Sintering Optimisation and Recovery of Aluminium and Sodium From

1 Article 2 Sintering Optimisation and Recovery of Aluminium 3 and Sodium from Greek Bauxite Residue

4 Pritii W.Y. Tam 1,2*, Dimitrios Panias 2 and Vicky Vassiliadou 1

5 1 Department of Continuous Improvement and Systems Management, Aluminium of Greece plant, 6 Metallurgy Business Unit, Mytilineos S.A., Agios Nikolaos, 32003 Boeotia, Greece; [email protected]; 7 [email protected] 8 2 School of Mining and Metallurgical Engineering, National Technical University of Athens, 9, Iroon 9 Polytechniou, 15780, Zografou Campus, Athens, Greece; [email protected] 10 * Correspondence: [email protected], Tel.: +61 4 8109 0871

11 Received: date; Accepted: date; Published: date

12 Abstract: Bauxite residue is treated for the recovery of aluminium and sodium by sintering with the 13 addition of soda, metallurgical coke and other reagents such as CaO, MgO and BaO. A thorough 14 thermodynamic analysis using Factsage™ software was completed together with XRD mineralogy 15 of sinters with different fluxes and reagents additions. Through both thermodynamic interpretation 16 and mineralogical confirmations, it was observed that the type of desilication product in bauxite 17 residue influences the total aluminium recovery through sintering process and formation of sodium 18 aluminium silicate exists in equilibrium with sodium aluminate, unless silica is consumed by 19 additives (such as CaO, MgO, BaO etc) forming other more thermodynamically favourable species 20 and liberating alumina. Addition of barium oxide improves the aluminium and sodium recovery to 21 75% and 94% respectively. Complex sinter product formation that are triggered due to high calcium 22 content in the Greek bauxite residue reduces aluminium recovery efficiency. Optimised and feasible 23 recovery of aluminium and sodium for Greek bauxite residue was proved to be 70% and 85% 24 respectively, when sintered with 50% excess stoichiometric soda. It was observed that stoichiometric 25 carbon addition in inert atmosphere only assisted recovery up to 75% of aluminium and 83% of 26 sodium, though there are benefits to pre-reduce iron from hematite for downstream recovery.

27 Keywords: Bauxite residue; sintering; aluminium; sodium; recovery 28

29 1. Introduction 30 The aluminium sector has seen an increase of demand and production within the recent years 31 with current annual estimate in 2017 being 132,390 thousand metric tonnes of alumina [1]. The global 32 market shows no signs of reduction with China’s competitive focus in alumina production estimated 33 surpassing more than half of annual global production in 2017 [1]. Meanwhile, West Europe reports 34 about 5,890 thousand metric tonnes of alumina in 2017 with Aluminium of Greece contributing up to 35 820 thousand metric tons of alumina in their yearly production [1,2]. However, as 1 tonne of alumina 36 often generates about 1-1.5 tonnes of bauxite residue (BR) during its production in the Bayer process 37 route, the uncertainty of effectively managing bauxite residue stockpiles in areas of increasingly 38 tightening regulations has encouraged further research efforts into total valorisation. Power, Klauber 39 and Gräfe consolidated a series of papers that reviewed extensively towards methods of disposal, 40 management and best practices as well as options in the interim that dealt with the reuse of bauxite 41 residue [3,4]. They have also reported that from the period of 1964 to 2008, patents relating to bauxite 42 residue filed in civil and building constructions consisted of about a third from the total amount, 43 followed by around equal amounts of catalyst and adsorbent patents; ceramics, plastics, coating or 44 pigments, and wastewater treatment. Steel and slag patents and the recovery of major metals were

Minerals 2019, 7, x; doi: FOR PEER REVIEW www.mdpi.com/journal/minerals Minerals 2019, 9, x FOR PEER REVIEW 2 of 20

45 around 7% and 9% respectively. With recent years sustainable economy focus, minor elements in 46 enriched secondary resource such as bauxite residue had gained larger traction in developing rare 47 earth recovery processes [5-8]. 48 It is reported that iron often found in the range of 5-60% Fe2O3 depending on the mineral types 49 influencing bauxite residue in Bayer processing, followed by 5-30% of Al2O3, 3-50% of SiO2 and 0.3- 50 15% of TiO2. Calcium and sodium averages about 2-14% of CaO and 1-10% of Na2O in bauxite 51 residue. Calcium’s addition also ranges from pre-desilication step, scales suppression up until 52 mineral retardations and transformations [9,10]. Mishra and Gostu published a recent review into 53 several different treatment options of the residue, with pyrometallurgical paths either involving 54 smelting or sintering-type processing [11]. High temperature smelting of bauxite residue for complete 55 iron reduction and removal and the subsequent conditioning of slag are commonly found in current 56 researches [6,12-16]. The sintering or reductive-sintering of bauxite residue has also been explored in 57 literature [7,11,14,17-19], but inefficient separation of varied magnetic iron phases often hinders iron 58 recovery. However, sintering and reductive-sintering are able to recover aluminium and sodium 59 without suffering sodium gaseous losses when smelted at high temperatures. 60 As bauxites contain an amount of reactive silica that is soluble in caustic conditions of the Bayer 61 process, desilication becomes necessary to reduce silica impurities and to avoid scale build-up within 62 the evaporator and digester units. A desilication product that is often formed is a type of sodium 63 aluminosilicate with a formula which may be represented as Na8(AlSiO4)6X2·nH2O where X could be 64 ½CO32-, Cl-, OH-, and ½SO42- [20,21]. The sodium aluminosilicate product often exists as sodalite and 65 cancrinite, expressed in similar stoichiometry, and sometimes substituting Na+ with K+, Ca2+, OH-, 66 SO42-, Cl-, CO32- and H2O in its framework [20]. If calcium is inherently abundant in bauxite or if lime 67 is added to aid the silica removal, tricalcium aluminate (Ca3Al2(OH)12) may also be formed within the 68 Bayer liquor, which transforms into hydrogrossular or hydrogarnet type species (Ca3Al2(SiO4)3- 69 x(OH)4x) [10]. Smith further characterises the reactions and phase transformations that occur with 70 lime’s presence particularly in high temperature Bayer digestion condition. He observed that lime 71 addition encourages sodium recovery, supressing the sodalite form in favour of hydrogrossular 72 products [10]. Rivera et al. adds that when bauxite residue is subjected to CO2 neutralisation, 73 cancrinitic and grossular phases is stabilised with carbonate ions, thus reducing alkalinity [22]. 74 Meanwhile, the simple aluminium-bearing minerals are often originated from the unreacted 75 boehmitic (γ-AlO(OH)), diasporic (α-AlO(OH)) and gibbsitic (γ-Al(OH)3) minerals in bauxite ores. 76 Through the addition of soda and metal oxides, coupled with also a possible reductive stage of iron 77 by adding carbon, the aluminium-bearing minerals is then transformed into leachable sodium 78 aluminate phase, while supressing reactive silica formation. This will also allow ease of introduction 79 of leachates back again into the Bayer circuit.

80 1.1. Previous studies in soda and lime-soda sintering processes 81 Soda sintering or lime-soda sintering process, began from Louis Le Chatelier’s work in 1855 in 82 obtaining aluminium hydroxide (Al(OH)3) from sintering bauxite with soda at around 1000 °C which 83 forms a leachable sodium aluminate species [23]. It was further improved by introducing a 84 calcination step to produce alumina (Al2O3) from Al(OH)3. Deville-Pechiney process also introduced 85 lime and coke into the sintering processes [24]. Therefore, soda often targets the aluminium-bearing 86 mineral species into sodium aluminate, as described in reaction (1), (2) and (3).

2AlOOH + Na2CO3 → 2NaAlO2 + CO2 + H2O (1)

2Al(OH)3 + Na2CO3 → 2NaAlO2 + CO2 + 3H2O (2)

Na6Al6Si6O24·2CaCO3.nH2O + Na2CO3 + Fe2O3 + 10CaO = 6NaAlO2 + 6Ca2SiO4 + (3) 2NaFeO2 + 3CO2+ nH2O 87 Kaussen et al. explored a wide range of methods in recovering aluminium from bauxite residue 88 in the Lünen landfill [14,25]. This expounded from secondary Bayer leaching at higher temperature 89 and pressures in concentrated caustic solution [14], smelting to produce leachable calcium aluminate 90 phases, sometimes coupled with concentrated caustic leaching again [13,14], and soda/lime-soda

