ISTANBUL TECHNICAL UNIVERSITY « GRADUATE SCHOOL OF SCIENCE ENGINEERING AND TECHNOLOGY
M.Sc. THESIS
FEBRUARY 2020
CHARACTERIZATION AND RECOVERY OF RARE EARTH ELEMENTS FROM IRON MINING SLUDGE
Azmat Fatima SIDDIQUI
Department of Environmental Engineering
Department of Environmental Engineering
Environmental Science, Engineering and Management Programme
FEBRUARY 2020
ISTANBUL TECHNICAL UNIVERSITY « GRADUATE SCHOOL OF SCIENCE ENGINEERING AND TECHNOLOGY
CHARACTERIZATION AND RECOVERY OF RARE EARTH ELEMENTS FROM IRON MINING SLUDGE
M.Sc. THESIS Azmat Fatima SIDDIQUI
501171736
Thesis Advisor: Prof. Dr. Ismail KOYUNCU Thesis Co-Advisor: Prof. Dr. Börte KÖSE MUTLU
Çevre Mühendisliği Anabilim Dalı Çevre Bilimleri ve Mühendisliği Programı
ŞUBAT 2020
ISTANBUL TEKNİK ÜNİVERSİTESİ « FEN BİLİMLERİ ENSTİTÜSÜ
DEMİR MADENCİLİĞİ ÇAMURUNDAN NADİR TOPRAK ELEMANLARININ KARAKTERİZASYONU VE GERİ KAZANIMI
YÜKSEK LİSANS TEZİ Azmat Fatima SIDDIQUI
501171736
Tez Danışmanı: Prof. Dr. Ismail KOYUNCU Eş Danışman: Doç. Dr. Börte KÖSE MUTLU
v
Thesis Advisor : Prof. Dr. Ismail KOYUNCU ... İstanbul Technical University
Co-advisor : Assist. Prof. Dr. Börte KÖSE MUTLU ... Yeditepe University
Jury Members : Assoc. Prof. Dr. Mustafa Evren ERŞAHIN... İstanbul Technical University
Prof. Dr. Mustafa KUMRAL ... İstanbul Technical University
Prof. Dr. Vedat UYAK ... Pamukkale University
Azmat Fatima Siddiqui, a M.Sc. student of İTU Graduate School of Science Engineering and Technology student ID 501171736, successfully defended the thesis/dissertation entitled “CHARACTERIZATION AND RECOVERY OF RARE EARTH ELEMENTS FROM IRON MINING SLUDGE”, which she prepared after fulfilling the requirements specified in the associated legislations, before the jury whose signatures are below.
Date of Submission : 3 February 2020 Date of Defense : 14 February 2020
vii
ix FOREWORD
No work is complete without giving proper appreciation to the other, often invisible, hands behind it.
I would first like to extend my gratitude to Prof. Dr. Ismail KOYUNCU, my advisor, for giving me this opportunity, for believing in me and pushing me to achieve my potential and to Dr. Börte MUTLU KÖSE MUTLU, my co-advisor, for her
instruction and support. I would also like to express my gratitude to Res. Asst. Ayşe YUKSEKDAĞ, for her guidance and never-ending patience for answering my
questions, and for her help, positivity. Thank you for always being there for me every step of the way. I also want to say a heartfelt thank you to Gizem TUNCAY for all her invaluable aid and making my time during my research interesting and engaging. From their advice and kindness, I have learned many valuable skills that I will carry with me. Thank you also to the MEMTEK family for all your support and
encouragement, for taking me under your wings and pushing me to grow and learn every day. I am immensely lucky to have worked with a team that was more than just my coworkers, and rather have become my friends.
I would also like to thank The Scientific and Technological Research Council of Turkey (TÜBITAK) for their financial support of this project (Project ID: 117Y357). Last but not least, to my family: my mother and father for their never-ending
patience and care, and my sister and brother, for their pep talks and calming
influence. I would not be standing here today if it were not for your sacrifices, your support, and your belief in me; you continue to push me to be the best version of myself in everything I do. Thank you for your love.
February 2020 Azmat Fatima SIDDIQUI
xi TABLE OF CONTENTS Page FOREWORD ... ix ABBREVIATIONS ... xv SYMBOLS ... xvii
LIST OF TABLES ... xix
LIST OF FIGURES ... xxi
SUMMARY ... xxv ÖZET ... xxix 1. INTRODUCTION ... 1 1.1 Purpose of Thesis ... 1 1.2 Scope of Thesis ... 2 1.3 Hypothesis ... 3 2. LITERATURE SUMMARY ... 5
2.1 General Information about REE ... 5
2.2 REE Composition ... 9
2.3 Uses, Economic Status and Global Trade ... 11
2.4 Red Mud ... 16
2.5 Recovery and Reuse ... 17
2.5.1 Determination of rare earth elements by spectroscopic methods ... 17
2.5.2 Pretreatment ... 19
2.5.3 Leach preparation ... 19
2.5.4 Concentration using membrane processes ... 20
2.5.5 Nanofiltration ... 20
2.5.5.1 Operating Pressure ... 21
2.5.5.2 pH ... 21
2.6 REE Separation Techniques with Membrane Processes ... 22
2.7 Liquid Membrane Strategies ... 25
2.8 Supported Liquid Membranes ... 27
3. MATERIAL AND METHODS ... 29
3.1 Instruments ... 29
3.2 Revealing REE potential in Red Mud ... 31
3.3 Preliminary Work with Red mud ... 31
3.3.1 Leaching using the microwave ... 31
xii
3.4 Concentration Processes ... 39
3.4.1 Pretreatment ... 39
3.4.2 Nanofiltration studies with red mud ... 41
3.5 Supported Liquid Membrane Studies ... 42
3.5.1 Preparations for supported liquid membrane studies ... 42
3.5.2 SLM works with red mud ... 43
4. EXPERIMENTAL RESULTS ... 47
4.1 Waste Characterization Studies ... 47
4.1.1 M1 waste characterization ... 47
4.2 Concentration Studies ... 48
4.2.1 NF studies ... 48
4.2.1.1 M1-1 ... 48
4.2.1.2 M1-3 ... 49
4.3 Filtration and Removal Performances ... 51
4.3.1 pH 1.5 ... 51 4.3.1.1 M1-1 ... 52 4.3.1.2 M1-3 ... 53 4.3.1.3 Discussion ... 54 4.3.2 pH 2.5 ... 57 4.3.2.1 M1-1 ... 57 4.3.2.2 M1-3 ... 58 4.3.2.3 Discussion ... 60 4.3.3 pH 3.5 ... 63 4.3.3.1 M1-1 ... 63 4.3.3.2 M1-3 ... 64 4.3.3.3 Discussion ... 66 4.3.4 Discussion ... 69
4.4 Membrane Surface Analysis for M1-1 ... 70
4.5 Membrane Surface Analysis for M1-3 ... 81
4.6 Optimization Results of Nanofiltration Studies ... 93
4.6.1 RSM results of NF studies performed with red mud leach (M1-1) ... 93
4.6.2 RSM results of NF studies with M1-3 ... 97
4.7 Separation Studies with Supported Liquid Membrane ... 102
4.7.1 SLM Works with Iron Mine Wastes ... 102
xiii
REFERENCES ... 109 CURRICULUM VITAE ... 115
xv ABBREVIATIONS
EU : European Union
hREE : Heavy REE
ICP-MS : Inductively Coupled Plasma Mass Spectrometry
ICP-OES : Inductively Coupled Plasma Optical Emission Spectrometry LCD : Liquid Crystal Display
LED : Light Emitting Diode
lREE : Light REE
MF : Microfiltration
MTA : Mineral Research and Exploration Institute of Turkey NAA : Neutron Activation Analysis
NF : Nanofiltration
REE : Rare Earth Element
REEO (REO) : Rare Earth Element Oxides
RO : Reverse Osmosis
SEM-EDS : Scanning Electron Microscopy - Energy Scattering Spectrometer SLM : Supported Liquid Membrane
UF : Ultrafiltration
USA : United States of America
USEPA : United States Environmental Protection Agency
UV : Ultraviolet
UV-VIS : Ultraviolet and Visible Light Absorption Spectroscopy
XRF : X-Ray Fluorescence
xvii SYMBOLS
J : Mass flux of each element, mol / m2.hr
Δ REE : Change in REE concentration in the stripping phase over time Δt, M V : Volume of stripping solution, L
xix LIST OF TABLES
Page Table 2.1 : Chemical structure properties of REEs (University of Liverpool, 2017). 7 Table 2.2 : Elemental composition of the Earth's crust (Charalampiddes et al., 2015).
