in
construction. In this way, the application contributed a significant part to the
reduction of the volume of red mud discharging into the environment. The red mud
bricks showed a good durability in severe climate conditions and high strength. More
importantly, such products could compete with other building materials since they
consume lower energy and production cost. Nevertheless, it is noticed that the
ionising radiation level from the red mud bricks was approximately 2-3 times higher
than that of the conventional concrete. This is a potential impact on human health.
Using red mud as catalysts in some chemical applications such as
hydrogenation, hydro-dechlorination, exhausted gas clean-up has been widely
published. However, this practice was not a promising solution as it had a poor
performance compared with the commercial ones. Also, red mud has been
successfully employed in agricultural application. Untreated red mud can improve
soil given its ability for fixation of heavy metals. Furthermore, bauxite residue is also
useful in improving P retention in agricultural areas having sandy soil with low
phosphate content. The ability of P retention can help reduce the damaging impacts
of eutrophication in the ecosystem. The use of red mud in environmental treatment is
47
the most successful application. Many studies have confirmed that red mud has an
effective capacity for adsorbing and precipitating heavy metals, inorganics and
organics, metalloids, phenolic compounds and bacteria alike in wastewater.
Additionally, the efficiency of adsorption may increase if the mud can be activated
by HCl or the solution pH is retained above 5 since the red mud contains Fe2O3,
Al2O3 and TiO2 that play an significant role in removal of heavy metals.
Apart from technologies of utilisation of red mud mentioned above,
neutralisation methods have been widely employed to enhance the reuse of
neutralised waste. Red mud can be neutralised with seawater to precipitate the
insoluble hydroxides and carbonates. Although this neutralisation method does not
eliminate hydroxide alkalinity in the mud, it helps to reduce the pH level of the
residue slurry, and to increase the long-term acid neutralisation capacity of RM. The
neutralisation of red mud can be done by the addition of gypsum as a soluble calcium
source to the bauxite residue. The neutralisation of RM by gypsum can improve
chemical properties and concentrations of nutrients, and enhance the growth of plant
at red mud impoundment sites.
The fastest way of reducing pH and alkalinity of bauxite residue for safer
storage is the neutralisation of red mud with strong acids (HCl, H2SO4, HNO3). The
advantage of this method is that any equilibrium pH levels can be achieved
depending the amount of acid added to the RM. Nevertheless, this practice has not
been employed at a plant-scale because of its high cost and producing a huge amount
of sulphate or chloride impurities. Alternatively, the use of fly ash solid waste from
coal plants to neutralise the RM can be possible. Fly ash was considered efficacy of
neutralisation because of its varieties of composition and acid-base characteristics
when exposure to water. However, this method would be unlikely an economically
48
feasible remedy because of the slow rate of neutralisation and the large quantity of
fly ash required for neutralisation of RM.
The neutralisation of RM using CO2 gas is the most promising solution. The
carbonation of RM is considered an inexpensive and safe treatment process since it
produces thermodynamically stable products. This practice could help reduce the risk
of underground water pollution in the storage pond. Moreover, it could lower the risk
of future classification of RM as a hazardous waste, and improve the usefulness of
RM in other purposes. Finally, the carbonation of RM can contribute a significant
part to the greenhouse gas reduction and global warming mitigation strategies.
49
CHAPTER 3 MATERIALS AND METHODS
3.1. Materials
The raw red mud (RM) sample used throughout the experimental program was
supplied by Rio Tinto Alcan Queensland, Australia in the form of slurry with 44%
solids by weight (44%wt). The sample was stored in a 20-litre plastic bucket and sent
from the Rio Tinto Alcan to the university laboratory. Then, the RM was split into
smaller plastic containers for easier storage. The RM slurry samples stored in plastic
containers were covered by a layer of Argon gas on the top to prevent any air or
other gases contacting with the mud. The mineral compositions of raw RM as
determined by Bruker D4 Endeavor Powder X-ray Diffractometer (XRD) and
quantified by TOPAS V4.2 are given in Table 3.1.
Table 3.1. Major mineral composition of raw RM
Component in
RM
Possible formula wt (%)(1)
Sodalite
Cancrinite
Hematite
Boehmite
Gibbsite
Quartz
Anatase
Na8(AlSiO4)6(OH)2.4H2O
Na6(AlSiO4)6(CaCO3)(H2O)2
Fe2O3
AlO(OH)
Al(OH)3
SiO2
TiO2
19.75
2.32
58.03
8.35
1.04
3.74
6.77
(1): Quantified by Diffracplus TOPAS software associated with Bruker D4 XRD instrument
50
Analytical grade hydrochloric acid (HCl) used in this study was available in the
lab. Research grade CO2 gas cylinder was purchased from Coregas Adelaide and
used without further purification.
3.2. Materials Preparation
All RM slurries (44%wt) contained in plastic containers were homogenised
first by stirring with an impeller for all experiments. Particle size of the raw RM was
in the range of 0.25-224.40µm as measured by Master Sizer 2000, version 5.60
Malvern, UK. For each set of batch experiment, 100g of RM slurry was weighed on
a 4-figure balance Ohaus Model AX324-Ohaus Corporation, USA. The mud
suspension for each batch of carbonation study used an initial solid: water ratio of
1:5 by mixing 100g of RM with 500mL of distilled water to create slurry suitable for
dissolution of soluble alkaline components in the carbonation reaction (Johnston et
al. 2010). All experiments were conducted in duplicates to minimise errors and
improve reproducibility. All carbonation experiments and measurement techniques
were performed at ambient conditions.
3.3. Methods
3.3.1. Acid Titration Procedures
Acid titration of RM slurry was done to determine acid neutralisation capacity
(ANC) of available species in the solid and liquid phases of the RM slurry. The ANC
represents the amount of mineral acid required to reach a specific pH endpoint
51
(Carter et al. 2008; Lin et al. 2004; Snars et al. 2004). The acid titration of RM was
performed by using 0.1N hydrochloric acid (HCl). The method of acid titration for
RM slurry was adapted from standard procedure for soil titration (Page et al. 1982).
Accordingly, 10g of RM slurry (44% solid by weight) was transferred to a glass
beaker and continuously stirred by a magnetic bar and titrated by adding aliquots of
0.1N HCl to target pH endpoint 4.5, where all hydroxide and carbonate alkalinity
were converted to bicarbonate. A calibrated pH meter Jenway 3510 was used to
record pH values during the titration process. Duplicate ANC measurements were
obtained by repeating the same procedure as above.
Long-term titration of RM slurry by HCl was performed at pH values of 4.5, 6,
8 and 10 to get an estimate of the total ANC of the RM. 10 grams of RM was titrated
against 0.1N HCl to target the final pH values as above on the daily basis until the
desired pH value was stable. Duplicate long-term titrations were made by using the
same procedure to get the average value. After long-term titration, the slurry was
filtered by using Whatman 40 filter paper to obtain the liquor. Then, this liquid was
diluted by distilled water for dissolved metal analysis using Inductively Coupled
Plasma Mass Spectrometry (ICP-MS).
To distinguish the difference in contributions to ANC between the liquid and
solid phases, only RM liquor was used to perform the rapid titration. In this titration,
10g of RM liquor, obtained by centrifugation from 50g of RM slurry and filtered by
Whatman 40 paper, was titrated by adding aliquots of 0.1N HCl within 2 minutes
allowed for reactions after each acid addition (Khaitan et al. 2009a). Duplicate
measurements were done to get the average value.
52
3.3.2. Determination of Total Alkalinity of Raw RM and Carbonated RM
Total alkalinity (TA) is a measure of the ability of a water sample to neutralize
strong acid. Unlike acid neutralisation capacity, the TA was determined on a filtered
sample (Rounds 2012), which means it is the property of the liquid phase only. The
TA determination of RM slurry was carried out based on the method detailed by
Rounds (2012). Accordingly, a 10g of RM sample (wet weight) was directly titrated
against 0.1N HCl from an initial pH of 12.5 to a final pH of 4.0. The process was
performed in two steps. Firstly, the slurry was titrated to a pH of 8.3 (carbonate
endpoint) at which all the dissolved aluminium hydroxide in the mud precipitated.
Next, the sample mixture was filtered using Whatman 40 filter paper to separate the
precipitated hydroxide. Then, the filtrate was further titrated against 0.1N HCl to the
pH of 4.0 (bicarbonate endpoint).
For carbonated RM, 10g quantities of sample (wet weight) were also titrated
against 0.1N HCl to the bicarbonate endpoint (pH 5), since all carbonated samples
have pH<8.1 and >5.0 (Rounds 2012). Next, the mixture was filtered by using
Whatman 40 filter paper. The filtrate was then further titrated against 0.1N HCl to
pH of 4.0. Finally, web-based alkalinity calculator V2.22 (USGS 2007) was used to
obtain alkalinity values for RM and carbonated RM. In the web-based alkalinity
calculator, two columns of data included titrant volume and pH values were entered
in the titration data area. Other fields in the program such as sample temperature,
sample volume, acid concentration were filled with their availabilities. Filtered
sample area was selected “yes”, acid correction factor for HCl was 1.0, and burette
titration was selected in titration type area. In the fixed endpoint section, for raw RM
samples, carbonate (pH 8.3) and bicarbonate (pH 4.0) endpoints were entered, but for
53
carbonated RM carbonate field was left blank, and bicarbonate field was entered 5.0.
Finally, click “calculate” button for running the program.
3.3.3. X-ray Diffraction (XRD)
X-ray Diffraction analysis was used to characterise the crystalline materials
and provide the information about the structure of minerals in RM. The RM samples
were oven dried at 650C in 24h before their analyses. The identification of the
minerals and quantitative phase analysis were undertaken using the Bruker D4
Endeavor Powder X-ray Diffractometer with a Co-Kα radiation source generated at
35kV and 30mA (λ= 1.788970Å). Diffraction data in the 2θ range from 10 to 80°
were collected and matched with ICSD (Inorganic Crystal Structure Database)
reference patterns (ICSD 2012) using standard software associated with the
instrument. Mineral composition quantitative analysis of different phases was
determined by using TOPAS V4.2 software, Bruker AXS GmbH, Germany.
3.3.4. Scanning Electronic Microscopy and Energy Dispersive X-ray (SEM-EDX)
Scanning Electronic Microscopy (SEM) Quanta 450 was used to investigate
crystalline structures and spatial variations in chemical compositions of red mud
before and after treatment. In addition, element and chemical compositions of major
oxides of the samples were identified by using Energy Dispersive X-ray (EDX)
coupled with the SEM. Powder samples were placed on the mounting adhesive stubs.
As analysed by the EDX, the mounted samples were coated with a very thin carbon
to minimise sample-charging problems (Willis et al. 2002). The samples were then
analysed in a very high vacuum mode chamber. For EDX analysis, a carbon
coefficient of 14 (radioactive carbon) as default in the program was turned on, and
the element and major oxides results were obtained from spot analysis of SEM
54
images and pallet of the samples. All measurements were done in triplicates to get an
average value.
3.3.5. Carbon-Hydrogen-Nitrogen Elemental Analyser
Carbon content of the carbonated RM samples was determined by using
PerkinElmer® 2400 Series II CHNS/O Elemental Analyser (EA), in CHN mode.
Carbonated RM finely powdered samples were accurately weighed to 5mg (± 2mg)
on a 6-figure balance in tin capsules, folded, and were placed in the EA carousel.
Analysis was conducted with a furnace temperature of 9250C in Helium gas used as
carrier and provided inert atmosphere. S2 standards were run after every 10 samples
to check the calibration of the EA and validate the sample readings, i.e. S2: 29.99 ±
0.3% C; and CaCO3: 12.00 ± 0.3% C. If the standards were not within range, the
samples were labelled as invalid and were analysed again. The carbon content in
solid phase was estimated in percentage (%) and converted to gCO2/100g of RM.
Total inorganic carbon (TIC) content in raw red mud slurry was determined by
the difference between total carbon (TC) and total organic carbon (TOC) content
(TIC=TC-TOC) (Pansu & Gautheyrou 2007). To identify the TC, 50g of RM sample
was heated to 9250C for 5-10 minutes to convert all carbon to CO2 and the TC was
obtained. Duplicates were performed to get the average value. To identify the TOC,
two samples 44g & 46g of RM slurry were transferred to the two glass vials. Next,
0.2ml of phosphoric acid (H3PO4) were added to each vial to lower the pH of
samples for expelling the inorganic carbon at 2500C, then the samples were heated at
9250C to get the TOC. The TIC of samples was identified by the difference.
For carbonated RM liquor samples, the carbon content was determined by
using Total Organic Carbon (TOC) Analyser, model TOC-VCSH/CSN + TNM-1,
55
Shimadzu Corporation, Japan. Carbonated RM liquor samples filtered through 0.45µ
before analysis. To determine total carbon, the samples were injected into a catalyst
packed combustion tube, at a furnace temperature of 720°C. Organic carbon is
oxidised to CO2 and inorganic carbon is decomposed to CO2, which is measured by a
nondispersive infrared detector. Inorganic carbon is determined by injection of the
sample into a reaction chamber where it is acidified and the CO2 generated was
carried to the detector and measured. The carbon content in liquid phase was
estimated in moles C/L RM liquor and converted to gCO2/100g of RM liquor.
3.3.6. Thermal Analysis (TGA-DSC)
The Thermogravimetric Analysis and Differential Scanning Calorimetry
(TGA-DSC) was used to determine changes in physical or chemical composition of
substances in both raw RM and carbonated RM. The TGA-DSC analysis of samples
was performed using TGA-DSC 2 STAReSystem, METTLER TOLEDO, USA. To
do this analysis, 5-10 mg of sample was transferred to crucible and alumina was used
as a reference. Measurements were carried out in N2 atmosphere at a heating rate of
100C/min from 250 to 9100C.
3.3.7. Fourier Transform Infrared Spectroscopy (FT-IR)
The FT-IR spectra study of carbonated and noncarbonated red mud samples
helps to understand the changes in chemical bonds or stretching vibrations in a
molecule. The spectra of samples were collected by using Nicolet 6700 FT-IR
Spectrometer (USA). The spectrum was registered from 4000 to 400cm-1 because the
absorption radiation of most organic compounds and inorganic ions is within this
56
region (Thermo Electron Corporation 2004). Firstly, background spectrum was
collected without samples. It is a measurement of the response of the spectrometer
alone without absorption due to samples. Then, a small amount of dried sample was
placed in the sample holder for scanning in the registered spectrum 4000-400cm-1,
and data collection for samples was obtained.
3.4. Carbonation Experiments
3.4.1. Construction of Reaction Chamber
The carbonation reaction chamber as shown in Figure 3.1 was designed to
safely operate over the likely range of relevant parameters for the RM carbonation
process (e.g. materials of construction, temperature, pressure, residence time, etc.).
The absorption reactor used in the project was a 1000mL stainless steel cylinder
vessel with a height of 128mm and a diameter of 100mm. A stainless impeller
controlled by a rotor IKA® RW20 digital was placed at port no.1 in the middle of the
reactor to attain a uniform agitation. The port no.2 was designed up to 3 bar for
pressure vent to the environment if the reactor was used in high pressure. Other
apparatus include pH probe and temperature probe, which were specifically designed
for the reactor. The temperature and pH probes from calibrated pH meter Jenway 350
were inserted in port no. 3 and 4 to monitor the temperature and record pH values of
the solution. Air and CO2 gas controlled by mass flow rate meters Brooks MFCs
were mixed together and injected into the chamber via the inlet-gas port no.5. The
mass flow rate controller Brooks® 4800 series model 4850ABC, Brooks Instrument
USA, has the following characteristics:
57
- Flow ranges in full scale (FS): 25mL/min - 40L/min
- Accuracy: ±1.0% of FS
- Repeatability: ± 0.15% of FS
- Response time: Flow signal <0.3 second, flow control <0.5 second
- Pressure: 0 - 10 bar
- Temperature: 0 - 500C and humidity of 5 - 95% (ambient)
The gas was distributed to the solution by a circle-shape gas diffuser placed at
the bottom inside the reactor. The outlet-gas port no.6 was used for excessive gas in
the chamber escaping to the environment. The whole system of experimental
apparatus is described in Figure 3.2.
Figure 3.1. Carbonation reaction chamber
58
Figure 3.2. The experimental apparatus system for carbonation of RM
3.4.2. Carbonation of RM
Both red mud slurry and liquor samples were carbonated in the stainless steel
chamber as described in section 3.4.1. The samples were contacted with the mixture
of air and CO2 gas in a range of different operating conditions such as CO2
concentrations, total flow rate of gas, stirring speeds and solids concentrations in
RM. The specific different experimental conditions were used in the project as
follows:
- Total flow rate (TF) of gas: 100mL/min, 200mL/min, 300mL/min, and
400mL/min.
- CO2 concentrations ranging from 10% to 100% corresponding to the above TF.
- Stirring speeds: 250rpm, 350rpm, 500rpm, and 700rpm.
- Solids concentrations in RM: 35%, 40%, and 44% solids by weight.
59
All carbonation experiments were performed in the reaction chamber under
ambient temperature and pressure. For each experimental batch, 100g of RM was
mixed with 500mL of distilled water (solid: water ratio 1:5 by mass) to attain a
sufficient volume of slurry for carbonation. Air and CO2 gas from cylinders
controlled by mass flowrate controllers (Brooks MFCs) were mixed together then
passed through the samples via a circle-shape gas diffuser to create the small bubbles
increasing the surface area available for reactions to take place. The pH of the
solution as a function of time was recorded during the carbonation process until it
reached the equilibrium. Then, the neutralised RM was oven-dried at 650C in 24h,
ground to a powder for analysis. The carbonation experiments were repeated in a
change of different conditions.
3.4.3. pH Rebound of the Carbonated Red Mud
The increase in pH after the carbonation of red mud due to caustic soda
leaching after its initial leaching from the mud is commonly known as pH reversion
or rebound (Rai et al. 2013). Carbonated red mud samples obtained at different
operating conditions were used to evaluate pH rebound upon the equilibrium with the
atmospheric CO2 level. The pH values were recorded every 24 hours in the period of
35 days.
3.5. Chemical Equilibrium Modelling
The modelling was performed in this study using the chemical equilibrium
program MINEQL+ 5.0 (Schecher & McAvoy 2015). This software and its
precursor, REDEQL (Morel & Morgan 1972), were developed to solve the
60
expressions of mass balance by using equilibrium constants. Today, the MINEQL+
program incorporates the combination of numerical structure and the up-to-date
thermodynamic database in chemical speciation from WATEQ3 (Ball et al. 1981)
program through the minimisation of Gibbs free energy matrix.
The principles and capabilities of MINEQL+ 5.0 exist on three different levels:
modelling of first principles, secondary and conceptual modelling, and systems
modelling (Schecher & McAvoy 1992). The first level aims to calculate mass
balance and possible electroneutrality conditions. This can be done by using all
thermodynamic data such as stoichiometric coefficients, equilibrium constants, and
enthalpy values to make a specific chemical species of complexes or solids. The
second level considers addressing the model characteristics such as surface
complexation models, corrections for temperature and ionic strength, models for
calculation of pH and conductivity. The final level focuses on the definition of a
chemical system via database modelling. This level allows users to change
thermodynamic data of a system, and insert or delete chemical species, which are not
available in a database or problem set or not essential for chemical equilibrium
model (Schecher & McAvoy 1992).
In this study, the MINEQL+ 5.0 program was used to model the metal
concentrations from long-term titration in order to evaluate dissolved species and
their solid phases controlling the aqueous chemistry system. In addition, the
simulation was carried out to validate the final equilibrium pH values obtained from
the carbonation of RM in different CO2 concentrations.
All information on the chemical components in both RM slurry and liquor as
shown in Table 3.2 was used in the MINEQL+ program. The simulation of heavy
metals in long-term titration was performed at various pH values of 4.5, 6, 8, 10, and
61
12.5 with fixed carbonate content (TOTCO3) value as given in Table 3.2. In contrast,
the modelling for carbonation of RM was conducted at different partial pressure
(PCO2) of CO2. The ionic strength (I) corrections setting was kept on, and the method
of ionic strength calculated by the program was selected as this method is more
likely to give the accurate results (Schecher & McAvoy 2015).
Table 3.2. Concentration of raw RM and liquor
Constituent
Concentration
in dried solids
(g/kg solid)(1)
Concentration
in RM slurry
(g/L liquor)(2)
Concentration
in slurry (M)
Concentration
in liquor (M)
Carbonate
Na
Al
Si
Ca
Ti
Fe
5.1(3)
98.2
125
85.6
8
41.5
200
4.01(3)
77.2
98.2
67.3
6.3
32.6
157.1
0.76
3.35
3.64
2.40
0.16
0.68
2.81
0.011(4)
0.198(5)
0.076(5)
8.58E-05
0.00017(5)
1.42E-05
5.36E-06(5)
(1): Determined by EDX. (2): Converted to RM liquor volume basis (from RM 44% solid by weight as received). (3): Determined in this study (5.1mgTIC/gRM) and converted to liquor volume basis (5.1*1000/0.56). (4): Measured by Total Organic Carbon Analyser, Model TOC-VCSH/CSN + TNM-1, Shimadzu Corporation, Japan. (5): Determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS), Model Agilent 7500cs, Agilent Technologies,
USA.
62
CHAPTER 4 RESULTS AND DISCUSSIONS
4.1. Acid Neutralising Capacity (ANC) of the raw RM
Acid neutralisation capacity (ANC) reflects the amount of mineral acid
available in the red mud for neutralisation to reach a preselected pH endpoint, and it
was determined by the titration of an aqueous solution against a strong acid (Carter et
al. 2008; Lin et al. 2004; Liu et al. 2006; Stumm & Morgan 1981). In this study, the
ANC of raw RM slurries and liquors was performed in order to assess the
contribution of mineral solids and liquor chemical species to the ANC. The rapid
ANC was performed for RM liquor only, and the long-term ANC was done for RM
slurries because of the complex chemical composition in RM and the slow acid-base
reactions with the solid phases.
4.1.1. Rapid Titration of RM slurry and RM liquor
Figure 4.1 shows rapid titration results of RM samples carried out to a pH level
of 4.5 in order to determine the ANC of the RM. The initial pH of both RM slurry
and its liquor was recorded at around 12.5 and 12.4, respectively. It can be seen that
rapid titration of the RM slurry (0.79g solids and 1g RM liquor) yielded an ANC of
0.79 milliequivalents per gram (meq/g) of RM for titration to pH of 4.5. The ANC of
1 gram of RM liquor was about 0.22meq.
As illustrated in Figure 4.1, there was no difference between the two curves at
the initial phase of titration from pH of 11.5-12.5. However, at the later part of
titration, the gap between the two curves was more visible at pH 11 or lower. This
63
confirmed that the initial phase of titration was controlled by RM liquor, and the later
titration was controlled by the dissolution of solids in the RM. Similar observations
were reported by Khaitan et al. (2009a).
Figure 4.1. Rapid RM liquor titration compared with that of RM slurry (44%wt)
Furthermore, based on the titration curves in Figure 4.1 it can be concluded
that when acid was injected in the samples, H+ reacted with soluble aluminium in the
form of Al(OH)4-, NaOH and NaCO3
- in the liquid phase causing the resistance to pH
changes of RM liquor and buffering OH- (Khaitan et al. 2009a). All the titration data
are presented in Tables A.1-A.2 in the Appendix. The key controlling reactions are
as below.
Al(OH)4-(aq) + H+ = Al(OH)3
0 + H2O
NaOH(aq) + H+ = Na+ + H2O
NaCO3-(aq) + H+ = Na+ + HCO3
-
4
5
6
7
8
9
10
11
12
13
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
pH
meq/g liquor or meq/g liquor in 1.79g RM
1g RM liquor
1g RM liquor in RM
64
4.1.2. Long Term Titration of Red Mud
Figure 4.2 illustrates the results of long-term titration of RM slurry samples to
different pH levels of 4.5, 6, 8, and 10. The measurements were performed daily by
adding acid 0.1M HCl to adjust to the desired pH values until they remain
unchanged. It took over 40 days to complete the titration to specific pH levels of 4.5,
6, and 8. In long-term titration, the ANC recorded for the RM at pH 4.5 was about
1.91 meq/g RM. The long term ANC is double the ANC obtained from rapid titration
at the same pH value. It can be seen that the long neutralising titration time (45 days)
associated with the higher ANC obtained at pH levels of 4.5, 6 and 8 endpoints under
slow titration conditions suggest that solid dissolution occurs at these endpoints,
leading to the relatively high ANC values. Furthermore, there was an over 50%
difference between rapid and long term ANC as indicated in Table 4.1. This was due
to the complex chemical composition in RM and the slow acid-base reactions of
solid phases.
Figure 4.2. Long-term titration of RM
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0 4 8 12 16 20 24 28 32 36 40 44
AN
C (
meq
/g R
M)
Time (day)
pH 4.5
pH 6
pH 8
pH 10
65
Table 4.1. Comparison between rapid and long term ANC for RM
pH value Rapid ANC
(meq/g RM)
Long term ANC
(meq/g RM) % Difference
12.5
10.0
8.0
6.0
4.5
0
0.12
0.2
0.43
0.79
0
0.34
0.69
1.03
1.91
0
65
71
58
59
In comparison between rapid liquor titration (Fig. 4.1) and long-term RM
titration (Fig. 4.2), it can be seen that solid phase plays a significant role in long-term
ANC titration process. The titration of 1 gram of RM liquor required 0.22meq or the
liquor (0.56g) contributes about 0.12meq in 1 gram of RM (44% solid by weight).
Therefore, it is estimated that the solid phase contributes approximately to 81% of
ANC, whereas this number was 19% for the liquid phase in the final long-term ANC
determination. The ANC contribution of liquid phase in this study agrees with that
reported by Liu et al. (2006), who stated that the rapid ANC accounted for less than
20% of the total ANC of red mud. Additionally, by comparing the percentage of
solid contribution to the ANC (81%, RM 44%wt) in this study with that of previous
study (76%, RM 40%wt) (Khaitan et al. 2009a), it could be confirmed that the higher
the concentration of RM, the higher percentage of solid phase contribution to the
ANC.
Based on the rapid and long term ANC for the RM (Fig. 4.1 & 4.2, Table 4.1),
it can be seen that most ANC was derived from the solids dissolution, and rapid
titration was not able to capture all the ANC of the RM. Moreover, the ANC is also
conditional upon the final pH of the system. As indicated in Figure 4.2, the ANC
(under 0.3meq/g RM) at pH 10 did not significantly increase after 45 days of
66
titration, and there was a little or no solids dissolution with time confirmed by the
results of metal concentrations by using ICP-MS (Table 4.2). However, when the
endpoint of pH was reduced from 10 to 8, 6, and 4.5, there was a significant increase
in the amounts of solids dissolved, and the greater ANC values obtained. This
collaborates with the observations of earlier researcher (Khaitan et al. 2009a), which
described the higher ANC observed at lower pH endpoints by chemical equilibrium
modelling.
Table 4.2. Metal concentrations in RM liquor at different pH values
Component Concentration (M)
Initial liquor pH 10 pH 8 pH 6 pH 4.5
Al
Na
Ca
Fe
0.076
0.198
0
0
0
0.229
0.0015
0
0
0.262
0.032
0
0
0.28
0.035
0
0
0.487
0.104
0
In long-term titration, key reactions were the dissolution of sodalite
Na8(AlSiO4)6(OH)2.4H2O and cancrinite Na6(Al6Si6O24(CaCO3)(H2O)2 in the lower
pH environment. Weber (2001) and Newson et al. (2006) reported that the solubility
of sodalite and cancrinite was pH dependent. In other words, these minerals are less
soluble at high pH environment, but at low pH their solubility becomes much more
significant. As shown in XRD patterns in Figure 4.3 & 4.4, cancrinite and sodalite
have several common peaks located at 16.10, 270, 390, 480, 580 and 780 2theta
because they have the same framework stoichiometry [AlSiO4]6 as reported by
Gerson & Zheng (1997) and Barnes et al. (1999a). At pH 6 sodalite and cancrinite
peaks decreased but were still present in the RM. However, these peaks virtually
disappeared at the pH 4.5. Hence, at pH 6 sodalite and cancrinite were found to be
partly dissolved, but they become completely dissolved at pH lower than 5.5 (Zhao et
al. 2004). The long-term titration data as a function of time for RM by 0.1M HCl are
67
reported in Tables A.3-A.6. The key reactions in solid phase are thus the dissolution
of sodalite and cancrinite as shown below (Schecher & McAvoy 2015):
Na8(AlSiO4)6(OH)2.4H2O + 18H+ ↔ 8Na+ + 6Al3+ + 6Si(OH)4 + 2H2O (4.1)
Na6(AlSiO4)6(CaCO3)(H2O)2+24H+ ↔ 6Na++6Al3++6Si(OH)4+Ca2++CO32-+2H2O (4.2)
Figure 4.3. XRD pattern of raw RM overlapped with titrated RM at pH 6
Figure 4.4. XRD pattern of raw RM overlapped with titrated RM at pH 4.5
68
4.2. Carbonation of Red Mud
Carbonation of red mud was performed at different operating conditions such
as CO2 concentrations, total gas flow rate (TF), agitation speeds, and concentration
of solid in red mud. The experiments aimed to evaluate the effect of these operating
conditions on the carbonation process. The phenomenon of pH rebound was also
examined under these conditions.
4.2.1. Effect of CO2 Concentration on Carbonation of RM
Carbonation was performed in both RM slurry and RM liquor alone in order to
evaluate the effect of CO2 concentrations on the rate of carbonation in solid and
liquid phases. The carbonation experiments were performed at different CO2
concentrations ranging from 10% to 100% depending on the total flow rate of gas.
The results showed that the equilibrium pH of the neutralised RM decreased with
increasing CO2 concentrations in both RM slurry and RM liquor. Figure 4.5 indicates
that at a fixed total gas flow rate of 200mL/min, the steady state pH was reached
from 7.5 to 6.6 for CO2 concentration values ranging from 15% to 100%.