Minerals 2019, 9, x FOR PEER REVIEW 3 of 20

91 sintering [14,19]. Here, along with many other studies [26-29], it is noted that sodium ferrite species 92 and sodium titanate species are potentially transformed after sintering and leaching, into hematite 93 and titanium oxide species respectively. The converted salts of other present impurities in bauxite 94 residue do not affect the sodium recovery, with exception to sodium silicate which increases silica 95 impurity into the leachate [25]. Bruckard et al. reports that a solid-solution series of sodium 96 aluminate-carnegeite species formed when bauxite residue is treated at high temperatures of 1400 °C 97 with lime, and though it is a leachable species, up to 55% Al and 90% Na recovery were achieved. 98 Previous attempts of the soda sintering process in Greek bauxite residue achieved the maximum 99 Al recovery of 76% [30], may be suggestive that the recovery of aluminium is inhibited greatly 100 through the bulk presence of hematite and aluminium-bearing mineral types existing in the system, 101 which will be followed up in another study. Tathavadkar et al. investigated soda sintering for three 102 different types of bauxite residues from Hungarian, Indian and UK origins [31]. By introducing soda 103 to bauxite residues, accounting for total silica, aluminium and titanate formations during the 104 sintering process at 550-850 °C, they reported a varied recovery of aluminium ranging at a low 55- 105 60% for MAL, Hungarian refinery bauxite residue, followed by 83-95% in ALCAN, UK refinery 106 bauxite residue and 77-84% in INDAL, Indian refinery bauxite residue [31]. 107 A major difference noted in Hungarian bauxite residue is the higher calcium content (5.7%) 108 compared to both ALCAN and INDAL bauxite residue (<1%), coinciding with the raised levels of 109 silica. Whittington [32] notes that Ajka’s refinery introduces lime addition to the causticisation step, 110 where Ca(OH)2 interacts with DSP to form calcium aluminium hydrosilicate or possibly hydrogarnet 111 and calcium aluminium hydrates intermediate which then transforms into calcite and recovered 112 soda. AoG’s Greek bauxite residue behaviour in the sintering process is more comparable with 113 MAL’s Hungarian bauxite residue due to similarities in the existing calcium-bound aluminium 114 bearing minerals i.e. hydrogarnet (Ca3AlFe(SiO4)(OH)8) and cancrinite (Na6Ca2Al6Si6(CO3)2·O24·2H2O) 115 [33]. Iron-to-aluminium ratio in INDAL bauxite residue (1.5) is lesser compared to MAL and ALCAN 116 bauxite residues (2.3-2.4), with sodalite phase as desilication product in both INDAL and ALCAN 117 bauxite residues. 118 Meher and coworkers investigated the effects of divalent alkaline earth metal oxides, such as 119 lime (CaO), barium oxide (BaO) and magnesium oxide (MgO), in soda sintering process using 120 NALCO’s bauxite residue from Damanjodi, India [27,28]. The introduction of these alkaline earth 121 metals was designed to capture the free and bound silica in bauxite residue, transforming them to its 122 divalent metal silicate (M2SiO4 or the less thermodynamically stable MSiO3 phase). NALCO’s bauxite 123 residue consisted of about 51-57 % Fe2O3, 16-18 % Al2O3 and 8-12% SiO2. In Meher et al.’s researches, 124 the addition of metal oxides was performed by weight basis, with 10, 15 and 20 g of either MgO, CaO 125 or BaO added to 100 g bauxite residue and fixed amount of Na2CO3 of 25 g. By adding minimal 126 amounts of metal oxide in soda sinter system, optimised levels of alumina extraction was achieved 127 at 97.64%, 98.70% and 99.50% using CaO, MgO and BaO respectively [27,28]. Calcium levels in 128 NALCO’s bauxite residue were reported below 2.3% with about 34% aluminium trapped as 129 muscovite KAl₂(Si₃AlO₁₀)(OH)₂, sillimanite Al₂SiO₅, and kaolinite Al₂Si₂O₅(OH)4, in bauxite residue, 130 as reported by Mohapatra et al. [34]. The rest of aluminium was found in gibbsite, boehmite and 131 alumo-goethite, indicative of the 14% alumina losses within NALCO’s bauxite extraction process [34].

2MO/MO + SiO2 → M2SiO4 /MSiO3 (where M represents metals such as Ba, Mg, Ca) (4)

Fe2O3 + 3C → 2Fe + 3CO (5) 132 The addition of lime, and subsequently exploration of other metal oxides into the sintering 133 process through Meher and coworkers’ efforts [27,28,35,36], focuses on capturing the free silica 134 during sintering (Reaction (4)). Adding metallurgical coke into the system, such as Reaction (5), can 135 encourage the recovery of iron downstream by pre-reducing hematite and other iron species. It may 136 also suppress potential unwanted product formation such as complex garnet-type product (e.g. 137 3(Ca,Fe,Mg)O(Al,Fe)2O3·3SiO2) that may entrap both iron and aluminium into its matrix.

Minerals 2019, 9, x FOR PEER REVIEW 4 of 20

-4000

Ba₃SiO₅

Thousands -3500 Ca₃SiO₅

-3000 BaSi₂O₅ CaSiTiO₅ Ba₂SiO₄ Ca₂SiO₄

Mg₂SiO₄ Na₂Si₂O₅ -2500 Na₄SiO₄ Fe₂SiO₄ NaAlSiO₄ -2000 BaSiO₃ CaSiO₃ Gibbs Energy (kJ/mol) Energy Gibbs MgSiO₃

-1500 Na₂SiO₃ FeSiO₃

500 700 900 1100 1300 1500 138 Temperature (ᵒC)

139 Figure 1. Comparison of Gibbs free energy of phase formation of several silicate species throughout 140 the temperature range by Factsage 7.0™ database, with silica notably favouring barium presence.

-4900 Ca₃Ti₂O₇

-4400 Ca₃Ti₂O₆

Thousands Na₂Ti₆O₁₃

-3900

Ca₂Ti₂O₅ -3400 Na₂Ti₂O₅

-2900 Ba₂TiO₄ FeTi₂O₅ Mg₂Ti₂O₅ Na₄TiO₄ -2400 Mg₂TiO₄ FeTi₂O₄ Fe₂TiO₄ Gibbs Energy (kJ/mol) Energy Gibbs FeTi₂O₅ -1900 CaTiO₃ Na₂TiO₃ MgTiO₃ FeTiO₃ -1400

500 700 900 1100 1300 1500 141 Temperature (ᵒC)

142 Figure 2. Comparison of Gibbs free energy of phase formation of several titanate species throughout 143 the temperature range by Factsage 7.0™ database, with calcium titanate being most favourable.

144 Figure 1 and 2 reports the Gibbs free energy of formation of different silicate phases and titanate 145 phases at a range of temperature. Silica is more likely to be bonded together with barium or calcium 146 species and predicted to form the more stable species of Ba3SiO5 or Ca3SiO5, whereas titania is more 147 likely to form calcium-type bonds such as Ca3Ti2O7.

Minerals 2019, 9, x FOR PEER REVIEW 5 of 20

148 2. Materials and Methods

149 2.1. Characterisation of bauxite residue 150 Bauxite residue (BR) from Aluminium of Greece’s (AoG) high temperature Bayer processing is 151 sampled, homogenised and ground to below 150 μm. Chemical analysis of BR is indicated in Table 152 1. Chemical analysis of BR was performed using XRF glass disk fusion method with lithium 153 tetraborate in AoG. X-ray diffraction (XRD) and mineralogical analysis was performed using Brucker 154 D6 XRD equipment with integrated EVA software for identification of minerals. SEM images were 155 obtained using Jeol6380LV Scanning Electron Microscope (SEM) with INCA software for Energy 156 Dispersive X-Ray Spectroscopy (EDS) identification. Figure 3 shows the XRD identification of 157 mineralogical phases of AoG’s bauxite residue. AoG’s high temperature digestion process that uses 158 quick lime to suppress silica levels, indicates similar phase transformations by the identification of 159 hydrogrossular products in the bauxite residue as confirmed by literature. Diaspore, boehmite and 160 gibbsite have been identified as the simple aluminium-bearing minerals, whereas cancrinite and 161 hydrogrossular forms of desilication product are also detected. Chamosite mineral that exists 162 originally from the bauxite ore, reacts slowly in Bayer cycle and remains in bauxite residue [37].

LEGEND H H: Hematite - Fe2O3 140 Gi: Gibbsite - Al2O3·3H2O Go: Goethite - Fe2O3·H2O D: Diaspore - α-AlOOH P: Perovskite- CaTiO 120 H 3 B: Boehmite - γ-AlOOH Q: Quartz - SiO2 100 Gr: Grossular (Ca₃FeAl(SiO₄)((OH)₄) R: Rutile - TiO2 Ca: Cancrinite (Na₈(Al,Si)₁₂O₂₄(OH)₂.2H₂O A: Anatase - TiO2 80 Ch: Chamosite (Fe1.8Mg0.2Al0.8(Si1.3Al0.7)O5(OH)4 Ca: Calcite - CaCO3 H Gr H 60 H

Intensity Ca H H D H 40 Cn Q Cn D Cn Gi K Gr P Go K Redmud Gi G R DGr H H 20 Ch B P A Gi D P A o D 0 163 5 15 25 35 45 55 65 75 2-θ

164 Figure 3. XRD mineralogical analysis of AoG bauxite residue.

165 Table 1. Chemical analysis of AoG bauxite residue.

Major metals Others LOI Oxides Fe2O3 Al2O3 SiO2 TiO2 Na2O CaO V2O5 SO3 wt % 43.51 19.26 6.50 5.49 2.81 9.59 0.17 0.47 9.40 166 The quantification of phases of AoG bauxite residue is given in Table 2, performed using XDB 167 software designed by Sajó that uses a full-profile fit method [38]. This result was also verified in 168 another study [39], where the quantification of minerals was compared at different stages of soda 169 sintering and leaching process. The mineral phases shown in Table 2 were utilized for the prediction 170 of phases in sinter products by using the thermodynamic software Factsage™. However, the 171 simplification of complex minerals with simpler ones existing in the software database was necessary.