... 9
Table 2.3 : REE content in the Earth's crust (USEPA, 2012). ... 10
Table 2.4 : List of various applications for REEs. ... 11
Table 2.5 : Comparison of prices of rare earth metals and rare earth oxides (Schüler, 2011) (Schüler, 2011). ... 15
Table 2.6 : REE composition of red mud worldwide (Akcil et al., 2018). ... 17
Table 3.1 : Red mud solubilization studies. ... 33
Table 3.2 : Red mud solubilization studies using lower acid ratios. ... 33
Table 3.3 : Main element concentrations of M1 waste and leach. ... 34
Table 3.4 : REE concentrations of M1 waste and Leach. ... 35
Table 3.5 : Chemicals used in synthetic Leach preparation. ... 37
Table 3.6 : Synthetic leaches prepared for M1 waste and their properties. ... 39
Table 3.7 : Test set for NF process. ... 41
Table 3.8 : SLM testing sets. ... 45
Table 3.9 : Chemicals used in SLM process. ... 45
Table 4.1 : M1 waste XRF results. ... 47
Table 4.2 : Statement of the osmotic pressure generated by the leach solution. ... 54
Table 4.3 : Statement of the osmotic pressure generated by the Leach solution. ... 60
Table 4.4 : Statement of the osmotic pressure generated by the Leach solution. ... 66
xxi LIST OF FIGURES
Page Figure 2.1 : Placement of rare earths in the periodic table (University of Liverpool,
2017). ... 5 Figure 2.2 : Characteristics of REEs (Zepf, 2013). ... 8 Figure 2.3 : Distribution of REE quantity between countries producing REE mines between 2010 and 2013 (Charalampiddes et al., 2015). ... 12 Figure 2.4 : Critical materials strategy for 2025 (left)-right for 2050 (USA, 2011). 12 Figure 2.5 : Worldwide REE production and use in 2016 (Zhou et al., 2017). ... 13 Figure 2.6 : The interoperable status of NTIs in their areas of use (Chegwidden and Kingsnorthi, 2010; Lynas, 2010). ... 13 Figure 2.7 : Future forecasts for the quantity of rare soil oxides used in clean technologies (Zhou et al., 2017). ... 14 Figure 2.8 : Comparison of REE prices with gold price (Reuters, 2010). ... 15 Figure 2.9 : Prices of rare earths that are critical and used in green technologies (Humphries, 2012). ... 16 Figure 2.10 : Effect of operating pressure on NF performance: (a) effect on flux with pH and (b) singular effect on flux and removal (Kose-Mutlu et al., 2018). ... 22 Figure 2.11 : Number of publications of each REE with membrane separation technology over the last 30 years (Chen et al., 2018). ... 23 Figure 2.12 : Number of publications in membrane strategy for separation / purification of REEs over the last thirty years. ... 24 Figure 2.13 : Supported Liquid Membranes (Chen et al., 2018). ... 25 Figure 2.14 : a) Co-transport of rare earth element ions in the membrane b) Opposite transport mechanism (Chen et al., 2018). ... 26 Figure 2.15 : The model of the transition kinetics in the membrane (Chen et al., 2018).
... 27 Figure 2.16 : (a) three-unit SLM processes (Yang et al., 2003), (b) two-unit SLM process (Pei et al., 2009). ... 28 Figure 3.1 : Standard production microbial fuel cell for the SLM process. ... 29 Figure 3.2 : (a) Microwave and (b) ICP-MS device. ... 30 Figure 3.3 : Filtration system that can work under high pressure conditions (Sterilitech). ... 30 Figure 3.4 : Samples of red sludge from the iron mine. ... 31 Figure 3.5 : M1 waste synthetic leach solution. ... 36
xxii
Figure 3.6 : a) Used pretreatment membranes, b) pretreatment filtration mechanism. ... 40 Figure 3.7 : Before and after pretreatment of synthetic acidic leach. ... 40 Figure 3.8 : Pre-treatment process flow diagram. ... 40 Figure 3.9 : NF process components (1-nitrogen gas, 2-upstream filtration unit, 3-NF membrane, 4-magnetic stirrer, 5-permeate outlet line, 6-permeate, 7- weighing device, 8-manometer, 9-computer). ... 42 Figure 3.10 : Scenario I- Dissolved Cyanex 272 in kerosene. ... 42 Figure 3.11 : Scenario II - Dissolved 18-Crown-6 in kerosene. ... 43 Figure 3.12 : Scenario III - Dissolved D2EHPA in kerosene. ... 43 Figure 4.1 : Concentrations of M1-1 a) REE b) trace elements (by ICP-MS). ... 49 Figure 4.2 : Obtained concentrations from trace elements. ... 50 Figure 4.3 : Comparison of the obtained results from red mud leach and improved red mud leach. ... 50 Figure 4.4 : Flux profiles at a pH of 1.5 M1-1. ... 52 Figure 4.5 : Permeability results at pH 1.5 M1-1. ... 52 Figure 4.6 : Flux profiles at a pH 1.5 M1-3. ... 53 Figure 4.7 : Permeability results at pH 1.5 M1-3. ... 53 Figure 4.8 : Obtained efficiency at ph 1.5 - 12 bar study. ... 55 Figure 4.9 : Obtained efficiency at ph 1.5 - 18 bar study. ... 56 Figure 4.10 : Obtained efficiency at ph 1.5 - 24 bar study. ... 57 Figure 4.11 : Flux profiles at pH 2.5 M1-1. ... 58 Figure 4.12 : Permeability results at pH 2.5 M1-1. ... 58 Figure 4.13 : Flux profile for pH 2.5 with M1-3. ... 59 Figure 4.14 : Permeability results at pH 2.5 M1-3. ... 59 Figure 4.15 : Obtained efficiency at ph 2.5 - 12 bar study. ... 61 Figure 4.16 : Obtained efficiency at ph 2.5 - 18 bar study. ... 62 Figure 4.17 : Obtained efficiency at ph 2.5 - 24 bar study. ... 63 Figure 4.18 : Flux profiles at pH 3.5 M1-1. ... 64 Figure 4.19 : Permeability results at pH 3.5 M1-1. ... 64 Figure 4.20 : Flux profile at 3.5 ph M1-3. ... 65 Figure 4.21 : Permeability results at ph 3.5 M1-3. ... 65 Figure 4.22 : Obtained efficiency at ph 3.5 - 12 bar study. ... 67 Figure 4.23 : Obtained efficiency at ph 3.5 - 18 bar study. ... 68 Figure 4.24 : Obtained efficiency at ph 3.5 - 24 bar study. ... 69 Figure 4.25 : Comparison of efficiencies for all pH and pressure value. ... 69
xxiii
Figure 4.26 : Appearance of MF membranes used in pretreatment of M1-1 synthetic leach (pH of the pretreatment from left to right: 1.5, 2.5 and 3.5). ... 70 Figure 4.27 : NF membranes used after pretreatment of M1-1 synthetic leach (pH of the pretreatment from left to right: 1.5, 2.5 and 3.5). ... 70 Figure 4.28 : SEM views of MF membranes used in pretreatment of M1-1 synthetic leach (5000 X magnification). ... 71 Figure 4.29 : SEM views of NF membranes used at M1-1 synthetic leach concentration (1000 X magnification). ... 72 Figure 4.30 : Results of mapping of MF membranes (pH 1.5). ... 73 Figure 4.31 : Results of mapping of MF membranes (pH 2.5). ... 74 Figure 4.32 : Results of mapping of MF membranes (pH 3.5). ... 76 Figure 4.33 : Results of mapping of NF membranes (pH 1.5). ... 77 Figure 4.34 : Results of mapping of NF membranes (pH 2.5). ... 78 Figure 4.35 : Results of mapping of NF membranes (pH 3.5). ... 79 Figure 4.36 : MF membranes for pre-treatment of M1-3 synthetic leach (pH from left to right, 1.5, 2.5 and 3.5). ... 81 Figure 4.37 : NF membranes used after pretreatment of M1-3 synthetic leach (pH from left to right, 1.5, 2.5 and 3.5. ... 81 Figure 4.38 : SEM views and EDX results of MF membranes. ... 82 Figure 4.39 : SEM views and EDX results of NF membranes. ... 83 Figure 4.40 : Results of mapping of MF membranes (pH 1.5). ... 84 Figure 4.41 : Results of mapping of MF membranes (pH 2.5). ... 86 Figure 4.42 : Results of mapping of MF membranes (pH 3.5). ... 88 Figure 4.43 : Results of mapping of NF membranes (pH 1.5). ... 88 Figure 4.44 : Results of mapping of NF membranes (pH 2.5). ... 90 Figure 4.45 : Results of mapping of NF membranes (pH 3.5). ... 91 Figure 4.46 : NF removal efficiency under Pressure and pH (a) combined and (b, c) individual. ... 94 Figure 4.47 : ANOVA results. ... 95 Figure 4.48 : Mathematic relationship equation. ... 95 Figure 4.49 : NF concentration ratio under Pressure and pH (a) combined and (b, c) individual. ... 96 Figure 4.50 : ANOVA results. ... 97 Figure 4.51 : Mathematic relationship equation. ... 97 Figure 4.52 : NF removal efficiency under ph and Pressure (a) combined and (b,c) individual. ... 98 Figure 4.53 : ANOVA results with M1-3. ... 99 Figure 4.54 : Mathematic relationship equation M1-3. ... 99
