A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES

Transcription

1 A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES Haiyan Zheng 1 & Toru H. Okabe 2 1 Graduate School of Engineering, the University of Tokyo, Japan 2 Institute of Industrial Science, the University of Tokyo, Japan 1

2 A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES 1. Introduction 2. Thermodynamic analysis 3. Experimental 4. Experimental results 5. Summary 2

3 1. Introduction A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES 1. Introduction Background Purpose of this study 2. Thermodynamic analysis 3. Experimental 4. Experimental results 5. Summary 3

4 1. Introduction The Kroll process Chlorination: Ti ore (s) + C (s) + 2 Cl 2 (g) TiCl 4 (l) + MCl x (s, g) + CO 2 (g) Reduction: TiCl 4 (l) + 2 Mg (s) Ti (s) + 2 MgCl 2 (l) M: Impurity element in the ore Electrolysis: MgCl 2 (l) Mg (s) + Cl 2 (g) Reduction reactor for the Kroll process Mg & TiCl 4 feed port Mg & MgCl 2 recovery port Metallic reaction vessel Titanium sponge Ti / Mg / MgCl 2 mixture Furnace The essential advantage: High-purity Ti available The critical disadvantage: Low productivity 4

5 1. Introduction Chlorine cycle in the Kroll process Cl 2 Additional Cl 2 supply to compensate for chlorine loss C Ti ore Ti feed Carbo-chlorination CO x Chloride wastes Chlorine loss (FeCl x, AlCl 3, ) TiCl 4 Mg Although major portion of chlorine in the Kroll process is recycled, chloride wastes are generated in the Kroll process. Reduction MgCl 2 Electrolysis Ti Product Ti The generation of chloride wastes causes not only chlorine loss but also environmental problems. Cl 2 Mg Chlorine recycle Magnesium recycle 5

6 1. Introduction Current Ti scrap recycle 1) Ti scrap is used for producing ferro-alloys for steel making. 2) Ti smelting factory Ingot Ti mill product manufacturer Scrap Scrap Mill product Ti product consumer In the future, amount of low-purity Ti scrap will increase, and a new recycling process of Ti scrap is required. 6

7 1. Introduction The purpose of this study Low-grade Ti ore (FeTiO X ) MCl x (Cl 2 ) FeCl x Ti scrap 1 Selective chlorination 2 Combined Today s recovery topic of Ti and Cl Upgraded Ti ore (TiO 2 ) FeCl x (+ AlCl 3 ) Fe TiCl 4 3 Ti Ti smelting (e.g., Kroll process or PRP process) MCl x (M = Fe, Al, Si ) A new Ti smelting process combined with iron removal from low-grade Ti ore by selective chlorination and efficient Ti scrap recovery by utilizing chlorine wastes is investigated with the objective of reducing the production cost and decreasing the environmental burden. 7

8 1. Introduction Today s topic Low-grade Ti ore (FeTiO X ) MCl x (Cl 2 ) FeCl x Ti scrap 1 Selective chlorination 2 Combined Today s recovery topic of Ti and Cl Upgraded Ti ore (TiO 2 ) FeCl x (+ AlCl 3 ) Fe TiCl 4 3 Ti Ti smelting (e.g., Kroll process or PRP process) MCl x (M = Fe, Al, Si ) Ti (s) + FeCl x (l, g) TiCl 4 (g) + Fe (s) Ti metal scrap can be recycled. Chlorine in the chloride wastes can be efficiently recovered. Low-grade Ti ore can be used as the feed material. Cost of waste treatment can re reduced. 8

9 2. Thermodynamics analysis A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES 1. Introduction 2. Thermodynamic analysis 3. Experimental 4. Experimental results 5. Summary 9

10 2. Thermodynamics analysis Chemical potential diagram for the Ti-Cl-O system Oxygen partial pressure, log p O2 (atm) Ti-Cl-O system, p Cl2 = 0.1 atm (high p Cl2 atmosphere) TiO 2 (s) C (s)/co 2 (g) eq. TiCl 4 (g) Temperature T / K CO (g) / CO 2 (g) eq. C (s) / CO (g) eq. When p O2 is high, TiCl 4 cannot be obtained even under a high p Cl2 atmosphere. When C or CO is introduced into the system, TiCl 4 is generated by the chlorination of TiO 2 under a high p Cl2 atmosphere. 10

11 2. Thermodynamics analysis Chemical potential diagram for the Fe-Ti-Cl system Fe-Ti-Cl system, T = 1100 K Cl 2 (g) Ti (s) + FeCl x (l, g) TiCl 4 (g) + Fe (or FeTi, s) log p Cl2 (atm) FeCl 3 (g) FeCl 2 (l) A TiCl 4 (g) TiCl 3 (s) B Fe (s) FeTi (s) Ti (s) log a Ti TiCl 2 (s) log a Fe -60 Ti present in the Ti scrap can be extracted by iron chlorides. or TiCl 4 can be obtained by reacting Ti scraps with chloride wastes. 11

12 2. Thermodynamics analysis Vapor pressure, log p i (atm) Fe (s,l) TiCl 2 (s) Ti (s,l) Temperature, T / K FeCl 2 (s,l) TiCl 3 (s) TiCl 4 (l) FeCl 3 (s,l) Vapor pressure of some selected chlorides and metals -6 Region suitable for vaporization chlorides The separation of chlorides and recovery of high-purity TiCl 4 are possible by controlling the deposition temperature Reciprocal temperature, 1000 T -1 /K -1 Vapor pressure of some chlorides and metals as a function of reciprocal temperature. 12

13 3. Experimental A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES 1. Introduction 2. Thermodynamic analysis 3. Experimental 4. Experimental results 5. Summary 13

14 3. Experimental Experiment apparatus (1) (Deposits after the experiment) Quartz tube Graphite crucible Vacuum pump Ar gas Silicone rubber plug Heating element Sample mixture e.g., FeCl 2 + Ti powder 5 cm Exp. # CC T = 1100 K; t = 1h or 3h; At atmosphere Ti (s) + FeCl 2 (l, g) TiCl 4 (g) + Fe (s) 14