Additionally, it can be seen that the rate of carbonation increased with the increase in
the concentration of CO2 as observed in Figure 4.5. When using concentration of 10-
15% CO2, it took about 60-75 minutes for establishing a pH of 7.5, whereas a steady
state pH of 6.9 and 6.6 was reached after 30 and 15 minutes of carbonation at 50%
and 100% CO2 concentration, respectively.
69
Figure 4.5. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
200mL/min and stirring speed of 250rpm
In contrast to RM slurry, the carbonation of RM liquor occurred very fast as
there is no solid reaction to slow down the rate. As illustrated in Figure 4.6, it took
more than half an hour to obtain an equilibrium pH of 7.0 when carbonated at 10-
15% of CO2. If the concentration of CO2 is increased to 50% and 100%, the
carbonation process took about 15 minutes to complete and reach the steady state pH
of 6.5 and 6.3, respectively. This indicates that the CO2 concentration had a large
positive influence on the carbonation process.
Figure 4.6. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
200mL/min and stirring speed of 250rpm
70
By comparing the carbonation of RM slurry and RM liquor, Figure 4.7 reveals
that at the same CO2 concentration values ranging from 15% to 100%, it took about 1
hour to reach the equilibrium pH of 7.5-6.6, respectively for RM slurry, while the
steady state pH values attained for the RM liquor were from 7.0 to 6.3 (0.3-0.5 pH
unit lower). In the carbonation of RM liquor, the pH curves in Figure 4.7
demonstrates three distinct inflexions reflecting the concentrations of hydroxide (OH-
), carbonate (CO32-), and bicarbonate (HCO3
-) that were previously observed in the
determination of ANC of RM as illustrated in Figure 4.1. These inflexions appeared
hard to distinguish in the pH curves of RM slurry carbonation. The inflexion of
converting carbonate to bicarbonate in RM slurry at pH<8.3 seemed to disappear in
Figure 4.7. This might be due to some carbonates precipitating and adhering to the
RM slurry causing the absence of carbonate to bicarbonate conversion in RM slurry
carbonation. This suggested that the carbonation process in the short time of the
experiment is mainly due to the liquid phase neutralisation. The solid phase
carbonation reactions are much slower (Khaitan et al. 2009b).
Figure 4.7. Comparison of carbonation between RM slurry and RM liquor at some
different CO2 concentrations, fixed TF of 200mL/min and stirring speed of 250rpm
71
Figure 4.8 describes the carbonation rate constant for both RM slurry and RM
liquor as a function of CO2 concentration at fixed total gas flow rate of 200mL/min
and stirring speed of 250rpm. The data of the pH taken from the carbonation process
was regressed to obtain the reaction model. Based on the R2 analysis, the best fit was
achieved using the first-order regression model given as follows (Frost & Pearson
1961):
ln c = ln co - kt
where c is the concentration of H+, co is the initial concentration of H+, t is the
reaction time in minutes and k is the rate constant. Using the first-order regression
model, the rate constant, k is selected such that the R2 accuracy is above 95%. This
approach for determining the rate constant, k was employed for all the carbonation
reactions in this work (Tables B.1 – B.23). From Figure 4.8 it can be seen that the
rate of carbonation of RM liquor were faster than that of RM slurry. The liquid
carbonation occurred so fast because of the absorption of CO2 to form aqueous
carbon dioxide and convert the hydroxide to bicarbonate and the reversibility of key
alkalinity reactions between hydroxide, carbonate, and bicarbonate (Eqns. 2.1-2.3).
Moreover, with the excesses of CO2 gas in the carbonation process, free hydroxide in
the form of NaOH, NaCO3- and Al(OH)4
- in the liquor reacts with the aqueous
carbon dioxide to precipitate as aluminate hydroxide (Eqns. 2.4 & 2.5). The other
data and plots of pH change as a function of time for the carbonation of both RM
slurry and RM liquor at different CO2 concentrations is presented in Figure B.1-B.6
and Tables B.1-B.8 and B.16-B.23.
72
Figure 4.8. Carbonation rate constant (k) for both RM slurry and RM liquor at
different CO2 concentration, total gas flow rate 200mL/min and speed 250rpm
4.2.2. Effect of Total Gas Flow Rate on Carbonation of RM
The carbonation of RM at different total gas flow rates (TF) was done in a
range of 100mL/min to 400mL/min. The air and CO2 gas from cylinders controlled
by Brooks flow rate meter were mixed together to create desired total flow rates of
gas mentioned above for each experiment. The kinetics of neutralisation of RM
slurry at different TF of gas is illustrated in Figure 4.9, and time-based data for the
residue carbonation is reported in Table B.9.
73
Figure 4.9. Carbonation of red mud by 30% of CO2, 250rpm and different TF of gas
Figure 4.9 indicates the dependence of carbonation process on the total flow
rate of gas. At the fixed 30% concentration of CO2, the equilibrium pH of the
carbonated residue decreased from 12.5 to 7.3 after an hour of carbonation at the
lowest TF of gas in the study (100mL.min-1). When the TF of gas was double the pH
values reached at 7.0 after half an hour and the carbonation time was only 4 times
less than at the highest TF of gas (400mL.min-1) in the study, indicating a strong
correlation between carbonation process and TF of gas. Figure 4.10 indicates that the
rate of carbonation increases with increasing total gas flow rate at a given CO2
concentration. This is because at higher total gas flow rate the absorption of CO2 gas
in the solution is perhaps complete (Çopur et al. 2007) leading to the sufficiency of
CO2 that speeds up the rate of carbonation process. The results of RM carbonation at
other CO2 concentrations at different total gas flow rate are reported in Figures B.7-
B.9.
74
Figure 4.10. Carbonation rate constant (k) for RM slurry at 30% CO2 concentration,
stirring speed 250rpm, and different total gas flow rate
4.2.3. Effect of Stirring Speed on Carbonation of RM
Plots and data of carbonation of RM slurry at different stirring speeds are
shown in Figure 4.11 and Table B.10. The stirring speeds investigated in the
carbonation of bauxite residue were at four different levels ranging from 250rpm to
700rpm. It can be seen from Figure 4.8 that while the impeller speed has no effect on
the final equilibrium pH, it has positive effect on the rate of pH reduction. When the
stirring speed was increased from 250rpm to 700rpm, the rate of reaction between
the gas and the mud increased, suggesting that at 250rpm the pH of the solution was
12.5, which decreased to a pH of 7.3 at 700rpm. The steady state pH level of 7.3 was
achieved after 30 minutes of carbonation at the stirring speed of 250rpm compared to
only 15 minutes at 700rpm.
75
Figure 4.11. Carbonation of red mud by 30% of CO2, TF of 200mL/min and
different stirring speeds
According to Figure 4.12, the rate of carbonation was proportional to the
agitation speeds. This is because the rate of CO2 absorption into the liquid phase is
enhanced by the agitation provided by the rotating impeller. If there was no stirring,
CO2 gas still absorbs into the liquid but at a very slow rate mainly due to molecular
diffusion. However, with the increase in stirring speeds the rate of CO2 absorption
was immediately boosted, indicating a physical process being involved in the
distribution of CO2 to the reaction site (Cardile et al. 1994). The results of
carbonation of RM at other CO2 concentrations and total flow rate of gas at different
agitation speeds were presented in Figures B.10-B.12.
76
Figure 4.12. Rate constant (k) for carbonation of RM slurry by 30% CO2
concentration, TF of 200mL/min and different stirring speeds
4.2.4. Effect of Solids Concentrations in RM on Carbonation of RM
Figure 4.13 shows the effect of solids concentrations in RM slurry on the
carbonation process. The experiments were performed in three different solids
concentrations of RM slurry, namely 35%, 40%, and 44% solids by weight. It can be
seen that the equilibrium pH was achieved at 7.5 after 30 minutes of carbonation in
all three concentrations of RM slurry. The pH curves suggest that there was no effect
of solids concentrations of slurry in the study on neutralisation reaction efficiency.
This would be due to the CO2 gas just reacts with soluble alkaline components in the
liquid and lowering pH of the solution, while the solid did not contribute to the
reduction of pH. In a previous systematic study, Cardile (1994) used the factorial
experimental approach to evaluate the effect of RM slurry density on the carbonation
process. This experimental approach concluded that RM slurry density was the factor
77
having a little significance for reaction efficiency of carbonation process. This
conclusion has well supported for the above discussions.
Figure 4.13. Carbonation of red mud by 30% of CO2, TF of 200mL/min and stirring
speed of 250rpm, and different solids concentrations in RM
From Figure 4.14, it can be seen that at a given 30% CO2 concentration, total
gas flow rate of 200mL/min and stirring speed of 250rpm, the rate of carbonation of
different solids concentrations in RM was virtually unchanged. Furthermore, all red
mud samples were diluted by adding distilled water in the carbonation experiments.
Hence, the variation in the amount of solids is very small for these red muds. This
evidence confirmed the no effect of solids concentrations in RM on the carbonation
process. The results of RM carbonation at other CO2 concentrations, total flow rate
of gas and rotating speeds at different solids concentrations in RM were illustrated in
Figures B.13-B.19 and carbonation data are given in Table B.11.
78
Figure 4.14. Rate constant (k) for carbonation of RM slurry by 30% CO2
concentration, TF of 200mL/min, and different solids concentrations in RM
4.2.5. pH Rebound in Carbonated RM
Figure 4.15 shows pH rebound of short term carbonated RM and carbonated
liquor at three CO2 concentrations of 15%, 50%, and 100%. It can be seen that
carbonated samples at different concentrations of CO2 have different rates of pH
rebound when exposing to atmospheric CO2 environment. For carbonated RM, the
pH rebound was very fast after one day of carbonation and gradually increased and
stabilised to a final value of about 9.7 after 20-25 days when equilibrated with the
atmosphere. In contrast, the pH rebound of carbonated liquor increased slowly and
took nearly a month to equilibrate with the atmosphere, which was a week slower
than that obtained with the carbonated RM slurry. The same phenomenon was
observed in previous study conducted by Rai (2013) when RM slurry was carbonated
in a multiple cycles of 7 weeks (pH rebound to 9.5). The basicity of the RM slurry
again increases as much caustic soda adhered to the red mud particles was slowly
leached in the solution. For carbonated liquor, the pH also rebound back with time
once CO2 gas added ceased. This was due to the carbonated liquors still contains the
79
mixture of CO32-/HCO3
- with pH values ranging from 8.3-10.6 (Cardile et al. 1994),
which drives the pH in the system back to 9.7. The plot of pH recovery for both
carbonated RM and carbonated liquor at other CO2 concentrations was illustrated in
Figures B.20.
Figure 4.15. pH rebound for both RM slurry and liquor at three CO2 concentrations,
TF of 200mL/min, stirring speed of 250rpm
Figure 4.16 describes the pH rebound of carbonated RM with different
concentrations of solids by weight (35%wt-44%wt). Accordingly, the experiments
suggested that the rate of pH rebound was influenced by the solids concentration
within two weeks, and then it reached to the same pH level of about 9.5-9.7. It can be
seen that carbonated RM with lower percentage of solid has a slower rate of pH
rebound compared with carbonated RM with a higher solids loading. Additionally,
while solids concentration has no effect on pH reduction by carbonation as discussed
previously, but it has some effects on pH rebound after carbonation. This may be
explained by the fact that pH reduction by carbonation was due to only the free soda
content in RM slurry was getting carbonated, whereas the bound soda concentration
80
adhered to solid phase was leached to the solution after carbonation driving the
phenomena of pH rebound. Previous study (Cardile et al. 1994) stated that total Na
content and RM slurry density were the main factors showing a significant effect on
the final pH value after carbonation. Therefore, it is thought that the pH rebound was
dependent of solids concentrations observed in this study because RM slurry with
higher % of solid (44%wt) may have more bound soda contents adhered to the RM
particles than RM with lower % of solid (35%wt).
Figure 4.16. pH rebound of carbonated RM slurries at different solids concentrations
4.2.6. Longer Carbonation of RM
Due to the time frame of experiment, longer carbonation of bauxite residue was
performed for 5 days. The carbonation aims to investigate the potential impact of
mineral transformation on the steady state pH level in the carbonation process.
Initially, the longer carbonation of RM slurry was done at the concentration of CO2
ranging from 15% - 60%, total gas flow rate of 200mL/min and stirring speed of
250rpm. However, 30% CO2 concentration was found to be effective because of a
81
greater amount of CO2 sequestration. Thus, the concentration of 30% CO2 was
selected for further carbonation experiments in the changes of different total gas flow
rate varying from 100mL/min-400mL/min, stirring speeds (250rpm-700rpm), and
solids concentrations in RM (35%wt-44%wt). The resulting pH values reached were
from 7.4 to 6.6 as indicated in Figure 4.17 and Tables B.12 - B.13.
Figure 4.17. Longer carbonation of RM slurry at different CO2 concentrations
Figure 4.17 shows that the concentration of CO2 still has effect on the
carbonation of RM in longer experiments. Accordingly, the different steady state pH
levels were obtained for different concentrations of injected carbon dioxide. When
using 15% of CO2 concentration, the final pH equilibrium level reached at 7.4,
whereas the pH was observed at 7.1 when the CO2 concentration was double, and 6.6
at the concentration of 60% CO2 after the first day of carbonation. From the first to
the last day of carbonation, there was a slight increase in the final pH equilibrium
level, for instance from 7.4 to 7.6 and 6.6 to 6.8 for the concentrations of 15% and
60% CO2, respectively. This observation was also found in previous study (Sahu et
82
al. 2010), which conducted the carbonation of RM slurry in three days. This may be
due to the dissolution of minerals buffering the pH values in the solution during the
carbonation process. The longer carbonation would suggest the dissolution of
sodalite as indicated by equation (4.1), and/or the break-up of cancrinite in the CO2
environment to release calcite (Sirbescu & Jenkins 1999; Yadav et al. 2009) as
below:
Na6(AlSiO4)6CaCO3(H2O)2 ↔ 6NaAlSiO4 + CaCO3 + H2O (4.3)
Figure 4.18. Longer carbonation of RM slurry at fixed 30% CO2, stirring speed of
250rpm and at different total gas flow rate
The final pH equilibrium results for long carbonation of RM slurry at a given
concentration of 30% CO2 at different total gas flow rate (100-400mL/min), stirring
speeds (250-700rpm) and solids concentrations in RM (35-44%wt) were plotted in
Figures 4.18-4.20. It can be seen that the final pH levels was constant at 7.0 for all
cases of carbonation in the period of 5 days. In other words, total gas flow rate,
stirring speed, and solid concentrations have no effect on the final pH of 5-day
carbonation of RM. This is because when CO2 gas was injected in RM slurry, it took
83
about an hour to reach the pH equilibrium of 7.0 as discussed previously. Further, if
the CO2 gas is consecutively added to the slurry in the period of 5 days, it will lead to
the excess CO2 amount absorbed into the system keeping the pH of the solution
constant.
Figure 4.19. Longer carbonation of RM slurry at fixed 30% CO2, total gas flow rate
of 200mL/min and different stirring speeds
Figure 4.20. Longer carbonation of RM slurry at fixed 30% CO2, total gas flow rate
of 200mL/min, stirring speeds of 250rpm and different solids concentrations of RM
84
4.3. Mineralogical Characterisation of Red Mud and Carbonated Red Mud
4.3.1. X-ray Diffraction Analysis
4.3.1.1. Solid phase composition in raw RM as quantified by XRD
Figure 4.21 shows the X-ray diffraction patterns of raw RM. The mineral
composition of this RM is comprised of predominantly hematite (Fe2O3), sodalite
(Na8(AlSiO4)6(OH)2.4H2O), and aluminium mineral compounds (Al2O3 &
AlO(OH)). Although the mineral analysis of RM has been reported in numerous
papers, the compositions of each RM sample differ because of the original
compositions of bauxite ore and the operating conditions used to extract alumina.
The elemental abundance in bauxite residue generally follow the order Fe > Si ~ Ti >
Al > Ca > Na (Gräfe et al. 2009; Liang et al. 2014). The mineral phase compositions
of the raw RM used in this study identified by XRD pattern consist of sodalite
(Na8(AlSiO4)6(OH)2.4H2O), cancrinite (Na6(AlSiO4)6(CaCO3)(H2O)2), boehmite
(AlO(OH)), gibbsite (Al(OH)3), anatase (TiO2), quartz (SiO2), and hematite (Fe2O3).
Figure 4.21. Variation of powder XRD pattern of raw RM
85
The quantification results determined by TOPAS V4.2 software in Figure 4.22
suggest that hematite occupied a large proportion (58%) as the major composition of
raw RM, followed by sodalite (~20%). Boehmite and anatase made up a minority of
~8.4% and ~6.8% of the total composition, respectively. Quartz accounted for only
~3.7%, while cancrinite and gibbsite accounted for even smaller amounts of ~2.3%
and ~1%, respectively. The broadness of phase peaks and quantifications in the XRD
patterns align with the following remarks from Grafe et al (2009), that approximately
70% (by weight) of bauxite phases are crystalline whereas the remaining 30% are
amorphous materials.
Figure 4.22. Phase composition quantification of raw RM
Figure 4.23 illustrates variation of carbonated XRD pattern compared to raw
RM. Changes to the XRD intensity associated to some mineral phases can be clearly
seen. This shows that these minerals are unstable in the CO2 environment during the
carbonation process. In particularly, Figure 4.23 indicates the dissolution of sodalite
(Na8(AlSiO4)6(OH)2.4H2O) and cancrinite-(Na6(AlSiO4)6CaCO3(H2O)2) and the
formation of calcite in the carbonation process. Both sodalite and cancrinite
86
structures have the same framework stoichiometry [AlSiO4]6, (Barnes et al. 1999a;
Gerson & Zheng 1997), the presence/absence of sodalite, therefore, is not easily
defined due to the high degree of overlap of the sodalite diffraction pattern by
cancrinite (Gerson & Zheng 1997). In Figure 4.23, there is no calcite peak in raw
RM pattern, but this peak is present in carbonated RM pattern at 330 2theta. Sodalite
and cancrinite peaks at 16.10, 270, 390 and 480 2theta show a decline but still present
after 5 days of carbonation, indicating the dissolution and/or break-up of these phases
caused the formation of calcite. Furthermore, the XRD pattern of treated red mud
revealed that the intensity of gibbsite at peak 210 2theta was significantly increased
while boehmite peaks at 170 and 310 2theta were decreased. The dissolution of
sodalite is responsible for the increase of gibbsite and over the long term carbonation
boehmite is converted to the more stable phase gibbsite (Khaitan et al. 2009b) as
indicated in Figure 4.23.
Figure 4.23. XRD pattern of carbonated RM compared with raw RM
87
4.3.1.2. Effect of CO2 concentrations on solid phase composition in carbonated RM
The mineral content of carbonated red mud at different concentrations of CO2
gas using TOPAS V4.2 was reported in Figures 4.24-4.27, and Table 4.3. It can be
seen that after carbonation at different CO2 concentrations, there were marked
changes in the content of certain minerals in the treated mud. In particularly, at 15%
of CO2, the amounts of sodalite and cancrinite were markedly reduced from 19.75%
and 2.32% to 16.03% and 1.53%, while the amounts of gibbsite and calcite rose from
1.04% and 0% to 3.47% and 0.81%, respectively. Other minerals also showed a
decrease after carbonation such as boehmite (from 8.35% to 8.3%), quartz (3.74% to
2.93%) and anatase (6.8% to 5.7%). However, when the concentration of CO2 was
doubled (30%), the percentages of sodalite and cancrinite decreased more
significantly to 15% and 0.82%, respectively. Similarly, the proportion of gibbsite
and calcite increased to 5.05% and 1.51%, respectively. It can be seen that the
increase in the calcite content (1.51%) corresponded to a similar decreased in
cancrinite (from 2.32% to 0.8%). This confirmed that the amount of calcite formed in
the carbonated RM was attributable to the breakdown of cancrinite in the CO2
environment (Yadav et al. 2009).
88
Figure 4.24. Phase composition quantification of carbonated RM at 15% CO2
concentration and total gas flow rate of 200mL/min
Table 4.3. Effect of CO2 concentrations on the composition of solid phase in
carbonated RM as quantified by XRD
Composition Raw
RM
Phase quantification (%) of carbonated RM at fixed total gas flow
rate 200mL/min, stirring speed 250rpm and different CO2
concentrations
15%CO2 20%CO2 30%CO2 40%CO2 50%CO2 60%CO2
Sodalite
Cancrinite
Hematite
Boehmite
Quartz
Anatase
Gibbsite
Calcite
19.75
2.32
58.03
8.36
3.74
6.77
1.04
0.00
16.03
1.53
61.23
8.30
2.93
5.70
3.47
0.81
16.27
1.20
61.15
8.26
2.53
6.07
3.39
1.13
15.01
0.82
61.24
8.21
2.53
5.64
5.05
1.51
15.83
1.10
61.10
8.24
2.24
6.17
4.09
1.22
17.25
1.45
61.00
8.17
2.64
5.97
2.62
0.91
16.89
1.54
61.60
8.27
2.39
6.42
2.07
0.81
89
According to Figure 4.26, when the concentration of CO2 increased from 30%
to 40%, the percentage of gibbsite and calcite formed began to decline from 5.05% to
4.09% and 1.51% to 1.22%, respectively. The quantity of sodalite and cancrinite
correspondingly showed a slight increase from 15% to 15.83%, and 0.8% to 1.1%,
respectively. This means that less amount of sodalite and cancrinite was dissolved. If
the carbonation was done at higher concentration of CO2, for instance at 60%, as
shown in Figure 4.27 and Table 4.3, it can be seen that the dissolution of sodalite and
cancrinite was observed even lower (by ~2.9% and ~0.8% compare with the raw
RM), leading to the decrease in the formation of gibbsite and calcite to 2.07% and
0.81%, respectively. The quantifications of the minerals in the carbonated RM at
other CO2 concentrations are reported in Figures C.1-C.2.
Figure 4.25. Phase composition quantification of carbonated RM at 30% CO2
concentration and total gas flow rate of 200mL/min
90
Figure 4.26. Phase composition quantification of carbonated RM at 40% CO2
concentration and total gas flow rate of 200mL/min
Figure 4.27. Phase composition quantification of carbonated RM at 60% CO2
concentration and total gas flow rate of 200mL/min
91
4.3.1.3. Effect of total gas flow rate on solid phase composition in carbonated RM
The mineral phase composition of carbonated RM at a given 30% CO2
concentration, stirring speed of 250rpm and solids concentration of 44%wt but at
different total gas flow rate (TF) was presented in Figures 4.28-4.30, and Table 4.4.
It can be seen that the proportion of sodalite, cancrinite, and boehmite decreased at
all TF of gas. The dissolution of these minerals contributes to the increase in gibbsite
and calcite as discussed before. However, the amounts of gibbsite and calcite formed
at TF of 200mL/min were higher than that of other TF (100mL/min, 300mL/min, and
400mL/min). This again confirmed the efficiency of carbonation at 30% CO2 and TF
of 200mL/min.
Figure 4.28. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 100mL/min
92
Table 4.4. Effect of total gas flow rate on the composition of solid phase in
carbonated RM as quantified by XRD
Composition Raw
RM
Phase quantification (%) of carbonated RM at fixed
30% CO2 concentrations, stirring speed 250rpm, and
different total gas flow rate
100mL/min 200mL/min 300mL/min 400mL/min
Sodalite
Cancrinite
Hematite
Boehmite
Quartz
Anatase
Gibbsite
Calcite
19.75
2.32
58.03
8.36
3.74
6.77
1.04
0.00
17.97
1.58
61.13
8.31
2.78
6.97
1.50
0.77
15.01
0.82
61.24
8.21
2.53
5.64
5.05
1.51
15.71
1.33
61.15
8.28
2.62
5.66
4.24
1.01
15.79
1.60
61.63
8.30
2.38
5.77
3.79
0.84
Figure 4.29. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 300mL/min
93
Figure 4.30. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and total gas flow rate of 400mL/min
4.3.1.4. Effect of stirring speed on solid phase composition in carbonated RM
Carbonated RM samples at fixed 30% CO2 concentration and TF of
200mL/min but at different stirring speeds were quantified the phase composition
and plotted in Figures 4.31-4.33 and Table 4.5. The results show that when the
stirring speed rose from 350rpm to 700rpm, the formation of calcite and gibbsite also
increased associated with the decrease in amounts of sodalite and cancrinite
suggesting that the impeller can boost the dissolution of sodalite and cancrinite.
94
Figure 4.31. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 350rpm
Table 4.5. Effect of stirring speed on the composition of solid phase in carbonated
RM as quantified by XRD
Composition Raw
RM
Phase quantification (%) of carbonated RM at fixed
30% CO2 concentrations, total gas flow rate
200mL/min, and different stirring speeds
250rpm 350rpm 500rpm 700rpm
Sodalite
Cancrinite
Hematite
Boehmite
Quartz
Anatase
Gibbsite
Calcite
19.75
2.32
58.03
8.36
3.74
6.77
1.04
0.00
15.01
0.82
61.24
8.21
2.53
5.64
5.05
1.51
17.02
0.80
60.92
8.13
3.27
6.56
1.73
1.55
16.81
0.71
60.65
8.23
3.37
6.52
2.12
1.59
16.71
0.69
60.57
8.22
3.22
6.45
2.49
1.66
95
Figure 4.32. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 500rpm
Figure 4.33. Phase composition quantification of carbonated RM at fixed 30% CO2
concentration and stirring speed of 700rpm
96
4.3.2. Micro-morphological Characterisation of raw RM and Carbonated RM by
SEM
Figure 4.34 describes the SEM images of raw red mud before treatment with
CO2. The SEM images showed that raw RM composed of the large rounded shape
aggregate particles. Sodalite’s SEM photomicrograph showing “cotton ball” type
morphology reported by others (Barnes et al. 1999a, 1999b) can be seen here in
Figure 4.34a. As previously discussed, sodalite and cancrinite structures have the
same framework stoichiometry [AlSiO4]6, (Barnes et al. 1999a; Gerson & Zheng
1997) and were identified by the same peaks in XRD pattern, but the “cotton ball”
type morphology of sodalite may be used to distinguish the two (Fig. 4.34b) (Deng et
al. 2006). However, due to its small concentration (2.32% as quantified by TOPAS),
cancrinite is morphologically indistinguishable from the surrounding sodalite crystals
in the SEM photomicrograph (Barnes et al. 1999a).
Figure 4.34. SEM imaging of raw RM: (a) Sodalite in “cotton ball” form, and (b)
Structure of crystalline sodalite
97
SEMs of carbonated red mud treated at different CO2 concentrations are shown
in Figure 4.35. In addition to the changes in the mineral content over the
experimental period by contacted with CO2, SEM showed a corresponding change in
crystalline morphology. Raw RM before treatment often has its large rounded shape
particles. However, after carbonation these large rounded shape objects were
decreased as indicated in Figure 4.35a, and more porous materials, which were
absent in raw RM, appeared in carbonated RM (Fig. 4.35a-f). This indicates that
some mineral phases such as sodalite, cancrinite containing in RM are more soluble
or broken-up in acidic environment to form porous materials (Huijgen et al. 2005;
Newson et al. 2006; Sahu et al. 2010; Yadav et al. 2009).
After carbonation, a small concentration of calcite (1.51%) was observed by
XRD. This phase was derived from the break-up of cancrinite during the period of
experiment. SEM image showed the presence of a hexagonal prismatic crystal in
carbonated RM (Fig. 4.35f) as cancrinite morphology (Barnes et al. 1999a), and the
porous coating adhering around was determined as calcite (Yadav et al. 2009).
Therefore, a conclusion that the formation of calcite in carbonated RM might be due
to the break-up of cancrinite in CO2 environment can be made.
98
Figure 4.35. SEM imaging of carbonated RM at different CO2 concentration, TF of
200mL/min and stirring speed of 250rpm (a) 15%CO2 (b) 20%CO2 (c) 30% CO2 (d) 40%CO2 (e) 50%CO2 (f) 60%CO2
99
4.3.3. Chemical Composition Changes by EDX
Major chemical compositions of solids in raw RM and carbonated RM
determined by Energy Dispersive X-ray (EDX) from spot analysis and pallet of the
samples at different concentrations of CO2 are given in Tables 4.6 and 4.7, and
Figure 4.36.