172 Table 2. XRD Mineralogical Quantification using XDB Software by Sajo [38,39].

Mineralogy Anatase Boehmite Gibbsite Hematite Goethite Perovskite Cancrinite Diaspor Calcite Rutile Grossular Chamosite

Phase 1.0 2.1 1.0 37.5 5.2 5.2 11.5 13.0 4.2 0.5 14.6 2.1 percentage %

173 2.2. Laboratorial methods in sintering optimisation

Minerals 2019, 9, x FOR PEER REVIEW 6 of 20

174

175 Figure 4. Procedure for sintering of AoG’s Greek bauxite residue.

176 Figure 4 indicates the procedure of sintering and leaching in this lab-scale study. Lab-grade 177 reagents of Na2CO3 (99.8%, Chembiotin), CaO (>98%, Merck), BaO (99.5%, Alfa Aesar), MgO (>98%, 178 Sigma Aldrich) and metallurgical coke (80.3% C) were used as reactive additives to the bauxite 179 residue in this study. Dried homogenised bauxite residue (<150 μm) were mixed together with the 180 additives and placed into alumina crucibles. For the lab-scaled experiments, the muffle furnace 181 (Tmax:1700 °C) shown in Figure 5 is as below. 1000 900 800

C) 700 o 600 500 400 300 Temperature ( Temperature -0.395x 200 y = 1758.4x - 1169.2 y = 2837.6e R² = 0.9978 R² = 0.9658 100 0 0 2 4 6 8 10 12 Time (Hours) 182 183 Figure 5. Temperature heating profile of muffle furnace.

184 The heating rate is kept to 30-35 °C per min, and after sintering within set holding time, indicated 185 in the plateau region of temperature profile in Figure 5, is then left to cool overnight. The cooling 186 kinetics are out of the scope of this study. Introduction of inert atmosphere for carbon-reduction 187 aided sintering process with 5 N L/min argon flowrate, where stoichiometric to 50% excess carbon 188 required to reduce iron were used. Current leaching conditions during sintering optimisation were 189 kept at 80 °C for 4 hours, at 240 rpm stirring rate, solid-to-liquid ratio of 1.5g/100 mL (1:66) in 0.1 M 190 NaOH solution. Leachates are then filtered, diluted and sent to Atomic Absorption Spectroscopy 191 (AAS) for chemical analysis. Leached residue is collected and sent for XRF analysis. Two 192 experimental series were performed. In the first the soda sintering process was studied in the 193 presence of 50% excess of soda over stoichiometric amount required for sodium aluminate conversion 194 and in the absence of any additives. In the second the effect of additives was studied. Particularly, 195 AoG’s bauxite residue was subjected to 20 g of additives (divalent metal oxides) to 100 g of bauxite 196 residue, with 25 g of Na2CO3 (~50% excess soda amounts required to form NaAlO2).

Minerals 2019, 9, x FOR PEER REVIEW 7 of 20

197 3. Results and discussion

198 3.1. Effect of sintering temperature and retention time on metals recoveries 199 Figure 6 shows the thermodynamic predictions of solid phase’s evolution during sintering 200 temperature increase. For performing this thermodynamic analysis, the real desilication phases 201 existing in AoG’s Bauxite residue (namely, cancrinite, sodalite and hydrogrossular) were replaced by 202 ones (such as Ferrocordierite, Grossularite and Albite) which was available in the FactsageTM 203 software. 100% (FeO)₂(TiO₂) CaFe₂O₄ 100% 90% CaFe₄O₇ CaFe₂O₄ CaFe₄O₇ NaFe₂O₃ 90% 80% NaFeO₂ 80% Fe₂O₃ 70% Fe₂O₃ 70% 60% CaTiO₃ CaTiO₃ Na₃Fe₅O₉ 60% Ca₂Fe₂O₅ 50% Na₄TiO₄ NaFeO₂ Ca₂Al₂SiO₇ CaAl₄O₇ 50% Na₄TiO₄ 40% 40% Ca₂Al₂SiO₇ NaAlSiO₄ NaAlSiO₄ CaAl₄O₇ 30% NaAlSiO₄ NaAl₉O₁₄ (Nepheline) 30%

Phase Distribution (%) (Carnegeite) 20% NaAl₉O₁₄ (Nepheline) NaAlSiO₄ CaCO₃ Na₂Al₁₂O₁₉ 20% Na₂Al₁₂O₁₉ (Carnegeite) 10% CaCO₃ NaCO₃ NaAlO₂ 10% 0% NaCO₃ NaAlO₂ 500 600 700 800 900 1000 1100 1200 1300 1400 1500 0% 500 600 700 800 900 1000 1100 1200 1300 1400 1500 Temperature (ᵒC) 204 Temperature (ᵒC) 205 (a) (b) 100% 100% CaFe₂O₄ CaFe₄O₇ (FeO)₂(TiO₂) Na₂Ca₂SiO₄ Na₂Ca₂Si₂O₇ 90% 90% Ca₂SiO₄ Na₃Fe₅O₉ NaFe₂O₃ 80% Fe₂O₃ 80% Na₄TiO₄ NaFeO₂ 70% 70% Fe₂O₃ CaTiO₃ Fe (s1) Fe (s2) 60% 60% Fe (s1) 50% Ca₂Al₂SiO₇ Na₄TiO₄ 50% Ca₂Al₂SiO₇ CaAl₄O₇ 40% 40% NaAlSiO₄ NaAlSiO₄ NaAlSiO₄ 30% 30% C (Carnegeite) NaAl₉O₁₄ (Nepheline) NaAlSiO₄ (Nepheline) Phase Distribution (%) NaAl₉O₁₄ 20% Na₂Al₁₂O₁₉ (Carnegeite) 20% CaCO₃ 10% CaCO₃ NaCO₃ NaAlO₂ 10% NaAlO₂ NaCO₃ 0% 0% 500 600 700 800 900 1000 1100 1200 1300 1400 1500 500 600 700 800 900 1000 1100 1200 1300 1400 1500 Temperature ( 206 ᵒC) Temperature (ᵒC) 207 (c) (d)

208 Figure 6. Thermodynamic predictions of sinter phase formations with different representatives of 209 desilication product such as (a) Ferrocordierite, Fe2Al4Si5O18 (b) Glossularite, Ca3Al2Si3O12 (c) Albite, 210 NaAlSi3O8 (d) Albite, NaAlSi3O8 with 50% excess of soda over stoichiometric amount.

211 As seen in Figure 6, at 50% stoichiometric excess of soda in sintering system achieves only partial 212 conversion into sodium aluminate species, and sodium aluminosilicate forms of nepheline and 213 carnegeite are still present in the system. NaAl9O14 and Na2Al12O19 represented unreacted β-alumina 214 and β''-alumina phases respectively. Silica presence is predicted to be a constraint for the total 215 transformation of aluminium bearing phases to leachable sodium aluminate which is in agreement 216 with previous studies on soda sintering at high temperatures [12]. It is interesting to note that despite 217 glossularite’s high calcium content when replaced as a desilication product, the result of the 218 predictive transformation is still the same. The calcium shows an affinity towards iron, with calcium 219 ferritic sinter products (CaFe2O4, CaFe4O7 and Ca2Fe2O5) formed instead of binding towards the 220 available silica. This agrees with Hodge et al., noting that calcium has the tendency to preferentially 221 react with iron phases instead of silica during bauxite residue soda sintering process [39]. In Figure 6 222 (d), albite was used as the extrapolative value of desilication product, and the addition of carbon as 223 reducing agent promotes iron reduction and is predicted to suppress the calcium-iron interaction,

Minerals 2019, 9, x FOR PEER REVIEW 8 of 20

224 though it presents only a slight increase of sodium aluminate phase over nepheline/carnegeite 225 species. 226 Figure 7 (a) shows the metal recoveries as a function of sintering temperature at 2 hours retention 227 time and 5% stoichiometric excess of soda in bauxite residue. At 900 °C it reaches at 55% due to 228 entrapment of part of aluminium in sodium aluminosilicate and calcium aluminosilicate species 229 which are by definition very difficult soluble in diluted NaOH solutions. This conclusion is supported 230 by the behaviour of silica which is recovered at a value lower than 20% showing that the dissolution 231 of silicate phases formed during sintering is limited in diluted NaOH solutions hindering the alumina 232 recovery. In Figure 7 (b), once excess soda levels were investigated and 50% stoichiometric excess of 233 soda in bauxite residue was used, it is noted that increased retention time of sintering from 2 to 4 234 hours displayed very slight change towards recovery. Aluminium recovery agrees with the 235 thermodynamic predictions. 100 100 90 %Al₂O₃-AAS %SiO₂-AAS 90 %Al₂O₃-AAS %SiO₂-AAS 80 80 70 70 60 60 50 50 40 40 Recovery (%) Recovery 30 (%) Recovery 30 20 20 10 10 0 0 800 900 1000 1100 1.5 2.5 3.5 4.5 Temperature (⁰C) Time (Hour of Sintering) 236 237 (a) (b)

238 Figure 7. Recovery of Al2O3 and SiO2 for (a) temperature-variable sintering experiments with 5% 239 stoichiometric excess of Na2CO3, and (b) time-variable with 50% stoichiometric excess of Na2CO3.