15 3. Experimental Experiment apparatus (2) (Deposits after the experiment) Vacuum pump Ar gas Quartz tube Graphite crucible Silicone rubber plug NaOH gas trap Heating element Sample mixture e.g., FeCl 2 + Ti powder 5 cm Exp. # CA, CB, CD, CF NaOH (s) + MCl x (g) NaCl (s) + M(OH) x (s) 15

16 3. Experimental Experiment conditions Exp. No. Mass of feed materials, w i / g Ti scraps FeCl2 (Powder) NaOH Mass Ratio w Ti / w FeCl2 Reaction temp., T / K Reaction time, t / h Atmosphere* CA a Ar CB a Ar CC a Ar CD b Ar CE b Ar CF c Ar a: Ti powder was used in this experiment. b. Ti shot was used in this experiment. c: Ti turning was used in this experiment. *: Reduced atmosphere (0.2 atm at room temperature). Ti (s) + 2 FeCl 2 (l, g) = TiCl 4 (g) + 2 Fe (s) Stoichiometric amount of Ti to FeCl 2 is 1:

17 4. Experimental results A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES 1. Introduction 2. Thermodynamic analysis 3. Experimental 4. Experimental results XRD, XRF, ICP-AES, and potentiometric titration method 5. Summary 17

18 4. Experimental results The distribution of the temperature Distribution of temperature in the horizontal furnace. (K) K K K (cm) Assembled quartz tube after experiment. 5 cm 18

19 4. Experimental results Observation Assembled quartz tube after experiment. 515 K 990 K 1100 K 25~28cm 5 cm The image of the obtained residue and deposit after experiment. Silicone plug Solid (White) Flake (Brown) Residue (Black) 5 cm Silicone plug after experiment 5 cm 2 cm 1 cm Deposit on the surface of the NaOH gas trap Deposit inside the quartz tube Residue in the graphite crucible Melting point of FeCl 2 : atm Melting point of TiCl 4 : atm 19

20 4. Experimental results Ti powder without NaOH gas trap: Composition Analytical results of the samples before and after heating, and the deposits obtained on the surface of silicone plug and inside the quartz tube after heating. Exp. CC Ti Concentration of element i, C i (mass%) Fe Cl Initial sample before heating Residue in the graphite crucible Deposit inside the quartz tube Deposit on the surface of silicone plug 15.7 a 8.95 b 0.33 c (16.7 b ) The value excludes carbon and gasous elements except Cl. a: Calculated. b: Determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES). c: Determined by X-ray fluorescence analysis (XRF). d: Determined by potentiometric titration method a 91.1 b 56.2 c (2.22 b ) C Ti : 15.7% 8.95%. C Fe : 37.2% 91.1%. TiCl x (TiCl 4 ) obtained. The silicone plug was damaged due to the reaction with TiCl a c (81.1 d ) 20

21 4. Experimental results Before experiment After experiment Ti powder without NaOH gas trap: XRD Intensity, I (a. u.) Intensity, I (a. u.) (a) (b) : α -Ti (JCPDS # ) : FeCl 2 2H 2 O (JCPDS # ) : α - Fe (JCPDS # ) : FeCl 2 2H 2 O (JCPDS # ) Angle, 2θ (deg.) XRD patterns of the sample before experiment (a) and the residue after experiment (b) (Exp. CC) Fe was generated at heating zone. Ti (s) + FeCl 2 (l, g) TiCl 4 (g) + Fe (s) 21

22 4. Experimental results Discussion (1) Fe-Ti-Cl system, T = 1100 K log p Cl2 (atm) FeCl 3 (g) FeCl 2 (l) Cl 2 (g) A TiCl 4 (g) TiCl 3 (s) Fe (s) FeTi (s) Ti (s) TiCl 2 (s) -60 Ti (s) + FeCl 2 (l, g) TiCl 4 (g) + Fe (s) Reaction proceeded at point A under Fe(s) / FeCl 2 (s) / TiCl 4 (g) equilibrium. The obtained experimental results are in good agreement with the thermodynamic analysis log a Ti log a Fe Chemical potential diagram for the Fe-Ti-Cl system at 1100 K 22

23 4. Experimental results Discussion (2) Ti in Ti scraps was recovered by FeCl 2 as the form of TiCl 4, but the silicone plug was damaged due to the reaction with the TiCl 4. Before experiment After experiment NaOH was introduced as a gas trap for recovering TiCl 4. 23

24 4. Experimental results Ti powder: Composition Analytical results of the samples before and after heating, and the deposits obtained on the surface of the NaOH gas trap and inside the quartz tube after heating. Exp. CB Ti Concentration of element i, C i (mass%) a Fe Cl Initial sample before heating 14.6 a 37.5 a 47.8 a Residue in the graphite crucible 4.90 b 95.1 b - Deposit inside the quartz tube 2.71 c 54.6 c 42.7 c Deposit on the surface of the NaOH gas trap (16.7 b ) (0.85 b ) (87.9 d ) The value excludes carbon and gasous elements except Cl. a: Calculated. b: Determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES). c: Determined by X-ray fluorescence analysis (XRF). d: Determined by potentiometric titration method. C Ti : 14.6% 4.90%. C Fe : 37.5% 95.1%. There was no damage on the silicone plug. The obtained TiCl 4 was recovered by NaOH successfully. 24

25 4. Experimental results Before experiment After experiment Intensity, I (a. u.) Intensity, I (a. u.) (a) (b) Ti powder: XRD : α -Ti (JCPDS # ) : FeCl 2 2H 2 O (JCPDS # ) : α - Fe (JCPDS # ) Angle, 2θ (deg.) XRD patterns of the sample before experiment (a) and the residue after experiment (b) (Exp. CB) Fe was generated at heating zone. Ti (s) + FeCl 2 (l, g) TiCl 4 (g) + Fe (s) 25