xxiv
Figure 4.55 : NF concentration ratio under ph and Pressure (a) combined and (b,c) individual. ... 100 Figure 4.56 : ANOVA results M1-3. ... 101 Figure 4.57 : Mathematic relationship equation M1-3. ... 101 Figure 4.58 : SLM system and components. ... 102 Figure 4.59 : Expressing the separation efficiencies obtained in SLM studies by mass flux. ... 104 Figure 4.60 : Comparison of SLM results for Ce, La and Nd elements. ... 106 Figure 4.61 : Comparison of SLM results for Y element. ... 106
xxv
CHARACTERIZATION AND RECOVERY OF RARE EARTH ELEMENTS FROM IRON MINING SLUDGE
SUMMARY
REEs are a vital and irreplaceable part of our modern technological and digital industries. The uses of the 17 elements regarded as the rare earth elements vary from use in magnets to ceramics, to use in nuclear industries. Although rare earth elements are called ‘rare,’ any metal that is less than 0.1% of the Earth’s crust can be described as being rare. The “rarity’ of REEs can be explained as they are often found in ores, in small quantities, and in conjunction with other metals and other REEs; they are also widely dispersed within the natural sources where they can be found, making them difficult to aggregate and extract using economically feasible techniques.
Although all REEs are valuable to different industries, a list has been comprised to categorize the value of the respective REEs, in relation to future predicted demand, and availability of supply. Among the REEs that are the most critical to recover are Ce, La, and particularly, Nd, and Y. These elements are in high demand and at a potential supply risk. To enable the continued use of these crucial REEs, innovative techniques must be considered to account for supply. Worldwide, natural REE sources are limited and spread unevenly between a few countries. Of these countries, China holds the lion’s share of these sources, equaling up to 90% of the global REE supply. The rest of the natural REE sources are divided amongst the US, India and a few other countries. This presents a challenge in ensuring the stability of the supply of REEs to other countries, such as Turkey. As every nation requires REEs to remain competitive in a global market and ensure production of useful technologies that encourage economic growth within the country, it is a precarious position to be in to rely entirely on REE supplies from other nations.
The removal of REEs from other metals is difficult in and of itself, it is even more difficult to remove individual REEs from an amalgamation of other REEs, due to their similar chemical and physical properties. As many countries such as Turkey do not possess large natural resources for REEs, alternative methods must be considered to reduce the disparity between the supply and demand, that do not rely on import of materials. One of the most promising techniques for this, is the use of waste from various industries to recover REEs. E-waste, thermal power plant waste, and the waste from mining activities present exciting possibilities for REE reuse and recovery. In this study, red mud or iron mining sludge was obtained from Central Anatolia of Turkey to discuss the possibility of the extraction of REE from this source.
The samples were initially dried overnight at 105oC to remove any moisture that may be present in the sample.
After the samples were dried, the next step was to leach the samples using a microwave, as microwaves produce efficient leaching results without excessive environmental impact through reduced chemical consumption. In general, higher temperatures result in higher leaching efficiency. 0.1 g of the red mud samples were weighed using precision weighing. Various acid ratios were added to the samples,
xxvi
before placing them into the microwave to be leached. A 5-step program was used, in which the main leaching occurred at 210oC for 30 minutes. After leaching, the samples that contained the higher acid ratio were submitted to a second stage with H3BO3 to counteract the effects of HF on the glassware and ICP-device. The REE and major element concentrations were obtained from the red mud waste and the leaching samples, showing that from the major elements present, Fe, as expected was at the highest concentration, along with Al, Mg, K, Ca, Li, Ni, Ti, and Mn. Of the REEs, Ce, La, Y, and Nd were selected as they showed the highest percentage presence and because they are categorized as near critical and critical REEs. A synthetic solution was then prepared according to the concentrations seen in the original and leached samples, metal salts were added in equivalence to the amounts described by the studies, whereas REEs were standardized to 60 ppb and added in liquid form from 1000 ppm standard solutions. Two different acid solutions were prepared, one with a higher acid content, M1-1 and one with lower acid content, M1-3. M1-1 used 120ml HCl and 80ml HNO3 per liter, whereas M1-3 used only 20ml HNO3 per liter.
After the preparation of the synthetic solutions, the solutions were pretreated and filtered using nanofiltration. Supported liquid membrane processes were also conducted, without pretreating the synthetic solutions.
In the initial pretreatment steps, NaOH was used for pH adjustment to 1.5, 2.5, and 3.5. The solution was run through a MF membrane and then directly used for NF processes. Before using the NF membranes, they were pressurized by passing pure water for 1 hour under 24 bar pressure. Nanofiltration was carried out at pH levels. 1.5, 2.5, and 3.5 and under 12, 18, and 24 bar operating pressure. SLM was carried out using a PVDF membrane, a polymeric membrane, with a pore diameter of 0.45 um and with 5M as the stripping solution. 5M was used as higher acid content in the stripping solution leads to increased chemical transport or increased separation efficiency. Three scenarios were used, I, using dissolved Cyanex 272 in kerosene, II, with dissolved 18-Crown-6 in kerosene, and lastly, III, dissolved D2EHPA in kerosene. Before using the SLM was used, the membrane was kept in the solution to be used overnight, to ensure complete saturation. 250 ml of the stripping and feed solutions were added, with stirring set to 600 rpm, with the reaction time varying between 3 hours and 24 hours.
The XRF was used for the initial characterization of the iron mining sludge, or red mud, but as the XRF cannot read values below a certain limit, often it is not a suitable source to determine the amount of REEs present. However, it was used to identify other chemicals within the red mud sample.
NF studies were carried out as described above, and it was found that between M1-1 and M1-3 acid leaching methods, M1-3 was preferred as it had significantly smaller environmental impact, as less acid was used and it obtained similar REE recovery to M1-1 methods. In fact, overall, M1-3 was recorded to have lower flux values and higher overall efficiency. M1-1 was unable to effectively recover REEs at the pH 3.5, as more NaOH was required to raise the pH from its initial pH of less than zero to 3.5. For M1-3, as the NaOH amount was less, the total yield obtained was above 98% at this pH.