240 3.2. Effect of Na2CO3 excess on metals recoveries 241 Figure 8 (a) shows the thermodynamic predictions of sinter phase formation when soda 242 (Na2CO3) is added in excess to stoichiometric amounts. The solid sinter phase distribution of sodium 243 aluminate when soda is introduced in excess is noted reach a plateau of equilibrium with nepheline 244 phase when the soda excess is equal or greater than 200%. The metastable states of Na2Ca3Al16O28 and 245 Na2Ca8Al6O18 were suppressed and excluded from the thermodynamic prediction in the software. 246 The calcium content is observed to partially form calcium aluminate and changing into sodium 247 calcium silicate instead beginning from 100% excess. Hematite is predicted to initially bind 248 preferentially towards calcium and then shift to sodium instead, forming potential soluble sodium 249 ferrite phases. Actual experimental results further confirms the linear increase until reaching plateau, 250 albeit at an earlier range with levels of 67-70% being reached when 50% excess of stoichiometric soda 251 required is added. 252 Figure 9 shows the XRD mineral identification of sinters and leached residue at 50% excess soda 253 addition. Jadeite (NaAlSi2O6) observations in leached residue and sinters confirms that the nepheline- 254 sodium aluminate equilibrium does exist. The consumption of soda reagent into iron products such 255 as sodium iron oxide (Na5FeO4) and calcium products such as gehlenite (Ca2Al2SiO7) are the 256 unwanted side reactions further fortifies the existence of nepheline-sodium aluminate equilibrium 257 despite soda excess levels during sintering. Though silica is observed to be within the range of 10- 258 20%, silica dissolution is still observed in the leaching system. Sodium recovery at 80 to 90% range 259 throughout the experimental data indicates that a variety of sodium-soluble products exists.

Minerals 2019, 9, x FOR PEER REVIEW 9 of 20

100% CaFe₂O₄ 100 90% 90 80% Na₃Fe₅O₉ NaFeO₂ 80 70% 70 Fe₂O₃ 60% 60 %Al₂O₃-AAS 50% Na₄TiO₄ 50 %SiO₂-AAS CaAl₄O₇ 40% Na₂CaSiO₄ 40 %Na₂O-XRF

30% (%) Recovery 30 NaAlSiO₄ Phase Distribution (%) 20 20% (Nepheline) 10% 10 NaAlO₂ 0% 0

0 5 0 50 100 150 200 10 15 25 50 100 200 300 400 500 600 700 800 900

1000 Excess Stoichiometric Addition of Na₂CO₃ (%) Excess Na₂CO₃ (%)

260 (a) (b)

261 Figure 8: (a) Thermodynamic predictions of sinter phase formation and (b) results of experimental recovery of

262 Al2O3, SiO2, and Na2O at the range of excesses of stoichiometric Na2CO3 additions at 900 °C for 2 hours.

0: Quartz (SiO₂) 10: Rutile (TiO₂) 1: Hematite (Fe₂O₃) 11: Perovskite (CaTiO₃) 2: Goethite (α-FeOOH) 12: Calcite (CaCO₃) 3: Cancrinite (Na₈(Al,Si)₁₂O₂₄(OH)₂.2H₂O 13: Sodium aluminate (NaAlO₂) 4: Chamosite 14: Jadeite (NaAlSi₂O₆) (Fe1.8Mg0.2Al0.8(Si1.3Al0.7)O5(OH)4) 15: Gehlenite (Ca₂Al₂SiO₇) 5: Boehmite (γ-AlOOH) 16: Wustite (FeO) 6: Gibbsite (Al(OH)₃) 17: Magnetite (Fe₃O₄) 7: Diaspore (α-AlOOH) 18: Sodium iron oxide (Na₅FeO₄) 11,17,1 8: Grossular (Ca₃FeAl(SiO₄)((OH)₄) 14 19: Lorenzenite (Na₂Ti₂SiO₉)

9: Anatase (TiO₂) 14 17 11 17 11 17,1 19 17 14 16 14,1 15 16 18 19

Leached Residue 13 11,1 13,14 Intensity 11 17,1 11 17 17 16 15 16 18 17 14,1 14 15 19

Sinter 1,8 1,11 7 1 7 12 5 10 8 1 3 7 7 3 6 1 1 1 0 3 10 8 7 8, 11,12 5 8 8 8 2 8 9 8 4 Bauxite Residue

263 10 20 30 40 50 60 2-θ

264 Figure 9. XRD of mineralogical phases for 50% stoichiometric excess Na2CO3 at 900 °C and for 2 hours.

265 3.3. Effect of carbon additions on metals recoveries 266 Fixing to 50% excess of soda which reduces the amount of soda required per aluminium 267 recovery, the study next focuses towards the suppression of side reactions through the reduction of 268 iron with metallurgical coke additions as a carbon source in inert conditions. The thermodynamic 269 prediction of sinter phases in Figure 10 (a) indicates that the equilibrium that exists between leachable 270 sodium aluminate and nepheline form still exists despite the conversion of hematite or goethite into 271 metallic iron. 272 Figure 11 shows the XRD confirmations of mineral phases present in the sinter product and 273 leached residue. Part of iron has formed sodium iron oxide while hematite is partially reduced to 274 magnetite and wustite. Though the calcium iron oxide species were now suppressed, as it can be seen 275 from Figure 10 (a) thermodynamic predictions, there is a secondary issue of gehlenite (Ca2Al2SiO7) 276 formation on top of nepheline throughout the range of stoichiometric carbon additions, causing 277 additional aluminium losses downstream.

Minerals 2019, 9, x FOR PEER REVIEW 10 of 20

100% CaFe₂O₄ Na₂CaSiO₄ 90% Na₄TiO₄ 80% 100 Na₃Fe₅O₉ 70% 90 NaFe₂O₃ Fe 80 60% 70 50% Ca₂Al₂SiO₇ 60 %Al₂O₃-AAS 40% CaAl₄O₇ 50 %SiO₂-AAS 30% NaAlSiO₄ 40

Recovery Recovery (%) %Na₂O-XRF

Phase Distribution (%) (Nepheline) 30 20% 20 10% NaAlO₂ 10 0% 0 0 1 2

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 0 10 20 30 40 50

Stoichiometric carbon-to-iron (C:Fe2O3) ratio addition Excess Carbon (%) Fe2O3 + 3C → 2Fe + 3CO 278 (a) (b)

279 Figure 10. (a) Thermodynamic predictions of sinter phase formation and (b) results of experimental 280 recovery of Al2O3, SiO2, and Na2O at the range of excesses of stoichiometric carbon additions (based 281 on Carbon:Fe2O3 ratio from bauxite residue) with 50% excess soda, 900 °C for 2 hours.

282 Figure 1 shows the mineralogy of sinter and leached residue, confirming the theoretical 283 formations of sodum aluminium silicates and gehlenite, though latter phases were minor. The 284 suppression of soluble sodium-iron products occurs past 0.7 stoichiometric ratio indicates that carbon 285 addition is favourable at stoichiometric ratio onwards. This is also confirmed in the experimental 286 results in Figure 10 (b), and an increase from 68% to 75% is observed in aluminium recovery, though 287 sodium recovery reduces to 3% lesser amount that the previous soda excess additions in atmospheric 288 environment. Silica amount remains relatively unchanged within the range of 18% recovery. 0: Quartz (SiO₂) 10: Rutile (TiO₂) 1: Hematite (Fe₂O₃) 11: Perovskite (CaTiO₃) 2: Goethite (α-FeOOH) 12: Calcite (CaCO₃) 3: Cancrinite (Na₈(Al,Si)₁₂O₂₄(OH)₂.2H₂O 13: Sodium aluminate (NaAlO₂) 4: Chamosite 14: Jadeite (NaAlSi₂O₆) (Fe1.8Mg0.2Al0.8(Si1.3Al0.7)O5(OH)4) 15: Gehlenite (Ca₂Al₂SiO₇) 5: Boehmite (γ-AlOOH) 16: Wustite (FeO) 6: Gibbsite (Al(OH)₃) 17: Magnetite (Fe₃O₄) 7: Diaspore (α-AlOOH) 18: Sodium iron oxide (Na₅FeO₄) 11,17,1 8: Grossular (Ca₃FeAl(SiO₄)((OH)₄) 14 19: Lorenzenite (Na₂Ti₂SiO₉)

9: Anatase (TiO₂) 14 17 11 17 11 17,1 19 17 14 16 14,1 15 16 18 19

Leached Residue 13 11,1 13,14 Intensity 11 17,1 11 17 17 16 15 16 18 17 14,1 14 15 19

Sinter 1,8 1,11 7 1 7 12 5 10 8 3 1 7 7 3 6 1 1 1 0 3 10 8 7 8, 11,12 5 8 8 8 2 8 9 8 4 Bauxite Residue

289 10 20 30 40 50 60 2-θ

290 Figure 11. XRD of mineralogical phases for stoichiometric carbon addition needed to produce Fe from 291 reaction (5), with 50% stoichiometric excess Na2CO3 at inert conditions at 900 °C and for 2 hours..