Table 4.6. Major elemental composition (%w/w in average) of RM and carbonated
RM at different concentrations of CO2, TF of gas 200mL/min, stirring speed 250rpm
Major
element
Raw
RM
Red mud carbonated at different concentrations of CO2
15% 20% 30% 40% 50% 60% 75% 100%
C
O
Na
Al
Si
Ca
Ti
Fe
1.84 ±0.04
42.34 ±1.50
9.82 ±0.27
12.50±0.50
8.56 ±0.57
0.80 ±0.14
4.15 ±0.30
19.99±1.37
4.37 ±0.75
37.77±1.07
7.14 ±0.69
11.14±0.96
6.39 ±0.29
0.88 ±0.21
4.91 ±0.80
27.40±1.65
4.55 ±0.34
39.97±1.52
9.46 ±0.28
11.65±0.34
8.14 ±0.16
0.95 ±0.04
4.36 ±0.08
20.91±1.85
5.68 ±0.54
39.01±2.40
9.09 ±0.83
12.51±0.72
8.86 ±0.55
1.03 ±0.55
4.21 ±0.41
19.62±1.76
4.68 ±0.08
41.70±2.02
8.65 ±0.89
12.70±0.27
7.56 ±0.23
0.83 ±0.14
3.97 ±0.67
19.92±1.83
4.28 ±0.05
41.83±1.57
9.92 ±0.52
11.91±0.27
8.13 ±0.39
0.75 ±0.12
3.64 ±0.06
19.54±2.83
4.57 ±0.20
38.77±2.38
10.05±0.33
13.89±0.33
10.06±0.59
0.81 ±0.25
3.93 ±0.34
17.93±1.07
4.15 ±0.58
40.20±1.84
8.85 ±0.49
11.56±0.70
8.12 ±0.15
0.96 ±0.16
4.35 ±0.38
21.80±2.41
3.89 ±0.04
41.40±3.56
9.02 ±1.29
11.65±0.53
7.98 ±0.35
0.95 ±0.06
4.48 ±0.39
20.63±1.15
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
The results of elemental composition and chemical composition of major
oxides determined by EDX were obtained from short-term (2 hours) carbonation
experiments. The results showed that the raw RM used in this research has a
particularly high content of aluminium, indicating that the operating conditions for
aluminium extraction were not optimised. The extent of carbonation was found to be
100
dependent on concentration of CO2 gas. After carbonation, carbonated RM samples
were rich in C at all concentrations of CO2, and reached the highest amount (5.68%
w/w) at 30% CO2 concentration, over 3 times higher than before treatment. The
amount of CO2 (%w/w) absorbed in carbonated RM increased with the increasing of
CO2 concentration, from 14.2%w/w at 15% CO2 to 14.87%w/w at 50% CO2, and
reached the highest amount of nearly 18%w/w at 30% CO2 concentration, 2.5 times
higher than raw RM. After that, the amount of CO2 absorbed by RM decreased
gradually with increasing CO2 concentration, and the lowest extent of carbonation
occurred at 100% CO2, where the amounts of both C and CO2 absorbed by RM was
recorded at 3.89% and 13.57%, respectively. This proved that there was a positive
effect of CO2 concentration on the amount of CO2 captured by RM in the
carbonation process. For the short-term carbonation, the highest extent of
carbonation would take place at 30% CO2 concentration with nearly 18% of CO2
absorbed by RM as illustrated in Figure 4.36. To confirm this result, further
experiments of RM carbonation at CO2 concentrations ranging from 15% to 60%
were carried out in 5 days, and then other carbonation experiments at fixed 30% CO2
concentration were investigated at different total flow rate of gas, stirring speeds, and
solids concentrations in RM.
101
Table 4.7. Major compound composition (%w/w in average) of RM and carbonated
RM at different concentrations of CO2, TF of gas 200mL/min, stirring speed 250rpm
Major
compound
Raw
RM
Red mud carbonated at different concentrations of CO2
15% 20% 30% 40% 50% 60% 75% 100%
CO2
Na2O
Al2O3
SiO2
CaO
TiO2
Fe2O3
6.89 ±0.37
13.41±0.58
23.98±1.15
18.62±1.25
1.14 ±0.18
7.00 ±0.38
28.95±1.38
14.21±0.78
8.93 ±1.37
19.44±1.82
12.52±1.86
1.14 ±0.22
7.56 ±0.88
36.19±1.86
15.13±0.43
12.00±0.33
20.57±0.62
16.16±0.23
1.24 ±0.04
6.84 ±0.16
28.06±2.22
17.77±0.17
11.15±0.82
21.24±0.47
16.81±0.61
1.32 ±0.77
6.35 ±0.58
25.36±2.11
16.01±0.85
11.20±0.47
22.84±1.47
15.34±0.70
1.10 ±0.16
6.31 ±0.90
27.21±2.05
14.87±0.04
12.94±0.86
21.68±0.85
16.69±1.08
1.02 ±0.17
5.86 ±0.14
26.95±1.47
14.39±0.48
12.43±0.79
23.75±0.32
19.20±0.32
1.02 ±0.28
5.93 ±0.33
23.27±0.59
13.97±0.50
11.35±0.95
20.67±0.79
16.34±0.84
1.27 ±0.22
6.89 ±0.51
29.52±1.57
13.57±0.84
11.82±0.14
21.27±1.78
16.44±1.48
1.29 ±0.05
7.23 ±0.34
28.38±1.81
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Figure 4.36. The amounts of C and CO2 absorbed by RM after 2-hour carbonation at
different CO2 concentration, TF of 200mL/min and stirring speed of 250rpm
102
The results of elemental and chemical compositions of carbonated red mud in
5-day carbonation are demonstrated in Figure 4.37 and Tables 4.8 & 4.9. It can be
seen that the amounts of C and CO2 absorbed rose significantly after longer
carbonation (5 days), reaching 5.39%w/w for C, and 16.02%w/w for CO2,
respectively at 15% CO2 concentration. The longer carbonation confirmed the
highest extent of carbonation again occurred at 30% CO2, where the amounts of C
and CO2 absorbed permanently by the solid were 9.07% and 23.93%w/w,
respectively. The extent of carbonation started going down at 40% CO2 with
7.64%w/w for C and 21.77%w/w for CO2. The study showed that during carbonation
process while the CO2 concentration increased, equilibrium pH of the solution
decreased but the amount of CO2 absorbed was found to reduce. This means that the
extent of carbonation would be effective at a particular CO2 concentration, and it was
at 30% CO2 in this research.
Table 4.8. Major elemental composition (%w/w in average) of RM and carbonated
RM at 15%-60% CO2, TF of 200mL/min, 250rpm in 5 days of carbonation
Major
element
Raw
RM
Red mud carbonated at different concentrations of CO2
15%
CO2
20%
CO2
30%
CO2
40%
CO2
50%
CO2
60%
CO2
C
O
Na
Al
Si
Ca
Ti
Fe
1.84 ±0.04
42.34 ±1.50
9.82 ±0.27
12.50 ±0.50 8.56 ±0.57 0.80 ±0.14 4.15 ±0.30
19.99 ±1.37
5.39 ±0.23
35.99 ±0.69
10.47 ±0.28
12.87 ±0.41 9.57 ±0.36 0.85 ±0.13 4.10 ±0.36
20.75 ±0.39
6.68 ±0.32
40.65 ±0.19 8.46 ±0.73
10.49 ±0.54 6.99 ±0.29 0.89 ±0.09 4.38 ±0.22
21.47 ±1.15
9.07 ±0.58
32.38 ±0.48 8.01 ±0.68
11.49 ±0.12 8.23 ±0.56 0.99 ±0.25 4.63 ±0.30
25.21 ±1.54
7.64 ±1.12
36.10 ±1.46 9.40 ±0.61
12.36 ±0.65 8.57 ±0.90 1.34 ±0.57 4.12 ±0.12
20.41 ±0.73
6.27 ±0.19
40.76 ±0.34
10.54 ±0.31
12.25 ±0.08 8.11 ±0.21 0.40 ±0.01 3.04 ±0.26
18.63 ±0.30
4.83 ±0.35
39.33 ±1.46 7.81 ±0.18
11.57 ±0.77 8.04 ±0.39 1.19 ±0.20 4.74 ±0.10
22.47 ±0.76
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
103
Table 4.9. Major compound composition (%w/w in average) of RM and carbonated
RM at 15%-60% CO2, TF of 200mL/min, 250rpm in 5 days of carbonation
Major
compound
Raw
Red
Mud
Red mud carbonated at different CO2 concentrations
15%
CO2
20%
CO2
30%
CO2
40%
CO2
50%
CO2
60%
CO2
CO2
Na2O
Al2O3
SiO2
CaO
TiO2
Fe2O3
6.89 ±0.37
13.41 ±0.58
23.98 ±1.15
18.62 ±1.25 1.14 ±0.18 7.00 ±0.38
28.95 ±1.38
16.02 ±0.69
12.47 ±0.27
21.09 ±0.51
17.46 ±0.46 1.03 ±0.14 5.98 ±0.54
25.96 ±0.73
21.59 ±0.86
10.54 ±0.98
18.16 ±1.05
13.57 ±0.61 1.14 ±0.13 6.73 ±0.32
28.27 ±1.37
23.93 ±0.69 8.76 ±0.36
17.17 ±0.76
13.60 ±1.78 1.10 ±0.23 6.24 ±0.38
29.20 ±1.48
21.77 ±0.66
10.93 ±0.47
19.72 ±0.22
15.17 ±0.69 1.61 ±0.79 5.91 ±0.21
24.88 ±0.69
20.19 ±0.44
13.17 ±0.45
21.20 ±0.03
15.74 ±0.36 0.52 ±0.01 4.66 ±0.37
24.52 ±0.51
15.76 ±1.54 9.77 ±0.38
20.13 ±1.07
15.71 ±0.47 1.54 ±0.24 7.32 ±0.07
29.76 ±0.66
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Figure 4.37. Amounts of C and CO2 absorbed by RM after 5-day carbonation
104
Tables 4.10 & 4.11 and Figure 4.38 describe the major element and
compound composition of carbonated RM at a given 30% CO2 concentration and
different total gas flow rate. As discussed previously, total flow rate (TF) of gas has a
positive effect on the carbonation process, but the most effective extent of
carbonation occurred at total gas flow rate of 200mL/min (Fig. 4.37). At TF of
100mL/min, the amounts of C and CO2 absorbed by RM was about 5.24% and
17.24%w/w, respectively. When the TF of gas was doubled, the extent of
carbonation increased significantly with the absorbed C and CO2 amounts of 9.07%
and 23.93%w/w. The amounts of C and CO2 absorbed by RM started reducing at
higher total gas flow rate (300mL/min), with 7.05%w/w for C and 21%w/w for CO2.
Unfortunately, the higher the total flow rate of gas used the less effective is the
carbonation process and the smaller the amounts of C and CO2 captured by the RM.
The lowest amounts of C and CO2 (4.88% and 16.78%w/w, respectively) was
observed at TF of 400mL/min in the study.
Table 4.10. Major elemental composition (%w/w in average) of RM and carbonated
RM at 30% CO2 concentration, 250rpm and different total gas flow rate
Major
element
Raw Red
Mud
Red mud carbonated at different TF of gas
100mL/min 200mL/min 300mL/min 400mL/min
C
O
Na
Al
Si
Ca
Ti
Fe
1.84 ±0.04
42.34 ±1.50
9.82 ±0.27
12.50 ±0.50
8.56 ±0.57
0.80 ±0.14
4.15 ±0.30
19.99 ±1.37
5.24 ±0.04
39.66 ±1.65
12.78 ±0.53
11.36 ±1.31
7.23 ±1.20
0.93 ±0.11
3.88 ±0.65
18.93 ±0.19
9.07 ±0.58
32.38 ±0.48
8.01 ±0.68
11.49 ±0.12
8.23 ±0.56
0.99 ±0.25
4.63 ±0.30
25.21 ±1.54
7.05 ±0.54
37.70 ±0.60
8.25 ±0.29
12.20 ±0.28
8.21 ±0.35
1.21 ±0.20
4.74 ±0.39
20.65 ±0.68
4.88 ±0.16
41.23 ±0.31
14.32 ±1.29
11.29 ±1.58
6.81 ±0.64
1.11 ±0.21
4.20 ±0.32
16.15 ±0.36
Total 100.0 100.0 100.0 100.0 100.0
105
Table 4.11. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2 concentration, 250rpm and different total gas flow rate
Major
compound
Raw Red
Mud
Red mud carbonated at different TF of gas
100mL/min 200mL/min 300mL/min 400mL/min
CO2
Na2O
Al2O3
SiO2
CaO
TiO2
Fe2O3
6.89 ±0.37
13.41 ±0.58
23.98 ±1.15
18.62 ±1.25
1.14 ±0.18
7.00 ±0.38
28.95 ±1.38
17.24 ±0.34
16.15 ±0.92
19.90 ±1.37
14.26 ±1.55
1.22 ±0.14
6.02 ±1.01
25.20 ±1.93
23.93 ±0.69
8.76 ±0.36
17.17 ±0.76
13.60 ±1.78
1.10 ±0.23
6.24 ±0.38
29.20 ±1.48
20.99 ±1.15
9.80 ±0.23
20.01±0.55
15.00 ±0.50
1.47 ±0.23
6.91 ±0.67
25.82 ±1.07
16.78 ±0.71
18.61 ±1.83
20.39 ±1.69
13.86 ±1.24
1.49±0.28
6.72 ±0.45
22.15 ±0.70
Total 100.0 100.0 100.0 100.0 100.0
Figure 4.38. Amounts of C and CO2 absorbed by RM at a given 30% CO2
concentration, 250rpm and different TF of gas
106
The large positive effect of stirring speeds on the extent of carbonation was
verified by results in Figure 4.39 and Tables 4.12 & 4.13. The results showed that the
extent of carbonation presented by the absorbed CO2 content was directly
proportional to the stirring speeds. It can be seen that there was a steady increase in
the CO2 content from nearly 24%w/w at 250rpm to 26%w/w at 350rpm, followed by
27.5%w/w at 500rpm, respectively, then, remarkably rose to 32.2%w/w at 700rpm.
The higher stirring speeds may help increase the solubility of CO2 gas in the RM
solution and boost the surface area of reaction between CO2 and the mud leading to
the greater extent of carbonation (Jones et al. 2006; Sahu et al. 2010).
Figure 4.39. Amounts of C and CO2 absorbed by RM at given 30% CO2, TF of
200mL/min and different stirring speeds
107
Table 4.12. Major element composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min and different stirring speeds
Major
element
Raw Red
Mud
Red mud carbonated at different stirring speeds
250rpm 350rpm 500rpm 700rpm
C
O
Na
Al
Si
Ca
Ti
Fe
1.84 ±0.04
42.34 ±1.50
9.82 ±0.27
12.50 ±0.50
8.56 ±0.57
0.80 ±0.14
4.15 ±0.30
19.99 ±1.37
9.07 ±0.58
32.38 ±0.48
8.01 ±0.68
11.49 ±0.12
8.23 ±0.56
0.99 ±0.25
4.63 ±0.30
25.21 ±1.54
9.51 ±0.31
35.48 ±1.87
9.21 ±1.18
12.60 ±1.49
8.79 ±1.67
1.03 ±0.07
3.93 ±0.17
19.45 ±1.14
10.34 ±0.54
34.99 ±1.89
8.28 ±0.60
12.29 ±1.13
8.23 ±0.73
0.99 ±0.19
4.50 ±0.07
20.40 ±0.70
10.98 ±0.48
40.47 ±1.65
7.79 ±0.16
10.43 ±0.50
6.72 ±0.38
0.85 ±0.21
3.86 ±0.45
18.91 ±0.89
Total 100.00 100.00 100.00 100.00 100.00
Table 4.13. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min and different stirring speeds
Major
compound
Raw Red
Mud
Red mud carbonated at different stirring speeds
250rpm 350rpm 500rpm 700rpm
CO2
Na2O
Al2O3
SiO2
CaO
TiO2
Fe2O3
6.89 ±0.37
13.41 ±0.58
23.98 ±1.15
18.62 ±1.25
1.14 ±0.18
7.00 ±0.38
28.95 ±1.38
23.93 ±0.69
8.76 ±0.36
17.17 ±0.76
13.60 ±1.78
1.10 ±0.23
6.24 ±0.38
29.20 ±1.48
25.88 ±1.46
10.38 ±0.97
19.38 ±1.35
14.88 ±0.95
1.18 ±0.02
5.39 ±0.53
22.91 ±1.29
27.46 ±0.97
9.19 ±0.46
18.65 ±1.33
13.80 ±0.43
1.11 ±0.18
6.08 ±0.29
23.71 ±1.08
32.18 ±0.65
9.12 ±0.41
16.83 ±0.85
12.08 ±0.60
1.02 ±0.27
5.53 ±0.55
23.23 ±0.55
Total 100.00 100.00 100.00 100.00 100.00
108
The solids concentrations in RM show its effect on the amounts of C and CO2
absorbed as given in Tables 4.14 & 4.15 and Figure 4.40. Accordingly, the contents
of C increased steadily but the amount of CO2 demonstrated a remarkable rise in all
treatments. Bauxite residue with 35% of solids by weight was carbonated yielding an
extent of about 12.6% of CO2 absorbed, doubled that before treatment. The extent of
CO2 absorbed was rising to 17.8% when the solids concentration in RM increased by
5%. From Figure 4.40, it is noticeable that if the solids concentration rose by ~10%,
from 35%wt to 44%wt, the extent of CO2 absorbed increased twofold. The
experiments suggest that the more CO2 can be captured if the RM slurry contains
more content of solids in the composition. This was due to RM with higher solids
concentrations may contain more cancrinite mineral that is responsible for reacting
with CO2 to form carbonates (Yadav et al. 2009).
Table 4.14. Major element composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min, 250rpm and different solids concentrations
Major
element
Raw Red
Mud
Red mud carbonated at different solids
concentrations
35%wt 40%wt 44%wt
C
O
Na
Al
Si
Ca
Ti
Fe
1.84 ±0.04
42.34 ±1.50
9.82 ±0.27
12.50 ±0.50
8.56 ±0.57
0.80 ±0.14
4.15 ±0.30
19.99 ±1.37
3.37 ±0.40
44.29 ±0.23
10.69 ±0.60
13.02 ±0.41
8.26 ±0.74
0.82 ±0.22
3.23 ±0.18
16.31 ±1.24
5.15 ±0.37
42.70 ±1.16
9.30 ±0.35
11.49 ±0.26
7.72 ±0.16
0.64 ±0.11
3.86 ±0.89
19.14 ±1.08
9.07 ±0.58
32.38 ±0.48
8.01 ±0.68
11.49 ±0.12
8.23 ±0.56
0.99 ±0.25
4.63 ±0.30
25.21 ±1.54
Total 100.0 100.0 100.0 100.0
109
Table 4.15. Major compound composition (%w/w in average) of RM and carbonated
RM at 30% CO2, TF of 200mL/min, 250rpm and different solids concentrations
Major
element
Raw Red
Mud
Red mud carbonated at different solids
concentrations
35%wt 40%wt 44%wt
CO2
Na2O
Al2O3
SiO2
CaO
TiO2
Fe2O3
6.89 ±0.37
13.41 ±0.58
23.98 ±1.15
18.62 ±1.25
1.14 ±0.18
7.00 ±0.38
28.95 ±1.38
12.56 ±0.40
14.54 ±0.84
24.86 ±0.86
17.87 ±1.63
1.17 ±0.30
5.44 ±0.32
23.55 ±1.76
17.80 ±0.93
12.12 ±0.81
20.88 ±0.67
15.82 ±0.31
0.86 ±0.13
6.17 ±0.21
26.36 ±0.87
23.93 ±0.69
8.76 ±0.36
17.17 ±0.76
13.60 ±1.78
1.10 ±0.23
6.24 ±0.38
29.20 ±1.48
Total 100.0 100.0 100.0 100.0
Figure 4.40. Amounts of C and CO2 captured by RM in different solids
concentrations
110
By comparing the extent of carbonation between 2-hour and 5-day
carbonation processes, it can be seen that in the longer carbonation the degree of
absorption of C and CO2 was significantly increased. In Figure 4.41, at 15% CO2
concentration, there was not much difference in the amounts of C (1%) and CO2
(1.8%) captured by solids in RM between 2-hour and 5-day carbonations. However,
the prominent difference between the short-term and long-term carbonations was
identified at CO2 concentrations varying from 20%-50% in the study. It is estimated
that at 30% CO2 concentration, the amounts of C and CO2 absorbed after 5-day
carbonation were 3.4% and 6.2%, respectively, higher than that of the 2-hour one.
The degree of carbonation at the higher CO2 concentrations also illustrated
that at the 40% CO2 concentration, this difference was dropped to 3% and 5.7% for C
and CO2, respectively, then continued to decrease to 2% and 5% for C and CO2 at the
50% CO2 concentration. Interestingly, the difference in the amounts of C and CO2
between short term and long-term carbonations seemed to be disappeared at the 60%
CO2 concentration. The comparison shows that the most optimal extent of
carbonation occurred at 30% CO2 concentration for both 2-hour and 5-day
carbonations because of the greater amount of CO2 absorbed by the RM. The study
suggests that at higher CO2 concentration, for instance 60% CO2, the extent of CO2
absorbed by the mud remained unchanged regardless of short term or long-term
carbonation.
111
Figure 4.41. Comparison of amounts of C and CO2 captured between 2-hour and 5-
day carbonation at fixed TF of 200mL/min, 250rpm and different CO2 concentrations
4.3.4. Determination of Alkalinity of RM and Carbonated RM
Total alkalinity is defined as concentration of free and combined sodium
hydroxide (NaOH) plus concentration of sodium carbonate (Na2CO3) in the solution,
and expressed as mg/L of CaCO3 equivalent (Rounds 2012). In this study, the
variations of alkalinity of raw RM and carbonated RM are indicated in Figure 4.42.
Raw red mud at an initial pH of 12.5 yielded an estimated alkalinity of 232 meq/L or
11,610 mg/L as total carbonate. The major contributor to the alkalinity as determined
by using the advanced speciation method (USGS 2007) was hydroxide and carbonate
anions with the values of 309mg/L and 6381mg/L, respectively, whereas the
concentration of bicarbonate anions (72.2mg/L) was much lower. After carbonation
with different concentrations of CO2, total gas flow rate of 200mL/min and stirring
speed of 250rpm, total alkalinity dropped rapidly to 3,088 mg/L, having lost over
8,500mg/L of alkalinity (initial pH of 7.5 at 15% CO2), while all hydroxide alkalinity
was consumed (non-detectable), and carbonate alkalinity was almost consumed,
112
reducing to 2mg/L. In contrast to these two fractions, bicarbonate alkalinity increased
to 3,761mg/L after carbonation with CO2. These changes of alkalinity in both raw
RM and carbonated RM are also in accordance to investigations reported by Jones et
al. (2006).
In raw red mud, from Figure 4.43a three distinct contributors to the alkalinity
were observed, reflecting the concentration of hydroxide (OH-), carbonate (CO32-),
and bicarbonate (HCO3-). However, from Figure 4.43b after carbonation at 30% CO2
concentrations (pH~7.1), only peak of bicarbonate was observed, indicating that the
concentration of OH- (non-detectable) and CO32- (0.8-2 mg/L) were virtually absent
in all treatments. It can be seen from Figure 4.42, total alkalinity of carbonated RM
decreased to 2,104 mg/L as total carbonate at 30% CO2 concentration, confirming
that the carbonation process would be optimised at this condition.
Figure 4.42. Changes in HCO3-, CO3
2-, and OH- alkalinity in raw RM and
carbonated RM at different concentrations of CO2, TF of 200mL/min, 250rpm
113
Figure 4.43. Acid titration curves for a) Raw RM and b) Carbonated RM at 30%
CO2, TF of 200mL/min and stirring speed of 250rpm
4.3.5. Thermal Analysis using TGA-DSC
The TGA-DSC analysis aims to not only measure physical or chemical
changes in both raw RM and carbonated RM but also determine the composition of
substances after carbonation. Figure 4.44 shows that the weight loss of RM takes
place in several steps. The first step occurred between 600-2000C where the RM
undergoes a weight reduction of 0.84% and the DSC peak centred at 650C due to the
removal of physically adsorbed moisture (Sushil et al. 2010). Another weight loss of
1.5% taking place in the range of 2000C to 4000C with the broad DSC peak at 2480C
can be attributed to the loss of structural H2O (Liu et al. 2007), and also the removal
of H2O from Al(OH)3 (Sahu 2011). The third endothermic effect between 4000C and
5750C, which corresponds to a weight loss of 0.6%, and the final step occurring in
the range from 5750C to 7000C associated with the weight loss of 0.35%, can be
attributed to the dehydroxylation of Ca(OH)2 to lime and partial dehydration of
silicates that continues up to over 7000C (Navarro et al. 2010).
114
Figure 4.44. TGA-DSC plots indicating weight loss of RM
For the carbonated RM, the thermal decomposition behaviours also occurred in
four steps as shown in Figure 4.45. Between 500C and 1700C, the weight loss was
determined about 2.12% of the total weight. Probably, this loss can be attributed to
the evaporation of physically absorbed water in the carbonated RM. Next step taking
place in the range from 2200C to 3800C was found to be 1.36% of weight loss. This
was proposed to be the loss of loosely and strongly bound water H2O in the minerals
(Sahu 2011), which may be due to the removal of H2O from sodalite and cancrinite
in the carbonated RM. The third step occurred in the range of 4400C-6000C with the
weight loss of 1.13%, possibly corresponding to the release of CO2 during
calcination of CaCO3 to CaO (Liu et al. 2007; Sushil et al. 2010; Zhang & Pan
2005). Finally, in the range from 6000C-7100C may take place the remaining
endothermic evolution of H2O and CO2 in the carbonated RM, the weight loss
associated to this process is about 0.6%.
115
Figure 4.45. TGA-DSC plots indicating weight loss of carbonated RM
4.3.6. FT-IR Spectroscopy
Raw RM contains a complex chemical composition of different minerals. Thus,
after contacting with CO2 gas some minerals may react with CO2 to form chemical
bonds in a molecule or a group of products, such as carbonate products. FR-IR
Spectroscopy helps identify these chemical bonds or stretching vibrations of
products. The FT-IR spectra of pristine raw RM and carbonated RM are presented in
Figure 4.46. The peaks illustrated at ~3300 cm-1 and ~1650 cm-1 region for both fresh
RM and carbonated RM are due to the stretching vibrations of OH group and of
molecular H2O, respectively (Cardell et al. 2009; Gok et al. 2007). This form of
water hydroxyl-stretching vibrations derived from water adsorbed on the outer
surface and free water between layers of RM structures. These vibrations, therefore,
are more intense in an infrared spectrum because of the large change in dipole
moment (Palmer et al. 2009). The bands that appear at 1412 cm-1 and 1410 cm-1 were
ascribed to C=O stretching vibrations, which confirm the existence of carbonate
groups (Cardell et al. 2009; Haberko et al. 2006). This was to prove that CO2 has
been absorbed by the pristine RM during carbonation process. Additionally, the
116
intensity of this peak in carbonated RM was decreased, which to a certain extent
confirmed the break-up of carbonate minerals such as cancrinite, when the RM
contacts with CO2.
The absorptions between 966 cm-1 and 963 cm-1 in both RM and carbonated
RM were characteristics of Si-O, or O-Si-O stretching modes of silica and silicates.
The bands at 527 cm-1 and 440 cm-1 correspond to Si-O-Al stretching vibrations and
Fe-O bonds, respectively (Gok et al. 2007; Navarro et al. 2010).
Figure 4.46. Fourier Transform Infrared (FT-IR) spectra of RM and carbonated RM
4.4. Determination of CO2 Sequestration
The amount of CO2 sequestered by RM can be estimated based on the quantity
of CO2 consumed by both the solid phase and liquid phase in the carbonation.
4.4.1. Determination of CO2 sequestered in 2-hour carbonation of RM
The amounts of CO2 sequestered by RM when carbonated at different
concentrations of carbon dioxide in the period of 2 hours are presented in Figure
4.47. Overall, the quantities of CO2 captured by RM differed when the mud was
117
carbonated at different CO2 concentrations. The experiments indicated that the
significant amount of CO2 captured by RM was recorded in the range of 15%-60%
CO2 concentration. Total gas flow rate (TF) of 200mL/min was found to be the most
effective where the amount of CO2 captured was prominent.
It can be seen that the highest amount of CO2 captured by RM was estimated to
be 4.56g CO2/100g of RM (or 45.6g CO2/kg RM) at 30% CO2 concentration. In the
range of CO2 concentration from 15% to 30%, the amount of CO2 captured by RM
increased significantly then decreased when the concentration of CO2 was higher
than 30%. However, the experiments suggested that the amounts of CO2 sequestered
by RM (3.9g and 3.51g CO2/100g of RM or 39g and 35.1g CO2/kg RM) at the
concentration of CO2 40% and 50% were still higher than that (3.67g and 3.45g
CO2/100g of RM or 36.7g and 34.5g CO2/kg RM) at 20% and 15%. In summary, the
most effective CO2 concentration for the sequestration was 30%, where the CO2
amount captured was highest in this study.
Figure 4.47. Amounts of CO2 sequestered by RM after 2-hour carbonation at
different CO2 concentrations, stirring speed of 250rpm
118
4.4.2. Determination of CO2 sequestered in 5-day carbonation of RM
The amounts of CO2 sequestered by both RM solid and RM liquor in 5-day
carbonation process at CO2 concentrations 15%-60%, total gas flow rate of
200mL/min and stirring speed of 250rpm are presented in Figures 4.48.
The 5-day carbonation experiments show the same trend of CO2 capture as that
observed with the 2-hour carbonation process as indicated in Figure 4.47. It can be
seen from Figure 4.48b that the amounts of CO2 consumed by RM liquor were
insignificant. The maximum quantity of CO2 absorbed by the liquor was found to be
0.139g/100g of RM liquor or ~1.4g CO2/kg RM liquor at 30% CO2 concentration,
whereas this amount in RM solid accounted for 6.36g CO2/100g of RM solid or
63.6g CO2/kg RM solid. This means that the liquor contributed only 2.2% to the CO2
sequestered by RM, while the remainder (~98%) came from the solid.
Totally, the amounts of CO2 sequestered by the whole RM were considered
both the solid phase and the liquid phase. In Figure 4.48, the amount of CO2 captured
by RM increased slightly from 4.0g CO2/100g of RM (or 40g CO2/kg RM) at 15%
CO2 concentration to 4.2g (or 42g CO2/kg RM) at 20% CO2 concentration. Again,
this amount reached its maximum value of 6.5g CO2/100g of RM (or 65g CO2/kg
RM) at 30% CO2 concentration, then decreased to 4.7g (or 47g CO2/kg RM) at 40%
CO2 concentration. The experiments also confirmed that the significant amount of
CO2 captured by RM occurred in the concentration of CO2 from 15%-60% as
discussed above.
119
Figure 4.48. Amounts of CO2 captured by RM (A): solid, (B): liquor, after 5-day
carbonation at different CO2 concentrations, TF of 200mL/min and speed of 250rpm
By comparing the potential of CO2 sequestration between 2-hour and 5-day
carbonations, it can be seen from Figure 4.49 that the longer time carbonation would
have yielded higher amounts of CO2 captured than the shorter time carbonation. The
experiments suggested that the amount of CO2 captured in the longer carbonation
was nearly 40% higher than that in the shorter carbonation at the same concentration
of CO2 (30%). This did confirm that the long-time carbon sequestration was
associated with interaction with bauxite residues, whereas short-time capture was
associated to reaction with the liquor. Carbon sequestration by the RM liquor is
associated with the conversion of hydroxide alkalinity to carbonate and bicarbonate
alkalinity (Cardile et al. 1994; Rai 2013; Shi et al. 2000), while the longer-time
carbon sequestration involved the reaction, dissolution and/or break-up of sodalite
and cancrinite to form calcite (Yadav et al. 2009) as discussed previously.