26 4. Experimental results Ti granule and turning: Composition Exp. CD (Feed mateirial: Ti granule) Initial sample before heating Residue in the graphite crucible Deposit inside the quartz tube Deposit on the surface of NaOH gas trap Exp. CF (Feed material: Ti turning) Initial sample before heating Residue in the graphite crucible Deposit inside the quartz tube Deposit on the surface of the NaOH gas trap The value excludes carbon and gasous elements except Cl. a: Calculated. b: Determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES). c: Determined by the potentiometric titration method. d: Determined by X-ray fluorescence analysis (XRF) a Concentration of element i, C i (mass%) Ti 62.8 b 0.10 b 0.15 b 0.06 d (0.04 d ) Fe 37.4 a 37.2 b 49.7 b 0.38 b 50.3 d (1.29 d ) Black coat was formed on the surface of the residue. The residue was magnetic material. Fe element presents in the residue after heating. Cl 47.5 a c 99.5 c Concentration of element i, C i (mass%) a Ti 13.6 a 29.1 d Fe 38.1 a 64.7 d Cl 48.3 a 6.18 d 49.6 d (98.7 d ) 26

27 4. Experimental results Exp. # Form of Ti scraps Mass of feed materials, w i / g Ti scraps FeCl2 (Powder) Mass balance Mass of the obtained sample, w / g Concentration of Ti (mass%) Recovery Ratio of Ti, R (%) CA b Powder CB b Powder CC b Powder CD c Granule CE c Granule CF d Turning a: Experiment date. b: Ti powder was used in this experiment. c: Ti granules was used in this experiment. d: Ti turing was used in this experiment. When Ti powder was used as the feed material, recovery ratio of Ti was obviously higher than those when Ti granule or turning was used. The reaction speed was affected by the morphology of the Ti scraps. 27

28 5. Summary A NOVEL RECYCLING PROCESS OF TITANIUM METAL SCRAPS BY USING CHLORIDE WASTES Summary 1. Ti in Ti scraps was extracted by chloride wastes as the form of TiCl Fe was generated at heating zone. 3. The obtained experimental results are in good agreement with the thermodynamic analysis: Ti (s) + FeCl 2 (l, g) TiCl 4 (g) + Fe (s) 4. The recovery ratio of Ti and Cl as well as the reaction speed were largely dependent on the morphology of the Ti scraps: Ti scraps in the form of powder is easier to be recycled by FeCl 2 than Ti granule or Ti turning. 28

29 The purpose of this study Low-grade Ti ore (FeTiO X ) MCl x (Cl 2 ) FeCl x Ti scrap 1 Selective chlorination 2 Combined Today s recovery topic of Ti and Cl Upgraded Ti ore (TiO 2 ) FeCl x (+ AlCl 3 ) Fe TiCl 4 3 Ti Ti smelting (e.g., Kroll process or PRP process) MCl x (M = Fe, Al, Si ) A new Ti smelting process combined with iron removal from low-grade Ti ore by selective chlorination and efficient Ti scrap recovery by utilizing chlorine wastes is investigated with the objective of reducing the production cost and decreasing the environmental burden. 29

30 For Questions and Answers 30

31 Iron removal from low-grade Ti ore by selective chlorination Vacuum pump Analytical results of the obtained sample after selective chlorination. Chloride Condenser (Deposit obtained after exp.) Stainless steel susceptor Graphite crucible Exp. No. Ti ore b SCD Al 0.1 n.d. Concentration of element i, C i (mass%) a Si Ti V Cr n.d. n.d. Mn Fe Ni n.d. n.d. Chlorination Reactor Sample RF coil SCO SCP n.d. n.d n.d Quartz tube SCS n.d gas (N 2 + H 2 O) Ceramic tube 10 cm a: Determined by X-ray fluorescence analysis (XRF), the value excludes carbon and gaseous elements, n.d. = not detected (below 0.1 % ) b: Natural ilmenite ore produced in Vietnam. c: R = 100 {1- (C Fe, after / C Ti, after ) / (C Fe, before / C Ti, before )}. Experimental apparatus for the selective chlorination of titanium ore using radio frequency (RF) furnace. Experimental conditions: T = 973~1293 K t = 3h or 6 h N 2 + H 2 O atmosphere After experiment: C Fe : 51.3% 3.2% Fe was removed from Ti ore successfully. 31

32 Prefrom reduction process (PRP) Ti ore Flux Binder Mixing Slurry Ti ore: Rutile Flux: CaCl 2 Binder: Collodion Rutile+CaCl 2 +Binder T: Room temp., t : 6 h e.g. 40mm 20mm 8mm Preform fabrication Feed preform Ti ore + flux 0.8~1.4 mass % Fe T: 1273 K, t : 1 ~ 2 h T: 1273 K, t : 6 ~10 h Calcium vapor T: Room temp. 50% CH 3 COOH aq., t : 6 h 20% HCl aq., t : 1 h Calcination/iron removal FeCl x TiO 2 feed in flux Sintered feed preform Reduction Ti + CaO + Ca Reduced preform Leaching Waste solution Ti powder Vacuum drying Powder 90 % Fe removal ~0.13 mass % Fe 98 mass % up purity 0.14 mass % Fe 88 % yield 32

33 Production of Titanium Powder Directly from Titanium Ore by Preform Reduction Process (PRP) Ti ore Flux Binder Mixing Slurry Ti ore: Rutile Flux: CaCl 2 Binder: Collodion Rutile+CaCl 2 +Binder T: Room temp.; t : 6 hr e.g. 40mm 20mm 8mm Preform fabrication Feed preform Ti ore + flux 0.8~1.4 mass % Fe T: 1273 K; t : 1 hr~ 2hr T: 1273 K, t : 6 hr~9 hr Calcium vapor T: RT 50% CH3COOH aq., t : 6 hr 20% HCl aq., t : 1 hr Calcination/iron removal FeCl x TiO 2 feed in flux Sintered feed preform Reduction Ti + CaO + flux Reduced preform Leaching Waste solution Vacuum drying Ti powder Powder 90 % Fe removal ~0.13 mass % Fe 98 mass % up purity 0.14 mass % Fe 88 % yield 33