Membrane surface analysis was carried out using the SEM for both the MF and NF membranes that were used for M1-1 and M1-3. For M1-1, it can be seen that the MF membrane remains clean and the fibers are clear, as no precipitate has formed at pH 1.5. The EDX results show that for the MF, only the elements in the raw material of
xxvii
the membrane were seen, at pH 2.5, Mg, Al, K, Ca, and Fe can be obtained and at 3.5, only Fe is obtained because at pH 3.5, the dominant element is Fe, which is the main element in red mud. For M1-3, the fiber structure in the MF membranes is the clearest. There were only a small number of chemical precipitates, which prevented filter clogging. When EDX was examined, Na, Mg, Al, K, Ca, Fe and Zn elements are seen on the membrane surface used in pH 2.5 study. This means that in the MF step performed after the chemical precipitation applied, the respective elements are retained on the membrane surface. The surfaces of the NF membranes are smoother and contain pores with relatively small pore diameters rather than filamentous structures. The filtration of the solution at pH 1.5 contained a high proportion of iron on the surface of the membrane, but the amount of iron decreases at the pH values of 2.5. When the leach solution with a pH of 3.5 was filtered, it was observed that some chemical precipitates which were thought to escape to the MF membrane had spot clogging but there was no continuous scaling layer as in other low pH values.
Statistical analysis was conducted of the results of the NF steps to determine the effects of pH and pressure on removal. Through this analysis, it is seen that the optimum operating conditions are at pH 3.5 at 24 bar, using the M1-3 leaching solution. However, the effects of operating pressure are seen to be low, therefore, operation of the NF procedure could occur at 12 bar, as well.
7 SLM studies were carried out with red mud using the synthetic solutions, varying the chemical present in the organic phase solution, the concentration within the organic phase and the reaction time. SLM-3 using 0.3M D2EHPA, with a 3-hour reaction time showed the highest mass flux values for Ce, La, Nd, and Y.
Overall, through this study it can be seen that red mud presents a viable source for the recovery and reuse of REEs, the elements can be extracted using limited acid leaching and concentrated through the use of membranes.
xxix
DEMİR MADENCİLİĞİ ÇAMURUNDAN NADİR TOPRAK ELEMANLARININ KARAKTERİZASYONU VE GERİ KAZANIMI
ÖZET
NTE'ler modern, teknolojik ve dijital endüstrilerimizin hayati, yeri doldurulamaz bir parçasıdır. 17 element grubundan oluşan nadir toprak elementleri, birçok farklı endüstride de kullanılmaktadır. Nadir toprak elementleri "nadir" olarak adlandırılsa da, Dünya kabuğunun% 0.1' inden daha az olan herhangi bir metal nadir olarak tanımlanabilir; ancak, bazı çalışmalara göre birçok NTE altından daha fazla miktarda bulunmaktadır. Bununla birlikte, NTE'lerin “nadirliği”, diğer metaller ve diğer NTE'lerle birlikte cevherlerde, küçük miktarlarda sıklıkla bulundukları için açıklanabilir; aynı zamanda, bulunabilecekleri doğal kaynaklar içinde geniş bir alana yayılmışlardır, bu da ekonomik olarak uygulanabilir teknikler kullanarak toplanmalarını ve çıkarılmalarını zorlaştırmaktadır. Tüm NTE'ler farklı endüstriler için değerli olsa da, ilgili NTE'lerin geleceğe göre değerlerini kategorize etmek için bir liste kategorize edilmiştir. Geri kazanımı en kritik olan NTE'ler arasında Ce, La ve özellikle Nd ve Y vardır. Bu elementler yüksek talep görmektedir ve temin riski altındadır. NTE'lerin sürdürülebilirliği için, yenilikçi teknolojiler göz önünde bulundurulmalıdır. Dünya genelinde doğal NTE kaynakları sınırlıdır ve birkaç ülke arasında eşit olmayan bir şekilde yayılmıştır. Bu ülkelerden Çin, bu kaynakların en büyük payını elinde tutuyor ve küresel NTE talebinin % 90'ına denk geliyor. Doğal NTE kaynaklarının geri kalanı ABD, Hindistan ve diğer birkaç ülke arasında bölünmüştür. Bu da Türkiye gibi birçok ülkede talebin istikrarlılığının sağlanmasında bir zorluk oluşturmaktadır. Her ülke küresel markette rekabetçi konumda kalmak ve ülke içerisinde büyümeyi teşvik eden kullanışlı teknolojilerin üretiminden emin olmak için NTE’lere ihtiyaç duyar çünkü NTE için tamamen diğer üretici ülkelere güvenmek riskli ve güvencesiz bir durumdur. Ayrıca benzer fiziksel ve kimyasal özelliklere sahip olmalarından dolayı NTE’lerin gerek diğer elementlerden çıkartılmaları gerekse bireysel olarak saflaştırılmaları oldukça zordur. Türkiye gibi birçok ülkede çok sayıda doğal kaynak bulunmadığından, alternatif yöntemler göz önünde bulundurulmalıdır. Bunun için en umut verici tekniklerden biri, çeşitli endüstrilerden gelen atıkların örneğin e-atıkların, termik santral küllerinin ve madencilik faaliyetlerinden kaynaklanan atıkların NTE’lerin geri kazanımı için kullanılmasıdır.
Bu çalışmada Türkiye'nin İç Anadolu Bölgesi’nde bulunan demir madeni çamurundaki NTE’lerin potasiyeli ve geri kazanımı araştırılmıştır. Numune, öncelikle bir gece boyunca 105 ° C'de kurutularak nemi uzaklaştırılmıştır.
xxx
Kurutulan numuneler bir sonraki aşamada mikrodalga yöntemi ile parçalanmıştır. Mikrodalga daha az kimyasal tüketerek daha verimli bir çözünürleştirme gerçekleçtirdiği için daha az çevresel etkilere sebep düşünülmektedir. 0,1 g kuru numune hassas tartı ile tartılmıştış ve çeşitli asit türleri ilave edilmiştir. 5 aşamalı bir programla parçalamanın gerçekleşmesinde esas parçalama 210oC’de 30 dakika boyunca gerçekleştirilmiştir. HF ilave edilen çözünürleştirmelerde ikinci aşama olarak H3BO3 ilave edilerek, HF’in cam malzemeler ve özellikle ICP üzerindeki aşındırma etkileri elimine edilmiştir. Kızıl çamur liçinde elde edilen major ve iz element konsantrasyonlarına göre en yüksek konsantrasyon beklenildiği gibi Fe elementinden kaynaklanmaktadır. Sırası ile diğer yüksek konsantrasyonlara sahip elementler Al, Mg, K, Ca, Li, Ni, Ti ve Mn’dır. NTE’lerden ise en yüksek konsantrasyonlara sahip olan ve kritiğe yakın ve kritik olan Ce, La, Y ve Nd elementleri seçilmiştir. Ardından liç çözeltilerinden elde edilen konsantrasyonlara göre sentetik liç çözeltisi hazırlanmıştır. Burada seçilen dört adet NTE konsantrasyonları 60 ppb olarak standardize edilmiştir ve 1000 ppm’lik standart çözeltilerden sıvı olarak eklenmiştir. Diğer major elementler ise her bir elementin metal tuzları olarak çözeltiye ilave edilmiştir. İki farklı asit içeriğine sahip sentetik liç çözeltisi hazırlanmıştır. Birinci çözelti (M1-1) litre başına 120 mL HCl ve 80 mL HNO3 içeren yüksek asitli, ikinci çözelti (M1-3) ise litre başına sadece 20 mL HNO3 içeren düşük asit içerikli çözeltidir. Hazırlanan sentetik çözeltiler ön arıtmadan geçtikten sonra nanofiltrasyon ile konsantre edilip, ön arıtma yapılmamış hali ile de destekli sıvı membran prosesine tabi tutulmuştur.
Ön arıtma için çözelti pH’ı NaOH kullanılarak 1,5; 2,5 ve 3,5’e yükseltilip MF membranından süzülmüştür. Ön arıtmadan çıkan besleme çözeltisi daha sonra NF prosesine tabi tutulmuştur. NF membranları kullanılmadan önce 24 bar basınç altında saf su geçirilerek 1 saat boyunca sıkıştırılmıştır. pH 1,5; 2,5 ve 3,5 ön arıtma çıkışları 12, 18 ve 24 bar altında NF membranında süzülerek konsantre edilmiştir. DSM prosesi için polimerik ve gözenek çapı 0,45 um PVDF membranı ile sıyırma fazı olarak 5 M konsantrasyona sahip HNO3 çözeltisi kullanılmıştır. Asidik liç çözeltisinden daha yüksek asit konsantrasyonuna sahip sıyırma çözeltisiyle, kimyasal taşınımın ve ayırma işleminin verimi arttırılmıştır. DSM için üç farklı senaryo kullanılmıştır. Senaryo I: kerosen içerisinde Cyanex 272, Senaryo II: kerosen içerisinde 18-Crown-6 ve Senaryo IIO: kerosen içerisinde D2EHPA içeren organik fazlardan oluşmaktadır. Membranlar kullanılmadan önce tüm gözeneklerin organik çözelti ile doldurulduğundan emin olmak için bu organik çözeltiler içerisinde bekletilmiştir. Besleme ve sıyırma fazların 250 mL çözelti alınarak 600 rpm’de karıştırma sağlanmıştır. Reaksiyon süreleri ise 3 saatten 24 saate kadar değişmiştir.