120
Figure 4.49. Comparison of CO2 amounts captured between 2-hour and 5-day
carbonations
Stirring speed had a positive effect on the carbonation efficiency as discussed
earlier. This also contributed to the carbon sequestration efficiency. The experiments
show that there was an increase in CO2 captured by RM when the stirring speed
increased. The CO2 amount captured rose by 0.34g/100g of RM (or 3.4g CO2/kg
RM) from the stirring speed of 250rpm to 700rpm as illustrated in Figure 4.50.
Figure 4.50. Amounts of CO2 captured by RM carbonated 30% CO2 concentration,
TF of 200mL/min and at different stirring speeds
121
With respect to carbon sequestration, the CO2 capture capacity also varied with
solids concentrations in RM studied in this research. Figure 4.51 demonstrates that
RM with higher concentration of solids by weight (44%wt) sequestered more CO2
(65g CO2/kg RM) than the RM with lower concentration of solids (40%wt and
35%wt, 39.6g and 33.2g CO2/kg of RM, respectively). This provides further
evidence that the carbon sequestration potential involved reaction with the solids of
red mud.
Figure 4.51. Amounts of CO2 captured by RM with different solids concentrations
carbonated at 30% CO2 concentration, TF of 250mL/min and 250rpm
In comparison of CO2 sequestration capacity of RM in this study with previous
researchers’ estimated results reported such as 23g CO2/kg of RM (Shi et al. 2000),
11.9g CO2/kg of RM (Khaitan et al. 2009b), 53g CO2/kg of RM (Yadav et al. 2009),
70.2g CO2/kg of RM (Sahu et al. 2010), 41.5g CO2/kg of RM (Bonenfant et al.
2008) as summarised earlier in Table 2.4, this study produced 65g CO2/kg of RM. In
122
fact, it is quite inappropriate to make the direct comparison of CO2 sequestration
capacity amongst RM from different studies due to vastly different mineralogy of the
residues.
Based on the annual production of 135 million tonnes of RM and future
inventory prediction of 5 billion tonnes by 2030 worldwide (Dentoni et al. 2014), it
is estimated that the amounts of CO2, which could be potentially captured by RM
were about 7.8 million tonnes per year and 325 million tonnes for the cumulation,
respectively. Similarly, the annual production of RM in Australia is about 30 million
tonnes (Sutar et al. 2014), which could potentially be used to capture ~2 million
tonnes of CO2. Presently, a full-scale carbonation facility has been established in
Western Australia by Alcoa. This facility helps to enhance the storage capacity of
RM by neutralising the RM before disposal, and to capture up to 4% of CO2 from the
refinery (Cooling et al. 2002). It is expected that the utilisation of red mud waste can
potentially remove large amounts of CO2 produced from industrial activities.
Moreover, the process of capturing CO2 or other forms of carbon by using the RM
waste would lock up large amounts of greenhouse gas released to the atmosphere and
to either solve environmental problems or mitigate and defer global warming.
4.5. Modelling of Carbonation Process
Modelling was performed on two processes: metal concentrations from long-
term titration and carbonation of RM. The chemical equilibrium modelling system
MINEQL+ 5.0 (Schecher & McAvoy 2015) was used to predict the heavy metal
species concentrations from long-term titration data and final pH equilibrium values
obtained from carbonation of both RM slurry and RM liquor.
123
4.5.1. Modelling of potentially dissolved metals
The general chemical equilibrium constants were used in the program to
calculate the metal ions concentrations during that accounts for the CO2 absorption
behaviour observed in the long-term titration at pH values of 4.5, 6, 8, 10, and 12.5.
The selection of solids governing aqueous chemistry as input data in the model was
based on the measured concentration of dominant constituents as indicated in Table
4.16. Additionally, the rapid and long-term titration data as discussed previously in
Figures 4.1&4.2 illustrated pH region where the dissolution of these solids occur, and
this information was used to choose the solids. The selection of solids was also based
on the measured metal concentrations at different pH values as given in Figure 4.52
(dots only). This information identifies the metals dissolving or precipitating when
red mud was titrated from pH of 12.5 to 4.5 corresponding to bicarbonate endpoint.
Table 4.16. Concentration of raw RM and liquor
Constituent
Concentration
in dried solids
(g/kg solid)(1)
Concentration
in RM slurry
(g/L liquor)(2)
Concentration
in slurry (M)
Concentration
in liquor (M)
Carbonate
Na
Al
Si
Ca
Ti
Fe
5.1(3)
98.2
125
85.6
8
41.5
200
4.01(3)
77.2
98.2
67.3
6.3
32.6
157.1
0.76
3.35
3.64
2.40
0.16
0.68
2.81
0.011(4)
0.198(5)
0.076(5)
8.58E-05
0.00017(5)
1.42E-05
5.36E-06(5)
(1): Determined by EDX. (2): Converted to RM liquor volume basis (from RM 44% solid by weight as received). (3): Determined in this study (5.1mgTIC/gRM) and converted to liquor volume basis (5.1*1000/0.56). (4): Measured by Total Organic Carbon Analyser, Model TOC-VCSH/CSN + TNM-1, Shimadzu Corporation, Japan. (5): Determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS), Model Agilent 7500cs, Agilent Technologies,
USA.
124
Figure 4.52. Metal concentrations in RM liquor as a function of pH
Chemical components for the RM slurry tabulated in Table 4.16 were used in
the MINEQL+ 5.0 program. As the MINEQL+ 5.0 focuses on concentrations
normalised to the liquid phase volume, the dried solid concentration of each
constituent in Table 4.16 was therefore input on a liquid volume basic. For carbonate
content, the TOTCO3 setting was changed to simulate a “Closed to the atmosphere”
system in equilibrium with a fixed value as given in Table 4.16. Because the system
does not contain sodalite and cancrinite, it is essential to include these minerals in the
system. In the runtime manager, the ionic strength (I) corrections setting was kept on,
and the method of ionic strength calculated by the program from the final
equilibrium composition was selected as this method is more likely to give the
accurate results (Schecher & McAvoy 2015). For each metal considered in the
simulation, it is important to choose metal compound in the solids that may
potentially dissolve. Finally, the metal ions species simulations were conducted at the
125
selected pH values of 4.5, 6, 8, 10 and 12.5 to produce outputs in terms of metal
concentrations and dissolution/precipitation reactions using logK values for the
reactions from the database in the system. A list of solid dissolution/precipitation
reactions associated with their logK values was presented in Table 4.17. The metal
compounds in the RM solids were also inputted in the simulation, as these metal ions
will control cation concentrations in the suspending medium or liquid. The logK
values were obtained from the MINEQL+ 5.0 database except for sodalite and
cancrinite.
Table 4.17. Solid precipitation/dissolution reactions in red mud model
Solid
formed
Reactions LogK(1)
Sodalite
Cancrinite
Boehmite
Hematite
Calcite
2H2O + 6Al3+ + 8Na+ + 6Si(OH)4 =
Al6Na8Si6O26H2(s) + 26H+
6Al3+ + Ca2+ + CO32- + 6Na+ + 6Si(OH)4 =
Al6CaNa6Si6O24CO3(s) + 24H+
2H2O + Al3+ = AlOOH(s) + 3H+
3H2O + 2Fe3+ = Fe2O3(s) + 6H+
Ca2+ + CO32- = CaCO3(s)
-34.2
-36.9
-8.578
1.418
8.48
(1): LogK values were obtained from MINEQL+ 5.0 database, except for sodalite (fitted), cancrinite (fitted)
4.5.1.1. Analysis of solids controlling Al
Al exists in bauxite residue slurry in the forms of soluble Al (predominantly
Al(OH)4- at high pH ~11-13) in RM liquor and Al solid (boehmite-AlO(OH) and
gibbsite-Al(OH)3). Additionally, previous work (Akitt et al. 1972; Mesmer & Baes
1971; Sposito 1989) confirmed that Al at high pH existed in various forms including
Al2(OH)24+, Al3(OH)4
5+, Al(H2O)63+, Al2(OH)2(H2O)8
4+, Al13O4(OH)24(H2O)127+ and
Al8(OH)20(H2O)x4+. However, it is important to determine which form of Al from
126
these sources can control Al in bauxite residue and the carbonation process. As
discussed previously, Figure 4.1 indicated that the initial phase of titration was
controlled by RM liquor, and the later titration was controlled by the dissolution of
solids in the RM. This was confirmed by Khaitan et al. (2009a) stating that soluble
Al (predominantly Al(OH)4- at high pH ~11-13) was responsible for the initial buffer
region. In this study, it is not necessary to simulate all various forms of Al because
one of these species Al3(OH)45+ has been modelled by Khaitan et al. (2009a), and
stated that the results did not change significantly. The ANC in long-term titration of
RM (Fig. 4.2, Table 4.1) indicates that only 19% of ANC comes from liquid phase.
Thus, it is concluded that soluble Al in RM liquor would not affect the ANC, and Al-
containing solids in RM such as boehmite and gibbsite may control Al concentration.
Al solid was found to be present in RM in the form boehmite (AlO(OH)(s)) and
gibbsite (Al(OH)3(s)) as reported by numerous papers (Hanahan et al. 2004; Liang et
al. 2014; Rai 2013; Yadav et al. 2009). These solids were also identified in XRD
analysis and they might control Al concentration in the RM. Nevertheless, when
choosing potentially dissolved solids for simulation of Al metal, gibbsite and
boehmite cannot co-exist in a Gibb’s phase matrix for entire system because it leads
to a phase rule violation. This problem was also struggled by previous simulator
(Khaitan et al. 2009a), and concluded that gibbsite was found not to control Al in red
mud as it is the more stable phase and less soluble than boehmite. Therefore, in this
study, boehmite was selected in the Gibb’s phase matrix for entire system and it
worked smoothly with a logK of -8.578 by default matching well with initial Al
concentration of 0.076M.
The simulation of Al metal concentrations measured at different pH values in
Figure 4.52 describes the dissolution behaviour of boehmite in the system. Based on
127
the data from Figure 4.2 as discussed earlier and Table A.7 in the appendix, there
was no boehmite dissolution observed in the long-term titration process. It is because
if boehmite could have dissolved during titration process, its dissolution would have
resulted in the exceedance of the ANC values. This was confirmed by Al
concentrations in RM liquor at different pH values (both experiment and simulation
data) from Figure 4.52, where Al remains unchanged at pH level of 4.5. Further, this
was well supported by the XRD analysis and quantitative results of carbonated RM
at different CO2 concentrations (Fig. 4.24-4.27), different total gas flow rate (Fig.
4.28-4.30), and different stirring speeds (Fig. 4.31-4.33), showing no change in the
proportion of boehmite.
4.5.1.2. Analysis of Na controlling solids
XRD results showed the presence of Na in two solid phases called sodalite
Na8(AlSiO4)6(OH)2.4H2O and cancrinite Na6(AlSiO4)6(CaCO3)(H2O)2. Previous
investigators (Barnes et al. 1999b; Gerson & Zheng 1997) reported that in Bayer
process plant, NaAlSiO4 is present mainly in the form of sodalite and to a lesser
extent cancrinite. The quantification results in this study also confirmed the
proportion of sodalite (19.75%) is many times higher than that of cancrinite (2.32%).
Thus, simulation focuses on the Na concentration controlled by sodalite. Simulation
was tried with two equilibrium constant values (logK) of sodalite, the first logK=34.2
reported by Wannenmacher et al. (2005), and the second logK=39.2, which was 5
orders of magnitude higher, reported by Deng et al. (2006). The results indicate that
formation of logK=34.2 was the best fitted modelled corresponding to initial Na
concentration 0.198M in Table A.7. The simulated results in Figure 4.52 show that at
pH 10, Na simulated concentration matched well with Na experimental value, but
128
from pH 9 or lower, Na concentration increased quickly (0.32M at pH 8 and 0.36M
at pH 6) compared with experimental data (0.26M at pH 8 and 0.28M at pH 6) as
indicated in Tables A.7&A.8. Therefore, the simulation suggested that NaAlSiO4(s),
which is present mainly in the form of sodalite (Barnes et al. 1999b; Gerson &
Zheng 1997), dissolves at pH 9 rather than pH 8 as reported by Khaitan et al.
(2009a). The increase of Na concentration as a function of pH was attributable to the
dissolution of sodalite as illustrated by equation (4.1). In addition, the dissolution of
sodalite was confirmed by quantitative results in Figures 4.24-4.33 illustrating the
decrease of sodalite in carbonated RM at different CO2 concentrations, total gas flow
rate, and stirring speeds.
4.5.1.3. Analysis of Ca controlling solids
Although XRD analysis in Figure 4.21 did not show the presence of calcite in
raw RM, this pattern was prominently observed at 330 2theta in carbonated RM as
indicated in Figure 4.23. Further, the concentration of Ca in RM slurry as shown in
Table 4.16 was 0.16M, while the initial Ca concentration in RM liquor was virtually
zero (6.7mg/L or ~0.00017M). Therefore, it was concluded that Ca concentration
(0.16M) cannot be controlled by calcite in RM, but it should be controlled by a Ca-
bearing solid in RM. Many studies have noticed that Ca-bearing solid such as
tricalcium aluminate (Ca3Al2O6) existing in raw RM is often the main source to
produce Ca in porewater of RM and control the Ca concentration (Cardile et al.
1994; Khaitan et al. 2009a). However, in this study there was no tricalcium
aluminate peak to be observed in the bauxite residue pattern (Fig. 4.21) instead of
having the peak of another Ca-containing solid, namely cancrinite
Na6(AlSiO4)6(CaCO3)(H2O)2 at 16.10, 270, 390 and 480 2theta. This was used to
129
confirm that no tricalcium aluminate to be present in RM used in this study. Thus, it
was evident that cancrinite is the only source of controlling the Ca concentration.
XRD analysis in Figure 4.21 shows the possibility of the presence of cancrinite
at multi-peaks overlapping with sodalite. Sirbescu (1999) reported that there were
numerous types of cancrinite in the crystal structure database, but the cancrinite with
the structural formula expressed as Na6(AlSiO4)6(CaCO3)(H2O)2 is the only one that
can form or release calcite and nepheline. Although the amount of cancrinite in RM
was quite small (2.3%), it acts as a solid that controls Ca concentration in the system.
The formation logK= -36.9 of cancrinite as reported by Deng et al. (2006) was
applied in the system and worked well at different pH levels. The shape of the
simulated curve as shown in Figure 4.52 matches well with the experimental data.
Accordingly, the concentration of Ca increased as the pH decreased to 4.5, and this
was predicted accurately by the model. This confirmed that cancrinite dissolved to
control Ca concentration. This was further substantiated by the decrease in the
proportion of cancrinite in carbonated red mud at different operating conditions as
indicated by quantitative results in Figures 4.24-4.33.
4.5.1.4. Analysis of Fe controlling solids
The bauxite residue pattern in Figure 4.21 confirmed the presence of hematite
(Fe2O3(s)) in the RM. Therefore, it was included in the Gibb’s phase matrix for entire
system of the model. The dissolution of hematite was studied by modelling the
measured Fe concentration at different pH levels as shown in Figure 4.52. Data from
long-term titration in Figure 4.2 demonstrated that no hematite could dissolve during
the titration process. This was confirmed by the zero concentration of Fe being
measured at different pH levels in Table A.7. The equilibrium constant of Fe
130
(logK=1.418) provided by the program did not constrain any simulated results of Fe
at any pH levels. Therefore, simulation result of Fe concentration, which was plotted
in Figure 4.52 and tabulated in Table A.8, matches well with Fe experiment
concentration at all pH levels. The model predicted no dissolution of hematite which
was also observed by previous investigator (Khaitan et al. 2009a).
4.5.2. Modelling of RM carbonation
The chemical equilibrium program MINEQL+ 5.0 (Schecher & McAvoy 2015)
was used to simulate the final equilibrium pH values from RM carbonation. The
calculation of equilibrium pH in the model was done for both RM slurry and RM
liquor. The concentration of all components from RM slurry and RM liquor in Table
4.16 was used throughout the program. The amount of total inorganic carbon (TIC)
of RM was determined by the difference between the total carbon (TC) and total
organic carbon (TOC). The average TC and TOC were 6.5mg/g of RM and 1.4mg/g
of RM, respectively, and the TIC achieved from RM was 5.1mg of C/g of RM. Based
on the solids concentration of RM (44%wt), the amount of TIC was converted to
9107mgC/L or 0.76M, and reported as carbonate in Table 4.16. The carbonate
content obtained from RM liquor was 0.011M. This indicates that carbonate content
in RM liquor occupied a very small proportion (1.5%), while the remainder (98.5%)
was in solid phase. The carbonate content in RM liquor was mainly in the forms of
CO32- and NaCO3
- predicted by the MINEQL+ as shown in Table 4.18.
131
Table 4.18. Solid dissolution/precipitation and liquid reactions in RM simulation
Solid
formed
Potential Precipitation/Dissolution Reactions LogK(1)
Sodalite
Cancrinite
Calcite
Boehmite
2H2O + 6Al3+ + 8Na+ + 6Si(OH)4 =
Al6Na8Si6O26H2(s) + 26H+
6Al3+ + Ca2+ + CO32- + 6Na+ + 6Si(OH)4 =
Al6CaNa6Si6O24CO3(s) + 24H+
Ca2+ + CO32- = CaCO3(s)
2H2O + Al3+ = AlOOH(s) + 3H+
-34.2
-36.9
8.48
-8.578
Aqueous Species reactions
H2CO3 = 2H+ + CO32-
H+ + CO32- = HCO3
-
Al(OH)4- + 4H+ = 4H2O + Al3+
CO32- + Na+ = NaCO3
-
16.681
10.329
-22.688
1.270
(1): LogK values were obtained from MINEQL+ 5.0 database, except for sodalite (fitted) and cancrinite (fitted).
The procedures of modelling were similar to that of metal concentrations.
However, some adjustments were made in the program in order to achieve the good
results. For carbonate content, in the field labelled “Total CO3”, the TOTCO3 setting
was switched to an “Open to the atmosphere” system in that the partial pressure of
carbon dioxide (PCO2) was fixed. The concentrations of CO2 used for carbonation of
RM in this study were in percentage, so they were converted to partial pressure (PCO2)
to meet the requirement of the program. Because equilibrium pH values need to be
simulated, therefore the functions of “pH is calculated by MINEQL+” and “Base pH
calculation on Electroneutrality” in the calculation type of the Calculation Wizard
were selected in order to get more accurate results (Schecher & McAvoy 2015).
From the modelling performed, a list of solid and liquid phase reactions established
by MINEQL+ 5.0 was given in Table 4.18.
132
4.5.2.1. Simulation of RM liquor
The chemical equilibrium model MINEQL+ 5.0 was used to calculate final pH
equilibrium values at different partial pressures of CO2 (PCO2 ranging 0.1-1atm)
corresponding to different CO2 concentrations (10%-100%). The concentration of
components for the RM liquor in Table 4.16 was used as input data for the
simulation system. The simulated results were plotted in comparison with the
experimental data in Figures 4.53&4.54.
It can be seen from the Figures 4.53 & 4.54 that the simulated results in all
cases were higher than that of experiment. Although the shape of simulated curves
was correlative to the experiments, the difference between experimental and
simulated carbonation of red mud was observed greater in some points of PCO2 values.
Generally, the RM liquor simulation in most PCO2 values yielded 0.1-0.2 pH units
higher than experimental data. Particularly, the RM liquor simulation at other PCO2
values from 0.4 to 1.0 in Figure 4.54 yielded 0.3 pH units higher than experimental
data. This means that the difference between experiment and model prediction is 4%.
The difference was also observed to increase with increasing total gas flow rate. The
carbonation of RM liquor means no solid phase present during the carbonation.
Therefore, the solid phase did not resulted in this difference. The only explanation
provided to the difference between simulated and experimental data was due to the
final pH equilibrium values obtained from RM liquor carbonation was carried out
under stirring speeds of 250rpm and different total gas flow rate. Meanwhile the
chemical equilibrium-modelling program MINEQL+ 5.0 was merely performed with
the information of RM liquor from Table 4.16. By comparison with the difference
(0.5 pH units) for the bauxite residue pore water only in previous study (Khaitan et
133
al. 2009b), the difference in this study was smaller. The data for experimental and
simulated carbonation of RM liquor at different CO2 concentrations and total gas
flow rate was tabulated in Tables D.1-D.4.
Figure 4.53. Comparison of simulated and experimental carbonation of RM liquor at
different CO2 concentration and TF of (A): 100mL/min, (B): 200mL/min
Figure 4.54. Comparison of simulated and experimental carbonation of RM liquor at
different CO2 concentration and TF of (C): 300mL/min, (D): 400mL/min
4.5.2.2. Simulation of RM slurry
Simulation of RM slurry was performed similarly to RM liquor, and all
information about the concentration of solids in the RM slurry as shown in Table
4.16 was used in MINEQL+ 5.0. The results for experimental and simulated
134
carbonation of RM slurry at different CO2 concentrations and total gas flow rate are
illustrated in Figures 4.55 & 4.56.
Figure 4.55. Comparison of simulated and experimental carbonation of RM slurry at
different CO2 concentration and TF of (A): 100mL/min, (B): 200mL/min
Figure 4.56. Comparison of simulated and experimental carbonation of RM slurry at
different CO2 concentration and TF of (C): 300mL/min, (D): 400mL/min
Like RM liquor simulation results, it can be seen from Figures 4.55 & 4.56 that
the modelled carbonation of RM slurry was also predicted higher pH values than
experiments. The simulated carbonation of RM slurry showed the constraint of
135
relative solids in the slurry resulting in a difference of 6.0% higher than the
experimental pH values as given in Tables D.1-D.4. The slow dissolution of sodalite
and cancrinite as predicted by the model may be responsible for the higher pH values
compared with the experimental data. Furthermore, as discussed in simulation for
RM liquor, the experimental pH values in RM slurry were reduced under the effect
of other physical factors such as stirring speed and total gas flow rate. Such factors
cannot be added or adjusted in the MINEQL+ 5.0 program. Therefore, if the
carbonation of RM slurry were performed in the laboratory without the effect of
stirring speed and total flow rate of gas, the simulated values would match well with
the experimental data. The maximum difference between experimental pH values
and modelled pH values for carbonation of RM slurry in this study was observed
about 6.0%. This difference could be in an acceptable level as it is about 2 times
lower than that of previous work (Khaitan et al. 2009b). The data for experimental
and simulated carbonation of RM slurry at different CO2 concentrations and total gas
flow rate were given in Tables D.1-D.4.
The modelling results in this study provided the insights about the potential of
heavy metal leaching from RM at different pH values, and the final equilibrium pH
values in the carbonation of red mud. The metal leaching results predicted by the
model helps to determine the key metals controlling the aqueous chemistry of the red
mud carbonation process. Although the simulated equilibrium pH difference for both
RM liquor and RM slurry was from 4.0-6.0% higher than experiments, this
difference was still lower than that of previous work (6.5-11.2%) (Khaitan et al.
2009b). However, the simulation results in this study were higher than the
experiment data, while the simulations presented earlier (Khaitan et al. 2009b) were
lower than experiments. This can be explained that the RM carbonation experiments
136
in this study were performed with the effects of physical factors such as stirring
speeds, total gas flow rate, whereas the experiments by Khaitan (2009b) were
conducted without these factors. The chemical model developed in this study can be
used to calculate metal speciation, solubility equilibria for projects working with the
dissolution of heavy metals and pH equilibrium calculation in an aqueous solution.
4.6. Summary
Acid neutralisation of the whole red mud slurries was done by both rapid and
long-term titration to pH endpoints of 4.5, 6, 8, and 10. The acid neutralisation
capacity was 0.79meq/g RM for rapid and 1.91meq/g RM for long-term titration,
respectively. The analysis of the long-term titration data for the whole RM slurry
illustrated that solids in RM slurry contributed about 81% of the ANC to pH 4.5,
whereas, the contribution of the liquid phase was only 19% at the same pH value.
The carbonation of bauxite residue was found to be significant dependent on
the concentrations of CO2, total flow rate of gas, and agitation speeds, while the
solids concentration in RM had a little effect on the carbonation process. The
carbonation of RM liquor involves the conversion of hydroxide and carbonate
alkalinity. The carbonation of solid phase involves the reactions of sodalite and/or
cancrinite dissolution resulting in the increase of gibbsite and the formation of
calcite, respectively. After carbonation, the pH of carbonated RM slurry and RM
liquor rebound back to 9.7 within 20-30 days. The amount of CO2 sequestered by
RM liquor was 1.4g/kg RM, meanwhile this quantity in RM slurry occupied about
63.6g/kg RM. The carbonation process was found to be the most efficient at 30%
CO2 concentration, total gas flow rate of 200mL/min with 65g CO2/kg RM captured
by the RM slurry. The CO2 sequestration capacity implies that there would be about
137
7.8 million tonnes of CO2 captured per year worldwide in general, and approximately
2 million tonnes of CO2 sequestered in Australia in particular.
Four key metals Al, Na, Ca, and Fe were found to control the chemistry of the
carbonation of RM as predicted by the simulation. However, the results of measured
metal concentrations from long-term titration of RM indicated that both Fe and Al
were not responsible for the control of carbonation process because they did not
dissolved in the titration at different pH values. The major solids governing the
carbonation process were considered as Na and Ca, which resulted from the
dissolution of sodalite and cancrinite. For the simulation of carbonation process, the
final equilibrium pH modelled values in all cases of both RM slurry and RM liquor
were higher than that of experimental data. The difference between the experimental
data and simulated results in both RM slurry and RM liquor was from 4.0-6.0%. The
difference was also increased with increasing total flow rate of gas. The slow
dissolution of sodalite and cancrinite may contribute a part to this difference.
138
CHAPTER 5 FINDING OUTCOMES AND CONCLUSIONS
Global warming and climate change have been of special interest to the
alumina industry because of its big demand of energy for the aluminium refining
process resulting in large CO2 emissions to the atmosphere. In addition, an enormous
amount of RM generated annually from this sector has posed major environmental
concerns. To confront these problems, neutralisation of RM by using CO2 gas brings
a promising potential for the industry to deal with both the CO2 emissions and RM
disposal problems. The process of capturing CO2 and mixing it with RM would lock
up large amounts of the greenhouse gases that otherwise would be discharged
directly into the atmosphere. A number of feasibility studies have evaluated treating
bauxite residue with different acidic sources such as SO2 from flue gases, seawater,
hydrochloric acid, acidic fly ash, and even CO2 aqueous and gas, which were
discussed in the literature review. However, none of them suggests any particular
condition, which the carbonation process would be optimised. Thus, a lab-scale
experimental carbonation process has been performed in a range of different
conditions in order to work out the specific condition for the carbonation process.
5.1. Major Findings of This Research
5.1.1. Acid Neutralisation Capacity (ANC) of Red Mud
Red mud was titrated in rapid and long-term titration scheme to pH 4.5 in order
to determine the acid neutralisation capacity of RM. Rapid titration of RM to pH 4.5
yielded an ANC of 0.79meq/g RM, while long-term titration got the value of
1.91meq/g RM at the same pH level. The rapid titration did not give the actual ANC
139
of the RM as it skipped the contribution of the solids to ANC. The liquid phase key
reactions taking place in the rapid titration were attributable to the ANC of the RM
liquor.
Al(OH)4-(aq) + H+ ↔ Al(OH)3
0 + H2O (5.1)
NaOH(aq) + H+ ↔ Na+ + H2O (5.2)
NaCO3-(aq) + H+ ↔ Na+ + HCO3
- (5.3)
Results from long-term titration of RM slurry suggested that long term ANC is
double that of rapid titration at the same endpoint pH. Solid phase plays as a major
contributor to the ANC of the RM with an estimate of 81% compared to 19% from
the liquid phase. The key reactions in the solid phase responsible for the ANC of the
RM slurry were as follows:
The dissolution of sodalite and cancrinite:
Na8(AlSiO4)6(OH)2.4H2O + 18H+ ↔ 8Na+ + 6Al3+ + 6Si(OH)4 + 2H2O (5.4)
Na6(AlSiO4)6(CaCO3)(H2O)2 +24H+↔ 6Na+ +6Al3+ +6Si(OH)4+CaCO3 +2H2O (5.5)
5.1.2. Carbonation of Bauxite Residue
In this project, bauxite residue was carbonated in a range of different
conditions such as concentrations of CO2 (10%-100%), total flow rate of gas (TF)
(100mL/min-400mL/min), stirring speeds (250rpm-700rpm) and solids
concentrations in RM (35%wt-44%wt). The results show that the carbonation
process was significantly dependent on the first three conditions mentioned above,
whereas the concentration of solids in RM was observed having a little effect on the
140
carbonation process. The pH of the RM decreased with increasing concentrations of
CO2, total flow rate of gas and stirring speeds, indicating the physical behaviour of
mixing and distribution of CO2 being important factor in the carbonation process. In
other words, the rate and extent of carbonation process was directly proportional to
the concentrations of CO2, total flow rate of gas and stirring speeds, while the
concentration of solids in RM had little effects on the carbonation process.
In the carbonation process, the reactions occurred in both the liquid and solid
phases. The carbonation reaction in the liquid phase with hydroxide to form
carbonate involved the following species NaOH, NaCO3- and Al(OH)4
- and the
formation or the precipitation of gibbsite.