292 3.2. Effect of CaO additions on phase transformations during sintering process

Minerals 2019, 9, x FOR PEER REVIEW 11 of 20

293 Following soda and carbon additions, the addition of lime into the sintering process of 20 g CaO 294 to 100 g bauxite residue (80% excess to stoichiometric requirements to form Ca2SiO4) is necessary to 295 investigate the consumption of free and bound silica in desilication products, relying on higher 296 thermodynamic affinity with calcium. Figure 12 shows the thermodynamic predictions of sinter 297 phases at excess lime additions. The thermodynamic predictions made by Factsage™, however, 298 shows that calcium indeed reacts with iron to form CaFe2O4 and Ca2Fe2O5, when iron is available in 299 atmospheric conditions at 900 °C. 300 Compared to Figure 8 (a), thermodynamics of soda only addition into the sinter of bauxite 301 residue notes that sodium preferably reacts to form nepheline species and sodium titanate species at 302 lower than 50% soda excess levels, whereas, calcium is bound to either aluminium or iron initially. 303 At between 50% to 200% soda excess levels, sodium ferrite species of intermediate Na3Fe5O9 forms, 304 transforming into NaFeO2 at even higher soda concentrations. Both sodium ferrite then dissolves 305 back as sodium, hence phases do not inhibit system. Also at higher soda levels (Figure 8 (a)), soda 306 and calcium is predicted to preferentially react, still existing in equilibrium with nepheline species. 307 In Figure 12, the excess CaO predicted to be added in the sintering system is divided into two 308 amounts, Figure 12 (a), one that encourages monocalcium silicate formation and the other, Figure 12 309 (b), dicalcium silicate instead. Increased additions of lime only encourage calcium to complex into 310 Ca2Fe2O5 stable species instead, with nepheline phase still existing in equilibrium amounts to sodium 311 aluminate species. Grossular species in the form of Ca2Al2SiO7 and sodium calcium silicate species 312 (Na2CaSiO4) exists when CaO is added in excess. 100% 100% 90% Ca₂Fe₂O₅ 90% Ca₂Fe₂O₅ 80% 80% CaFe₂O₄ 70% 70% 60% 60% 50% CaFe₂O₄ 50% NaFeO2 NaFeO2 40% Na4TiO4 40% Na4TiO4 Ca₂Al₂SiO₇ 30% 30% Phase Distribution % Distribution Phase NaAlSiO₄ NaAlSiO₄ Na₂CaSiO₄ 20% 20% 10% NaAlO₂ 10% NaAlO₂ 0% 0% 0% 5% 10% 15% 25% 50% 100% 200% 0% 5% 10% 15% 25% 50% 100% Excess CaO Reagent % Excess CaO Reagent % 313 CaO + SiO₂ → CaSiO₃ 2CaO + SiO₂ → Ca₂SiO₄ 314 (a) (b)

315 Figure 12. Thermodynamic predictions of sinter phase formation (a) when monocalcium silicate is 316 expected or (b) when dicalcium silicate is expected, at excess of lime additions at 50% excess soda, 317 900 °C for 2 hours.

318 Figure 13 shows the XRD identification of the sinter phase and leached residue, whereas Figure 319 14 shows the SEM confirmations of minerals identified in the sinters and leached residue. The sinters 320 confirmed the presence of sodium aluminate, calcium ferrite and titanate peaks as well as dicalcium 321 silicate in both XRD and SEM. In Figure 13, both presence of harmunite (CaFe2O4) and srebrodolskite 322 (Ca2Fe2O5) were detected, where angular rods exist in both sinters and leached residue images in SEM 323 (Figure 14, (2) and (6), nano-rods, SEM empirical formula of CaFe6.9O10). Iron was found to be 324 entrained at varying levels throughout SEM-EDS element identification.

Minerals 2019, 9, x FOR PEER REVIEW 12 of 20

0: Quartz (SiO₂) 10: Rutile (TiO₂) 1: Hematite (Fe₂O₃) 11: Perovskite (CaTiO₃) 2: Goethite (α-FeOOH) 12: Calcite (CaCO₃) 3: Cancrinite (Na₈(Al,Si)₁₂O₂₄(OH)₂.2H₂O 13: Sodium aluminate (NaAlO₂) 4: Chamosite (Fe1.8Mg0.2Al0.8(Si1.3Al0.7)O5(OH)4) 14: Sodium aluminium silicate (NaAlSiO₄) 5: Boehmite (γ-AlOOH) 15: Sodium ferrotitanate (NaFeTiO₄) 6: Gibbsite (Al(OH)₃) 16: Harmunite (CaFeO₄) 7: Diaspore (α-AlOOH) 1,11 17:Srebrodolskite (Ca₂Fe₂O₅) 8: Grossular (Ca₃FeAl(SiO₄)((OH)₄) 16 18: Dicalcium silicate (Ca₂SiO₄) 9: Anatase (TiO₂) 12 1,16 11 18,8 8 8 11 8 16 8 8 8 1 12 16 8 16 14 8 1 1 18 8 1 17 Leached Residue 16 1,11 1,16 Intensity 8 13 18 11 1 8 13 11 8 1,16 13,14 17 8,0 17 16 16 1 18 8 18 16 1 17

Sinter 1,8 1, 11 7 11 7 5 12 10 8 3 1 7 7 3 6 1 1 1 0 3 8 7 8, 11,12 5 4 10 8 8 8 2 8 9 8

Bauxite Residue 2-θ 325 10 20 30 40 50 60

326 Figure 13. XRD of mineralogical phases for lime additions (20 g CaO/100 g bauxite residue, equivalent

327 to 80% excess CaO addition for Ca2SiO4 formation), and 50% excess soda at 900 °C and for 2 hours.

328 Calcium sodium aluminium silicate (grossular, SEM empirical formula of 329 Ca2.23Na1.11AlSi1.15FeO9.6) trace remainder in both XRD and SEM images (Figure 14, (5)) confirms the 330 thermodynamic inhibition of aluminium phase that converts from nepheline/sodium aluminium 331 silicate phases towards grossular products instead of sodium aluminate phases, indicating 332 incomplete recovery of aluminium. Meher et al. [27,40] also detected similar formations of minerals 333 in lime/calcite added soda sinter system, with calcium aluminium silicate detected instead of sodium 334 aluminium silicate type products, though alumina extraction was more favourable at 10 g CaO ratio 335 at 900 °C [40]. Kirschsteinite (CaFeSiO4) was also detected in their paper [40] and it may exists in 336 minimal amounts that remained undetected in our sinter system.

6 2

4 5

337 338 (a) (b)

339 Figure 14. SEM mineral identifications in CaO, Na2CO3 and bauxite residue (a) Sinter product; (1) 340 Sodium aluminate, (2) Calcium ferrite, Calcium titanate, (3) Calcium silicate, sodium calcium silicate; 341 and (b) Leached residue; (4) Calcium ferrite, calcium silicate, (5) Calcium ferrite, grossular product, 342 (6) Calcium ferrite.

Minerals 2019, 9, x FOR PEER REVIEW 13 of 20

343 3.3. Effect of MgO additions on phase transformation during sintering process 344 Figure 15 shows the thermodynamic prediction of the effects of MgO added in excess of 345 stoichiometric amounts to form magnesium silicate product.

100% 100% CaFe₂O₄ CaFe₂O₄ 90% 90% Na₃Fe₅O₉ Na₃Fe₅O₉ 80% 80% 70% Fe₂O₃ 70% Fe₂O₃ 60% 60% Na₄TiO₄ Na₄TiO₄ 50% 50% 40% NaAlSiO₄ 40% NaAlSiO₄ 30% (Nepheline) 30% (Nepheline) Phase Distribution (%) 20% 20% 10% NaAlO₂ 10% NaAlO₂ MgO MgO 0% 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Excess MgO reagent % Excess MgO reagent % MgO + SiO → MgSiO₃ 2MgO + SiO → Mg₂SiO₃ 346 (a) (b)

347 Figure 15. Thermodynamic predictions of sinter phase formation (a) when monomagnesium silicate 348 is expected or (b) when dimagnesium silicate is expected at excess of MgO additions at 50% excess 349 soda, 900 °C for 2 hours.

350 Figure 16 and Figure 17 shows the XRD identification and SEM confirmations of minerals 351 identified in the sinters and leached residue respectively. The effect of the addition of magnesium 352 oxide in the soda sintering system of 20 g MgO to 100 g bauxite residue (130% excess to stoichiometric 353 requirements to form Mg2SiO4) when thermodynamically interpreted, is limited and not able to push 354 the nepheline product reaction towards sodium aluminate. Any calcium existing in the system is said 355 to change into CaFe2O4 product in Figure 15, and sodium reacts with iron and titanium to form 356 Na3Fe5O9 and Na4TiO4 preferentially.