34 Titanium (Ti)? 1. Light and high-strength 2. Corrosion resistance 3. Biocompatibility 4. Special alloy (Shape memory alloy, Superelastic alloy ) 5. Ninth most abundant element Aerospace industry Japan Aerospace Exploration Agency Ocean industry Implant Buildings TMinato-Machi River Place (Osaka Japan) The JAPAN TITANIUM SOCIETY Photo: 34

35 Gregor: Discovery of titanium element 1795 Klaproth: Denominated as Titanium 1887 Nilson, Petterson: Reduction of TiCl 4 by Na Year Hunter: Reduction of TiCl 4 by Na Ti sponge production in Japan [kt] 1925 van Arkel, DeBore: Thermal dissociation of TiCI Kroll: Reduction of TiCl 4 by Mg Current titanium production process based on the Kroll process. 35

36 History of Titanium 1791 First discovered by William Gregor, a clergyman and amateur geologist in Cornwall, England 1795 Klaproth, a German chemist, gave the name titanium to an element re-discovered in Rutile ore Nilson and Pettersson produced metallic titanium containing large amounts of impurities 1910 M. A. Hunter produced titanium with 99.9% purity by the sodiothermic reduction of TiCl 4 in a steel vessel. (119 years after the discovery of the element) 1946 W. Kroll developed a commercial process for the production of titanium: Magnesiothermic reduction of TiCl 4.. Titanium was not purified until 1910, and was not produced commercially until the early 1950s. 36

37 Titanium is the 10 th most abundant element in the earth s crust Rank Element Clark #. 1 8 O Si Al Fe Ca Na K Mg H Ti Cl Mn P C S 0.03 The tenth most abundant element Rank Element Clark # N F Rb Ba Zr Cr Sr V Ni Cu W Li Ce Co Sn Exhausting element 37

38 Comparison of Ti with common metal Symbol Density (g/cm C) Specific strength [(kgf/mm 2 )/(g/cm 3 )] Clarke No. Metal Melting point ( C) Price ( /kg) Production volume (t/world@2004) Iron Fe (Pure) 6.7(SUS304) x 10 8 Aluminum Titanium Al Ti (Pure) 5.1(Pure) 8.9(0.5Mg0.5Si) 24.6(6Al4V) ~ x / / ~ x 10 4 Although Ti is the ninth most abundant element in Earth s crust, its production volume is very small. 38

39 Current status of Ti production (a) Production of Ti sponge in the world (2004) (b) Transition of production volume of Ti mill products in Japan USA 8 kt Kazakhstan 13 kt China 5 kt Total 76.5 kt Japan 23.5 kt (31% share) Russia 27 kt Amount of titanium mill products [kt] kt (2004) Year 2000 Japan has about 30% world market share, and its titanium industry is growing steadily. 39

40 Amount of titanium products, w i / kt Sponge Ti Mill product Year Amount of titanium products, w i / kt Ref (1) Ref (2) Sponge Ti Mill product Ref (1) Ref (1) Ref (2) Year Ref (2) Transition of production volume of titanium sponge and mill products in China. Ref(1): China Titanium Association (Courtesy of Mr. Akiyama, JTS) Ref(2): China Titanium Association (H. Z., Private communication) 40

41 Current status of Ti production (a) Production of Ti sponge in the world (2004) China 5 kt (7% share) USA 8 kt Kazakhstan 13 kt Total 76.5 kt Japan 23.5 kt Russia 27 kt (b) Transition of production volume of Ti mill products in Japan Amount of titanium products, w i / kt Sponge Ti Mill product Ref (1) Ref (2) Ref (2) Ref (2) Ref (1) Ref (1) Year Although China has only about 7% world market share, its titanium industry is growing sharply in recent years. Transition of production volume of titanium sponge and mill products in China. Ref(1): China Titanium Association (Courtesy of Mr. Akiyama, JTS) Ref(2): China Titanium Association (H. Z., Private communication) 41

42 Ocean engineering 0.03 kt (0.28%) Glasses 0.08 kt (0.79%) Medical 0.09 kt (0.97%) Ship 0.19 kt (1.98%) Metallurgy 0.20 kt (2.06%) Other sports leisure 0.20 kt (2.14%) Salt industry 0.23 kt (2.43%) Electric power 0.42 kt (4.49%) Watch 0.47 kt (4.99%) Export 0.82 kt (8.70%) Aerospace 0.91 kt (9.61%) Others 0.92 kt (9.77%) Chemical industry 4.13 kt (43.67% ) Total 9.45 kt Golf 1.00 kt (10.60%) Shipments of titanium mill product in various field s application in China (2004). 42

43 The Kroll process Ti feed (TiO 2 ) Reductant (C) Chlorine (Cl 2 ) Carbo-chlorination Crude TiCl 4 CO 2 FeCl x, AlCl 3 Distillation H 2 S etc. Pure TiCl 4 Other compounds Mg Reduction Electrolysis Sponge Ti + MgCl 2 + Mg MgCl 2 Vacuum distillation Sponge Ti MgCl 2 + Mg TiCl 4 (g) + 2 Mg (l) Ti (s) + 2 MgCl 2 (l) Crushing / Melting Ti Ingot The essential advantage: High-purity titanium available. 43

44 The Kroll process Chlorination: Ti ore (s) + C (s) + 2 Cl 2 (g) TiCl 4 (l) + MCl x (s, g) + CO 2 (g) Reduction: TiCl 4 (l) + 2 Mg (s) Ti (s) + 2 MgCl 2 (l) Electrolysis: MgCl 2 (l) Mg (s) + Cl 2 (g) Reduction reactor for the Kroll process Mg & TiCl 4 feed port Mg & MgCl 2 recovery port Metallic reaction vessel Titanium sponge Ti / Mg / MgCl 2 mixture Furnace The essential advantage: High-purity titanium available. M: Impurity element in the ore 44