2OH-(aq) + CO2(aq) → CO3
2- + H2O (5.6)
H2O + CO32- + CO2(aq) → HCO3
-(aq) + H+
(aq) (5.7)
[Al(OH4)-](aq) + CO2(aq) + Na+
(aq) → Al(OH)3(s) + Na+(aq) + HCO3
-(aq) (5.8)
The longer carbonation experiments resulted in the dissolution of sodalite
and/or cancrinite in the acidic environment. The dissolution of sodalite contributed to
the increase of gibbsite, while the dissolution of cancrinite helped to release calcite,
lowering the pH of the slurry. The longer carbonation reactions are expressed as
equations (5.4) & (5.5) above.
It was found that at the same CO2 concentrations ranging from 10%-100%, it
took about 1 hour to reach the equilibrium pH of 7.5-6.6, respectively for RM slurry,
while the steady state pH for the RM liquor was lower 7.0-6.3 (0.3-0.5 pH unit
lower) in the short term carbonation reaction. This suggested that the liquor
141
chemistry had an influence on the short-term carbonation and that there was no solid
dissolution occurred in this process.
In both carbonated RM slurry and RM liquor, the pH recovery phenomena
were found to rebound back to 9.7, but in different times and rates. The pH from
carbonated RM slurry rebound very fast and took about 20-25 days to stabilise at 9.7,
while the carbonated liquor showed a lower rate of pH recovery, and needed a month
(about one week slower) to equilibrate with the atmosphere. For carbonated RM
slurry, the pH rebound occurs due to bound soda adhered to the RM particles slowly
leached in the solution (Rai 2013), while for the RM liquor, the CO32-/HCO3
- mixture
was found existing in the liquor (Cardile et al. 1994) causing the reverse reactions as
expressed above.
In this work, the maximum quantity of CO2 absorbed by the liquor was found
to be ~1.4g/kg of RM liquor, whereas this amount in RM solid accounted for 63.6g
CO2/kg of RM solid. This means the liquor contributed only 2.2% to the CO2
sequestration, while the remainder (~98%) came from the solid. Furthermore, the
carbonation of RM in different concentrations of CO2 yielded the different amount of
CO2 captured, and the longer carbonation (5 days) gave more CO2 sequestration
potential than the short term (2 hours). The other physical factors such as total gas
flow rate, stirring speeds and solids concentrations in RM also had different effects
on the CO2 sequestration in carbonation process. However, the carbonation process
was found to be most efficient at 30% CO2 concentration, total gas flow rate of
200mL/min, where the captured CO2 amount obtained at the maximum value of 65g
CO2/kg of RM. At this condition, gibbsite and calcite concentration were observed to
142
be highest 5.05% and 1.51%, respectively as the result of dissolution of sodalite and
cancrinite during the carbonation process (Eqns. (5.4) & (5.5)). In addition, the
alkalinity decreased to 2,104mg/L as total carbonate at the CO2 concentration of 30%
and TF of 200mL/min, confirming the carbonation process was optimised at this
condition. The stirring speed was observed to have a very positive effect on the
carbonation efficiency and CO2 capture as well. The CO2 amount sequestered by RM
increased by 3.4g/kg RM when stirring speed rose from 250rpm to 700rpm.
Based on the CO2 sequestration potential values of 65g/kg of RM, it is
estimated about 7.8 million tonnes of CO2 would be captured per year from the
annual RM waste generation, and approximate 325 million tonnes of CO2
sequestered by the cumulative residue. These estimates were made using an annual
production of 135 million tonnes of RM and global inventory prediction of 5 billion
tonnes by 2030. For Australian alumina industry, it is estimated that the annual
production of red mud is about 30 million tonnes, which could be potentially used to
sequestered approximately 2 million tonnes of CO2. Thus, CO2 sequestration using
RM waste would be considered as an environmentally friendly technology for
strategies of mitigating the most challenging problem of global warming and climate
change.
5.1.3. Modelling of the Carbonation Process
The general chemical equilibrium modelling system MINEQL+ 5.0 was used
to simulate the heavy metal concentrations dissolved in long-term titration of red
143
mud at different pH levels of 4.5, 6, 8, 10, and 12.5 and final pH equilibrium values
obtained from carbonation of both RM slurry and RM liquor. The modelling
suggested that four key dominant metals Al, Na, Ca, and Fe were found to govern the
aqueous chemistry of the red mud carbonation process due to their presence in both
soluble and solid forms in red mud. Measured metal concentration from long-term
titration at various pH values indicated that boehmite (AlO(OH)) and hematite
(Fe2O3) did not dissolve in the system, therefore, both Al and Fe were not
responsible for the control of carbonation process as their concentrations remained
unchanged. However, Na and Ca were considered the major solids controlling the
process. The dissolution of sodalite (Na8(AlSiO4)6(OH)2.4H2O) and cancrinite
(Na6(AlSiO4)6(CaCO3)(H2O)2) as discussed earlier were attributable to Na and Ca
concentrations in the system. The key reactions are illustrated in equations (5.4) and
(5.5) above.
The chemical model was formulated in MINEQL+ 5.0 to calculate the final
equilibrium pH values for both carbonation of RM slurry and RM liquor at different
partial pressures of CO2 (PCO2 ranging 0.1-1atm) corresponding to different CO2
concentrations (10%-100%). In both RM solid and RM liquor simulation, the final
equilibrium pH simulated results in all cases were higher than that of experimental
data. Specifically, the modelled data in most PCO2 values yielded from 0.1-0.4 pH
units higher than experimental data. The difference was also observed to increase
with increasing total gas flow rate. The slow dissolution of sodalite and cancrinite as
predicted by the model may be a part of the higher pH values compared with the
experimental data. Additionally, the experimental pH values in RM slurry were
reduced lower because of the effect of physical factors such as stirring speed and
total gas flow rate in the carbonation process. Such these factors cannot be added or
144
adjusted in the MINEQL+ 5.0 program. The simulated values would match well with
the experimental data in the case of the carbonation of RM slurry performed without
the effect of stirring speed and total flow rate of gas. However, the difference would
be in an acceptable threshold as it is about 2 times lower than that of previous work
done by Khaitan (2009b).
5.2. Conclusions
Today, the use of industrial wastes as a waste remediation technology to solve
another environmental problem has become a promising solution worldwide. This
method not only helps to improve the activities of pollution control, but to provide an
economic incentive to reuse waste as a resource. Aluminium industry is one of the
industrial sectors producing large amounts of waste red mud and high levels of CO2
emissions in the environment. Therefore, the disposal of red mud is a major problem
for most alumina plants. One of the viable solutions, which could be developed and
applied for disposing of the waste economically and safely to the environment, is the
carbonation of RM. Neutralisation of red mud using carbon dioxide (CO2) will
convert the highly caustic waste to the less hazardous state, and reducing the release
of greenhouse gas to the environment.
This research has performed the carbonation of RM over a range of different
operating conditions in order to optimise the carbonation process and determine the
potential amount of CO2 captured by the red mud. From the lab-scale results of the
study, it is concluded that this research would provide alumina refineries with
145
promising solutions to the problem involving in red mud storage and utilisation.
Furthermore, the research also provide data and information about optimal conditions
of carbonation process that can be implemented for practical scale up and design of
red mud carbonation plant. Finally, the outcomes of this research could support other
industries that can use wastes from red mud as valuable materials in the efforts of
reducing CO2 emissions, and of contributing to the global warming and climate
change mitigation strategies.
146
CHAPTER 6 RECOMMENDATIONS FOR THE FUTURE
WORK
Annually, the enormous quantities of red mud have been quickly generated
worldwide posing an alarming environmental problem. Future predictions suggest
approximately 5 billion tonnes of red mud will be produced by 2030. It is essential to
develop more appropriate management methods and utilisation practices to
ameliorate this challenging problem. This research reveals only the optimal condition
for the carbonation of red mud in limited considerations. Therefore, future research
needs to take into account the following suggestions:
1. Carbonation process should be carried out under more different operating
conditions such as temperature, content of sodium in red mud, higher
stirring speeds, pressure of gas, and density of red mud in order to
determine the most involved factors controlling the efficiency of the
process.
2. Red mud can be carbonated by using flue gas (emissions) from other
sources instead of pure CO2 gas as done in this research, or the mix of flue
gas with other acid gas sources under different conditions. Then, the results
achieved from the most optimal condition can compare with those when the
red mud is carbonated by pure CO2.
3. A pilot plant of large-scale carbonation of red mud using CO2 gas should
be highly recommended for the sustainable development and cleaner
production of alumina industry.
147
REFERENCES
Agrawal, A., Sahu, K. K. & Pandey, B. D. 2004, "Solid waste management in non-
ferrous industries in India", Resour. Conserv. Recy., 42, (2), 99-120.
Akhurst, D. J., Jones, G. B., Clark, M. & McConchie, D. 2006, "Phosphate removal
from aqueous solutions using neutralised bauxite refinery residues (BauxsolTM)",
Environ. Chem., 3, (1), 65-74.
Akitt, J. W., Greenwood, N. N., Khandelwal, B. L. & Lester, G. D. 1972, "27Al
nuclear magnetic resonance studies of the hydrolysis and polimerization of the
hexa-aquo-aluminum (III) cation", J. Chem. Soc. Dalton Trans., 604-610.
Alcoa 2011, Long term residue management strategy, Alcoa, Australia, viewed 02
June 2012, <www.alcoa.com/australia/en/pdf/2011_ltrms_pinjarra_refinery.pdf>.
Altundogan, H. S., Altundogan, S., Tumen, F. & Bildik, M. 2000, "Arsenic removal
from aqueous solutions by adsorption on red mud", Waste Manag., 20, (8), 761-
767.
Altundogan, H. S., Altundogan, S., Tumen, F. & Bildik, M. 2002, "Arsenic
adsorption from aqueous solutions by activated red mud", Waste Manag., 22, (3),
357-363.
Altundoğan, H. S. & Tümen, F. 2003, "As(V) removal from aqueous solutions by
coagulation with liquid phase of red mud", J. Environ. Sci. Health A: Tox.
Hazard. Subst. Environ. Eng., 38, (7), 1247-1258.
Amritphale, S. S., Anshul, A., Chandra, N. & Ramakrishnan, N. 2007, "A novel
process for making radiopaque materials using bauxite-Red mud", J. Eur. Ceram.
Soc., 27, (4), 1945-1951.
Ball, J. W., Jenne, E. A. & Cantrell, M. W. 1981, WATEQ3: a geochemical model
with uranium added, Report, Open-File Report, no. 81-1183.
Barnes, M. C., Addai-Mensah, J. & Gerson, A. R. 1999a, "The mechanism of the
sodalite-to-cancrinite phase transformation in synthetic spent Bayer liquor",
Microporous and Mesoporous Materials, 31, (3), 287-302.
Barnes, M. C., Addai-Mensah, J. & Gerson, A. R. 1999b, "A methodology for
quantifying sodalite and cancrinite phase mixtures and the kinetics of the sodalite
148
to cancrinite phase transformation", Microporous and Mesoporous Materials, 31,
(3), 303-319.
Barrow, N. 1982, "Possibility of using caustic residue from bauxite for improving the
chemical and physical properties of sandy soils", Austr. J. Agric. Res., 33, (2),
275-285.
Bernard, C. J., Charmaine, D. W. G., Glen, H., William, R. G. & Gerald, R. 2007,
Method of reducing calcia concentrations of green bayer process liquors, Alcoa
Australia, PCT/AU2007/000460, WO 2007115365 A1, Australia.
Bertocchi, A. F., Ghiani, M., Peretti, R. & Zucca, A. 2006, "Red mud and fly ash for
remediation of mine sites contaminated with As, Cd, Cu, Pb and Zn", J. Hazard.
Mater. B, 134, (1-3), 112-119.
Bonenfant, D., Kharoune, L., Sauves, S., Hausler, R., Niquette, P., Mimeault, M. &
Kharoune, M. 2008, "CO2 sequestration by aqueous red mud carbonation at
ambient pressure and temperature", Ind. Eng. Chem. Res., 47, (20), 7617-7622.
Borges, A. J. P., Hauser-Davis, R. A. & Ferreira de Oliveira, T. 2011, "Cleaner red
mud residue production at an alumina plant by applying experimental design
techniques in the filtration stage", Journal of Cleaner Production, 19, (15), 1763-
1769.
Bott, R., Langeloh, T. & Hahn, J. 2002, "Dry bauxite residue by hi-barR steam
pressure filtration", in Chadrashekar, S. (ed.), Proc. 6th Int. Alumina Quality
Workshop, AQW Incorporation, Brisbane, pp. 24-32.
Brunori, C., Cremisini, C., Massanisso, P., Pinto, V. & Torricelli, L. 2005, "Reuse of
a treated red mud bauxite waste: studies on environmental compatibility", J.
Hazard. Mater., 117, (1), 55-63.
Cardell, C., Guerra, I., Romero-Pastor, J., Cultrone, G. & Rodriguez-Navarro, A.
2009, "Innovative analytical methodology combining micro-X-ray diffraction,
scanning electron microscopy-based mineral maps, and diffuse reflectance
infrared fourier transform spectroscopy to characterise archeological artifacts",
Analytical Chemistry, 81, (2), 604-611.
Cardile, C. M., Hughes, C. A. & Summers, K. J. 1994, "Evaluation of carbon dioxide
treatment of red mud", Proceedings of the 6th AusIMM Extractive Metallurgy
Conference, Brisbane, pp. 317-320.
149
Carter, C. M., van der Sloot, H. A., Cooling, D., van Zomeren, A. & Matheson, T.
2008, "Characterization of untreated and neutralized bauxite residue for improved
waste management", Environ. Eng. Sci., 25, (4), 475-488.
Castaldi, P., Silvetti, M., Santona, L., Enzo, S. & Melis, P. 2008, "XRD, FTIR, and
thermal analysis of bauxite ore-processing waste (red mud) exchanged with heavy
metals", Clays and Clay Minerals, 56, (4), 461-469.
Cengeloglu, Y., Kir, E., Ersoz, M., Buyukerkek, T. & Gezgin, S. 2003, "Recovery
and concentration of metals from red mud by Donnan dialysis", Colloids and
Surfaces A: Physico-chemical Engineering Aspects, 223, (1-3), 95-101.
Cengeloglu, Y., Tor, A., Ersoz, M. & Arslan, G. 2006, "Removal of nitrate from
aqueous solution by using red mud", Separation and Purification Technology, 51,
(3), 374-378.
Clark, M. W., McConchie, D. M. & Johnston, M. 2009, "Bauxite refinery residues,
potential for waste utilisation", in J.R. Lavelle (ed.), Waste management:
Research, Technology and Developments, Nova Science Publishers, Inc., New
York, pp. 289-310.
Collazo, A., Fernandez, D., Izquierdo, M., Novoa, X. R. & Perez, C. 2005,
"Evaluation of red mud as surface treatment for carbon steel prior painting",
Progress in Organic Coatings, 52, (4), 351-358.
Cook, G. W. 1970, Removal of iron from domestic bauxite liquor, Reynolds Metals
Co., US Patent, US3493327 A.
Cooling, D. J. 1989, "Developments in the disposal of residue from the alumina
refining industry", in Campbell, P.R. (ed.), Light Metals, TMS, Halifax, pp. 49-
54.
Cooling, D. J. 2007, "Improving the sustainability of residue management practices-
Alcoa world Alumina", in A.B. Fourie & R.J. Jewell (eds.), Past and thickened
tailings: A guide, Australian Centre for Geomechanics, Perth, pp. 3-16.
Cooling, D. J., Hay, P. S. & Guilfoyle, L. 2002, "Carbonation of bauxite residue", in
Chadrashekar, S. (ed.), Proc. 6th Int. Alumina Quality Workshop, AQW
Incorporation, Brisbane, pp. 185-190.
Çopur, M., Pekdemir, T., Çolak, S. & Künkül, A. 2007, "Industrial symbiosis: High
purity recovery of metals from Waelz sintering waste by aqueous SO2 solution", J.
Hazard. Mater., 149, (2), 303-309.
150
Costa, G., Baciocchi, R., Polettini, A., Pomi, R., Hills, C. D. & Carey, P. J. 2007,
"Current status and perspectives of accelerated carbonation processes on
municipal waste combustion residues", Environmental Monitoring and
Assessment, 135, (1), 55-75.
Couillard, D. 1982, "Use of red mud, a residue of alumina production by the Bayer
process, in water treatment", Sci. Total Environ., 25, (2), 181-191.
Courtney, R. G. & Timpson, J. P. 2005, "Reclamation of Fine Fraction Bauxite
Processing Residue (Red Mud) Amended with Coarse Fraction Residue and
Gypsum", Water, Air, and Soil Pollution, 164, (1), 91-102.
Deng, Y., Flury, M., Harsh, J. B., Felmy, A. R. & Qafoku, O. 2006, "Cancrinite and
sodalite formation in the presence of cesium, potassium, magnesium, calcium and
strontium in Hanford tank waste simulants", Appl. Geochem., 21, (12), 2049-
2063.
Dentoni, V., Grosso, B. & Massacci, G. 2014, "Environmental Sustainability of the
Alumina Industry in Western Europe", Sustainability, 6, 9477-9493.
Dlugokencky, E. & Tans, P. 2015, Trends in Atmospheric Carbon Dioxide, National
Oceanic & Atmospheric Admistration-Earth System Research Laboratory
(NOAA/ESRL), USA, viewed 28 December 2015,
<http://www.esrl.noaa.gov/gmd/ccgg/trends/global.html>.
Dreybrodt, W. & Buhmann, D. 1991, "A mass transfer model for dissolution and
precipitation of calcite from solutions in turbulent motion", Chemical Geology,
90, (1-2), 107-122.
Dreybrodt, W., Lauckner, J., Zaihua, L., Svensson, U. & Buhmann, D. 1996, "The
kinetics of the reaction CO2 + H2O → H+ + HCO3− as one of the rate limiting
steps for the dissolution of calcite in the system H2O-CO2-CaCO3", Geochimica et
Cosmochimica Acta, 60, (18), 3375-3381.
Enick, R. M., Beckman, E. J., Shi, C. & Xu, J. 2001, "Remediation of metal-bearing
aqueous waste streams via direct carbonation", Energy & Fuels, 15, (2), 256-262.
Enserink, M. 2010, "After red mud flood, scientists try to halt wave of fear and
rumors", Science, 330 (6003), 432-433.
Fois, E., Lallai, A. & Mura, G. 2007, "Sulfur dioxide absorption in a bubbling reactor
with suspensions of Bayer red mud", Ind. Eng. Chem. Res., 46, (21), 6770-6776.
151
Friesl, W., Friedl, J., Platzer, K., Horak, O. & Gerzabek, M. H. 2006, "Remediation
of contaminated agricultural soils near the former Pb/Zn smelter in Austria: batch,
pot and field experiments", Environ. Pollut., 144, (1), 40-50.
Friesl, W., Lombi, E., Horak, O. & Wenzel, W. W. 2003, "Immobilisation of heavy
metals in soils using inorganic amendments in a greenhouse study", Journal of
Plant Nutrition and Soil Science, 166, (2), 191-196.
Frost, A. A. & Pearson, R. G. 1961, Kinetics and mechanism-A study of
homogeneous chemical reactions, John and Wiley & Sons, Inc., USA.
Garau, G., Castaldi, P., Santona, L., Deiana, P. & Melis, P. 2007, "Influence of red
mud, zeolite and lime on heavy metal immobilization, culturable heterotrophic
microbial populations and enzyme activities in a contaminated soil", Geoderma,
142, (1–2), 47-57.
Genc-Fuhrman, H., Bregnhoj, H. & McConchie, D. 2005, "Arsenate removal from
water using sand–red mud columns", Water Res., 39, (13), 2944-2954.
Genc-Fuhrman, H., Mikkelsen, P. S. & Ledin, A. 2007, "Simultaneous removal of
As, Cd, Cr, Cu, Ni and Zn from stormwater- Experimental comparison of 11
different sorbents", Water Res., 41, (3), 591-602.
Genc-Fuhrman, H., Tjell, J. C. & McConchie, D. 2004a, "Adsorption of Arsenic
from water using activated neutralized red mud", Environ. Sci. Technol., 38, (8),
2428-2434.
Genc-Fuhrman, H., Tjell, J. C. & McConchie, D. 2004b, "Increasing the arsenate
adsorption capacity of neutralized red mud (Bauxsol)", J. Colloid Interface Sci.,
271, (2), 313-320.
Genç, H., Tjell, J. C., McConchie, D. & Schuiling, O. 2003, "Adsorption of arsenate
from water using neutralized red mud", J. Colloid Interface Sci., 264, (2), 327-
334.
Gerdemann, S. J., O'Connor, W. K., Dahlin, D. C., Penner, L. R. & Rush, H. 2007,
"Ex situ aqueous mineral carbonation", Environ. Sci. Technol., 41, (7), 2587-
2593.
Gerrise, B. & Thomas, G. 2008, "Transport of bauxite residue leachate through clay
liners of storage impoundments: A synthesis of experimental and simulated data",
Proc. 8th Int. Alumina Quality Workshop, Darwin, Australia, pp. 154-161.
152
Gerson, A. R. & Zheng, K. 1997, "Bayer process plant scale: transformation of
sodalite to cancrinite", Journal of Crystal Growth, 171, (1-2), 209-218.
Glenister, D. J. & Thornber, M. R. 1985, "Alkalinity of red mud and its application
for the management of acid water", paper presented at Chemeca 85: Innovation in
the Process and Resource Industry-The 13th Australian Chemical Engineering
Conference, Perth, 25-28 August.
Gok, A., Omastova, M. & Prokes, J. 2007, "Synthesis and characterisation of red
mud/polyaniline composites: Electrical properties and thermal stability",
European Polymer Journal, 43, (6), 2471-2480.
Gong, C. & Yang, N. 2000, "Effect of phosphate on the hydration of alkali-activated
red mud–slag cementitious material", Cement Concrete Res., 30, (7), 1013-1016.
Gräfe, M. & Klauber, C. 2011, "Bauxite residue issues: IV. Old obstacles and new
pathways for in situ residue bioremediation", Hydrometallurgy, 108, (1-2), 46-59.
Gräfe, M., Power, G. & Klauber, C. 2009, Review of Bauxite Residue Alkalinity and
Associated Chemistry, CSIRO Document DMR-3610, Project ATF-06-3:
"Management of Bauxite Residues", Department of Resources, Energy and
Tourism (DRET).
Gräfe, M., Power, G. & Klauber, C. 2011, "Bauxite residue issues: III. Alkalinity and
associated chemistry", Hydrometallurgy, 108, (1-2), 60-79.
Greenberg, S. A. & Chang, T. 1965, "Investigation of colloidal calcium silicates. II-
Solubility relationships in the calcium oxide-silica-water system at 250C", J. Phys.
Chem., 69, (1), 182-188.
Greenberg, S. A., Chang, T. & Anderson, E. 1960, "Investigation of colloidal
hydrated calcium silicates. I-Solubility products", J. Phys. Chem., 64, (9), 1151-
1157.
Guilfoyle, L., Hay, P. S. & Cooling, D. J. 2005, "Use of flue gas for carbonation of
bauxite residue", Proc. 7th Int. Alumina Quality Worhkshop, AQW Ltd., Perth,
Australia, pp. 218-220.
Gupta, V. K., Ali, I. & Saini, V. K. 2004, "Removal of chlorophenols from
wastewater using red mud: an aluminum industry waste", Environ. Sci. Technol.,
38, (14), 4012-4018.
153
Gupta, V. K., Gupta, M. & Sharma, S. 2001, "Process development for the removal
of lead and chromium from aqueous solutions using red mud-an aluminium
industry waste", Water Resources, 35, (5), 1125-1134.
Gupta, V. K. & Sharma, S. 2002, "Removal of Cadmium and Zinc from Aqueous
Solutions Using Red Mud", Environ. Sci. Technol., 36, (16), 3612-3617.
Gupta, V. K. & Suhas 2009, "Application of low-cost adsorbents for dye removal-A
review", J. Environ. Manag., 90, (8), 2313-2342.
György, B. & Tran, M. H. 2008, Dewatering, disposal and utilisation of red mud,
viewed on 25 February 2012, <www.redmud.org/files/banvolgyi040110.pdf>.
Haberko, K., Bucko, M. M., Brzezinska-Miecznik, J., Haberko, M., Mozgawa, W.,
Panz, T., Pyda, A. & Zarebski, J. 2006, "Natural hydroxyapatite-Its behaviour
during heat treatment", J. Eur. Ceram. Soc., 26, (4-5), 537-542.
Haerter, M. & Shefer, B. 1975, "A new red mud disposal system", Proc. 9th Int.
Manifestation of ICSOBA, Dubrovnik, Yogoslavia.
Hanahan, C., McConchie, D., John, P., Creeiman, R., Clark, M. & Stocksiek, M.
2004, "Chemistry of seawater neutralisation of bauxite refinery residues (red
mud)", Environ. Eng. Sci., 21, (2), 125-138.
Ho, G., Mathew, K. & Newman, P. G. 1989, "Leachate quality from gypsum
neutralized red mud applied to sandy soils", Water, Air, and Soil Pollution, 47, (1-
2), 1-18.
Ho, G. E., Newman, P. W. G., Mathew, K. & De Potter, H. 1985, "Neutralisation of
red mud residue with copperas", paper presented at Chemeca 85: Innovation in the
Process and Resource Industry-The 13th Australian Chemical Engineering
Conference, Perth, 25-28 August.
Huijgen, W. J., Witkamp, G. & Comans, R. N. J. 2005, "Mineral CO2 sequestration
by steel slag carbonation", Environ. Sci. Technol., 39, (24), 9676-9682.
ICSD 2012, ICSD for WWW-Inorganic Crystal Structure Database, viewed 5th
August 2014, <http://icsd.iqfr.csic.es/icsd/>.
International Marine Organisation 2005, Report of the twenty-eighth meeting of the
scientific group, Scientific Group Meeting-28th Meeting, IMO, London, UK.
154
IPCC 2005, IPCC Special Report on Carbon Dioxide Capture and Storage.
Prepared by Working Group III of the Intergovernmental Panel on Climate
Change, Metz, B., Davidson, O., de Coninck, H.C., Loos, M. & Meyer, L.A.
(eds.), Cambridge University Press, United Kingdom and New York.
IPCC 2014, Climate Change 2014: Synthesis Report. Contribution of Working
Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel
on Climate Change, R.K. Pachauri, R.K. & Meyer, L.A. (eds.), IPCC, Geneva,
Switzerland.
Johnston, M., Clark, M. W. & McConchie, D. M. 2008, "Geochemistry of bauxite
refinery residues neutralised by different methods", Proc. 8th Int. Alumina Quality
Workshop, AQW Inc., Darwin, Australia, pp. 145-157.
Johnston, M., Clark, M. W., McMahon, P. & Ward, N. 2010, "Alkalinity conversion
of bauxite refinery residues by neutralization", J. Hazard. Mater., 182, (1-3), 710-
715.
Jones, G., Joshi, G., Clark, M. & McConchie, D. 2006, "Carbon capture and the
Aluminium Industry:Preliminary studies", Environ. Chem., 3, (4), 297-303.
Kalkan, E. 2006, "Utilisation of red mud as a stabilisation material for the
preparation of clay liners", Eng. Geol., 87, (3-4), 220-229.
Khaitan, S., Dzombak, D. A. & Lowry, G. V. 2009a, "Chemistry of the acid
neutralisation capacity of bauxite residues", Environ. Eng. Sci., 26, (5), 873-881.
Khaitan, S., Dzombak, D. A. & Lowry, G. V. 2009b, "Mechanisms of neutralisation
of bauxite residue by carbon dioxide", J. Environ. Eng., 135, (6), 433-438.
Khaitan, S., Dzombak, D. A. & Lowry, G. V. 2009, "Neutralization of Bauxite
Residue with Acidic Fly Ash", Environ. Eng. Sci., 26, (2), 431-440.
Kirkpatrick, D. B. 1996, "Red mud production development", in Hale, W. (ed.),
Light Metals, TMS, Anaheim, pp. 75-80.
Kirwan, L. J., Hartshorn, A., McMonagle, J. B., Fleming, L. & Funnell, D. 2013,
"Chemistry of bauxite residue neutralisation and aspects to implementation", Int.
J. Miner. Process., 119, 40-50.
Klauber, C., Gräfe, M. & Power, G. 2009, Review of Bauxite Residue "Re-use"
options, CSIRO Document DMR-3609, Project ATF-06-3: "Management of
Bauxite Residues", Department of Resources, Energy and Tourism (DRET).
155
Klauber, C., Gräfe, M. & Power, G. 2011, "Bauxite residue issues: II. Options for
residue utilization", Hydrometallurgy, 108, (1-2), 11-32.
Kopittke, P. M., Menzies, N. W. & Fulton, I. M. 2004, "Gypsum solubility in
seawater, and its application to bauxite residue amelioration", Aust. J. Soil Res.,
42, (8), 953-960.
Koumanova, B., Drame, M. & Popangelova, M. 1997, "Phosphate removal from
aqueous solutions using red mud wasted in bauxite Bayer's process", Resour.
Conserv. Recy., 19, (1), 11-20.
Kumar, S., Kumar, R. & Bandopadhyay, A. 2006, "Innovative methodologies for the
utilisation of wastes from metallurgical and allied industries", Resour. Conserv.
Recy., 48, (4), 301-314.
Li, Y., Liu, C., Luan, Z., Peng, X., Zhu, C., Chen, Z., Zhang, Z., Fan, J. & Jia, Z.
2006, "Phosphate removal from aqueous solutions using raw and activated red
mud and fly ash", J. Hazard. Mater., 137, (1), 374-383.
Li, Y., Wang, J., Luan, Z. & Liang, Z. 2010, "Arsenic removal from aqueous
solution using ferrous based red mud sludge", J. Hazard. Mater., 177, (1-3), 131-
137.
Liang, W., Couperthwaite, S. J., Gurkiran, K., Cheng, Y., Johnston, D. W. & Millar,
G. J. 2014, "Effect of strong acids on red mud structural and flouride adsorption",
J. Colloid Interface Sci., 423, 158-165.
Lin, C., Maddocks, G., Lin, J., Lancaster, G. & Chu, C. 2004, "Acid neutralising
capacity of two different bauxite residues (red mud) and their potential
applications for treating acid sulfate water and soils", Aust. J. Soil Res., 42, (6),
649-657.