8: Grossular (Ca₃FeAl(SiO₄)((OH)₄) 1,11,16 9: Anatase (TiO₂) 1 14,15 11 14 17 14 16 14 1 14 16 11

Leached Residue 1 15 1 Intensity 14 13,17 1,11,16 13 14 11 14 16 16 14 1 14 11 Sinter 1,8 1,11 7 11 7 5 12 10 8 3 1 7 7 3 6 1 1 1 0 3 8 7 8, 10 11,12 5 8 8 8 2 8 9 8 4 Bauxite Residue 357 10 20 30 40 50 60 2-θ

358 Figure 16. XRD of mineralogical phases for MgO additions (20 g MgO/100 g bauxite residue) at 359 atmospheric conditions at 900 °C and for 2 hours.

Minerals 2019, 9, x FOR PEER REVIEW 14 of 20

2 6

3 5

360 361 (a) (b)

362 Figure 17. SEM mineral identifications in MgO, Na2CO3 and bauxite residue (a) sinter products; (1) 363 Calcium iron aluminium sodium oxide, (2) Magnesium oxide, (3) Calcium magnesium iron 364 aluminium sodium oxide (4) Magnesium iron oxide, Calcium titanate, Calcium magnesium silicate, 365 and (b) Leached residue; (5) Magnesium oxide coating, (6) Calcium magnesium iron oxide (7) 366 Calcium magnesium iron aluminium sodium oxide.

367 This contrasts with the XRD of experimental results of 20 g MgO additions in 100 g bauxite 368 residue in Figure 16, where complex sinter mineral formations such as magnesioferrite (Mg(Fe3+)2O4), 369 bredigite (Ca2Mg2SiO4)8) and brownmillerite (Ca2(Mg,Al,Fe)2O5) were detected in addition to 370 unreacted magnesia. Leached residue indicates that excess MgO as well as these complex phases 371 remains as part of the leftover residue and that aluminium remains trapped still from the conversion 372 towards these complexes. 373 SEM mineral identifications in Figure 17 confirms a series of complex compound containing 374 calcium iron aluminium sodium oxide and calcium magnesium iron aluminium sodium oxide (SEM 375 empirical formulas of CaFe2.4AlNa1.6O9 and CaFe4.5Mg7Al1.2Na1.6O13). Complex sinter area that 376 contains calcium magnesium iron aluminium and silicon (Total SEM empirical formula of 377 Ca1.8Fe11.6SiTi1.48Mg6.7Al2.6O44.7) indicates that the sinter product becomes a heterogenous mix of 378 elements that are harder to distinguish from topographical SEM-EDS method. In Figure 16, they 379 appear as bredigite and brownmillerite species instead. 380 A combination of calcium magnesium iron oxide (SEM empirical formula of CaMg8.9Fe11O16) 381 contains minute traces of distributed calcium, and empirical formula indicates that particles such as 382 Figure 17 (6) suggests that magnesioferrite species are present in the system, with partial calcium 383 substitution in the species. Overall, adding MgO does not aid the sinter reactions as they still contain 384 unfavourable complex garnet-type products. Meher and Padhi reports cancrinite and bayerite 385 mineral identifications in their XRDs of MgO and Na2CO3 sintered bauxite residue system, though 386 magnesium was found to be bound with ferrite, titanate and silica forms [36].

387 3.4. Effect of BaO additions on phase transformations during sintering process 388 By introducing barium oxide into the soda sinter system, the thermodynamic prediction of 389 minerals changes to improve the recovery of aluminium via sodium aluminate phases. Figure 18 390 shows that in the thermodynamic predictions, barium favourably reacts towards the available silica 391 and preferentially binds them into either Ba2Ca2Si4O16 or Ba2SiO4 forms, pushing also the equilibrium 392 of reaction from sodium aluminium silicate product towards sodium aluminate product, especially 393 when added in excess of 100% stoichiometric requirements for Ba2SiO4 formation.

Minerals 2019, 9, x FOR PEER REVIEW 15 of 20

100% Ba₂Ca₆Si₄O₁₆ 100% Ba₂Ca₆Si₄O₁₆ 90% 90% 80% Ba₂SiO₄ 80% Ba₂SiO₄ 70% Na₃Fe₅O₉ 70% 60% 60% Na₃Fe₅O₉ Fe₂O₃ 50% 50% Fe₂O₃ 40% Na₄TiO₄ 40% 30% NaAlSiO₄ 30% Na₄TiO₄

Phase Distribution (%) (Nepheline) NaAlSiO₄ 20% 20% (Nepheline) 10% NaAlO₂ 10% NaAlO₂ 0% 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Excess BaO reagent % Excess BaO reagent % BaO + SiO → BaSiO₃ 2BaO + SiO → Ba₂SiO₃ (a) (b)

394 Figure 18. Thermodynamic predictions of sinter phase formation (a) when monobarium silicate is 395 expected or (b) when dibarium silicate is expected at excess of BaO additions at 50% excess soda, 900 396 °C for 2 hours.

397 Unlike calcium-added sinter system, where the addition of excess causes potential unwanted 398 formation of grossular-type products that inhibit aluminium recovery, barium preferentially reacts 399 with calcium during the sintering, leaving aluminium component to bind with sodium instead. 400 Meher et al. reduced the amount of barium-to-silica mol ratio requirements to 1, and encouraging the 401 formation of sodium barium silicate to allow for sodium and aluminium extraction (99 % recovery) 402 with decreased BaO reagent usage [35]. 403 They also detected varying degrees of barium ferrite, Ba2Fe2O5 and BaFe12O19, and barium 404 silicate products, as calcium is low in NALCO’s bauxite residue system. In AoG bauxite residue 405 system, the addition of 20 g BaO to 100 g bauxite residue (25% excess to stoichiometric requirements 406 to form BaSiO3), is investigated. XRD mineral identifications in Figure 19 indicates that batiferrite 407 (Ba(Fe10Ti2)O19), ankangite (BaTi8O16) and trasicite (Ba9Fe2Ti2Si12O3(OH)) were the favourable phases 408 that were formed experimentally, indicating that experimentally barium prefers to react with 409 titanium instead. 410 In Figure 20, the SEM identification of barium titanate is visibly large tetragonal-type particles 411 (1) with SEM empirical formula of Ba3.2TiO10, confirming barium affinity to titanium species. More 412 complex agglomerates such as Figure 20 (2) that SEM empirical formula of the area is distributed as 413 BaTi1.2Fe5.9Ca1.5AlSi1.5O24.3 and (3) as Ba3.7Fe3Ca1.2AlSi1.4O20.9 shows that there are partial losses of 414 aluminium within the sinter product, though XRD only picks up trace amounts of existing sodium 415 aluminium silicate as opposed to the complex product types.

Minerals 2019, 9, x FOR PEER REVIEW 16 of 20

0: Quartz (SiO₂) 9: Anatase (TiO₂) 1: Hematite (Fe₂O₃) 10: Rutile (TiO₂) 2: Goethite (α-FeOOH) 11: Perovskite (CaTiO₃) 3: Cancrinite (Na₈(Al,Si)₁₂O₂₄(OH)₂.2H₂O 12: Calcite (CaCO₃) 4: Chamosite (Fe1.8Mg0.2Al0.8(Si1.3Al0.7)O5(OH)4) 13: Sodium aluminate (NaAlO₂) 5: Boehmite (γ-AlOOH) 14: Sodium aluminium silicate (NaAlSiO₄) 6: Gibbsite (Al(OH)₃) 15: Batiferrite (Ba(Fe₁₀Ti₂)O₁₉) 7: Diaspore (α-AlOOH) 16: Ankangite (BaTi₈O₁₆) 8: Grossular (Ca₃FeAl(SiO₄)((OH)₄) 1,11 17: Trasicite (Ba₉Fe₂Ti₂Si₁₂O₃₆(OH)) 14 15,17 7 15, 17 15 11 15,14 15 15,14 14 15 1 14 16 16 17

Leached Residue 14 1,11 Intensity 13 15 15,16 15 16 15 11 14 15 13 16 15,14 15,14 14 1

Sinter 1,8 1,11 7 11 7 5 12 10 8 1 3 7 7 3 6 1 1 1 0 3 8 7 8, 11,12 5 10 8 8 8 4 8 2 9 8 Bauxite Residue 2-θ 416 10 20 30 40 50 60 417 Figure 19: XRD of mineralogical phases for BaO additions (20 g BaO/100 g bauxite residue) at atmospheric 418 conditions at 900 °C and for 2 hours.

2 1

419 (a) (b)

420 Figure 20. SEM mineral identifications in BaO, Na2CO3 and bauxite residue (a) sinter products; (1) 421 Barium titanate, and (b) Leached residue; (2) Barium titanium iron calcium aluminium silicon oxide 422 (3) Barium iron calcium aluminium silicon oxide.