XRF, demir madeni çamuru veya kızıl çamuru karakterize etmek için kullanılmıştır fakat XRF belirli bir limitin altındaki konsantrasyona sahip elementleri belirleyemediğinden genellikle NTE analizi için uygun değildir. Fakat numunedeki diğer major elementlerin belirlenmesi için XRF analizi yapılmıştır.
Yukarıda belirtilen NF çalışmalarına göre M1-3 asit liçi ile daha düşük çevresel etkiler ile M1-1 liçine yakın geri kazanım oranları sunmuştur. Genel olarak, M1-3 liçi ile daha düşük akı değerleri fakat daha yüksek konsantrasyon oranları elde edilmiştir. M1-1 için 3,5 ön arıtma ile NF çalışması yapıldığında verim oldukça düşmüştür. Bunun sebebi, asit liçinin negatif değerlerdeki başlangıç pH’sının 3,5’ e kadar çıkartılması için gerekli yüksek miktarda NaOH’tır. M1-3 için gerekli NaOH miktarı daha az olup toplam konsantrasyon oranı %98’in üzerindedir.
xxxi
M1-1 ve M1-3 asidik liçleri için kullanılan MF ve NF membranlarının SEM analizleri ile yüzey morfolojileri incelenmiştir. M1-1 çözeltisinin 1,5 pH’da ön arıtımı için kullanılan MF membranın temiz kaldığı ve herhangi bir çökeltinin olmadığı yüzey analizi sonucunda görülmektedir. EDX sonuçlarında pH 1,5 membranında sadece membran malzemesinde bulunan elementler görünürken, pH 2,5’te Mg, Al, K, Ca ve Fe gözlemlenmiştir. pH 3,5’te ise sadece Fe elementi görülmektedir, çünkü kızıl çamurdaki en yüksek konsantrasyona sahip element Fe’dir. M1-3 asit liçi ile yapılan çalışmalarda ön arıtma sonrasında daha az kimyasal çökelek oluştuğundan MF membranları görece daha temiz kalmışlardır. pH 2,5 ön arıtması için kullanılan MF membranları EDX analizi yapıldığında membran yüzeyinde Na, Mg, Al, K, Ca, Fe ve Zn elementleri görülmüştür. Bu da, kimyasal çökeltim ile elementlerin MF membran yüzeyinde tutulduğunu göstermektedir. NF membran yüzeyi daha pürüzsüz olup gözenek çapı oldukça düşüktür. pH değeri 1,5 olan asidik liçin filtrasyonundan yüksek oranda Fe gelirken, pH değeri 2,5’ yükseldiğinde çökelen demirin azaldığı gözlenmiştir. pH’ı 3,5 olan çözeltinin NF’ten süzülmesi durumunda, MF membranında tutunamayan kimyasal çökeleklerin NF yüzeyinde tıkanmaya yol açtığı fakat diğer düşük pH değerlerindeki gibi bunun sürekli bir tıkanmaya sebep olmadığı görülmüştür.
pH ve basıncın giderim üzerindeki etkilerini belirlemek için NF sonuçlarına istatistiksel analiz yapılmıştır. Bu analiz sayesinde, M1-3 liç çözeltisi için optimum işletme parametrelerinin pH 3,5 ve 24 bar olduğu görülmüştür. Fakat, işletme basıncının etkisinin düşük olduğu görüldüğünden, NF prosesi 12 bar basınç altında da işletilebilir.
Farklı mobil taşıyıcılara sahip organik fazlar, bu taşıyıcıların farklı konsantrasyonları ve farklı reaksiyon süreleri denenerek 7 farklı DSM çalışması yapılmıştır. 0,3 M D2EHPA içeren DSM-3’ün 3 saatlik reaksiyon süresi sonunda en yüksek Ce, La, Nd ve Y kütlesel akıları elde edilmiştir.
Genel olarak bu çalışma ile NTE’lerin geri kazanımı ve tekrar kullanımı için kızıl çamurun kullanılabilir bir kaynak olduğu görülmektedir. NTE’ler az asit ile suya geçirilip, membran prosesler ile konsantre edilebilmektedirler.
PH ve basıncın çıkarılma üzerindeki etkilerini belirlemek için NF adımlarının sonuçlarının istatistiksel analizi yapıldı. Bu analiz sayesinde M1-3 liç çözeltisi kullanılarak optimum çalışma koşullarının 24 bar'da pH 3.5'te olduğu görülmektedir. Bununla birlikte, çalışma basıncının etkilerinin düşük olduğu görülmektedir, bu nedenle, NF prosedürünün çalışması da 12 bar'da gerçekleşebilir.
Konsantrasyondan sonra organik çamur çözeltisinde bulunan kimyasal maddeyi, organik faz içindeki konsantrasyonu ve reaksiyon süresini değiştirerek kırmızı çamurla 7 SLM çalışması gerçekleştirilmiştir. 3 saatlik reaksiyon süresi ile 0.3M D2EHPA kullanan SLM-3, Ce, La, Nd ve Y için en yüksek kütle akısı değerlerini gösterdi. Genel olarak, bu çalışma yoluyla, kırmızı çamurun NTE'lerin geri kazanımı ve yeniden kullanımı için uygun bir kaynak sunduğu görülebilir, elementler sınırlı asit liçi kullanılarak ekstrakte edilebilir ve membranlar kullanılarak konsantre edilebilir.
1 1. INTRODUCTION
1.1 Purpose of Thesis
Rare earth elements (REE) are a group of elements that include the elements of Yttrium and Scandium, in addition to lanthanides and have a variety of important applications in the fields of modern and green technology. As the People’s Republic of China currently has hold over 97% of the world’s REE resources, which are vital in the production of superconducting materials, fiber optics and laser products, many of the world’s authorities have mobilized to create strategies to ensure independence in obtaining REE raw materials through recovery from natural resources, waste and wastewater. According to recent strategy reports from the European Union Commission and the United States Department of Energy, REE recovery is at the top of the list of critical elements, and the interest in its procurement is rapidly increasing. The aim of the project is to ensure the recovery of REEs from red mud, which is usually disposed of, using membrane technology, an emerging technology that is gaining importance every day. As there is a lack of a comprehensive studies of the various waste and wastewater sources in Turkey, in terms of REE contents, this only reveals the national relevance of this project. Additionally, this project also holds importance on a global scale as REE recovery has not been comprehensively examined.
The objectives planned to be achieved in order to study REE recovery from waste and to establish the most feasible method with an integrated approach can be summarized as follows:
After the collection of waste from the designated source (red mud), the main elements as well as REE contents are determined in full detail
To increase efficiency by reducing the feed flow using nanofiltration after pretreatment and reducing the field/volume needs and energy costs of supported liquid membrane
2
In the recovery step, the selective separation of REE from other elements at high yields and the purity of various recovered solutions containing REE are revealed and evaluated
In conclusion, the aim is to recover REEs from red mud, through using membrane technologies, following the characterization of REEs within the scope of this research. In this way, awareness of the potential REE resources and methods of their recovery in Turkey will be made. Through these experiments, REE recovery can be optimized. In addition, these experiments can provide examples and add to the research being conducted about REEs.
1.2 Scope of Thesis
REEs from red mud in Turkey are to be recovered. There is a growing need in the energy, electronics and defense industries for REEs, and growing uncertainty of REE supply has encouraged nations worldwide to establish a stable REE resource from within their own countries. In view of the current monopoly held by China and the likely increase in demand, the introduction of national REE resources has become essential. Although the exploration of new REE mines is a step that can be considered, there are many disadvantages to REE mining in terms of the economy and the environment; it leads researchers to unconventional, unusual, and more environmentally sustainable resources.