Liu, C., Li, Y., Luan, Z., Chen, Z., Zhang, Z. & Jia, Z. 2007, "Adsorption removal of
phosphate from aqueous solution by active red mud", J. Environ. Sci., 19, (10),
1166-1170.
Liu, W., Yang, J. & Xiao, B. 2009, "Review on treatment and utilization of bauxite
residues in China", Int. J. Miner. Process., 93, (3-4), 220-231.
Liu, Y., Lin, C. & Wu, Y. 2006, "Characterization of red mud derived from a
combined Bayer process and bauxite calcination method", J. Hazard. Mater., 146,
(1-2), 255-261.
156
Liu, Y., Naidu, R. & Ming, H. 2011, "Red mud as an amendment for pollutants in
solid and liquid phases", Geoderma, 163, (1-2), 1-12.
Lombi, E., Zhao, F. J., Wieshammer, G., Zhang, G. & McGrath, S. P. 2002a, "In-situ
fixation of metals in soils using bauxite residue: biological effects", Environ.
Pollut., 118, (3), 445-452.
Lombi, E., Zhao, F. J., Zhang, G., Sun, B., Fitz, W., Zhang, H. & McGrath, S. P.
2002b, "In-situ fixation of metals in soils using bauxite residue: chemical
assessment", Environ. Pollut., 118, (3), 435-443.
Martcheck, K. J. 2003, "Measuring, Reporting and Verifying greenhouse emissions
gas from aluminum reduction operations", Light Metals, 227-231.
Martinet-Catalot, V., Lamerant, J. M., Tilmant, G., Bacou, M. S. & Ambrost, J. P.
2002, "Bauxaline: a new product for various applications of Bayer process red
mud", in Schneider, W. (ed.), Light Metals, TMS, Columbus, pp. 125-131.
McCarthy, S., Armour-Brown, V. S., Iyer, V. S., Desu, S. B., Kander, R. G. &
Vaseashta, A. 1992, "Utilisation of Jamaica bauxite tailings as a building material
and its socio-economic considerations", in Glenister, D.J. (ed.), International
Bauxite Tailings Workshop, Australian Bauxite and Alumina Producers, Perth,
Western Australia.
McConchie, D., Clark, M. & Davies-McConchie, F. 2002, "New strategies for the
management of bauxite refinery residues (red mud)", in Chadrashekar, S. (ed.),
Proc. 6th Int. Alumina Quality Workshop, AQW Incorporation, Brisbane, pp. 327-
332.
Menzies, N. W., Fulton, I. M. & Morrell, W. J. 2004, "Seawater Neutralization of
Alkaline Bauxite Residue and Implications for Revegetation", J Environ Qual, 33,
(5), 1877-1884.
Mesmer, R. E. & Baes, C. F. 1971, "Acidity measurements at elevated temperatures.
V. Aluminum ion hydrolysis", Inorg. Chem., 10, (10), 2290-2296.
Morel, F. & Morgan, J. 1972, "Numerical method for computing equilibriums in
aqueous chemical systems", Environ. Sci. Technol., 6, (1), 58-67.
Namasivayam, C. & Arasi, D. J. S. E. 1997, "Removal of congo red from wastewater
by adsorption onto waste red mud", Chemosphere, 34, (2), 401-417.
157
Namasivayam, C., Yamuna, R. T. & Arasi, D. J. S. E. 2001, "Removal of acid violet
from wastewater by adsorption on waste red mud", Environ. Geol., 41, (3-4), 269-
273.
Navarro, C., Diaz, M. & Villa-Garcia, M. A. 2010, "Physico-chemical
characterisation of steel slag. Study of its behavior under simulated environmental
conditions", Environ. Sci. Technol., 44, (14), 5383-5388.
Newson, T., Dyer, T., Adam, C. & Sharp, S. 2006, "Effect of structure on the
geotechnical properties of bauxite residue", Journal of Geotechnical and
Geoenvironmental Engineering, 132, (2), 143-151.
Nguyen, Q. D. & Boger, D. V. 1998, "Application of rheology to solving tailings
disposal problems", Int. J. Miner. Process., 54, (1-4), 217-233.
Page, A. L., Miller, R. H. & Keeney, D. R. 1982, Methods of Soil Analysis- Part 2:
Chemical and Microbiological Properties, Agronomy, Soil Science Society of
America, Madison, Wisconsin, USA.
Palmer, S. J., Frost, R. L. & Nguyen, T. 2009, "Hydrotalcites and their role in
coordination of anions in Bayer liquors: Anion binding in layered double
hydroxides", Coordination Chemistry Reviews, 253, (1), 250-267.
Palmer, S. J., Nothling, M., Bakon, K. H. & Frost, R. L. 2010, "Thermally activated
seawater neutralised red mud used for the removal of arsenate, vanadate and
molybdate from aqueous solutions", J. Colloid Interface Sci., 342, (1), 147-154.
Pan, S. Y., Chang, E. E. & Chiang, P. C. 2012, "CO2 capture by accelerated
carbonation of alkaline wastes: A review on its principles and applications",
Aerosol and Air Quality Research, 12, (5), 770-791.
Pansu, M. & Gautheyrou, J. 2007, "Organic and Total C, N (H, O, S) Analysis",
Handbook of Soil Analysis, Springer-Verlag, Berlin Heidelberg, New York, pp.
327-370.
Paradis, M., Duchesne, J., Lamontagne, A. & Isabel, D. 2007, "Long-term
neutralisation potential of red mud bauxite with brine amendment for the
neutralisation of acidic mine tailings", Appl. Geochem., 22, (11), 2326-2333.
Paradis, R. D. 1992, "Disposal of red mud using wet stacking technology", Proc. Int.
Bauxite Tailings Workshop, Perth, Western Australia, pp. 179-186.
158
Paramguru, R. K., Rath, P. C. & Misra, V. N. 2005, "Trends in red mud utilisation -
A review", Mineral Processing & Extractive Metallurgy Review, 26, (1), 1-29.
Park, A.-H. A., Jadhav, R. & Fan, L.-S. 2003, "CO2 mineral sequestration:
chemically enhanced aqueous carbonation of Serpentine", The Canadian Journal
of Chemical Engineering, 81, (3-4), 885-890.
Park, S. J. & Jun, B. R. 2005, "Improvement of red mud polymer-matrix
nanocomposites by red mud surface treatment", J. Colloid Interface Sci., 284, (1),
204-209.
Piga, L., Pochetti, F. & Stoppa, L. 1993, "Recovering metals from red mud generated
during alumina production", J. Miner. Metals Mater. Soc., 45, (11), 54-59.
Pinnock, W. R. 1991, "Measurements of radioactivity in Jamaican building materials
and gamma dose equivalents in a prototype red mud house", Health physics, 61,
(5), 647-651.
Pinnock, W. R. 1999, "Radon levels and related doses in a prototype Jamaican house
constructed with bauxite waste blocks", Radiation Protection Dosimetry, 81, (4),
291-299.
Pontikes, Y., Nikolopoulos, P. & Angelopoulos, G. N. 2007, "Thermal behaviour of
clay mixtures with bauxite residue for the production of heavy-clay ceramics", J.
Eur. Ceram. Soc., 27, (2-3), 1645-1649.
Power, G., Gräfe, M. & Klauber, C. 2009, Review of Current Bauxite Residue
Management, Disposal and Storage: Practices, Engineering and Science, Asia-
Pacific Partnership Project ATF-06-3: "Management of Bauxite Residues",
National Research Flagshps Light Metal-CSIROs.
Power, G., Gräfe, M. & Klauber, C. 2011, "Bauxite residue issues: I. Current
management, disposal and storage practices", Hydrometallurgy, 108, (1-2), 33-45.
Pradhan, J., Das, J., Das, S. & Thakur, R. S. 1998, "Adsorption of Phosphate from
Aqueous Solution Using Activated Red Mud", J. Colloid Interface Sci., 204, (1),
169-172.
Purnell, B. G. 1986, "Mud disposal at the Burntisland alumina plant", Light Metals,
pp. 157-159.
Rai, S., Lataye, D. H., Chaddha, M. J., Mishra, R. S., Mahendiran, P.,
Mukhopadhyay, J., Yoo, C. & Wasewar, K. L. 2013, "An Alternative to Clay in
159
Building Materials: Red Mud Sintering Using Fly Ash via Taguchi’s
Methodology", Advances in Materials Science and Engineering, 2013, 1-7.
Rai, S. B. 2013, "Sequestration of carbon dioxide in red mud", Desalination and
Water Treatment, 51, (10-12), 2185-2192.
Renforth, P., Mayes, W. M., Jarvis, A. P., Burke, I. T., Manning, D. A. & Gruiz, K.
2012, "Contaminant mobility and carbon sequestration downstream of the Ajka
(Hungary) red mud spill: The effects of gypsum dosing", Sci. Total Environ., 421-
422, (1), 253-259.
Rounds, S. A. 2012, "Alkalinity and acid neutralising capacity (version 4.0)", in
Wilde, F.D. & Radtke, D.B. (eds.), Natural field manual for the collection of
water - quality data, U.S. Geological Survey Techniques of Water-Resources
Investigations, Book 9, Chapter A6, Section 6.6, p. 45.
Sahu, R. C. 2011, "Neutralisation of red mud using CO2 sequestration and their
utilisations", PhD thesis, National Institute of Technology, Orissa, India.
Sahu, R. C., Patel, R. K. & Ray, B. C. 2010, "Neutralization of red mud using CO2
sequestration cycle", J. Hazard. Mater., 179, (1-3), 28-34.
Santini, T. & Fey, M. 2015, "Fly ash as a permeable cap for tailings management:
pedogenesis in bauxite residue tailings", Journal of Soils and Sediments, 15, (3),
552-564.
Schecher, W. D. & McAvoy, D. C. 1992, "MINEQL+: A software environment for
chemical equilibrium modeling", Computers, Environment and Urban Systems,
16, (1), 65-76.
Schecher, W. D. & McAvoy, D. C. 2015, MINEQL+ A chemical Equilirium
Modeling System: Version 5.0 for Windows, User's Manual, Environmental
Research Software, Hallowell, Maine.
Shah, R. P. & Gararia, S. N. 1995, "Upgradation of alumina refinery at Hindalco,
Renukoot (India)", in Ivan, J. (ed.), Light Metals, TMS, Las Vegas, pp. 25-29.
Shannon, E. E. & Verghese, K. I. 1976, "Utilization of Alumized Red Mud Solids for
Phosphorus Removal", J. Water Pollut. Control Fed., 48, (8), 1948-1954.
Shi, C., Xu, J., Beckman, E. J. & Enick, R. M. 2000, "Carbon dioxide sequestration
via pH reduction of red mud using liquid CO2", ACS Division of Fuel Chemistry,
45, (4), 703-705.
160
Shiao, S. J. & Akashi, K. 1977, "Phosphate Removal from Aqueous Solution from
Activated Red Mud", J. Water Pollut. Control Fed., 49, (2), 280-285.
Singh, M., Upadhayay, S. N. & Prasad, P. M. 1996, "Preparation of special cements
from red mud", Waste Manag., 16, (8), 665-670.
Singh, M., Upadhayay, S. N. & Prasad, P. M. 1997, "Preparation of iron rich cements
using red mud", Cement Concrete Res., 27, (7), 1037-1046.
Sirbescu, M. & Jenkins, D. M. 1999, "Experiments on the stability of Cancrinite in
the system Na2O-CaO-Al2O3-SiO2-CO2-H2O", American Mineralogist, 84, (11-
12), 1850-1860.
Smith, P. G., Pennifold, R. M., Davies, M. G. & Jamieson, E. J. 2003, "Reactions of
carbon dioxide with Tri-calcium aluminate", in Young, C.A., Alfantazi, A.M.,
Anderson, C.G., Dreisinger, D.B., Harris, B. & James, A. (eds.), Hydrometallurgy
2003: Proc. 5th Int. Symp. Honoring Professor Ian M. Ritchie, Vancouver,
Cananda, vol. 2, pp. 1705-1715.
Snars, K. E., Gilkes, R. J. & Wong, M. T. F. 2004, "The liming effects of bauxite
processing residue (red mud) on sandy soils", Aust. J. Soil Res., 42, (3), 321-328.
Sofrá, F. & Boger, D. V. 2002, "Environmental rheology for waste minimisation in
the minerals industry", Chem. Eng. J., 86, (3), 319-330.
Soner Altundogan, H., Altundogan, S., Tumen, F. & Bildik, M. 2000, "Arsenic
removal from aqueous by adsorption on red mud", Waste Manag., 20, (8), 761-
767.
Sparks, T. 2010, "Alumina: Filtration in the alumina production process", Filtration
and Separation, 47, (3), 20-23.
Sposito, G. 1989, The Environmental Chemistry of Aluminum, Boca Raton, Fla: CRC
Press.
Stumm, W. & Morgan, J. J. 1981, Aquatic Chemistry: An introduction emphasizing
chemical equilibria in natural water, John Wiley & Sons Inc., USA.
Summers, R., Guise, N., Smirk, D. & Summers, K. 1996a, "Bauxite residue (red
mud) improves pasture growth on sandy soils in Western Australia", Soil
Research, 34, (4), 569-581.
161
Summers, R., Smirk, D. & Karafilis, D. 1996b, "Phosphorus retention and leachates
from sandy soil amended with bauxite residue (red mud)", Soil Research, 34, (4),
555-567.
Summers, R. N., Bolland, M. D. A. & Clarke, M. F. 2001, "Effect of application of
bauxite residue (red mud) to very sandy soils on subterranean clover yield and P
response", Aust. J. Soil Res., 39, (5), 979-990.
Summers, R. N., Guise, N. R. & Smirk, D. D. 1993, "Bauxite residue (red mud)
increases phosphorus retention in sandy soil catchments in Western Australia",
Fertiliser Research, 34, (1), 85-94.
Summers, R. N. & Pech, J. D. 1997, "Nutrient and metal content of water, sediment
and soils amended with bauxite residue in the catchment of the Peel Inlet and
Harvey Estuary, Western Australia", Agriculture, Ecosystems & Environment, 64,
(3), 219-232.
Summers, R. N., Van Gool, D., Guise, N. R., Heady, G. J. & Allen, T. 1999, "The
phosphorus content in the run-off from the coastal catchment of the Peel Inlet and
Harvey Estuary and its associations with land characteristics", Agriculture,
Ecosystems & Environment, 73, (3), 271-279.
Sushil, S., Alabdulrahman, A. M., Balakrishnan, M., Batra, V. S., Blackley, R. A.,
Clapp, J., Hargreaves, J. S., Monaghan, A., Pulford, I. D., Rico, J. L. & Zhou, W.
2010, "Carbon deposition and phase transformations in red mud on exposure to
methane", J. Hazard. Mater., 180, (1-3), 409-418.
Sushil, S. & Batra, V. S. 2008, "Catalytic applications of red mud, an aluminium
industry waste: A review", Applied Catalysis B: Environmental, 81, (1-2), 64-77.
Sushil, S. & Batra, V. S. 2012, "Modification of red mud by acid treatment and its
application for CO removal", J. Hazard. Mater., 203-204,, 264-273.
Sutar, H., Mishra, S. C., Sahoo, S. K., Chakraverty, A. P. & Maharana, H. S. 2014,
"Progress of Red Mud Utilization: An Overview", American Chemical Science
Journal, 4, (3), 255-279.
Szirmai, E., Babusek, S., Balogh, G., Nedves, A., Horvath, G., Lebenyi, Z. & Pinter,
J. 1991, Method for the mutistage , waste-free processing of red mud to recover
basic materials of chemical industry, US Patent, 5053144.
162
Thakur, R. S. & Sant, B. R. 1983a, "Utilisation of red mud. 1-Analysis and utilisation
as raw material for absorbents, building-materials, catalysts, filler, paints and
pigments", J. Sci. Ind. Res., 42, (2), 87-108.
Thakur, R. S. & Sant, B. R. 1983b, "Utilisation of red mud. 2-Recovery of alkali,
iron, aluminum, titanium, and other constituents and the pollution problems", J.
Sci. Ind. Res., 42, (8), 456-469.
Theis, T. L. & Wirth, J. L. 1977, "Sorptive behavior of trace metals on fly ash in
aqueous systems", Environ. Sci. Technol., 11, (12), 1096-1100.
Thermo Electron Corporation 2004, Nicolet FT-IR User's guide, Madison WI 53711,
USA.
Thornber, M. R. & Hughes, C. A. 1986, "The mineralogical and chemical properties
of red mud waste from the Western Australian alumina industry", Proceedings of
the International Conference on Bauxite Residue Tailings, Kingston, Jamaica, pp.
1-18.
Tsakiridis, P. E., Agatzini-Leonardou, S. & Oustadakis, P. 2004, "Red mud addition
in the raw meal for the production of Portland cement clinker", J. Hazard. Mater.
B, 116, (1-2), 103-110.
UNIDO 1985, Environmental Aspects of Alumina Production, UNEP, Nairobi.
USGS 2007, Alkalinity Calculation Methods, Oregon Water Science Centre, viewed
5 June 2012, <http://or.water.usgs.gov/alk/methods.html>.
Vieillard, P. & Rassineux, F. 1992, "Thermodynamic and geochemical modelling of
the alteration of two cement matrices", Appl. Geochem., 7, Supplement 1, 125-
136.
Wang, Q., Luan, Z., Wei, N., Li, J. & Liu, C. 2009, "The color removal of dye
wastewater by magnesium chloride/red mud (MRM) from aqueous solution", J.
Hazard. Mater., 170, (2-3), 690-698.
Wang, S., Ang, H. M. & Tade, M. O. 2008, "Novel applications of red mud as
coagulant, adsorbent and catalyst for environmentally benign processes",
Chemosphere, 72, (11), 1621-1635.
Wannenmacher, N., Kimura, S. & Frederick, J. W. 2005, "Solubility of
Aluminosilicates in Kraft Green and White Liquors", Ind. Eng. Chem. Res., 44,
(21), 9316-9322.
163
Weber, C. F. 2001, Phase equilibrium studies of Savannah River tanks and feed
streams for the salt waste processing facility, ORNL/TM-2001-109 Oak Ridge
National Laboratory Report, Computational Physics anf Engineering Division 1-
62.
Whittington, B. I. 1996, "The chemistry of CaO and Ca(OH)2 relating to the Bayer
process", Hydrometallurgy, 43, (1–3), 13-35.
Willis, R. D., Blanchard, F. T., Conner, T. L. & Morris, C. 2002, Guidelines for the
application of SEM/EDX analytical techniques to particulate matter samples,
National Exposure Research Laboratory, U.S. Environmental Protection Agency,
NC 27711.
WISE 2015, Chronology of major tailings dam failures from 1960, World
Information Service on Energy Uranium Project (WISE), viewed 25th December
2015, <http://www.wise-uranium.org/mdaf.html>.
Wong, J. W. C. & Ho, G. E. 1993, "Use Of Waste Gypsum In The Revegetation On
Red Mud Deposits- A Greenhouse Study", Waste Manag. Res., 11, (3), 249-256.
Wong, J. W. C. & Ho, G. E. 1994, "Effectiveness of acidic industrial waste for
reclaiming fine bauxite refining residue (red mud)", Soil Science, 158, (2), 115-
123.
Woodard, H. J., Hossner, L. & Bush, J. 2008, "Ameliorating caustic properties of
aluminum extraction residue to establish a vegetative cover", Journal of
Environmental Science and Health, Part A, 43, (10), 1157-1166.
Yadav, V. S., Prasad, M., Khan, J., Amritphale, S. S., Singh, M. & Raju, C. B. 2009,
"Sequestration of carbon dioxide (CO2) using red mud", J. Hazard. Mater., 176,
(1-3), 1044-1050.
Yalcin, N. & Sevinc, V. 2000, "Utilization of bauxite waste in ceramic glazes",
Ceram. Int., 26, (5), 485-493.
Yang, J. & Xiao, B. 2008, "Development of unsintered construction materials from
red mud wastes produced in the sintering alumina process", Constr. Buil. Mater.,
22, (12), 2299-2307.
Zhang, S., Liu, C., Luan, Z., Peng, X., Ren, H. & Wang, J. 2008, "Arsenate removal
from aqueous solutions using modified red mud", J. Hazard. Mater., 152, (2),
486-492.
164
Zhang, Y., Qu, Y. & Wu, S. 2001, "Engineering geological properties and
Comprehensive utilisation of solid waste in aluminium industry", Environ. Geol.,
41, (3-4), 249-256.
Zhang, Y. N. & Pan, Z. H. 2005, "Characterisation of red mud thermally treated at
different temperature", Journal of Jinan University (Science and Technology), 19,
(4), 293-297.
Zhao, H., Deng, Y., Harsh, J. B., Flury, M. & Boyle, J. S. 2004, "Alteration of
Kaolinite to Cancrinite and Sodalite by simulated hanford tank waste and its
impact on cesium retention", Clays and Clay Minerals, 52, (1), 1-13.
Zhu, C., Luan, Z., Wang, Y. & Shan, X. 2007, "Removal of cadmium from aqueous
solutions by adsorption on granular red mud (GRM)", Separation and Purification
Technology, 57, (1), 161-169.
165
APPENDIX
Table A-1. Rapid titration of RM by 0.1N HCl
Volume of 0.1N
HCl (mL)
Meq/g
RM(*)
Meq/g
PW(**)
pH
Rep 1
pH
Rep 2
pH
Average
Std Dev
pH
0
4
8
12
15
20
25
28
31
34
37
40
43
46
50
54
57
60
64
68
72
74
76
78
79
0.00
0.04
0.08
0.12
0.15
0.20
0.25
0.28
0.31
0.34
0.37
0.40
0.43
0.46
0.50
0.54
0.57
0.60
0.64
0.68
0.72
0.74
0.76
0.78
0.79
0.00
0.07
0.14
0.21
0.27
0.36
0.45
0.50
0.55
0.61
0.66
0.71
0.77
0.82
0.89
0.96
1.02
1.07
1.14
1.21
1.29
1.32
1.36
1.39
1.41
12.42
11.6
10.44
9.55
8.86
7.83
7.23
7.01
6.74
6.56
6.38
6.08
5.98
5.74
5.49
5.38
5.25
5.1
5
4.8
4.7
4.63
4.6
4.55
4.48
12.49
11.66
10.53
9.71
9.04
8.03
7.4
7.09
6.85
6.62
6.42
6.3
6.1
5.85
5.6
5.36
5.29
5.19
5.05
4.91
4.78
4.72
4.67
4.6
4.54
12.46
11.63
10.49
9.63
8.95
7.93
7.32
7.05
6.80
6.59
6.40
6.19
6.04
5.80
5.55
5.37
5.27
5.15
5.03
4.86
4.74
4.68
4.64
4.58
4.51
0.05
0.04
0.06
0.11
0.13
0.14
0.12
0.06
0.08
0.04
0.03
0.16
0.08
0.08
0.08
0.01
0.03
0.06
0.04
0.08
0.06
0.06
0.05
0.04
0.04
(*) Mass of RM used: 10g (**) Normalised to RM liquor basis, divide by 0.56 as RM contains 44% solids.