423 3.4. Experimental results of leaching tests from 50% excess soda sinters with CaO, MgO and BaO additions

424 The results of the leaching tests of Na2CO3 only system, CaO + Na2CO3 system, MgO + Na2CO3 425 system and BaO + Na2CO3 system as shown in Figure 21, further confirms the outcomes of the 426 thermodynamic and experimental analysis. Aluminium recovery of the systems was in the range of 427 64 to 75% in total, with the highest to lowest recovery coming from BaO>Na2CO3>MgO>CaO added 428 systems. The co-recovery of silica was higher in both Na2CO3 only and MgO added system, at about 429 20%, indicating that existence of partially leachable silica phases was formed during sintering. These 430 phases were most like Na2SiO3 phase, as part of the soda added reacted preferentially with the 431 available silica instead. In the CaO and BaO systems, the thermodynamically favourable formations 432 of Ca2SiO4 and Ba2SiO4, and, the formation of a myriad of components such as sodium calcium

Minerals 2019, 9, x FOR PEER REVIEW 17 of 20

433 silicate, grossular and garnet complex phases, locked silica into the matrix and hindered silica 434 leaching from sinters. 435 The sodium recovery increases to 94% in systems that managed to stop silicon from turning into 436 leachable products, whereas Na2CO3 only system exhibited the least amount of soda recovery. 437 Barium oxide addition in the soda sintering system assisted the best recovery for both aluminium 438 and sodium while suppressing silica in system, however, the price of barium oxide (US $300- 439 $500/metric tonne) when compared to lime (US $50-200/metric tonne), magnesium oxide (US $80- 440 $200/metric tonne) and soda (US $210-$250/metric tonne), becomes considerably more expensive, for 441 the aluminium and sodium recovered in bauxite residue. 100 94.9 94.0 SiO₂ Al₂O₃ Na₂O 88.7 90 84.6 80 75.2 70.5 69.6 70 64.2 60

30 Percent ofPercent % Recovery 20.1 21.0 20

10 1.9 5.0 0 Na₂CO₃ CaO + Na₂CO₃ MgO + Na₂CO₃ BaO + Na₂CO₃ Reactants added to Bauxite Residue 442 Weight Ratio of BR:Na₂CO₃:Metal Oxide is 1:0.25:0.2

443 Figure 21. Results of experimental recovery of Al2O3, SiO2, and Na2O with CaO, MgO and BaO added 444 with 50% excess soda, compared to only 50% excess soda, Na2CO3, at 900 °C for 2 hours.

445 A simplified mass balance of the laboratory scale experiment of optimised recovery at 50% 446 excess soda (Soda only) and Stoichiometric carbon with 50% excess soda (Carbon + Soda) is presented 447 in the flowchart (Figure 22) and mass balance (Table 3).

448

449 Figure 22. Total proposed flowsheet for the recovery of aluminium and sodium.

450 Table 3. Mass balance of input and output from sintering system (Soda only vs Carbon + Soda).

Leached Metallurgi Mass Balance Bauxite Na₂CO₃ Sinter CO₂/CO NaOH (kg) Residue Residue cal Coke Soda only 1 0.3 1.08 0.22 0.288 0.737 - Carbon+ Soda 1 0.3 0.905 0.318 0.288 0.6325 0.122

Minerals 2019, 9, x FOR PEER REVIEW 18 of 20

451 4. Conclusions 452 The recovery of aluminium from Greek bauxite residue was found to reach a plateau of 70-75% 453 when excess soda, retention time of sintering and temperature of sintering were investigated. This 454 was confirmed in both thermodynamic and experimental results, whereas sodium recovery levels 455 were optimised at 85-90%. It is recommended that the minimum amount of soda reagent required to 456 convert sodium and aluminium into leachable products, through this optimised study, should be 457 used at 50% excess soda levels, aiding sodium-aluminium recoveries at feasible amounts and thus, 458 processing and conditioning residue for use downstream. 459 Thermodynamic study of the bauxite residue notes that the type of complex desilication 460 products that aluminium remains trapped in bauxite residue, such as sodium-based cancrinite or 461 sodalite, or calcium-based grossular and garnet type product or cancrinite with calcium inclusions, 462 transforms into equilibrium amounts of sodium aluminate and sodium aluminium silicates 463 converted during sintering. Additional additives such as CaO, MgO and BaO are also tested to 464 consume silica from adhering to sodium and aluminium, forming other thermodynamically 465 favourable and stable silica-bound species and liberating the main elements for recovery. 466 The addition of barium oxide aids the recovery to achieve 75% of recovered aluminium and 94% 467 recovered sodium respectively. The presence of calcium existing in the Greek bauxite residue 468 prevents the recovery of aluminium due to NaAlSiO4/NaAlO2 sintered equilibrium formations. 469 Increased lime additions will preferentially react with iron in this particular system. Addition of 470 carbon within inert conditions only assisted aluminium recovery to 75% and sodium recovery to 83%. 471 The benefit of addition of carbon and in inert system, however, is to assist pre-reduction of iron and 472 its recovery downstream. Optimal recovery of aluminium and sodium for Greek bauxite residue is 473 recommended to be 70% and 85% respectively, when sintered with 50% excess stoichiometric 474 Na2CO3.

475 Acknowledgments: Research leading to these results has received funding from the European Community’s 476 Horizon 2020 Program ([H2020/2014–2019]) under Grant Agreement no. 636876 (MSCA-ETN REDMUD). Special 477 thanks to NTUA researchers Ioanna Giannoupoulou and Alexandra Alexandri; and AoG’s team of helpers, 478 Athina Dritsa and Athina Fillipou for experimental and analytical support and efforts and to the rest of the 479 REDMUD team for the countless support. This publication reflects only the author’s view, exempting the 480 Community from any liability. Project website: http://www.etn.redmud.org.

481 References

482 1. IAI. Current IAI Statistics: Alumina Production In Total for 2017, Available online: http://www.world- 483 aluminium.org/statistics/alumina-production/ (2nd July 2018). 484 2. Mytilineos S.A. Metallurgy, Available online: http://www.mytilineos.gr/en-us/metallurgy-and- 485 mining/activities (3rd July 2018). 486 3. Power, G.; Gräfe, M. and Klauber, C.; Bauxite residue issues: I. Current management, disposal and storage 487 practices. Hydromet 2011, No. 108, (1–2), pp. 33-45. http://dx.doi.org/10.1016/j.hydromet.2011.02.006. 488 4. Klauber, C.; Gräfe, M. and Power, G.; Bauxite residue issues: II. options for residue utilization. Hydromet 489 2011, No. 108, (1–2), pp. 11-32. http://dx.doi.org/10.1016/j.hydromet.2011.02.007. 490 5. Alkan, G.; Schier, C.; Gronen, L.; Stopic, S. and Friedrich, B.; A Mineralogical Assessment on Residues after 491 Acidic Leaching of Bauxite Residue (Red Mud) for Titanium Recovery. Metals 2017, No. 7, (11), pp. 458. 492 6. Borra, C. R.; Blanpain, B.; Pontikes, Y.; Binnemans, K. and Van Gerven, T.; Recovery of Rare Earths and 493 Major Metals from Bauxite Residue (Red Mud) by Alkali Roasting, Smelting, and Leaching. J Sustainable 494 Metall 2017, No. 3, (2), pp. 393-404. 10.1007/s40831-016-0103-3. 495 7. Borra, C. R.; Blanpain, B.; Pontikes, Y.; Binnemans, K. and Van Gerven, T.; Comparative Analysis of 496 Processes for Recovery of Rare Earths from Bauxite Residue. JOM 2016, No. 68, (11), pp. 2958-2962. 497 10.1007/s11837-016-2111-y. 498 8. Avdibegovic, D.; Regadio, M. and Binnemans, K.; Efficient separation of rare earths recovered by a 499 supported ionic liquid from bauxite residue leachate. RSC Advances 2018, No. 8, (22), pp. 11886-11893. 500 10.1039/C7RA13402A. 501 9. Evans, K.; The history, challenges, and new developments in the management and use of bauxite residue. 502 J Sustainable Metall 2016, pp. 1-16. 10.1007/s40831-016-0060-x.