45 Advantages of the Kroll process Reduction: TiCl 4 (l) + 2 Mg (s) Ti (s) + 2 MgCl 2 (l) High-purity Ti can be obtained. The separation of Ti and salt is easy. Chlorine and Mg can be efficiently recycled in the system. MgCl 2 electrolysis with high efficiency can be utilized. Reduction process and electrolysis process are separated completely. 45

46 Problems of the Kroll process The critical disadvantage of the Kroll process is its low productivity because of the following factors: Complex, labor consuming, and multi-step batch type Slow production speed (~ 1 ton / day reactor) Huge heat generated during the reduction step It takes time for removing MgCl 2 and cooling the reactor. Contamination of iron from reaction vessel unavoidable Chloride wastes generated although expensive titanium concentrates with high TiO 2 was used The usage of expensive titanium concentrates causes high Ti production cost. The generation of chloride wastes causes chlorine loss in the process. Chloride wastes causes environmental burden. Disposal cost of chloride wastes is high. 46

47 (a) FFC process (Fray et al.) e TiO 2 powder e Carbon anode TiO 2 preform CaCl 2 molten salt (b) OS process (Ono & Suzuki) Direct reduction of TiO 2 (1) Electrolysis Cathode: TiO 2 + 4e- Ti + 2O 2- (a1) Anode: C + x O 2- CO x + 2x e - (a2) TiO 2 + 2Ca Ti + 2O 2- + Ca 2+ (b1) Electrolysis Cathode: Ca e - Ca (b2) Carbon anode Anode: C + x O 2- CO x + 2x e - (b3) Ca CaCl 2 molten salt Several new smelting processes for producing Ti by direct reduction of TiO 2 are under investigation recently. 47

48 (c) EMR/MSE process (Okabe et al.) e e Direct reduction of TiO 2 (2) Current monitor/ controller Carbon anode CaCl 2 -CaO molten salt TiO 2 Ca-X alloy Cathode: TiO 2 + 4e - Ti + 2O 2- (c1) Anode: 2Ca 2Ca e - (c2) Electrolysis Cathode: Ca e - Ca (c3) Anode: C + x O 2- CO x + 2x e - (c4) Overall reaction TiO 2 + C Ti + CO 2 (c5) Problems: The control of the contamination from the system is difficult; The process for producing high purity TiO 2 feed is high cost; The electric efficiency is low; The reduction speed is slow; The separation of salt and Ti is difficult; A large amount of molten salt is used. 48

49 Direct reduction of TiO 2 (3) (d) Preform reduction process (PRP) Feed preform (TiO 2 feed + flux) Reductant vapor Reductant (R = Ca or Ca-X alloy) Calciothermic reduction TiO 2 + 2Ca Ti + 2CaO PRP will be stated in the Chapter 5. 49

50 Features of reductant and feed materials in metallothermic reduction process. Mg Na Ca TiCl 4 Possible to remove Mg and MgCl 2 by distillation. Possible to efficiently eletrolysis MgCl 2 Easy to control purity (strong contamination of carbon) Difficult to remove Na Difficult to control the temperature Easy to purity control (strong resistance to Ni contamination) High energy loss Difficult to remove Ca or CaCl 2 Cost of the reductant production TiO 2 Impossible to remove oxygen Impossible to remove oxygen Difficult to purity control Difficult to remove Ca or CaCl 2 Cost of the reductant production Process with strong resistance to Oxygen 50

51 Kroll Comparison of various processes Advantages High purity titanium available Easy metal / salt separation Established chlorine circulation Utilizes efficient Mg electrolysis Reduction and electrolysis operation can be carried out independently Complicated process Slow production speed Batch type process Disadvantages FFC OS EMR / MSE Simple process Semi-continuous process Simple process Semi-continuous process Resistant to iron and carbon contamination Semi-continuous process Reduction and electrolysis operation can be carried out independently Difficult metal / salt separation Reduction and electrolysis have to be carried outsimultaneously Sensitive to carbon and iron contamination Low current efficiency Difficult metal / salt separation Sensitive to carbon and iron contamination Low current efficiency Difficult metal / salt separation when oxide system Complicated cell structure Complicated process PRP This study Effective control of purity and morphology Flexible scalability Resistant to contamination Small amount of fluxes necessary High speed reduction process Semi-continuous process Titanium scrap enable Facilities for Kroll process can be utilized Difficult recovery of reductant Environmental burden by leaching Difficulty of TiCl 2 handling Multiple reduction process 51

52 1. Introduction The concept of the combined recovery of Ti scraps and chloride wastes Resource scraps Titanium scraps + Costly waste disposal Chloride wastes (FeCl x ) Valuable material Titanium feed (TiCl 4 ) + Cheap wastes disposal Iron scraps (Chlorine free wastes) Titanium in the scraps and chlorine in the wastes are recovered simultaneously. Therefore, not only is chlorine loss in the Kroll process reduced but also waste treatment cost can be reduced. 52

53 Upgrading Ti ore for minimizing chloride wastes FeO x Others TiO x Upgrade FeO x TiO x Others Chloride wastes Ti ore (eg. Ilmenite) Upgraded Ilmenite (UGI) Discarded When low-grade ore is used, a large amount of chloride wastes (e.g., FeCl x ) are generated in the Kroll process. Disposal cost of chloride wastes Environmental issues Causes chlorine loss in the process Currently expensive upgraded ilmenite ore (UGI) is used for reducing chloride waste and environmental burden. 53

54 Ilmenite Becher process. Coal (low ash) Air Reduction (in kiln) Reduced ore Gas + particle Particle -1 mm Screen +1 mm Cyclone Gas Mag. separator Reduced ilmenite Waste NH 4 Cl (Non. mag.) Air Leaching TiO 2 H 2 SO 4 aq. Iron oxide + Sol. Acid Leaching Thickener TiO 2 Iron oxide Sol. Filtering / Drying TiO 2 (Synthetic rutile) TiO 2 92~93%; TiFe 2.0~3.5% 54