166
Table A-2. Rapid titration of RM liquor to pH 4.5 by 0.1N HCl
Volume of 0.1N
HCl (mL)
Meq/g
PW(*)
pH(**)
Rep 1
pH(**)
Rep 2
pH
Average
Std Dev
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
17.5
18
18.5
19
19.5
20.0
20.5
21.0
21.5
21.7
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.18
0.19
0.19
0.20
0.20
0.21
0.21
0.22
0.22
12.42
12.28
12.21
12.14
12.05
11.95
11.81
11.62
11.32
10.73
10.15
9.96
9.83
9.74
9.6
9.35
8.97
7.85
6.85
5.89
5.8
5.57
5.32
5.2
5.12
4.62
4.55
4.38
12.46
12.35
12.27
12.18
12.09
11.97
11.83
11.63
11.32
10.71
10.16
10.02
9.92
9.79
9.68
9.47
9.13
8.31
7.3
6.43
6.2
5.8
5.71
5.56
5.27
4.97
4.72
4.43
12.44
12.32
12.24
12.16
12.07
11.96
11.82
11.63
11.32
10.72
10.16
9.99
9.88
9.77
9.64
9.41
9.05
8.08
7.08
6.16
6.00
5.69
5.52
5.38
5.20
4.80
4.64
4.41
0.03
0.05
0.04
0.03
0.03
0.01
0.01
0.01
0.00
0.01
0.01
0.04
0.06
0.04
0.06
0.08
0.11
0.33
0.32
0.38
0.28
0.16
0.28
0.25
0.11
0.25
0.12
0.04
(*) Mass of RM liquor used: 10g
(**) Titration to pH 4.5
167
Table A-3. Long-term titration of RM to pH 4.5 by 0.1N HCl
Time
(day)
Acid
added(*)
0.1N HCL
mL, Rep 1
Acid added
meq/g
RM(**) Rep
1
Acid
added(*)
0.1N HCL
mL, Rep 2
Acid added
meq/g
RM(**) Rep 2
Acid added
Average
(Rep1,
Rep2)
Acid
added
Std
Dev
0
1
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
45
90.5
116.0
137.0
160.5
171.5
179.1
182.8
184.9
186.2
187.0
187.5
188.1
188.7
189.3
189.9
190.6
191.2
191.5
191.7
191.9
192.1
192.3
192.4
192.4
192.4
0.91
1.16
1.37
1.61
1.72
1.79
1.83
1.85
1.86
1.87
1.87
1.88
1.89
1.89
1.90
1.91
1.91
1.92
1.92
1.92
1.92
1.92
1.92
1.92
1.92
87.0
107.5
124.0
143.4
155.5
162.9
168.7
173.2
176.6
179.0
181.1
182.6
183.8
184.7
185.7
186.2
186.9
187.4
187.7
187.9
188.1
188.3
188.5
188.7
188.8
0.87
1.08
1.24
1.43
1.56
1.63
1.69
1.73
1.77
1.79
1.81
1.83
1.84
1.85
1.86
1.86
1.87
1.87
1.88
1.88
1.88
1.88
1.89
1.89
1.89
0.89
1.12
1.31
1.52
1.64
1.71
1.76
1.79
1.81
1.83
1.84
1.85
1.86
1.87
1.88
1.88
1.89
1.89
1.90
1.90
1.90
1.90
1.91
1.91
1.91
0.02
0.06
0.09
0.12
0.11
0.11
0.10
0.08
0.07
0.06
0.05
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
(*) pH was kept at 4.5 after each titration. (**) Mass of RM used: 10g
168
Table A-4. Long-term titration of RM to pH 6.0 by 0.1N HCl
Time
(day)
Acid
added(*)
0.1N HCL
mL, Rep 1
Acid added
meq/g
RM(**) Rep
1
Acid
added(*)
0.1N HCL
mL, Rep 2
Acid added
meq/g
RM(**) Rep 2
Acid added
Average
(Rep1,
Rep2)
Acid
added
Std
Dev
0
1
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
45
49.5
60.2
66.7
72.1
74.7
77.0
79.3
81.9
84.1
86.1
87.4
88.4
89.6
90.8
92.1
93.3
94.4
95.6
96.6
97.7
98.7
99.8
100.9
101.6
101.8
0.50
0.60
0.67
0.72
0.75
0.77
0.79
0.82
0.84
0.86
0.87
0.88
0.90
0.91
0.92
0.93
0.94
0.96
0.97
0.98
0.99
1.00
1.01
1.02
1.02
50.2
61.4
68.2
74.2
77.1
79.7
82.2
84.6
87.0
88.9
90.3
91.4
92.6
93.8
94.9
95.9
97.1
98.2
99.2
100.2
101.2
102.0
103.0
104.0
104.5
0.50
0.61
0.68
0.74
0.77
0.80
0.82
0.85
0.87
0.89
0.90
0.91
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
1.02
1.03
1.04
1.04
0.50
0.61
0.67
0.73
0.76
0.78
0.81
0.83
0.86
0.87
0.89
0.90
0.91
0.92
0.93
0.95
0.96
0.97
0.98
0.99
1.00
1.01
1.02
1.03
1.03
0.00
0.01
0.01
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.02
0.02
(*) pH was kept at 6.0 after each titration. (**) Mass of RM used: 10g
169
Table A-5. Long-term titration of RM to pH 8.0 by 0.1N HCl
Time
(day)
Acid
added(*)
0.1N HCL
mL, Rep 1
Acid added
meq/g
RM(**) Rep
1
Acid
added(*)
0.1N HCL
mL, Rep 2
Acid added
meq/g
RM(**) Rep
2
Acid added
Average
(Rep1,
Rep2)
Acid
added
Std
Dev
0
1
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
45
24.7
31.2
34.7
39.6
42.4
44.8
47.2
48.7
50.5
52.1
53.6
54.8
56.2
57.4
58.4
59.7
61.0
62.1
63.3
64.7
65.8
66.7
67.3
67.9
68.2
0.25
0.31
0.35
0.40
0.42
0.45
0.47
0.49
0.51
0.52
0.54
0.55
0.56
0.57
0.58
0.60
0.61
0.62
0.63
0.65
0.66
0.67
0.67
0.68
0.68
25.3
32.1
35.8
41.3
44.6
47.4
50.2
51.9
53.6
55.5
57.0
58.5
60.1
61.3
62.4
63.4
64.4
65.2
66.2
67.3
68.3
69.0
69.6
70.2
70.5
0.25
0.32
0.36
0.41
0.45
0.47
0.50
0.52
0.54
0.56
0.57
0.59
0.60
0.61
0.62
0.63
0.64
0.65
0.66
0.67
0.68
0.69
0.70
0.70
0.71
0.25
0.32
0.35
0.40
0.44
0.46
0.49
0.50
0.52
0.54
0.55
0.57
0.58
0.59
0.60
0.62
0.63
0.64
0.65
0.66
0.67
0.68
0.68
0.69
0.69
0.00
0.01
0.01
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.03
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
(*) pH was kept at 8.0 after each titration. (**) Mass of RM used: 10g
170
Table A-6. Long-term titration of RM to pH 10 by 0.1N HCl
Time
(day)
Acid
added(*)
0.1N HCL
mL, Rep 1
Acid added
meq/g
RM(**) Rep
1
Acid
added(*)
0.1N HCL
mL, Rep 1
Acid added
meq/g
RM(**) Rep
1
pH
Average
Std
Dev
pH
0
1
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
45
11.2
13.9
15.3
17.0
18.4
19.8
21.0
22.3
23.8
25.1
26.1
27.3
28.8
29.8
30.9
32.0
32.9
33.6
34.3
35.0
35.7
36.3
36.7
37.1
37.3
0.11
0.14
0.15
0.17
0.18
0.20
0.21
0.22
0.24
0.25
0.26
0.27
0.29
0.30
0.31
0.32
0.33
0.34
0.34
0.35
0.36
0.36
0.37
0.37
0.37
9.2
11.6
13.2
14.5
15.6
17.0
18.1
19.1
20.3
21.7
22.7
23.5
24.5
25.5
26.6
27.7
28.6
29.2
29.7
30.2
30.6
30.9
31.1
31.3
31.4
0.09
0.12
0.13
0.15
0.16
0.17
0.18
0.19
0.20
0.22
0.23
0.24
0.25
0.26
0.27
0.28
0.29
0.29
0.30
0.30
0.31
0.31
0.31
0.31
0.31
0.10
0.13
0.14
0.16
0.17
0.18
0.20
0.21
0.22
0.23
0.24
0.25
0.27
0.28
0.29
0.30
0.31
0.31
0.32
0.33
0.33
0.34
0.34
0.34
0.34
0.01
0.02
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.04
0.04
0.04
0.04
0.04
(*) pH was kept at 10 after each titration. (**) Mass of RM used: 10g
171
Table A-7. Metal concentrations in RM liquor measured at different pH values
Component
Concentration (M)
Initial RM
liquor pH 10 pH 8 pH 6 pH 4.5
Al
Na
Ca
Fe
0.076
0.198
0
0
0
0.229
0.0015
0
0
0.262
0.032
0
0
0.280
0.035
0
0
0.487
0.104
0
Table A-8. Simulated metal concentrations in RM liquor at different pH values
Component
Concentration (M)
Initial RM
liquor pH 10 pH 8 pH 6 pH 4.5
Al
Na
Ca
Fe
0.052
0.23
0
0
0
0.24
0.0055
0
0
0.32
0.035
0
0
0.36
0.04
0
0
0.53
0.17
0
172
Figure B-1. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
100mL/min and stirring speed of 250rpm
Figure B-2. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
300mL/min and stirring speed of 250rpm
Figure B-3. Carbonation of RM slurry at different CO2 concentrations, fixed TF of
400mL/min and stirring speed of 250rpm
173
Figure B-4. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
100mL/min and stirring speed of 250rpm
Figure B-5. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
300mL/min and stirring speed of 250rpm
Figure B-6. Carbonation of RM liquor at different CO2 concentrations, fixed TF of
400mL/min and stirring speed of 250rpm
174
Figure B-7. Carbonation of RM by 25% CO2 at different TF of gas
Figure B-8. Carbonation of RM by 40% CO2 at different TF of gas
Figure B-9. Carbonation of RM by 50% CO2 at different TF of gas
175
Figure B-10. Carbonation of red mud by 30% of CO2, TF of 100mL/min at different
stirring speeds
Figure B-11. Carbonation of red mud by 30% of CO2, TF of 300mL/min at different
stirring speeds
Figure B-12. Carbonation of red mud by 30% of CO2, TF of 400mL/min at different
stirring speeds
176
Figure B-13. Carbonation of red mud by 30% of CO2, TF of 100mL/min, stirring
speed 250rpm at different solid concentrations of RM
Figure B-14. Carbonation of red mud by 30% of CO2, TF of 300mL/min, stirring
speed 250rpm at different solid concentrations of RM
Figure B-15. Carbonation of red mud by 30% of CO2, TF of 400mL/min, stirring
speed 250rpm at different solid concentrations of RM
177
Figure B-16. Carbonation of red mud by 40% of CO2, TF of 200mL/min, stirring
speed 250rpm at different solid concentrations of RM
Figure B-17. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 350rpm at different solid concentrations of RM
Figure B-18. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 500rpm at different solid concentrations of RM
178
Figure B-19. Carbonation of red mud by 30% of CO2, TF of 200mL/min, stirring
speed 700rpm at different solid concentrations of RM
Figure B-20. pH rebound for both RM slurry and liquor at some CO2 concentrations,
TF of 200mL/min, stirring speed of 250rpm
179
Table B-1. Carbonation of RM at different CO2 concentrations and total gas flow rate of 100mL/min, stirring speed of 250rpm
Time
(min)
25% CO2 30% CO2 40% CO2 50% CO2 60% CO2
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.4
11.77
10.88
10.3
10.01
9.37
9.42
9.07
8.70
8.44
8.21
8.05
7.97
7.89
7.83
7.77
7.73
7.69
7.65
7.63
7.59
7.57
7.55
7.53
7.51
12.59
12.02
10.81
10.24
9.95
9.52
9.01
8.48
8.15
7.97
7.83
7.77
7.69
7.63
7.61
7.57
7.55
7.51
7.49
7.47
7.45
7.45
7.45
7.45
7.45
12.50
11.90
10.85
10.27
9.98
9.45
9.22
8.78
8.43
8.21
8.02
7.91
7.83
7.76
7.72
7.67
7.64
7.60
7.57
7.55
7.52
7.51
7.50
7.49
7.48
0.13
0.18
0.05
0.04
0.04
0.11
0.29
0.42
0.39
0.33
0.27
0.20
0.20
0.18
0.16
0.14
0.13
0.13
0.11
0.11
0.10
0.08
0.07
0.06
0.04
12.42
11.76
10.65
9.94
9.45
9.04
8.57
8.27
7.91
7.66
7.52
7.43
7.38
7.33
7.27
7.24
7.21
7.21
7.20
7.20
7.20
7.20
7.20
7.20
7.20
12.28
11.62
10.45
10.22
10.07
9.64
8.99
8.25
7.87
7.6
7.46
7.35
7.28
7.25
7.23
7.22
7.21
7.21
7.22
7.22
7.22
7.22
7.22
7.22
7.22
12.35
11.69
10.55
10.08
9.76
9.34
8.78
8.26
7.89
7.63
7.49
7.39
7.33
7.29
7.25
7.23
7.21
7.21
7.21
7.21
7.21
7.21
7.21
7.21
7.21
0.10
0.10
0.14
0.20
0.44
0.42
0.30
0.01
0.03
0.04
0.04
0.06
0.07
0.06
0.03
0.01
0.00
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.32
10.6
9.68
9.01
8.21
7.51
7.31
7.25
7.15
7.09
7.06
7.04
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
12.42
11.00
10.21
9.61
8.57
8.07
7.63
7.37
7.27
7.21
7.18
7.16
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
7.10
12.37
10.8
9.95
9.31
8.39
7.79
7.47
7.31
7.21
7.15
7.12
7.1
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
0.07
0.28
0.37
0.42
0.25
0.40
0.23
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
12.58
10.52
9.38
8.06
7.45
7.17
7.07
6.99
6.95
6.93
6.93
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
12.51
10.76
9.69
8.40
7.64
7.33
7.17
7.09
7.03
6.99
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
12.55
10.64
9.54
8.23
7.55
7.25
7.12
7.04
6.99
6.96
6.95
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
0.05
0.17
0.22
0.24
0.13
0.11
0.07
0.07
0.06
0.04
0.03
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.5
10.31
8.74
7.60
7.20
7.14
6.97
6.94
6.92
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
12.34
10.17
8.62
7.50
7.10
6.84
6.89
6.86
6.84
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
12.42
10.24
8.68
7.55
7.15
6.99
6.93
6.9
6.88
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
0.11
0.10
0.08
0.07
0.07
0.21
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06 (*) Mass of red mud used: 100g
180
Table B-2. Carbonation of RM at different CO2 concentrations and total gas flow
rate of 100mL/min, stirring speed of 250rpm
Time
(min)
75% CO2 100% CO2
pH(*)
Rep1
pH(*)
Rep2 Avg pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2 Avg pH
Stdev
pH
0
2.5
5
7.5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.64
11.27
10.1
9.17
8.06
7.2
6.93
6.85
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
6.79
12.65
11.39
10.2
9.36
8.23
7.3
7.01
6.89
6.85
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
12.65
11.33
10.15
9.27
8.15
7.25
6.97
6.87
6.82
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
0.01
0.08
0.07
0.13
0.12
0.07
0.06
0.03
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
12.52
10.85
9.28
8.22
7.35
7.05
6.85
6.76
6.69
6.66
6.64
6.62
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
12.40
10.75
9.24
8.10
7.30
7.02
6.79
6.70
6.67
6.65
6.63
6.61
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
12.45
10.80
9.25
8.15
7.32
7.03
6.82
6.72
6.68
6.66
6.64
6.62
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
0.08
0.07
0.03
0.08
0.04
0.02
0.04
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
(*) Mass of red mud used: 100g
181
Table B-3. Carbonation of RM at different CO2 concentrations and total gas flow rate of 200mL/min, stirring speed of 250rpm
Time
(min)
15% CO2 20% CO2 25% CO2 30% CO2 40% CO2
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.42
11.69
10.65
10.20
9.82
9.54
9.13
8.64
8.26
8.02
7.85
7.75
7.68
7.62
7.60
7.56
7.54
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
12.30
11.60
10.55
10.11
9.74
9.46
9.04
8.58
8.21
7.94
7.80
7.69
7.62
7.56
7.54
7.50
7.48
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
12.36
11.65
10.60
10.16
9.78
9.50
9.09
8.61
8.24
7.98
7.83
7.72
7.65
7.59
7.57
7.53
7.51
7.49
7.49
7.49
7.49
7.49
7.49
7.49
7.49
0.08
0.06
0.07
0.06
0.06
0.06
0.06
0.04
0.04
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.3
11.41
10.36
9.77
9.35
8.84
8.28
8.02
7.70
7.55
7.47
7.45
7.44
7.38
7.37
7.35
7.33
7.32
7.3
7.3
7.28
7.28
7.28
7.28
7.28
12.50
11.60
10.61
10.14
9.67
9.20
8.60
8.20
8.00
7.80
7.61
7.57
7.50
7.48
7.46
7.44
7.42
7.40
7.38
7.38
7.38
7.38
7.38
7.38
7.38
12.40
11.51
10.49
9.96
9.51
9.02
8.44
8.11
7.85
7.68
7.54
7.51
7.47
7.43
7.42
7.40
7.38
7.36
7.34
7.34
7.33
7.33
7.33
7.33
7.33
0.14
0.13
0.18
0.26
0.23
0.25
0.23
0.13
0.21
0.18
0.10
0.08
0.04
0.07
0.06
0.06
0.06
0.06
0.06
0.06
0.07
0.07
0.07
0.07
0.07
12.57
11.46
10.47
9.98
9.44
8.81
8.32
7.93
7.69
7.55
7.45
7.38
7.33
7.3
7.28
7.26
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
12.46
11.01
10.10
9.47
8.75
8.17
7.75
7.55
7.43
7.35
7.29
7.25
7.23
7.21
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.52
11.24
10.29
9.73
9.10
8.49
8.04
7.74
7.56
7.45
7.37
7.32
7.28
7.26
7.24
7.23
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
0.08
0.32
0.26
0.36
0.49
0.45
0.40
0.27
0.18
0.14
0.11
0.09
0.07
0.06
0.06
0.05
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.34
10.81
9.91
9.11
8.20
7.70
7.52
7.40
7.32
7.25
7.22
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.43
10.60
9.73
8.90
7.92
7.50
7.39
7.25
7.21
7.15
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
12.39
10.71
9.82
9.01
8.06
7.60
7.46
7.33
7.27
7.20
7.18
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
0.06
0.15
0.13
0.15
0.20
0.14
0.09
0.11
0.08
0.07
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.35
10.43
9.12
7.82
7.30
7.13
7.06
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
12.48
10.26
8.95
7.67
7.15
7.00
6.95
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
12.42
10.35
9.04
7.75
7.23
7.07
7.01
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
6.98
0.09
0.12
0.12
0.11
0.11
0.09
0.08
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09
0.09 (*) Mass of red mud used: 100g
182
Table B-4. Carbonation of RM at different CO2 concentrations and total gas flow rate of 200mL/min, stirring speed of 250rpm
Time
(min)
50% CO2 60% CO2 75% CO2 100% CO2 pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
20
25
30
35
40
45
50
60
70
80
90
100
110
120
12.48
11.35
10.68
9.7
8.8
8.01
7.5
7.24
7.13
7.11
7.07
7.05
7.03
7.03
7.03
7.03
7.03
7.03
7.03
7.03
7.03
12.4
11.20
10.00
9.54
8.70
7.48
6.86
6.80
6.78
6.76
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
12.44
11.28
10.34
9.62
8.75
7.75
7.18
7.02
6.96
6.94
6.91
6.90
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
0.06
0.11
0.48
0.11
0.07
0.37
0.45
0.45
0.31
0.25
0.25
0.23
0.22
0.21
0.21
0.21
0.21
0.21
0.21
0.21
0.21
12.45
10.83
9.81
8.85
7.67
7.15
6.95
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
12.40
10.92
9.82
8.90
7.68
7.18
7.00
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
12.43
10.88
9.82
8.88
7.68
7.17
6.98
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
0.04
0.06
0.01
0.04
0.01
0.02
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.42
10.37
8.98
7.93
6.95
6.62
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
12.43
10.29
8.72
7.87
6.89
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
12.43
10.33
8.85
7.90
6.92
6.59
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
0.01
0.06
0.18
0.04
0.04
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
12.36
10.38
7.84
7.13
6.8
6.65
6.6
6.58
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
12.47
10.33
7.78
7.07
6.81
6.67
6.61
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
12.42
10.36
7.81
7.10
6.81
6.66
6.61
6.58
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
6.56
0.08
0.04
0.04
0.04
0.01
0.01
0.01
0.00
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
(*) Mass of red mud used: 100g
183
Table B-5. Carbonation of RM at different CO2 concentrations and total gas flow rate of 300mL/min, stirring speed of 250rpm
Time
(min)
10% CO2 15% CO2 20% CO2 25% CO2 30% CO2
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.48
11.02
10.16
9.48
8.70
8.10
7.71
7.63
7.54
7.52
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
7.50
12.36
10.82
10.00
9.30
8.40
7.84
7.65
7.48
7.44
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
7.42
12.42
10.92
10.08
9.39
8.55
7.97
7.68
7.56
7.49
7.47
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
7.46
0.08
0.14
0.11
0.13
0.21
0.18
0.04
0.11
0.07
0.07
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
12.38
11.14
10.03
9.29
8.54
8.05
7.68
7.55
7.43
7.34
7.32
7.28
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
7.26
12.32
10.4
9.87
9.51
8.36
7.83
7.6
7.37
7.33
7.30
7.3
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
12.35
10.77
9.95
9.4
8.45
7.94
7.64
7.46
7.38
7.32
7.31
7.26
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
0.04
0.52
0.11
0.16
0.13
0.16
0.06
0.13
0.07
0.03
0.01
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.34
10.75
9.93
9.23
8.36
7.82
7.63
7.52
7.38
7.32
7.27
7.25
7.23
7.20
7.17
7.17
7.17
7.17
7.17
7.17
7.17
7.17
7.17
7.17
7.17
12.34
10.09
9.51
8.97
8.08
7.88
7.63
7.44
7.38
7.28
7.25
7.23
7.21
7.2
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.34
10.42
9.72
9.10
8.22
7.85
7.63
7.48
7.38
7.30
7.26
7.24
7.22
7.20
7.18
7.18
7.18
7.18
7.18
7.18
7.18
7.18
7.18
7.18
7.18
0.00
0.47
0.30
0.18
0.20
0.04
0.00
0.06
0.00
0.03
0.01
0.01
0.01
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.34
10.42
9.91
9.1
8.22
7.85
7.63
7.42
7.21
7.18
7.15
7.13
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
7.11
12.54
10.58
9.49
8.84
7.78
7.23
7.05
7.06
7.15
7.14
7.13
7.11
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
12.44
10.5
9.7
8.97
8.0
7.54
7.34
7.24
7.18
7.16
7.14
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
0.14
0.11
0.30
0.18
0.31
0.44
0.41
0.25
0.04
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.40
10.34
9.42
8.13
7.69
7.32
7.19
7.17
7.13
7.09
7.07
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
12.38
10.22
9.10
7.71
7.13
7.10
7.03
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
12.39
10.28
9.26
7.92
7.41
7.21
7.11
7.07
7.05
7.03
7.02
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
0.01
0.08
0.23
0.30
0.40
0.16
0.11
0.14
0.11
0.08
0.07
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05 (*) Mass of red mud used: 100g
184
Table B-6. Carbonation of RM at different CO2 concentrations and total gas flow rate of 300mL/min, stirring speed of 250rpm
Time
(min)
40% CO2 50% CO2 60% CO2
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
20
25
30
35
40
45
50
60
70
80
90
100
110
120
12.57
11.33
10.35
9.67
8.82
7.67
7.28
7.07
6.95
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
12.52
11.05
10.15
9.39
8.52
7.65
7.2
7.02
6.98
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
12.55
11.19
10.25
9.53
8.67
7.66
7.24
7.05
6.97
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
0.04
0.20
0.14
0.20
0.21
0.06
0.04
0.02
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.49
10.90
9.96
8.99
8.15
7.39
7.09
6.97
6.91
6.89
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
12.4
10.74
9.69
8.56
7.86
7.2
6.96
6.86
6.82
6.80
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
12.45
10.82
9.83
8.78
8.01
7.30
7.03
6.92
6.87
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
0.06
0.11
0.19
0.30
0.21
0.13
0.09
0.08
0.06
0.06
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.52
10.52
9.26
8.3
7.39
7.11
6.95
6.88
6.84
6.8
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
12.44
10.31
9.19
8.1
7.26
6.85
6.82
6.76
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
12.48
10.42
9.23
8.20
7.33
6.98
6.89
6.82
6.79
6.77
6.75
6.75
6.75
6.75
6.75
6.75
6.75
6.75
6.75
6.75
6.75
0.06
0.15
0.05
0.14
0.09
0.18
0.09
0.08
0.07
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01 (*) Mass of red mud used: 100g
185
Table B-7. Carbonation of RM at different CO2 concentrations and total gas flow rate of 400mL/min, stirring speed of 250rpm
Time
(min)
10% CO2 15% CO2 20% CO2 25% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.43
11.05
10.1
9.65
9.05
8.57
8.02
7.8
7.7
7.65
7.6
7.57
7.55
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
7.52
12.33
10.65
10.05
9.55
8.95
8.23
7.89
7.72
7.6
7.5
7.46
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
7.43
12.38
10.85
10.08
9.60
9.00
8.40
7.96
7.76
7.65
7.58
7.53
7.50
7.49
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
0.07
0.28
0.04
0.07
0.07
0.24
0.09
0.06
0.07
0.11
0.10
0.10
0.08
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
12.36
11.12
10.01
9.52
8.78
8.16
7.91
7.79
7.66
7.54
7.48
7.42
7.4
7.39
7.39
7.39
7.39
7.39
7.39
7.39
7.39
7.39
7.39
7.39
7.39
12.42
10.24
9.75
9.18
8.42
7.90
7.55
7.41
7.38
7.38
7.34
7.32
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
7.29
12.39
10.68
9.88
9.35
8.6
8.03
7.73
7.6
7.52
7.46
7.41
7.37
7.345
7.34
7.34
7.34
7.34
7.34
7.34
7.34
7.34
7.34
7.34
7.34
7.34
0.04
0.62
0.18
0.24
0.25
0.18
0.25
0.27
0.20
0.11
0.10
0.07
0.08
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
12.32
10.42
9.39
7.96
7.4
7.29
7.15
7.13
7.11
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
7.09
12.42
10.48
9.27
8.66
8.0
7.51
7.47
7.43
7.41
7.37
7.32
7.27
7.23
7.19
7.19
7.15
7.15
7.15
7.15
7.15
7.15
7.15
7.15
7.15
7.15
12.37
10.45
9.33
8.31
7.70
7.40
7.31
7.28
7.26
7.23
7.21
7.18
7.16
7.14
7.14
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
0.07
0.04
0.08
0.49
0.42
0.16
0.23
0.21
0.21
0.20
0.16
0.13
0.10
0.07
0.07
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.3
10.1
8.71
7.68
7.21
7.05
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
12.46
9.72
8.41
7.42
7.03
7.00
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
6.96
12.38
9.91
8.56
7.55
7.12
7.03
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
6.99
0.11
0.27
0.21
0.18
0.13
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04 (*) Mass of red mud used: 100g
186
Table B-8. Carbonation of RM at different CO2 concentrations and total gas flow rate of 400mL/min, stirring speed of 250rpm
Time
(min)
30% CO2 40% CO2 50% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
20
25
30
35
40
45
50
60
70
80
90
100
110
120
12.38
11.30
9.80
9.05
8.10
7.32
6.98
6.92
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
12.32
11.12
9.75
8.89
7.80
7.02
6.95
6.88
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
12.35
11.21
9.78
8.97
7.95
7.17
6.97
6.90
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
0.04
0.13
0.04
0.11
0.21
0.21
0.02
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
12.55
10.9
9.88
8.82
7.94
7.26
6.97
6.93
6.91
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
12.5
10.7
9.65
8.46
7.78
7.10
6.86
6.82
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
12.53
10.80
9.77
8.64
7.86
7.18
6.92
6.88
6.86
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
0.04
0.14
0.16
0.25
0.11
0.11
0.08
0.08
0.08
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
12.36
10.85
9.74
8.86
8.01
7.31
7.03
6.95
6.89
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
12.49
10.55
9.55
8.31
7.51
7.05
6.89
6.85
6.84
6.84
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
6.80
12.43
10.70
9.65
8.59
7.76
7.18
6.96
6.90
6.87
6.86
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
0.09
0.21
0.13
0.39
0.35
0.18
0.10
0.07
0.04
0.02
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05 (*) Mass of red mud used: 100g
187
Table B-9. Carbonation of RM by 30% CO2 concentrations, stirring speed of 250rpm and different total gas flow rate
Time
(min)
TF of 100mL/min TF of 200mL/min TF of 300mL/min TF of 400mL/min
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.42
11.76
10.65
9.94
9.45
9.04
8.57
8.27
7.91
7.66
7.52
7.43
7.38
7.33
7.27
7.24
7.21
7.21
7.20
7.20
7.20
7.20
7.20
7.20
7.20
12.28
11.62
10.45
10.22
10.07
9.64
8.99
8.25
7.87
7.6
7.46
7.35
7.28
7.25
7.23
7.22
7.21
7.21
7.22
7.22
7.22
7.22
7.22
7.22
7.22
12.35
11.69
10.55
10.08
9.76
9.34
8.78
8.26
7.89
7.63
7.49
7.39
7.33
7.29
7.25
7.23
7.21
7.21
7.21
7.21
7.21
7.21
7.21
7.21
7.21
0.10
0.10
0.14
0.20
0.44
0.42
0.30
0.01
0.03
0.04
0.04
0.06
0.07
0.06
0.03
0.01
0.00
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.34
10.81
9.91
9.11
8.20
7.70
7.52
7.40
7.32
7.25
7.22
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.43
10.60
9.73
8.90
7.92
7.50
7.39
7.25
7.21
7.15
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
12.39
10.71
9.82
9.01
8.06
7.60
7.46
7.33
7.27
7.20
7.18
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
0.06
0.15
0.13
0.15
0.20
0.14
0.09
0.11
0.08
0.07
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.40
10.34
9.42
8.13
7.69
7.32
7.19
7.17
7.13
7.09
7.07
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
7.04
12.38
10.22
9.10
7.71
7.13
7.10
7.03
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
6.97
12.39
10.28
9.26
7.92
7.41
7.21
7.11
7.07
7.05
7.03
7.02
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
7.01
0.01
0.08
0.23
0.30
0.40
0.16
0.11
0.14
0.11
0.08
0.07
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
12.38
11.30
9.80
9.05
8.10
7.32
6.98
6.92
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
12.32
11.12
9.75
8.89
7.80
7.02
6.95
6.88
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
12.35
11.21
9.78
8.97
7.95
7.17
6.97
6.90
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
6.88
0.04
0.13
0.04
0.11
0.21
0.21
0.02
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03 (*) Mass of red mud used: 100g
188
Table B-10. Carbonation of RM by 30% CO2 concentrations, total gas flow rate of 200mL/min and different stirring speeds
250rpm 350rpm Time
(min)
500rpm 750rpm
Time
(min)
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
Time
(min)