Two methods will be analyzed for concentration and separation, based on membrane technology: 1) Direct nanofiltration and 2) Supported Liquid Membranes. Two different methods were preferred to discuss the effect on the overall yield of recovery. In Chapter 2, we will discuss REEs in general, looking into their chemical and physical properties, their uses, their market values, and why they are important for our future. We will also briefly discuss the source of REEs being used in this paper, as well as difficulties involved in the measurement and extraction of these REEs from the source, followed by a brief description of the steps used in the scope of this thesis.
3
Chapter 3 is related to the materials and methods utilized in the course of this thesis study, discussing, in some detail, leaching processes, concentration steps and supported liquid membranes.
Chapter 4 presents the results of the experiments carried out: the initial waste characterization, filtration and removal performances of NF membranes, as well as mathematical analysis of this procedure, SEM-EDS results of the NF and MF membranes used, and SLM results.
Chapter 5 provides a discussion and conclusion to the study conducted.
1.3 Hypothesis
Red mud from Turkey holds the potential to remove various REEs using SLM and nanofiltration techniques.
5 2. LITERATURE SUMMARY
2.1 General Information about REE
Rare Earth elements (REEs) consist of 14 elements, known also as lanthanides. These lanthanides are: Cerium (Ce), Praseodymium (Pr), Neodymium (Nd), Promethium (Pm), Samarium (Sm), Europium (Eu), Gadolinium (Gd), Terbium (Tb), Dysprosium (Dy), Holmium (Ho), Erbium (Er), Thulium (Tm), Ytterbium (Yb) and Lutetium (Lu). In addition to these lanthanides, Lanthanum (La) and Yttrium (Y) are also classified as REE (Charalampiddes et al., 2015). Whether Scandium (Sc) can be classified as a rare earth element is under debate; however, Schüler et al. prepared a report including the element Scandium in the REE group, as Sc tends to be found in the same ore deposits as the other lanthanides and exhibits similar chemical properties to the other REEs (Schüler et al., 2011, Zepf, 2013). Therefore, in this article, it has been included under the category of REEs.
Figure 2.1 : Placement of rare earths in the periodic table (University of Liverpool, 2017).
6
REEs can be divided into two classes: light REEs (light REE, LREE, lREE) and heavy REEs (heavy REE, HREE, hREE). In some cases, the middle REEs (medium-middle REE, MREE) class is also used. There is no distinct delineation by which the elements are sorted into these classifications, and throughout a range of articles, the same element can be seen to be sorted between the different classes. All REEs have similar chemical and physical properties which makes it possible for REEs to be used in various application areas. For example, Gd, Dy, Er, Nd and Sm are used to make magnets with their magnetic properties, whereas other REEs are used in lighting and laser applications due to sharp energy level differences (Mortimer and Müller, 2001; RÖMPP, 2011). Examples of grouping REEs according to various sources, based on their properties, are given in Table 2.1 and Figure 2.2.
7
Table 2.1 : Chemical structure properties of REEs (University of Liverpool, 2017). Element
Z A
Electron
Configuration Radius / pm
Crystal
Structure Lattice Constant
Ionic Metallic a/pm c/pm c/a
Scandium Sc 21 45 (3d4s) 3 78.5 164.1 Hcp 330.9 526.8 1.592 Yttrium Y 39 89 (4d5s) 3 88.0 180.1 Hcp 364.8 573.2 1.571 Lanthanum La 57 139 4f0(5d6s)3 106.1 187.9 Dhcp 377.4 1217.1 3.225 Cerium Ce 58 140 4f1(5d6s)3 103.4 182.5 Fcc 56.1 - - Praseodymium Pr 59 141 4f2(5d6s)3 101.3 182.8 Dhcp 367.2 1183.3 3.222 Neodymium Nd 60 144 4f3(5d6s)3 99.5 182.1 Dhcp 365.8 1179.7 3.225 Promethium Pm 61 145 4f4(5d6s)3 97.9 181.1 Dhcp 365 1165 3.19 Samarium Sm 62 150 4f5(5d6s)3 96.4 180.4 Rhom 362.9 2620.7 7.222 Euphemism Eu 63 152 4f7(5d6s)2 95.0 204.2 Bcc 458.3 - - Gadolinium Gd 64 157 4f7(5d6s)3 93.8 180.1 Hcp 363.4 578.1 1.591 Terbium Tb 65 159 4f8(5d6s)3 92.3 178.3 Hcp 360.6 569.7 1.580 Dysprosium Dy 66 163 4f9(5d6s)3 80.8 177.4 Hcp 359.2 565.0 1.573 Holmium Ho 67 165 4f10(5d6s)3 89.4 176.6 Hcp 357.8 561.8 1.570 Erbium Private 68 167 4f11(5d6s)3 88.1 175.7 Hcp 355.9 558.5 1.569 Thulium Tm 69 169 4f12(5d6s)3 86.9 174.6 Hcp 353.8 555.4 1.570 Ytterbium Yb 70 173 4f14(5d6s)2 85.8 193.9 Fcc 548.5 - - Lutetium Lu 71 175 4f14(5d6s)3 84.8 173.5 Hcp 350.5 554.9 1.583 Neodymium Nd 60 144 4f3(5d6s)3 99.5 182.1 Dhcp 365.8 1179.7 3.225 Promethium Pm 61 145 4f4(5d6s)3 97.9 181.1 Dhcp 365 1165 3.19
8
9 2.2 REE Composition
The 12 dominant elements in the Earth's crust are: O, Si, Al, Fe, Ca, Mg, Na, K, Ti, H, Mn and P. These elements make up 99.23% of the earth's mass. Therefore, all metals with a mass ratio of less than 0.1% on earth can be considered as rare metals (Skinner, 1976). Thus, although, REEs can be considered “rare metals,” they are included within a large group of numerous other qualifiers.
The amounts of cerium, neodymium and lanthanum in the Earth's crust is about 66, 40 and 35 ppm respectively, while Thulium can be described as the rarest REE, with the lowest concentration present, 0.5 ppm, in the Earth’s crust. Most REEs, however, are not truly rare. Rather, they exist in various forms and are widely dispersed (Charalampiddes et al., 2015). Some sources even say they are more abundant than gold (Hurst, 2010, Rudnick and Gao, 2003). Table 2.2 and 2.3 shows the amounts of various elements in the Earth's crust throughout the different layers.
10
Table 2.3 : REE content in the Earth's crust (USEPA, 2012).
A main reason for the “rarity” of rare earths is that they do not exist alone in the Earth's crust; they are never found pure and are found in conjunction to other metals and REEs and are scattered around the world (Elsner, 2010). For example, when data from various studies is examined, it is seen that the amount of neodymium in the earth's crust is greater than the lanthanum (Allegre et al., 2001; Rudnick and Gao, 2003; Emsley, 1994). According to this data, while it is possible to obtain more neodymium in practice through mining, lanthanum is more readily available as neodymium remains quite dispersed and cannot be aggregated (Stosch, 2002).
In order to technically define whether an element is rare or not, it is necessary to look at whether its mining is feasible. Accordingly, Skinner coined the term "mineralogical barrier" (1976). This barrier determines the level at which the mining of the related element can be economically beneficial. Ore preparation processes for REE acquisition are quite complex and kept private.
Many physical and chemical processes are required for REEs to be separated from each other. All procedures have a significant negative impact on environmental health, for example, the leakage of hazardous chemicals, used in all stages of these processes, pose problems in toxicity (Zepf, 2013; Gupta and Krisnamurthy, 2005; Huang et al.,
Elements Crustal Abundance (parts per million) Elements Crustal Abundance (parts per million)
Nickel (38Ni) 90 Gadolinium 4.0
Zinc (30Zn) 79 Dysprosium 3.8
Copper (29Cu) 68 Tin 2.2
Cerium (58Ce) 60 Erbium 2.1
Lanthanum (57La) 30 Ytterbium 2
Cobalt (27Co) 30 Europium 1.3
Neodymium (60Nd) 27 Holmium 0.8
Yttrium (39Y) 24 Terbium 0.7
Scandium (21Sc) 16 Lutetium 0.4 Lead (82Pb) 10 Thulium 0.3 Praseodymium (59Pr) 6.7 Silver 0.08 Thorium (90Th) 6 Gold 0.0031 Samarium (62Sm) 5.3 Promethium 10-18
11
2005). Economic mining also depends on the technology to be used and market prices for the retained minerals.