55 Benilite process. Ilmenite Reductant (Heavy oil etc.) Fe 2+ / TFe = 80~95% Reduction (in kiln) (18~20% HCl) Reduced ore HCl aq. HCl vapor Leaching (in digestor) 145C (2.5 kg/cm 2 ) *4 hr *2 step Leached ilmenite Water Spray acid Fuel Filtration Roasting TiO 2 Sol. Iron oxide (90% purity) HCl Calcination Absorber TiO 2 (Synthetic rutile) 95% TiO 2 1% TiFe HCl aq. 55

56 Metallothermic reduction process for producing Ti powder (a) Conventional Hunter Process Na TiCl 2 High purity & homogeneous powder can be produced Suitable for uniform reduction Molten salt Ti (s) High cost process Batch type process Large amount of salt is used, and large amount of waste solution generated Uses highly reactive reductant (b) Conventional Metallothermic Reduction Feed powder Reductant vapor Reductant (R = Ca, Mg) Simple and low cost process Flexible scalability Difficult morphology control Sensitive to contamination from reaction container 56

57 The concept of this study Low-grade Ti ore (FeTiO X ) MCl x (Cl 2 ) FeCl x Ti scraps 1 Selective chlorination 2 Chlorine recovery Upgraded Ti ore (TiO 2 ) FeCl x (+ AlCl 3 ) Fe TiCl 4 3 Ti Ti smelting (e.g., Kroll process or PRP process) MCl x (M = Fe, Al, Si ) A new Ti smelting process combined with iron removal from low-grade Ti ore by selective chlorination and efficient chlorine recovery by utilizing Ti scraps is investigated with the objective of reducing the production cost and decreasing the environmental burden 57

58 The objective of this study 1. Iron removal from low-grade Ti ore by selective chlorination method using metal chlorides as the resource of chlorine (Chapter 3) The concept of this study Low-grade Ti ore MCl x FeCl x Ti scraps 2. Chlorine recovery by utilizing Ti scraps (Chapter 4) 3. Ti powder production directly from Ti ore by calciothermic reduction (PRP, Chapter 5) 4. Thermodynamic analysis on the abovementioned processes (Chapter 2) 1 Selective chlorination 2 Upgraded Ti ore FeCl x 3 Ti smelting Ti MCl x Chlorine recovery Fe TiCl 4 58

59 Table Gibbs energy change of formation and reaction in this study. Reactions Gibbs energy change, ΔG ο f or ΔG ο r (kj/mol) a 900 K 1000 K 1100 K 1200 K 1300 K TiO 2 (s ) + FeO(s ) = FeTiO 3 (s ) TiO 2 (s ) + CaO(s ) = CaTiO 3 (s ) TiO 2 (s ) + Fe 2 O 3 (s ) = Fe 2 TiO 5 (s ) TiO 2 (s ) + MgO(s ) = MgTiO 3 (s ) MgCl 2 as chlorine source FeO(s ) + MgCl 2 (s,l ) = FeCl 2 (l,g ) + MgO(s ) FeTiO 3 (s ) + MgCl 2 (s,l ) = FeCl 2 (l,g ) + TiO 2 (s ) + MgO(s ) Fe 2 O 3 (s ) + 3MgCl 2 (g ) = 2FeCl 3 (g ) + 3MgO(s ) TiO 2 (s ) + 2MgCl 2 (s,l ) = TiCl 4 (g ) + 2MgO(s ) MgCl 2 (s,l ) + H 2 O(g ) = MgO(s ) + 2HCl (g ) FeO(s ) + 2HCl(g ) = FeCl 2 (g ) + H 2 O(g ) Fe 2 O 3 (s ) + 6HCl(g ) = 2FeCl 3 (g ) +3 H 2 O(g ) TiO 2 (s ) + 4HCl(g ) = TiCl 4 (g ) + 2H 2 O(g ) FeTiO 3 (s ) + 2HCl(g ) = FeCl 2 (l,g ) + H 2 O(g ) + TiO 2 (s ) CaCl 2 as chlorine source FeO(s ) + CaCl 2 (s,l ) = FeCl 2 (g ) + CaO(s ) FeO(s ) + TiO 2 (s ) + CaCl 2 (s,l ) = FeCl 2 (l,g ) + CaTiO 3 (s ) FeTiO 3 (s ) + CaCl 2 (s,l ) = FeCl 2 (l,g ) + CaTiO 3 (s ) Fe 2 O 3 (s ) + 3CaCl 2 (g ) = 2FeCl 3 (g ) + 3CaO(s ) TiO 2 (s ) + 2CaCl 2 (s,l ) = TiCl 4 (g ) + 2CaO(s ) CaCl 2 (s,l ) + H 2 O(g ) = CaO(s ) + 2HCl (g ) FeTiO 3 (s ) + 2HCl(g ) + CaO(s ) = FeCl 2 (l,g ) + CaTiO 3 (s ) + H 2 O(g ) C(s ) + O 2 (g )= 2CO(g ) a: References [1] I. Barin, Thermochemical Data of Pure Substances, 3rd ed., (Weinheim, Federal Republic of Germany, VCH Verlagsgesellschaft mbh, 1997). 59