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.44
10.81
9.91
9.11
8.2
7.7
7.52
7.4
7.32
7.25
7.22
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.53
10.60
9.73
8.90
7.92
7.50
7.39
7.25
7.21
7.15
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
12.49
10.71
9.82
9.01
8.06
7.60
7.46
7.33
7.27
7.20
7.18
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
0.06
0.15
0.13
0.15
0.20
0.14
0.09
0.11
0.08
0.07
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0
2.5
5
7.5
10
12.5
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
100
110
120
12.56
11.28
10.38
9.7
8.92
8.35
7.99
7.62
7.42
7.34
7.28
7.26
7.24
7.24
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
12.64
11.3
10.43
9.8
9.05
8.46
8.06
7.67
7.48
7.36
7.32
7.28
7.26
7.26
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
12.60
11.29
10.41
9.75
8.99
8.41
8.03
7.65
7.45
7.35
7.30
7.27
7.25
7.25
7.23
7.23
7.23
7.23
7.23
7.23
7.23
7.23
7.23
7.23
7.23
0.06
0.01
0.04
0.07
0.09
0.08
0.05
0.04
0.04
0.01
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0
1
2
3
4
5
6
7
8
9
10
12
14
16
20
24
28
32
40
50
60
70
80
100
120
12.63
12.09
11.12
10.61
10.22
9.8
9.36
8.93
8.59
8.34
8.14
7.87
7.69
7.57
7.42
7.34
7.3
7.28
7.26
7.26
7.26
7.26
7.26
7.26
7.26
12.55
11.9
10.96
10.46
10.04
9.56
9.04
8.64
8.35
8.13
7.96
7.72
7.56
7.46
7.34
7.28
7.24
7.22
7.22
7.22
7.22
7.22
7.22
7.22
7.22
12.59
12.00
11.04
10.54
10.13
9.68
9.20
8.79
8.47
8.24
8.05
7.80
7.63
7.52
7.38
7.31
7.27
7.25
7.24
7.24
7.24
7.24
7.24
7.24
7.24
0.06
0.13
0.11
0.11
0.13
0.17
0.23
0.21
0.17
0.15
0.13
0.11
0.09
0.08
0.06
0.04
0.04
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
12.59
11.8
10.84
10.32
9.76
9.16
8.69
8.37
8.13
7.95
7.82
7.62
7.51
7.43
7.33
7.27
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
12.5
11.75
10.84
10.31
9.81
9.24
8.76
8.42
8.17
7.99
7.84
7.64
7.52
7.44
7.32
7.28
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
7.24
12.55
11.78
10.84
10.32
9.79
9.20
8.73
8.40
8.15
7.97
7.83
7.63
7.52
7.44
7.33
7.28
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
7.25
0.06
0.04
0.00
0.01
0.04
0.06
0.05
0.04
0.03
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01 (*) Mass of red mud used: 100g
189
Table B-11. Carbonation of RM by 30% CO2 concentrations, TF of 200mL/min, speeds of 250rpm and different solids concentrations in RM
Time
(min)
RM-35%wt by solid RM-40%wt by solid Time
(min)
RM-44%wt by solid
pH(*)
Rep 1
pH(*) Rep
2 Avg pH Stdev pH
pH(*)
Rep 1
pH(*) Rep
2 Avg pH Stdev pH
pH(*)
Rep 1
pH(*) Rep
2 Avg pH Stdev pH
0
2.5
5
7.5
10
12.5
15
17.5
20
25
30
35
40
45
50
55
60
65
70
75
80
90
100
110
120
12.5
11.72
10.92
10.52
10.22
10.01
9.72
9.41
9.02
8.12
7.76
7.43
7.26
7.18
7.12
7.1
7.08
7.08
7.08
7.08
7.08
7.08
7.08
7.08
7.08
12.43
11.6
10.8
10.42
10.2
9.84
9.64
9.28
8.78
8.01
7.42
7.28
7.18
7.12
7.07
7.04
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
12.47
11.66
10.86
10.47
10.21
9.93
9.68
9.35
8.90
8.07
7.59
7.36
7.22
7.15
7.10
7.07
7.05
7.05
7.05
7.05
7.05
7.05
7.05
7.05
7.05
0.05
0.08
0.08
0.07
0.01
0.12
0.06
0.09
0.17
0.08
0.24
0.11
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.48
11.52
10.86
10.52
10.14
9.86
9.54
9.12
8.68
8.06
7.62
7.44
7.36
7.28
7.22
7.18
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
12.45
11.32
10.62
10.3
10.06
9.8
9.45
9.0
8.52
7.92
7.56
7.38
7.3
7.22
7.18
7.15
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
7.12
12.47
11.42
10.74
10.41
10.10
9.83
9.50
9.06
8.60
7.99
7.59
7.41
7.33
7.25
7.20
7.17
7.14
7.14
7.14
7.14
7.14
7.14
7.14
7.14
7.14
0.02
0.14
0.17
0.16
0.06
0.04
0.06
0.08
0.11
0.10
0.04
0.04
0.04
0.04
0.03
0.02
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
12.44
10.81
9.91
9.11
8.2
7.7
7.52
7.4
7.32
7.25
7.22
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
7.19
12.53
10.60
9.73
8.90
7.92
7.50
7.39
7.25
7.21
7.15
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
7.13
12.49
10.71
9.82
9.01
8.06
7.60
7.46
7.33
7.27
7.20
7.18
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
7.16
0.06
0.15
0.13
0.15
0.20
0.14
0.09
0.11
0.08
0.07
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04 (*) Mass of red mud used: 100g
190
Table B-12. Longer carbonation of RM at 15% - 30% CO2 concentrations, TF of 200mL/min and stirring speed of 250rpm
Time
(day)
15% CO2 20% CO2 30% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
1
2
3
4
5
12.48
7.45
7.47
7.6
7.61
7.64
12.52
7.36
7.45
7.56
7.56
7.62
12.5
7.41
7.46
7.58
7.59
7.63
0.03
0.06
0.01
0.03
0.04
0.01
12.48
7.2
7.2
7.21
7.21
7.31
12.45
7.18
7.19
7.19
7.2
7.32
12.47
7.19
7.20
7.20
7.21
7.32
0.02
0.01
0.01
0.01
0.01
0.01
12.56
7.23
7.18
7.18
7.1
7.16
12.46
7.12
7.13
7.12
7.09
7.15
12.51
7.18
7.16
7.15
7.10
7.16
0.07
0.08
0.04
0.04
0.01
0.01
Table B-13. Longer carbonation of RM at 40% - 60% CO2 concentrations, TF of 200mL/min and stirring speed of 250rpm
Time
(day)
40% CO2 50% CO2 60% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
1
2
3
4
5
12.48
7.1
6.93
6.9
6.88
6.96
12.5
7
6.95
6.9
6.87
6.95
12.49
7.05
6.94
6.90
6.88
6.96
0.01
0.07
0.01
0.00
0.01
0.01
12.52
6.92
6.81
6.8
6.8
6.9
12.5
6.82
6.78
6.78
6.79
6.83
12.51
6.87
6.80
6.79
6.80
6.87
0.01
0.07
0.02
0.01
0.01
0.05
12.53
6.62
6.6
6.7
6.7
6.78
12.46
6.52
6.58
6.66
6.64
6.75
12.50
6.57
6.59
6.68
6.67
6.77
0.05
0.07
0.01
0.03
0.04
0.02
(*) Mass of red mud used: 100g
191
Table B-14. Longer carbonation of RM at by 30% CO2 concentrations, stirring speed of 250rpm and different total gas flow rate
Time
(day)
TF of gas 100mL/min TF of gas 200mL/min TF of gas 300mL/min TF of gas 400mL/min
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
1
2
3
4
5
12.44
7.17
7.21
7.36
7.34
7.36
12.45
7.18
7.2
7.35
7.34
7.33
12.45
7.18
7.21
7.36
7.34
7.35
0.01
0.01
0.01
0.01
0.01
0.01
12.56
7.23
7.18
7.18
7.1
7.16
12.46
7.12
7.13
7.12
7.09
7.15
12.51
7.18
7.16
7.15
7.10
7.16
0.07
0.08
0.04
0.04
0.01
0.01
12.48
7.13
7.2
7.27
7.1
7.31
12.47
7.15
7.19
7.26
7.12
7.33
12.48
7.14
7.20
7.27
7.11
7.32
0.01
0.01
0.01
0.01
0.01
0.01
12.48
7.22
7.17
7.25
7.34
7.34
12.5
7.21
7.19
7.26
7.35
7.36
12.49
7.22
7.18
7.26
7.35
7.35
0.01
0.01
0.01
0.01
0.01
0.01
Table B-15. Longer carbonation of RM at by 30% CO2 concentrations, stirring speed of 250rpm, TF of 200mL/min and different solids
concentrations in RM
Time
(day)
RM-35%wt RM-40%wt RM-44%wt
pH(*)
Rep 1
pH(*) Rep
2 Avg pH Stdev pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH Stdev pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH Stdev pH
0
1
2
3
4
5
12.52
7.08
7.12
7.05
7.08
7.12
12.49
7.1
7.1
7.09
7.1
7.08
12.51
7.09
7.11
7.07
7.09
7.10
0.02
0.01
0.01
0.03
0.01
0.03
12.48
7.18
7.12
7.16
7.15
7.12
12.55
7.1
7.11
7.13
7.1
7.09
12.52
7.14
7.12
7.15
7.13
7.11
0.05
0.06
0.01
0.02
0.04
0.02
12.56
7.23
7.18
7.18
7.1
7.16
12.46
7.12
7.13
7.12
7.09
7.15
12.51
7.18
7.16
7.15
7.10
7.16
0.07
0.08
0.04
0.04
0.01
0.01
(*) Mass of red mud used: 100g
192
Table B-16. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 100mL/min and stirring speed of 250rpm
Time
(min)
25% CO2 30% CO2 40% CO2 50% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.5
12.17
11.43
10.85
10.48
10.3
10.24
10.12
10.02
9.79
9.5
8.96
8.07
7.57
7.3
7.13
7
6.94
6.88
6.88
6.88
6.88
6.88
6.88
12.47
12.11
11.34
10.78
10.42
10.23
10.14
10.10
10.00
9.78
9.42
8.94
8.00
7.50
7.20
7.05
6.98
6.96
6.90
6.88
6.88
6.88
6.88
6.88
12.49
12.14
11.39
10.82
10.45
10.27
10.19
10.11
10.01
9.79
9.46
8.95
8.04
7.54
7.25
7.09
6.99
6.95
6.89
6.88
6.88
6.88
6.88
6.88
0.02
0.04
0.06
0.05
0.04
0.05
0.07
0.01
0.01
0.01
0.06
0.01
0.05
0.05
0.07
0.06
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
12.48
11.64
10.87
10.41
10.23
10.08
9.96
9.85
9.69
9.3
8.47
7.63
7.23
7.03
6.93
6.87
6.83
6.81
6.79
6.77
6.75
6.75
6.75
6.75
12.46
11.6
10.8
10.33
10.1
10.04
9.92
9.80
9.66
9.27
8.43
7.60
7.21
7.00
6.91
6.85
6.83
6.80
6.78
6.76
6.76
6.76
6.76
6.76
12.47
11.62
10.84
10.37
10.17
10.06
9.94
9.83
9.68
9.29
8.45
7.62
7.22
7.02
6.92
6.86
6.83
6.81
6.79
6.77
6.76
6.76
6.76
6.76
0.01
0.03
0.05
0.06
0.09
0.03
0.03
0.04
0.02
0.02
0.03
0.02
0.01
0.02
0.01
0.01
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.48
11.02
10.4
10.21
10.04
9.83
9.59
9.22
8.5
7.47
7.05
6.88
6.78
6.72
6.68
6.66
6.66
6.66
6.66
6.66
6.66
6.66
6.66
6.66
12.52
11.1
10.46
10.24
10.09
9.86
9.62
9.25
8.55
7.50
7.12
6.92
6.80
6.74
6.70
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
12.50
11.06
10.43
10.23
10.07
9.85
9.61
9.24
8.53
7.49
7.09
6.90
6.79
6.73
6.69
6.66
6.66
6.66
6.66
6.66
6.66
6.66
6.66
6.66
0.03
0.06
0.04
0.02
0.04
0.02
0.02
0.02
0.04
0.02
0.05
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.47
10.6
10.23
10.02
9.78
9.44
8.81
7.92
7.45
6.98
6.78
6.68
6.62
6.58
6.56
6.54
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
12.46
10.55
10.2
10.00
9.75
9.41
8.78
7.90
7.41
6.93
6.72
6.65
6.60
6.57
6.55
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
12.47
10.58
10.22
10.01
9.77
9.43
8.80
7.91
7.43
6.96
6.75
6.67
6.61
6.58
6.56
6.54
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
0.01
0.04
0.02
0.01
0.02
0.02
0.02
0.01
0.03
0.04
0.04
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
(*) Mass of red mud liquor used: 100g
193
Table B-17. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 100mL/min and stirring speed of 250rpm
Time
(min)
60% CO2 75% CO2 100% CO2
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.51
10.32
10.02
9.65
8.94
7.86
7.35
6.86
6.66
6.56
6.52
6.48
6.46
6.44
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
12.48
10.27
10.06
9.68
8.96
7.88
7.40
6.88
6.69
6.60
6.56
6.52
6.50
6.48
6.45
6.43
6.43
6.43
6.43
6.43
6.43
6.43
6.43
12.50
10.30
10.04
9.67
8.95
7.87
7.38
6.87
6.68
6.58
6.54
6.50
6.48
6.46
6.43
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
0.02
0.04
0.03
0.02
0.01
0.01
0.04
0.01
0.02
0.03
0.03
0.03
0.03
0.03
0.04
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
12.5
10.04
9.58
8.47
7.5
7.03
6.8
6.56
6.48
6.42
6.36
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.4
12.46
10
9.55
8.44
7.42
7.00
6.77
6.57
6.50
6.43
6.39
6.35
6.33
6.33
6.33
6.33
6.33
6.33
6.33
6.33
6.33
6.33
6.43
12.48
10.02
9.57
8.46
7.46
7.02
6.79
6.57
6.49
6.43
6.38
6.35
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.34
6.42
0.03
0.03
0.02
0.02
0.06
0.02
0.02
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.02
12.48
9.8
9.25
8.1
7.22
6.78
6.51
6.38
6.34
6.3
6.29
6.27
6.27
6.27
6.27
6.27
6.27
6.27
6.27
6.27
6.27
6.27
6.27
12.45
9.76
9.22
8.00
7.14
6.75
6.47
6.35
6.32
6.28
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
12.47
9.78
9.24
8.05
7.18
6.77
6.49
6.37
6.33
6.29
6.27
6.26
6.26
6.26
6.26
6.26
6.26
6.26
6.26
6.26
6.26
6.26
6.26
0.02
0.03
0.02
0.07
0.06
0.02
0.03
0.02
0.01
0.01
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01 (*) Mass of red mud liquor used: 100g
194
Table B-18. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 200mL/min and stirring speed of 250rpm
Time
(min)
15% CO2 20% CO2 25% CO2 30% CO2 40% CO2
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
pH(*)
Rep1
pH(*)
Rep2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.49
11.8
11
10.5
10.24
10.11
10
9.9
9.8
9.48
8.89
7.93
7.48
7.26
7.16
7.12
7.08
7.06
7.06
7.06
7.06
7.06
7.06
7.06
12.47
11.98
11.1
10.66
10.34
10.24
10.12
10.01
9.86
9.53
8.95
8.02
7.56
7.3
7.18
7.14
7.1
7.08
7.08
7.08
7.08
7.08
7.08
7.08
12.48
11.89
11.05
10.58
10.29
10.18
10.06
9.96
9.83
9.51
8.92
7.98
7.52
7.28
7.17
7.13
7.09
7.07
7.07
7.07
7.07
7.07
7.07
7.07
0.01
0.13
0.07
0.11
0.07
0.09
0.08
0.08
0.04
0.04
0.04
0.06
0.06
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.52
11.63
10.73
10.26
10.14
10
9.85
9.65
9.01
7.93
7.46
7.22
7.08
7
6.98
6.96
6.94
6.94
6.94
6.94
6.94
6.94
6.94
6.94
12.43
11.76
10.75
10.27
10.17
10.02
9.87
9.66
9.02
8.01
7.55
7.24
7.1
7.03
7
7
6.98
6.96
6.96
6.96
6.96
6.96
6.96
6.96
12.48
11.70
10.74
10.27
10.16
10.01
9.86
9.66
9.02
7.97
7.51
7.23
7.09
7.02
6.99
6.98
6.96
6.95
6.95
6.95
6.95
6.95
6.95
6.95
0.06
0.09
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.06
0.06
0.01
0.01
0.02
0.01
0.03
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.49
11.48
10.6
10.17
10
9.82
9.61
9.28
8.57
7.56
7.19
6.99
6.91
6.87
6.85
6.83
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
12.46
11.47
10.57
10.13
9.96
9.78
9.66
9.23
8.36
7.29
7.05
6.95
6.91
6.89
6.87
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.85
12.48
11.48
10.59
10.15
9.98
9.80
9.64
9.26
8.47
7.43
7.12
6.97
6.91
6.88
6.86
6.84
6.83
6.83
6.83
6.83
6.83
6.83
6.83
6.83
0.02
0.01
0.02
0.03
0.03
0.03
0.04
0.04
0.15
0.19
0.10
0.03
0.00
0.01
0.01
0.01
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
12.47
10.84
10.13
9.92
9.67
9.32
8.81
7.91
7.29
6.99
6.85
6.79
6.77
6.77
6.77
6.77
6.77
6.77
6.77
6.77
6.77
6.77
6.77
6.77
12.45
10.76
10.1
9.88
9.77
9.35
8.86
7.94
7.34
7
6.86
6.82
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
6.74
12.46
10.80
10.12
9.90
9.72
9.34
8.84
7.93
7.32
7.00
6.86
6.81
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
6.76
0.01
0.06
0.02
0.03
0.07
0.02
0.04
0.02
0.04
0.01
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
12.48
10.42
10
9.75
9.41
8.62
7.72
7.27
7.01
6.88
6.74
6.66
6.64
6.62
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
12.5
10.46
10.04
9.8
9.44
8.67
7.82
7.34
7.03
6.85
6.75
6.67
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
12.49
10.44
10.02
9.78
9.43
8.65
7.77
7.31
7.02
6.87
6.75
6.67
6.65
6.64
6.63
6.63
6.63
6.63
6.63
6.63
6.63
6.63
6.63
6.63
0.01
0.03
0.03
0.04
0.02
0.04
0.07
0.05
0.01
0.02
0.01
0.01
0.01
0.02
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
(*) Mass red mud liquor used: 100g
195
Table B-19. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 200mL/min and stirring speed of 250rpm
Time
(min)
50% CO2 60% CO2 75% CO2 100% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.5
10.25
9.77
9.3
8.35
7.51
7.07
6.83
6.73
6.67
6.59
6.55
6.53
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
12.51
10.24
9.89
9.48
8.44
7.77
7.13
6.86
6.7
6.62
6.58
6.54
6.52
6.5
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
12.51
10.25
9.83
9.39
8.40
7.64
7.10
6.85
6.72
6.65
6.59
6.55
6.53
6.51
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
0.01
0.01
0.08
0.13
0.06
0.18
0.04
0.02
0.02
0.04
0.01
0.01
0.01
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
12.49
9.89
9.58
9.04
7.77
7.18
6.9
6.7
6.63
6.57
6.54
6.52
6.5
6.48
6.48
6.46
6.46
6.46
6.46
6.46
6.46
6.46
6.46
6.46
12.44
9.85
9.57
9.05
7.74
7.1
6.8
6.69
6.6
6.58
6.56
6.48
6.44
6.44
6.44
6.44
6.44
6.44
6.44
6.44
6.44
6.44
6.44
6.44
12.47
9.87
9.58
9.05
7.76
7.14
6.85
6.70
6.62
6.58
6.55
6.50
6.47
6.46
6.46
6.45
6.45
6.45
6.45
6.45
6.45
6.45
6.45
6.45
0.04
0.03
0.01
0.01
0.02
0.06
0.07
0.01
0.02
0.01
0.01
0.03
0.04
0.03
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.46
9.66
9.38
8.65
7.61
7
6.74
6.59
6.51
6.46
6.44
6.42
6.4
6.4
6.38
6.38
6.38
6.38
6.38
6.38
6.38
6.38
6.38
6.38
12.5
9.71
9.45
8.63
7.58
6.98
6.72
6.56
6.48
6.45
6.43
6.41
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
12.48
9.69
9.42
8.64
7.60
6.99
6.73
6.58
6.50
6.46
6.44
6.42
6.40
6.40
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
6.39
0.03
0.04
0.05
0.01
0.02
0.01
0.01
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.51
9.5
9.22
7.92
6.98
6.64
6.49
6.41
6.37
6.33
6.31
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
12.47
9.47
9.28
7.9
6.94
6.6
6.40
6.37
6.34
6.30
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
6.28
12.49
9.49
9.25
7.91
6.96
6.62
6.45
6.39
6.36
6.32
6.30
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
6.29
0.03
0.02
0.04
0.01
0.03
0.03
0.06
0.03
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01 (*) Mass of red mud liquor used: 100g
196
Table B-20. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 300mL/min and stirring speed of 250rpm
Time
(min)
10% CO2 15% CO2 20% CO2 25% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.47
12.08
11.64
11.16
10.64
10.4
10.26
10.14
10
-
9.9
9.74
9.44
8.9
8.18
7.7
7.4
7.3
7.26
7.26
7.26
7.26
7.26
7.26
12.5
12.14
11.7
11.25
10.8
10.49
10.4
10.26
10.2
-
10.06
9.8
9.45
8.91
8.2
7.74
7.46
7.34
7.3
7.28
7.25
7.25
7.25
7.25
12.49
12.11
11.67
11.21
10.72
10.45
10.33
10.20
10.10
-
9.98
9.77
9.45
8.91
8.19
7.72
7.43
7.32
7.28
7.27
7.26
7.26
7.26
7.26
0.02
0.04
0.04
0.06
0.11
0.06
0.10
0.08
0.14
-
0.11
0.04
0.01
0.01
0.01
0.03
0.04
0.03
0.03
0.01
0.01
0.01
0.01
0.01
12.5
11.82
11.14
10.54
10.34
10.2
10.01
9.88
9.64
-
9.17
8.13
7.54
7.29
7.17
7.11
7.07
7.05
7.03
7.03
7.03
7.03
7.03
7.03
12.43
11.78
11.17
10.4
10.16
10
9.86
9.7
9.5
-
9.12
8.13
7.4
7.13
7.01
6.97
6.95
6.95
6.93
6.93
6.93
6.93
6.93
6.93
12.47
11.80
11.16
10.47
10.25
10.10
9.94
9.79
9.57
-
9.15
8.13
7.47
7.21
7.09
7.04
7.01
7.00
6.98
6.98
6.98
6.98
6.98
6.98
0.05
0.03
0.02
0.10
0.13
0.14
0.11
0.13
0.10
-
0.04
0.00
0.10
0.11
0.11
0.10
0.08
0.07
0.07
0.07
0.07
0.07
0.07
0.07
12.44
11.52
10.58
10.18
10
9.8
9.61
9.25
8.7
7.94
7.39
7.05
6.97
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
12.48
11.57
10.64
10.25
10.08
9.96
9.78
9.36
8.78
8
7.43
7.1
7.01
6.94
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
12.46
11.55
10.61
10.22
10.04
9.88
9.70
9.31
8.74
7.97
7.41
7.08
6.99
6.93
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
6.91
0.03
0.04
0.04
0.05
0.06
0.11
0.12
0.08
0.06
0.04
0.03
0.04
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.47
11.24
10.4
10.12
9.91
9.7
9.29
8.36
7.71
7.3
7.01
6.89
6.84
6.82
6.8
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
12.5
11.3
10.54
10.13
9.84
9.5
9.1
8.44
7.62
7.22
7.04
6.91
6.88
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
6.86
12.49
11.27
10.47
10.13
9.88
9.60
9.20
8.40
7.67
7.26
7.03
6.90
6.86
6.84
6.83
6.82
6.82
6.82
6.82
6.82
6.82
6.82
6.82
6.82
0.02
0.04
0.10
0.01
0.05
0.14
0.13
0.06
0.06
0.06
0.02
0.01
0.03
0.03
0.04
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06 (*) Mass of red mud liquor used: 100g
197
Table B-21. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 300mL/min and stirring speed of 250rpm
Time
(min)
30% CO2 40% CO2 50% CO2 60% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.5
10.66
10.1
9.8
9.42
8.48
7.7
7.29
7.05
6.93
6.81
6.76
6.74
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
12.44
10.62
10.02
9.75
9.31
8.38
7.5
7.1
6.87
6.77
6.75
6.73
6.71
6.71
6.71
6.71
6.71
6.71
6.71
6.71
6.71
6.71
6.71
6.71
12.47
10.64
10.06
9.78
9.37
8.43
7.60
7.20
6.96
6.85
6.78
6.75
6.73
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
0.04
0.03
0.06
0.04
0.08
0.07
0.14
0.13
0.13
0.11
0.04
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.48
10.32
9.9
9.28
8.01
7.36
7
6.82
6.73
6.67
6.64
6.62
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
6.59
12.5
10.4
9.98
9.34
8.16
7.48
7.08
6.92
6.81
6.74
6.67
6.63
6.61
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
12.49
10.36
9.94
9.31
8.09
7.42
7.04
6.87
6.77
6.71
6.66
6.63
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
0.01
0.06
0.06
0.04
0.11
0.08
0.06
0.07
0.06
0.05
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.45
10.12
9.48
7.97
7.2
6.87
6.71
6.63
6.59
6.57
6.53
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
12.48
10.18
9.53
8.11
7.36
6.99
6.82
6.71
6.62
6.6
6.57
6.55
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
6.53
12.47
10.15
9.51
8.04
7.28
6.93
6.77
6.67
6.61
6.59
6.55
6.53
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
6.52
0.02
0.04
0.04
0.10
0.11
0.08
0.08
0.06
0.02
0.02
0.03
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
12.46
9.97
9.23
7.67
7.01
6.75
6.6
6.53
6.49
6.47
6.45
6.43
6.43
6.43
6.41
6.41
6.41
6.41
6.41
6.41
6.41
6.41
6.41
6.41
12.52
10.07
9.28
7.72
7.09
6.84
6.64
6.59
6.48
6.46
6.44
6.44
6.44
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
12.49
10.02
9.26
7.70
7.05
6.80
6.62
6.56
6.49
6.47
6.45
6.44
6.44
6.43
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
6.42
0.04
0.07
0.04
0.04
0.06
0.06
0.03
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
(*) Mass of red mud liquor used: 100g
198
Table B-22. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 400mL/min and stirring speed of 250rpm
Time
(min)
10% CO2 15% CO2 20% CO2
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2 Avg pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.63
12.28
11.67
10.98
10.55
10.27
10.2
10.12
10.04
9.94
9.71
9.42
8.93
8.11
7.65
7.45
7.35
7.31
7.27
7.27
7.27
7.27
7.27
7.27
12.55
12.26
11.36
10.53
10.3
10.18
10.08
10.02
9.92
9.85
9.58
9.29
8.44
7.84
7.52
7.34
7.22
7.2
7.17
7.15
7.15
7.15
7.15
7.15
12.59
12.27
11.52
10.76
10.43
10.23
10.14
10.07
9.98
9.90
9.65
9.36
8.69
7.98
7.59
7.40
7.29
7.26
7.22
7.21
7.21
7.21
7.21
7.21
0.06
0.01
0.22
0.32
0.18
0.06
0.08
0.07
0.08
0.06
0.09
0.09
0.35
0.19
0.09
0.08
0.09
0.08
0.07
0.08
0.08
0.08
0.08
0.08
12.57
11.72
10.74
10.31
10.16
10
9.8
9.56
9.21
8.58
7.61
7.27
7.13
7.07
7.04
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
12.46
11.84
10.86
10.33
10.01
9.9
9.85
9.58
9.19
8.34
7.48
7.22
7.1
7.06
7.05
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
12.52
11.78
10.80
10.32
10.09
9.95
9.83
9.57
9.20
8.46
7.55
7.25
7.12
7.07
7.05
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
7.02
0.08
0.08
0.08
0.01
0.11
0.07
0.04
0.01
0.01
0.17
0.09
0.04
0.02
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
12.42
11.18
10.35
10.12
9.9
9.62
9.19
8.35
7.75
7.41
7.08
6.98
6.92
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
12.54
11.22
10.27
10.17
9.88
9.56
9.03
8.28
7.94
7.37
7.07
6.88
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
12.48
11.20
10.31
10.15
9.89
9.59
9.11
8.32
7.85
7.39
7.08
6.93
6.88
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
6.87
0.08
0.03
0.06
0.04
0.01
0.04
0.11
0.05
0.13
0.03
0.01
0.07
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
(*) Mass of red mud liquor used: 100g
199
Table B-23. Carbonation of RM liquor at different CO2 concentrations, total gas flow rate of 400mL/min and stirring speed of 250rpm
Time
(min)
25% CO2 30% CO2 40% CO2 50% CO2
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
pH(*)
Rep 1
pH(*)
Rep 2
Avg
pH
Stdev
pH
0
2.5
5
7.5
10
12.5
15
17.5
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
12.45
10.78
10.25
9.96
9.59
8.91
7.89
7.42
7
6.86
6.82
6.8
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
6.78
12.5
11
10.3
10.01
9.68
9.05
8.16
7.65
7.15
6.96
6.92
6.88
6.86
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
6.84
12.48
10.89
10.28
9.99
9.64
8.98
8.03
7.54
7.08
6.91
6.87
6.84
6.82
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
6.81
0.04
0.16
0.04
0.04
0.06
0.10
0.19
0.16
0.11
0.07
0.07
0.06
0.06
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
12.53
10.72
10.15
9.83
9.36
8.26
7.55
7.19
6.9
6.78
6.74
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
6.72
12.42
10.66
10.18
9.82
9.3
8.08
7.38
7.06
6.9
6.74
6.69
6.67
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
6.65
12.48
10.69
10.17
9.83
9.33
8.17
7.47
7.13
6.90
6.76
6.72
6.70
6.69
6.69
6.69
6.69
6.69
6.69
6.69
6.69
6.69
6.69
6.69
0.08
0.04
0.02
0.01
0.04
0.13
0.12
0.09
0.00
0.03
0.04
0.04
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
12.53
10.41
10.01
9.59
8.24
7.37
6.98
6.81
6.72
6.65
6.63
6.61
6.61
6.61
6.61
6.61
6.61
6.61
6.61
6.61
6.61
6.61
6.61
12.49
10.33
9.97
9.45
8.12
7.31
6.92
6.81
6.69
6.62
6.6
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
6.58
12.51
10.37
9.99
9.52
8.18
7.34
6.95
6.81
6.71
6.64
6.62
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
6.60
0.03
0.06
0.03
0.10
0.08
0.04
0.04
0.00
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
12.51
10.05
9.66
8.9
7.55
6.96
6.68
6.61
6.56
6.53
6.51
6.49
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
6.48
12.55
10.1
9.74
9.02
7.86
7.06
6.78
6.7
6.65
6.61
6.58
6.55
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
6.54
12.53
10.08
9.70
8.96
7.71
7.01
6.73
6.66
6.61
6.57
6.55
6.52
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
6.51
0.03
0.04
0.06
0.08
0.22
0.07
0.07
0.06
0.06
0.06
0.05
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
(*) Mass of red mud liquor used: 100g
200
Figure C-1. Phase composition quantification of carbonated RM at 20% CO2 concentration, total gas flow rate 200mL/min
201
Figure C-2. Phase composition quantification of carbonated RM at 50% CO2 concentration, total gas flow rate 200mL/min
202
Table D-1. Simulated carbonation of RM at different CO2 concentrations and total gas flow rate of 100mL/min
CO2 concentration
(%)
PCO2 (atm) -Log(PCO2) Experimental
RM liquor pH
Simulated RM
liquor pH
Experimental
RM slurry pH
Simulated RM
slurry pH
25
30
40
50
60
75
100
0.25
0.3
0.4
0.5
0.6
0.75
1
0.602
0.522
0.397
0.301
0.221
0.124
0
7.0
6.8
6.7
6.5
6.4
6.3
6.3
7.20
7.13
6.93
6.71
6.55
6.52
6.40
7.35
7.20
7.14
6.94
6.85
6.81
6.60
7.45
7.37
7.24
7.15
7.07
6.90
6.87
Table D-2. Simulated carbonation of RM at different CO2 concentrations and total gas flow rate of 200mL/min
CO2 concentration
(%)
PCO2 (atm) -Log(PCO2) Experimental
RM liquor pH
Simulated RM
liquor pH
Experimental
RM slurry pH
Simulated RM
slurry pH
15
20
30
40
50
60
75
100
0.15
0.2
0.3
0.4
0.5
0.6
0.75
1
0.823
0.698
0.522
0.397
0.301
0.221
0.124
0
7.3
7.09
6.85
6.7
6.5
6.4
6.4
6.3
7.4
7.22
7.13
6.93
6.71
6.55
6.52
6.4
7.5
7.4
7.2
7.0
6.9
6.8
6.6
6.6
7.67
7.54
7.37
7.24
7.15
7.07
6.90
6.87
203
Table D-3. Simulated carbonation of RM at different CO2 concentrations and total gas flow rate of 300mL/min
CO2 concentration
(%) PCO2 (atm) -Log(PCO2)
Experimental
RM liquor pH
Simulated RM
liquor pH
Experimental
RM slurry pH
Simulated RM
slurry pH
10
15
20
25
30
40
50
60
0.1
0.15
0.2
0.25
0.3
0.4
0.5
0.6
1.0
0.823
0.698
0.602
0.522
0.397
0.301
0.221
7.43
7.1
6.99
6.9
6.78
6.7
6.6
6.5
7.56
7.4
7.22
7.2
7.13
6.93
6.71
6.55
7.5
7.3
7.18
7.1
7.0
6.9
6.8
6.7
7.78
7.67
7.54
7.45
7.37
7.24
7.15
7.07
Table D-4. Simulated carbonation of RM at different CO2 concentrations and total gas flow rate of 400mL/min
CO2 concentration
(%) PCO2 (atm) -Log(PCO2)
Experimental
RM liquor pH
Simulated RM
liquor pH
Experimental
RM slurry pH
Simulated RM
slurry pH
10
15
20
25
30
40
50
0.1
0.15
0.2
0.25
0.3
0.4
0.5
1.0
0.823
0.698
0.602
0.522
0.397
0.301
7.3
7.1
6.9
6.8
6.7
6.6
6.5
7.56
7.4
7.22
7.2
7.13
6.93
6.71
7.5
7.3
7.14
7.0
6.9
6.8
6.8
7.78
7.67
7.54
7.45
7.37
7.24
7.15