Minerals 2019, 9, x FOR PEER REVIEW 19 of 20

503 10. Smith, P.; Reactions of lime under high temperature Bayer digestion conditions. Hydromet 2017, No. Vol. 504 170, pp. 16-23. http://dx.doi.org/10.1016/j.hydromet.2016.02.011. 505 11. Mishra, B. and Gostu, S.; Materials sustainability for environment: Red-mud treatment. Frontiers of Chemical 506 Science and Engineering 2017, No. 11, (3), pp. 483-496. 10.1007/s11705-017-1653-z. 507 12. Bruckard, W. J.; Calle, C. M.; R.H. Davidson, R. H.; Glenn, A. M.; Jahanshahi, S.; Somerville, M. A.; Sparrow, 508 G. J. and Zhang, L.; Smelting of bauxite residue to form a soluble sodium aluminium silicate phase to 509 recover alumina and soda. Trans Inst Min Metall, Sect C 2010, No. 119, (1), pp. 18-26. 510 10.1179/037195509X12518785461760. 511 13. Kaußen, F. M.; Sofras, I. and Friedrich, B.; Carbothermic reduction of red mud in an EAF and subsequent 512 recovery of aluminum from the slag by pressure leaching in caustic solution. Bauxite Residue Valorisation 513 and Best Practices (BR 2015), Leuven, Belgium 5-7 October 2015; Pontikes, Y.; ACCO, 2015, No. 1, pp. 185- 514 189. 515 14. Kaußen, F. M. and Friedrich, B.; Methods for Alkaline Recovery of Aluminum from Bauxite Residue. J 516 Sustainable Metall 2016, No. 2, (4), pp. 353-364. 10.1007/s40831-016-0059-3. 517 15. Tam, P. W. Y.; Xakalashe, B.; Friedrich, B.; Panias, D. and Vassiliadou, V.; Carbothermic Reduction of 518 Bauxite Residue for Iron Recovery and Subsequent Aluminium Recovery from Slag Leaching. Proceedings 519 of 35th International Conference and Exhibition of The International Committee for Study of Bauxite, 520 Alumina & Aluminium (ICSOBA), Hamburg, Germany, 2-5 October 2017; ICSOBA, 2017, No. 42, pp. 603- 521 613. 522 16. Dobos, G.; Horváth, G. and Felföldi, Z.; Karšulin, M. C. E. and Marušić, R. E. Complex utilization of red 523 mud including the production of pig iron, sodium hydroxide, alumina and building material. In Travaux 524 ICSOBA. Hungary, International Colloquium on Alumina Production from Low Grade Bauxites Banská 525 Bistrica - Žiar nad Hronom. 1974. No. 12, pp. 151-158. 526 17. Li, G.; Jiang, T.; Liu, M.; Zhou, T.; Fan, X. and Qiu, G.; Beneficiation of High-Aluminium-Content Hematite 527 Ore by Soda Ash Roasting. Miner Process Extr Metall Rev 2010, No. 31, (3), pp. 150-164. 528 10.1080/08827501003727030. 529 18. Li, G.; Liu, M.; Rao, M.; Jiang, T.; Zhuang, J. and Zhang, Y.; Stepwise extraction of valuable components 530 from red mud based on reductive roasting with sodium salts. J Hazard Mater 2014, No. 280, pp. 774-780. 531 https://doi.org/10.1016/j.jhazmat.2014.09.005. 532 19. Kaußen, F. M. and Friedrich, B.; Soda Sintering Process for the Mobilisation of Aluminium and Gallium in 533 Red Mud Bauxite Residue Valorisation and Best Practices Conference (BR 2015) Leuven, Belgium Pontikes, 534 Y.; ACCO, 2015, No. 1, pp. 157-164. 535 20. Zheng, K.; Gerson, A. R.; Addai-Mensah, J. and Smart, R. S. C.; The influence of sodium carbonate on 536 sodium ahminosilicate crystallisation and solubility in sodium aluminate solutions. Journal of Crystal 537 Growth 1997, No. 171, pp. 197-208. https://doi.org/10.1016/S0022-0248(96)00480-0. 538 21. Bánvölgyi, G.; Scale Formation in Alumina Refineries. Proceedings of 34th International Conference and 539 Exhibition of The International Committee for Study of Bauxite, Alumina & Aluminium (ICSOBA), Quebec, 540 Canada, 3-6 October 2016; ICSOBA, 2016, No. 45, pp. 101-114. 541 22. Rivera, R. M.; Ounoughene, G.; Borra, C. R.; Binnemans, K. and Van Gerven, T.; Neutralisation of bauxite 542 residue by carbon dioxide prior to acidic leaching for metal recovery. Miner Eng 2017, No. 112, pp. 92-102. 543 https://doi.org/10.1016/j.mineng.2017.07.011. 544 23. Habashi, F. Karl Josef Bayer (1847-1904). In International Committee For Study of Bauxite, Alumina and 545 Aluminium (ISCOBA), Movement of Scientists and the Production of Aluminium, ICSOBA Newsletter. 546 2015. No. 13, pp. 9-13. 547 24. Dennis, W. Metallurgy: 1863-1963. In, Aldine Transaction. 548 25. Kaußen, F. M. and Friedrich, B.; Phase characterization and thermochemical simulation of (landfilled) 549 bauxite residue (“red mud”) in different alkaline processes optimized for aluminum recovery. Hydromet 550 2018, No. 176, pp. 49-61. https://doi.org/10.1016/j.hydromet.2018.01.006. 551 26. Smith, P.; The processing of high silica bauxites — Review of existing and potential processes. Hydromet 552 2009, No. 98, (1), pp. 162-176. https://doi.org/10.1016/j.hydromet.2009.04.015. 553 27. Meher, S. N.; Rout, A. K. and Padhi, B. K.; Extraction of Alumina from Red Mud by Divalent Alkaline Earth 554 Metal Soda Ash Sinter Process. Light Metals 2011 2011, pp. 231-236. 10.1002/9781118061992.ch41.

Minerals 2019, 9, x FOR PEER REVIEW 20 of 20

555 28. Meher, S. N. and Padhi, B. K.; A novel method for extraction of alumina from red mud by divalent alkaline 556 earth metal oxide and soda ash sinter process Int J Environ Waste Manage 2014, No. 13, (3), pp. 231-245. 557 10.1504/IJEWM.2014.059932. 558 29. Raghavan, P. K. N.; Kshatriya, N. K. and Wawrynink, K.; Recovery of Metal Values from Red Mud. In Light 559 Metals 2011; Lindsay, S. J.; Springer International Publishing: Cham, 2016, pp. 103-106, 10.1007/978-3-319- 560 48160-9_18. 561 30. Hertel, T.; Blanpain, B. and Pontikes, Y.; A Proposal for a 100 % Use of Bauxite Residue Towards Inorganic 562 Polymer Mortar. J Sustainable Metall 2016, No. 2, (4), pp. 394-404. 10.1007/s40831-016-0080-6. 563 31. Tathavadkar, V.; Jha, A.; Fülöp, T.; Török, T. I. and Rédey, A.; A Comparison of the Mineralogical 564 Characteristics and Alkali Roasting Behaviour of Red Mud of Different Origins. Global Symposium on 565 Recycling, Waste Treatment and Clean Technology (REWAS 2004), Madrid, Spain Gaballah, I., Mishra, B., 566 Solozabal, R. and Tanaka, M.; TMS Publications, 2004, No. 1, pp. 401-410. 567 32. Whittington, B. I.; The chemistry of CaO and Ca(OH)2 relating to the Bayer process. Hydromet 1996, No. 43, 568 (1), pp. 13-35. https://doi.org/10.1016/0304-386X(96)00009-6. 569 33. Gelencsér, A.; Kováts, N.; Turóczi, B.; Rostási, Á.; Hoffer, A.; Imre, K.; Nyirő-Kósa, I.; Csákberényi- 570 Malasics, D.; Tóth, Á.; Czitrovszky, A.; Nagy, A.; Nagy, S.; Ács, A.; Kovács, A.; Ferincz, Á.; Hartyáni, Z. 571 and Pósfai, M.; The Red Mud Accident in Ajka (Hungary): Characterization and Potential Health Effects of 572 Fugitive Dust. Environ Sci Technol 2011, No. 45, (4), pp. 1608-1615. 10.1021/es104005r. 573 34. Mohapatra, B. K.; Mishra, B. K. and Mishra, C. R.; Investigation of Alumina Discharge into the Red Mud 574 Pond at NALCO's Alumina Refinery, Damanjodi, Orrisa, India. In Light Metals 2011 Lindsay, S. L.; The 575 Minerals, Metals and Materials Society 2011, pp. 93-96, 10.1002/9781118061992.ch16. 576 35. Meher, S. N.; Route, A. K. and Padhi, B. K.; Recovery of Al and Na Values from Red Mud by BaO-Na2CO3 577 Sinter Process. E-Journal of Chemistry 2011, No. 8, (3), pp. 1387-1393. 10.1155/2011/528134. 578 36. Meher, S. N. and Padhi, B. K.; Effects of MgO and Na2CO3 Additives, Sintering Temperature and Leaching 579 Conditions for Extraction of Alumina from Bayer’s Process Waste Residue (Red Mud). Chemical Science 580 Transactions 2012, No. 1, (2), pp. 456-462. 10.7598/cst2012.180. 581 37. Vind, J.; Malfliet, A.; Bonomi, C.; Paiste, P.; Sajó, I. E.; Blanpain, B.; Tkaczyk, A. H.; Vassiliadou, V. and 582 Panias, D.; Modes of occurrences of scandium in Greek bauxite and bauxite residue. Miner Eng 2018, No. 583 123, pp. 35-48. https://doi.org/10.1016/j.mineng.2018.04.025. 584 38. Sajó, I. E. XDB Powder Diffraction Phase Analytical System, Version 3.0, User's Guide. In. Budapest. 2005. 585 39. Hodge, H.; Tam, P. W. Y.; Vaughan, J. and Panias, D.; Bauxite Residue Sinter Phase Transformations. 5th 586 International Slag Valorisation Symposium Leuven, Belgium, 3-5 April 2017; Iacobescu, R. I. and Malfliet, 587 A.; Color Image, 2017, No. 5. 588 40. Meher, S. N.; Alumina extraction from red mud by CaCO3 and Na2CO3 sinter process. International Journal 589 of Chemical Studies 2016, No. 4, (1), pp. 122-127.

590 © 2019 by the authors. Submitted for possible open access publication under the 591 terms and conditions of the Creative Commons Attribution (CC BY) license 592 (http://creativecommons.org/licenses/by/4.0/).