60 Reactions Gibbs energy change, ΔG ο f or ΔG ο r (kj/mol) a 900 K 1000 K 1100 K 1200 K 1300 K TiO 2 (s ) + FeO(s ) = FeTiO 3 (s ) TiO 2 (s ) + CaO(s ) = CaTiO 3 (s ) TiO 2 (s ) + Fe 2 O 3 (s ) = Fe 2 TiO 5 (s ) TiO 2 (s ) + MgO(s ) = MgTiO 3 (s ) MgCl 2 as chlorine source FeO(s) + MgCl 2 (s,l ) = FeCl 2 (l,g ) + MgO(s ) FeTiO 3 (s ) + MgCl 2 (s,l ) = FeCl 2 (l,g ) + TiO 2 (s ) + MgO(s ) Fe 2 O 3 (s ) + 3MgCl 2 (g ) = 2FeCl 3 (g ) + 3MgO(s ) TiO 2 (s ) + 2MgCl 2 (s,l ) = TiCl 4 (g ) + 2MgO(s ) MgCl 2 (s,l ) + H 2 O(g ) = MgO(s ) + 2HCl (g ) FeO(s ) + 2HCl(g ) = FeCl 2 (g ) + H 2 O(g ) Fe 2 O 3 (s ) + 6HCl(g ) = 2FeCl 3 (g ) +3 H 2 O(g ) TiO 2 (s ) + 4HCl(g ) = TiCl 4 (g ) + 2H 2 O(g ) FeTiO 3 (s ) + 2HCl(g ) = FeCl 2 (l,g ) + H 2 O(g) + TiO 2 (s ) CaCl 2 as chlorine source FeO(s ) + CaCl 2 (s,l ) = FeCl 2 (g ) + CaO(s ) FeO(s ) + TiO 2 (s) + CaCl 2 (s,l ) = FeCl 2 (l,g ) + CaTiO 3 (s) FeTiO 3 (s ) + CaCl 2 (s,l ) = FeCl 2 (l,g ) + CaTiO 3 (s) Fe 2 O 3 (s ) + 3CaCl 2 (g ) = 2FeCl 3 (g ) + 3CaO(s ) TiO 2 (s ) + 2CaCl 2 (s,l ) = TiCl 4 (g ) + 2CaO(s ) CaCl 2 (s,l ) + H 2 O(g ) = CaO(s ) + 2HCl (g ) FeTiO 3 (s ) + 2HCl(g ) + CaO(s ) = FeCl 2 (l,g ) + CaTiO 3 (s ) + H 2 O(g ) C(s ) + O 2 (g )= 2CO(g ) TiO 2 (s ) + 2Ca(s ) = Ti(s ) + 2CaO(s ) CaTiO 3 (s ) + 2Ca(s ) = Ti(s ) + 3CaO(s ) a: References [1] I. Barin, Thermochemical Data of Pure Substances, 3rd ed., (Weinheim, Federal Republic of Germany, VCH Verlagsgesellschaft mbh, 1997). 60

61 Gibbs energy change of formation and reaction in the Fe-Ti-O system. Reactions Gibbs energy change, ΔG o f or ΔG o r (kj/mol) Ref K 1200 K 1300 K 1273 K a Fe (s) O 2 (g) = FeO (s) Ti (s) + O 2 (g) = TiO 2 (s) Fe (s) + Ti (s) O 2 (g) = FeTiO 3 (s) Fe (s) + Ti (s) + 2 O 2 (g) = Fe 2 TiO 4 (s) TiO 2 (s) + Fe (s) O 2 (g) = FeTiO 3 (s) TiO 2 (s) + 2 Fe (s) + O 2 (g) = Fe 2 TiO 4 (s) TiO 2 (s) + FeO (s) = FeTiO 3 (s) , , , , TiO 2 (s) + 2 FeO (s) = Fe 2 TiO 4 (s) , , , , 8 References [1] I. Barin, Thermochemical Data of Pure Substances, 3rd ed., (Weinheim, Federal Republic of Germany, VCH Verlagsgesellschaft mbh, 1997). [2] Outokumpu HSC Chemistry for Windows, Version 5.0, (Finland, Outokumpu Research Oy Information Service, 2002). [3] O. Knacke, O. Kubaschewski, and K. Hesselmann, Thermochemical Properties of Inorganic Substances, 2nd ed., (Berlin, Federal Republic of Germany, Springer-Verlag, 1991). [4] NIST-JANAF Thermochemical Tables 4th ed., U.S. Bureau of Standards (1998). [5] S. Ito, Phase Equilibria of the Titanium-Iron-Oxygen system as 1,273 K on Titanium Extraction Processing, (Journal of the Mining and Materials Processing Institute of Japan (Vol. 112, p , 1996). [6] J. S. J. Van Devender, Kinetics of Selective Chlorination of ilmenite, (Thermochimica Acta, vol. 124, p , 1988). [7] Special Lecture for the Process Design of the Recycling Material, Distributed Documents, (Summer term, 2003) [8] O. Kubaschewski, High Temp. High pressures 4.1 (1972). a: Interpolated TiO x + n FeO = TiFe n O 1+x+n ΔG r = -30 ~ -9 kj / mol Ti ore is considered as the mixture of TiO x + FeO because ΔG r of TiFe x O y is not large as compared to those of the redox reactions in the related system. a: interpolated 61

62 Oxygen partial pressure, log p O2 (atm) Fe-Cl-O system, T = 1300 K Fe 2 O 3 (s) Fe 3 O 4 (s) FeO (s) Fe (s) FeCl (g) 3 FeCl (g) Chlorine partial pressure, log p Cl2 (atm) CaO (s) / CaCl 2 (l) eq. C (s) / CO (g) eq. H 2 O(g) / HCl (g) eq. MgO (s) / MgCl 2 (l) eq. Chemical potential diagram of the Fe-Cl-O system at 1300 K. A Mechanism of iron removal (FeO x chlorination) CO (g) / CO 2 (g) eq. Ⅰ: H 2 O (g) + CaCl 2 (l) HCl (g) + CaO (s) FeO x (FeTiO x, s) + HCl (g) FeCl 2 (l, g) + H 2 O (g) Ⅱ: FeO x (FeTiO x, s) + CaCl 2 (l) FeCl x (g) + CaO (CaTiO x, s) a CaO <<1 FeO x can be chlorinated using CaCl 2 + H 2 O. 62

63 Mechanism of iron removal (TiO x chlorination) Oxygen partial pressure, log p O2 (atm) Ti-Cl-O system, T = 1300 K TiO 2 (s) Ti 4 O 7 (s) Ti 2 O 3 (s) TiO (s) Ti (s) Ti 3 O 5 (s) TiCl 3 (g) TiCl 4 (g) Chlorine partial pressure, log p Cl2 (atm) Chemical potential diagram of the Ti-Cl-O system at 1300 K. B CaO (s) / CaCl 2 (l) eq. CO (g) / CO 2 (g) eq. C (s) / CO (g) eq. H 2 O(g) / HCl (g) eq. MgO (s) / MgCl 2 (l) eq. TiO x can not be chlorinated using CaCl 2, or CaCl 2 +H 2 O. 63