There is a shortage of REEs in the world as there are difficulties in economic mining (Zepf, 2013; WTO, 2010). Currently, REE mining is largely under Chinese control; this has been a large motivator to encourage the production of new solutions and alternatives, especially in the recent decades. As of 2010, worldwide reserve ownership and production are shared among countries as follows: China 97%, India 2% and Brazil 1% (USGS). Countries are working to find solutions because a possible scarcity of REEs in the future could present a huge problem, as they are vital and irreplaceable in the production of many modern technologies.
2.3 Uses, Economic Status and Global Trade
REEs are crucial for modern and traditional industries. Currently, the People's Republic of China has almost complete hold over the REEs market (Zhanheng, 2011). Table 2.4 lists the specific and important uses of REEs. Examples of uses include hybrid vehicle production, charger batteries, mobile phones, plasma televisions and disc drives. The industrial need for critical REEs is increasing day by day (Izatt et al., 2010).
12
Figure 2.3 : Distribution of REE quantity between countries producing REE mines between 2010 and 2013 (Charalampiddes et al., 2015).
Figure 2.4 : Critical materials strategy for 2025 (left)-right for 2050 (USA, 2011). Although REEs have similar chemical characteristics, each element is used to produce different materials. Their uses have also changed throughout history. Neodymium is a prime example; in the early 20th century, it was used almost exclusively in the glass industry. Today, it is better known for its role as the main component in magnet production. Although, each REE has a purpose and usage on its own, as is shown in Figure 2.5 and 2.6; there are also many important products that are formed with a combination of REEs. Knowing which REEs are used in conjunction with others is also a factor in determining which REEs are critical in terms of need and availability. Furthermore, when the uses of REEs are examined, it can be seen that they can play a
13
crucial role in the production of more environmentally friendly materials. For example, they already have a significant role in reducing energy consumption, as they enable the production of lamps, LEDs and electric-powered hybrid vehicles. Some REE alloys also reduce fuel consumption by reducing vehicle weight. The production of UV-protected glass, containing REE, reduces the energy use in buildings. Future forecasts for REE usage in green technologies are given in Figure 2.7.
Figure 2.5 : Worldwide REE production and use in 2016 (Zhou et al., 2017).
Figure 2.6 : The interoperable status of NTIs in their areas of use (Chegwidden and Kingsnorthi, 2010; Lynas, 2010).
14
Figure 2.7 : Future forecasts for the quantity of rare soil oxides used in clean technologies (Zhou et al., 2017).
While rare earths were previously used in more traditional industries, such as catalysis, glass making, lighting and metallurgy industries, today, they play key roles in new and rapidly growing industries, like the battery alloy, ceramic permanent magnet, and as previously discussed, clean technology industries. In traditional industries, lanthanum and cerium are mostly used, while new industries use dysprosium, neodymium and praseodymium (Charalampiddes et al., 2015). In 2008, 129,000 metric tons of rare earth element (in oxide form, rare earth oxide-REO) were used worldwide (Cordier and Hedrick, 2010).
The supply-demand balance of rare earths in the market is highly unstable. However, since 2007, prices of all REE varieties have increased (USCRS, 2010). REE price data and changes over the years are given in Figure 2.8 and 2.9. REE prices are currently on an upward trend. The increase in prices has led to a focus on two new strategies for REEs, particularly in the European Union and the United States. One of the strategies is to focus on the discovery of new sources of REEs within their own regions and the other is to recover and reuse products already containing REEs. REEs are usually sold in oxide forms in the market, as it can often be extremely difficult to obtain them as pure metals. For this reason, when researching prices, usually, the prices of rare soil oxides can be found. In Table 2.5, rare soil oxides and rare earth elements are
15
compared as of 2010. The prices of critical REEs that are used in green technologies are shown in Figure 2.10.
Figure 2.8 : Comparison of REE prices with gold price (Reuters, 2010). Table 2.5 : Comparison of prices of rare earth metals and rare earth oxides (Schüler,
2011) (Schüler, 2011). Rare Earth Metal Price ($/kg) Rare Earth Metal
Oxide Price ($/kg) Cerium 43-55 Cerium 59 -62 Dysprosium 372-415 Dysprosium 284-305 Europium 710-800 Europium 84-94 Gadolinium 53-56 Gadolinium 585-605 Lanthanum 42-46 Lanthanum 43-46 Neodymium 97-100 Neodymium 55-58 Praseodymium 84-106 Praseodymium 71-80 Samarium 44.50 – 53 Samarium 33-35 Terbium 750 – 792 Terbium 595-615 Yttrium 61-63 Yttrium 53-70
16
Figure 2.9 : Prices of rare earths that are critical and used in green technologies (Humphries, 2012).
For the scope of this project, Lanthanum, Cerium, Yttrium, and Neodymium have been selected as they have high value and range from critical to near-critical in demand-supply risk.
2.4 Red Mud
Rare earth elements are critical elements for our future.
China's monopoly on natural REE sources has encouraged all countries around the world to turn to REE recovery from waste sources within their own nation.
The type of products present in the source waste to be used in recovery, the characteristics of the geography in which the product and waste are formed, and the basic processes in the process of waste generation and collection are taken into consideration.
A mineral waste that can be selected in Turkey is the red mud formed in iron mines (Binnemans et al., 2013; MTA, 2018c).
Bauxite, or red mud as it is more commonly called, is an emerging resource for REE extraction. Approximately 2.7 billion tons of red mud is generated globally since 2007, with a yearly increase of 120 million tons (Ujaczki et al., 2017). In one location in Turkey, about 18000 tons are dumped annually (Erçağ and Apak, 1997). Red mud
17
contains a high percentage of Fe, among other elements, in addition to REEs, which range from 500 to 1700 ppm. (Akcil et al., 2018). According to Liu and Naidu (2014), scandium is present in double the quantity in red mud, as is found in their natural ores. A few studies have been conducted on the removal of these REEs from red mud in various countries, such as India, Greece, China and Turkey. Table 2.6 shows the composition of red mud from various countries (Akcil et al., 2018). Although red mud, produced in vast quantities every year, presents a valuable source for REE recovery, there is a lot of unexplored areas of study about this topic. This paper will attempt to discover the recoverability of REEs, specifically La, Ce, Nd, and Y from red mud from Central Anatolia in Turkey.
Table 2.6 : REE composition of red mud worldwide (Akcil et al., 2018).
2.5 Recovery and Reuse
2.5.1 Determination of rare earth elements by spectroscopic methods
The similarity between REEs can make the determination of these elements very difficult and complex. Serious problems arise when selected REEs need to be identified from a mixture of other REEs. The most frequently used instruments for determining REEs are the ICP-MS, ICP-OES, XRF (X-Ray Fluorescence) and NAA (Neutron Activation Analysis). In the case of ICP-OES, solid samples cannot be analyzed directly, but this method is often preferred because it enables rapid multi-element analysis over a wide range of concentrations. However, the concentration of REEs in samples is usually below the measurement limits, and large-share components such as inorganic salts and organic compounds can cause matrix effects.
Using an ICP-MS has become one of the most powerful techniques for the analysis of lanthanides, as it has high sensitivity, wide dynamic linear range, multi-element measurement and the possibility to measure isotopes. The ICP-MS provides simpler spectrums and lower measurement limits compared to the ICP-OES method, which generally prevents matrix separation and enables REE measurement in soil samples.
18
Interventions in ICP-MS can be eliminated in a variety of ways, including high separation power, reaction/collision cells and separation. As with most instruments, mathematical correction is also required.
REE analysis in challenging samples is difficult not only in terms of instrumental analysis, but also in terms of separation and subsequent pre-concentration steps. All methods can be applied for the separation of REEs from other non-lanthanide elements, including those using solid samples, such as XRF. The separation of REEs from other REEs requires more complex methods. The REE analysis method depends not only on the type of material, but also on the quantity of REEs present. Trace amounts (ppm) of REEs usually do not require separation and pre-concentration. REE analysis of very small amounts (ppb level), however, usually requires an appropriate pre-concentration method. Typically, this stage of analysis consumes the most chemicals, time and cost. Direct analysis of samples and REEs measurements are also becoming increasingly important. Through direct analysis, the use of aggressive reagents is eliminated, and micro-samples can be analyzed. This type of determination method can be applied for rapid scanning analysis.
The XRF can perform a direct analysis of the materials. This method is frequently used in the analysis of multiple elements containing REEs. XRF plays an important role in metal and ceramic industries, mineral and geological exploration and environmental monitoring. Many applications of XRF do not require any processing of materials, containing trace amounts of REE. Unfortunately, the typical determination limits of XRF for REE are not satisfactory for a variety of applications, such as analysis of environmental or biological samples, as in those cases, REEs are found in the ppm range.
In conclusion, ICP-MS, ICP-OES, and XRF techniques are the most commonly used techniques in REE analysis, as they allow multiple elemental analysis. However, complex matrices, and especially contamination during sampling, preservation and pretreatment of the sample, make it difficult to analyze trace amounts of REEs. Two approaches are used to solve this problem. In case of direct analysis, REE measurements are performed without any purification or sample dilution. The advantages of direct analysis are the minimization to sample preparation process and
19
a reduction in potential contamination. The second approach is to apply separation and enrichment steps prior to analysis to reduce or eliminate spectroscopic and non-spectroscopic interventions.
2.5.2 Pretreatment
Red mud samples are subjected to moisture removal at 105oC for one night and crushed to approximately 250 µg.
2.5.3 Leach preparation
The first step in element recovery from red mud is to remove the elements from the solid form and incorporate them into a liquid form. This is usually done by leaching. Leaching is carried out using large amounts of acid and/or base. When high temperature and pressure conditions are added, the efficiency of leaching process increases. Leaching can be carried out by conventional methods or by modern techniques such as using a microwave.
In recent years, the use of microwaves in various fields has been increasing due to its high efficiency and lack of environmental impact. Fast heating and reaction occur with the use of microwave (Ku et al., 2003). Particularly in the leaching of rare earth elements, it has been observed that microwave heating is more efficient than conventional heating (Gopalakrishnan et al., 2006). Lu et al. (2003), in their study, extracted REEs, EDTA, CH3COOH, HCl and CaCl2, from solid to liquid using a microwave. They aimed to optimize the heating time and microwave power and compared the results obtained with a power of 60% and extraction time of 30 minutes with the conventional extraction schemes in the literature. In this study, including soil samples, they showed that they can achieve similar results obtained by conventional methods, with microwaves, without environmental pollution and low cost. In 2007, Kulkarni et al. used HNO3, HF and H3BO3 to achieve efficient extraction, as well as trace and even ultra-trace concentrations of La, Ce, Pr, Nd, Sm, Gd, Eu and Dy elements by ICP-OES.
Published studies on leaching of REEs discuss mainly the leaching of electronic waste and mine waste. It is stated in the literature that the most optimal leaching method is the use of microwave, due to less chemical consumption. In addition to the low chemical consumption, leaching efficiency of the elements was also observed to be
20
higher (Haque, 1999). However, the conditions under which leaching of REEs are more efficient have not been established with definite conclusions and defined lines. In general, high temperature increases leaching efficiency. As the leaching kinetic rate increases, the process ends earlier. Due to the high leaching kinetic speed, less residue remains at the end of the process. Thermal stress on mineral surfaces increases the surface area (Al-Harahsheh and Kingman, 2004).
In a study performed with phosphogypsum, it was concluded that optimum REE extraction was at low power (600 W) and short time (5 minutes) or high power (1200 W) and long time (15 minutes). This technique is highly sustainable because it produces virtually zero waste. The only problem that researchers have focused on using microwave methodology is how to do this when a real-scale plant is installed. Bradshaw et al. also stated that the cost of heating a metal at room temperature by heating to 85 ° C was $ 8.3 / ton while the cost of microwave treatment was $ 0.18-0.94 / ton (2005).
2.5.4 Concentration using membrane processes Concentration is an important step in the recovery of critical metals. The concentration process means the separation of liquid from the mixture and the concentration of critical metals. Concentration processes can be physical, chemical or biologically based processes. The most preferred mechanism is the chemical separation / concentration mechanism. The most commonly used chemical concentration technique is ion exchange. However, in the case of the high operating and environmental costs of chemical methods, membrane processes stand out.
2.5.5 Nanofiltration
The nanofiltration (NF) process is preferred since it has higher flux than the reverse osmosis process and can be operated safely at lower pressure values. For example, Murthy and Gaikwad (2013) was able to separate the praseodymium element from an acidic solution with a high yield of 89% using an NF membrane. NF is a sustainable option for metal removal from mixed solutions due to lower energy requirements and no need for chelates such as supported liquid membranes (SLM) (Shon et al., 2013).
21 2.5.5.1 Operating Pressure
NF allows the separation of the main monovalent ions (such as nitrate, chloride and sodium) from acidic solutions. The main criteria that determines the performance of the NF process is that the filtrate flow can be maintained at the desired level continuously during the filtration of the solution (Fane et al., 2011). The applied operating pressure increases the permeate flux and accelerates clogging. The effect of increasing operating pressure on the removal of metals is quite complex. Yield is mostly dependent on the characterization of the solution (Kovacs and Samhaber, 2008; Aydıner et al., 2010; Cathie Lee et al., 2014). Figure 2.10 shows the effect of operating pressure on performance.
2.5.5.2 pH
The pH of the solution is one of the most important factors because it affects both flux and removal by changing the charge of functional groups on the membrane surface (Mullet et al., 2014; Tu et al., 2011). As the pH decreases, the number of functional groups with positive valence increases. This creates a positive zeta potential (ZP). Positive ZP increases the removal efficiency of cationic metal ions. Therefore, it is necessary to carry out an optimization study for the relevant solutions in the NF process operation and to determine the optimum conditions.
22
Figure 2.10 : Effect of operating pressure on NF performance: (a) effect on flux with pH and (b) singular effect on flux and removal (Kose-Mutlu et al., 2018). 2.6 REE Separation Techniques with Membrane Processes
Rare earth elements are a group of 17 elements with similar physical and chemical properties. According to Web of Science's classification results (Figure 2.11), the most studied REEs with membrane separation technologies in the last 30 years are Nd, Y, La and Pr, while the heavy REEs such as Tb and Tm, Yb, and Lu are the least investigated; these results exclude the radioactive element, Pm. The prices of heavy REEs are higher due to limited resources and difficult extraction processes from their mine origins.
(a)
23
Figure 2.11 : Number of publications of each REE with membrane separation technology over the last 30 years (Chen et al., 2018).
As mentioned in the previous sections, the rare earth elements are essential components of the modern industry, used in many devices that are closely related to human life. Catalysts, magnets, battery alloys, ceramics, among others, are listed as the most commonly used products by REE (Jordens et al., 2013). Demand for REEs continues to grow, as there are few alternatives to REEs, particularly in the low carbon and renewable energy industries such as in the creation of wind turbines and batteries of mobile electronics (Massari and Ruberti, 2013). Unfortunately, industrial separation of REEs from mineral ores does not seem as clean and environmentally friendly as their final use (Wang et al., 2013). Separation of REEs is listed as one of the seven chemical separation processes that will achieve great global benefits if improved (Sholl and Lively, 2016).
The most common treatment methods used so far, multi-stage solvent extraction, releases huge amounts of acidic and alkali wastewater and results in only a small amount of resources (Chen et al., 2018). To address these concerns, membrane separation techniques stand out as a hybrid process that combines extraction and stripping simultaneously without thermal heating processes such as distillation, drying or evaporation. In addition to energy savings, it does not require large amounts of volatile organic compounds, making it a promising environmental separation / separation process. Since Li invented liquid membranes (LM) for the separation of hydrocarbons in 1968, the development of liquid membranes in separation technology has continued to progress. Compared to conventional solvent extraction, liquid membranes are very efficient for pre-concentration and separation of metal ions thanks to their large mass transfer surface area (Tavlarides et al., 1987).