64 Mechanism of iron removal (Ti ore chlorination) Fe-Cl-O and Ti-Cl-O systems, T = 1300 K Oxygen partial pressure, log p O2 (atm) Chlorine partial pressure, log p Cl2 (atm) Combined chemical potential diagram of the Fe-Cl-O (dotted line) and Ti-Cl-O (solid line) systems at 1300 K. C CaO (s) / CaCl 2 (l) eq. CO (g) / CO 2 (g) eq. C (s) / CO (g) eq. H 2 O(g) / HCl (g) eq. MgO (s) / MgCl 2 (l) eq. Region for selective chlorination of iron FeO x (FeTiO x, s) + HCl (g) FeCl x (g) + H 2 O (g) FeO x (FeTiO x, s) + CaCl 2 (l) FeCl x (g) + CaO (CaTiO x, s) a CaO <<1 FeO x can be chlorinated using CaCl 2 +H 2 O. TiO x can not be chlorinated using CaCl 2, or CaCl 2 +H 2 O. 64

65 TiCl 4 (l) AlCl 3 (s,l) TiCl 3 (s) TiCl 2 (s) Al (s,l) FeCl 3 (s,l) FeCl 2 (s,l) Fe (s,l) Si (s,l) Ti (s,l) Reciprocal temperature, 1000 T -1 /K Vapor pressure, log pi (atm)

66 Si (s,l) Temperature, T / K SiCl 4 (l) -2 Vapor pressure, log p i (atm) Al (s,l) AlCl 3 (s,l) Reciprocal temperature, 1000 T -1 / K -1 Fig. Vapor pressure of aluminium and silicon chlorides as a function of reciprocal temperature. 66

67 Standard Gibbs energy of formation, ΔG f / kj mol Temperature, T / K Ellingham diagram of some selected oxides. Ellingham diagram of some selected oxides 3/2 Fe + O 2 =1/2 Fe 3 O 4 1/2 Fe + O 2 = 1/2 FeO 4 Na + O 2 = 2 Na 2 O Si + O 2 = SiO 2 Ti + O 2 = TiO 2 4/3 Al + O 2 = 2/3 Al 2 O 3 2 Mg + O 2 = 2 MgO 2 Ca + O 2 = 2 CaO TiO 2 (s) + Ca (g) Only Ca can be utilized as reductant for the production of metallic Ti with low oxygen content directly from Ti ore. Ti-1000 mass ppmo Ti-100 mass ppmo Ti-10 mass ppmo ~500ppmO Ti (s) + CaO (s) 67

68 Vapor pressure, log p i (atm) Vapor pressure of some selected metals and chlorides K Na TiCl 3 Mg Ca MgCl 2 KCl NaCl TiCl 2 CaCl K (Reaction temperature) Ti Range of vapor pressure feasible for supplying reductants in vapor form The vapor pressure of Ca: K Ca reductant can be supplied as vapor form for reducing Ti ore in the PRP Temperature, T / K Vapor pressure of selected metals and chlorides. 68

69 Oxygen partial pressure, log p O2 (atm) Impurity removal Si-Cl-O system, T = 1300 K SiO 2 (s) Si (s) SiCl 4 (g) CaO (s) / CaCl 2 (l) eq. CO (g) / CO 2 (g) eq. H 2 (g) / H 2 O(g) eq. C (s) / CO (g) eq. H 2 O(g) / HCl (g) eq. MgO (s) / MgCl 2 (g) eq. SiO 2 (in ore) + 2 MgCl 2 (Flux) SiCl 4 (g) + 2 MgO (s) Chlorine partial pressure, log p Cl2 (atm) 69

70 Ti-Ca binary phase diagram 70

71 Selective chlorination experiment Vacuum pump Chloride Condenser Chlorination Reactor (Deposit obtained after exp.) Stainless steel susceptor Graphite crucible Sample RF coil gas (N 2 + H 2 O) Quartz tube Ceramic tube 10 cm Experimental apparatus for the selective chlorination of titanium ore using radio frequency (RF) furnace. 71

72 Experimental apparatus part (1) (a) φ 30 φ 26.1 φ (b) φ φ Dimensions and appearance of stainless steel susceptor for the RF furnace used for the selective chlorination of titanium ore. 72

73 Experimental apparatus part (2) (a) Ni foil (b) Ni foil Sample mixture (Ti ore + MCl x (+ Carbon powder) ) Graphite crucible Graphite crucible Illustrations and appearance of the sample in the graphite crucible installed in the RF furnace: 73

74 Experiment conditions Exp. No. Mass of feed materials, w / g Ilmenite a CaCl 2 Mixture or layer sample (from bottom to top ) Reaction temp., T / K Reaction time, t / h Atmosphere Flow ratio, cc / min. SCD Mixture N2 + H2O full flow SCO Mixture N2 + H2O 50 SCP Mixture N2 + H2O 50 SCR Mixture N2 + H2O 50 SCS Mixture N2 + H2O 50 a: Natural ilmenite ore produced in Vietnam after pulverization. 74

75 Chapter 3 Results of selective chlorination (1): Observation (a) (b) (c) 5 cm 1 cm (a) Photo taken during the experiment: white smoke (b) The obtained sample in the graphite crucible: sintered sample (c) Deposit inside the quartz tube: white deposit 75

76 Results of selective chlorination (2): XRF Analytical results of the obtained sample after selective chlorination. Exp. No. Ti ore b SCD SCO SCP SCS Al 0.07 n.d. n.d. n.d. n.d. Concentration of element i, C i (mass%) a Si Ti V Cr n.d. n.d. n.d Mn Fe Ni Before Exp C Fe / C Ti a: Determined by X-ray fluorescence analysis (XRF), the value excludes carbon and gaseous elements, n.d. = not detected (below 0.01 % ) b: Natural ilmenite ore produced in Vietnam. c: R = 100 {1- (C Fe, after / C Ti, after ) / (C Fe, before / C Ti, before )}. After Exp Iron removal ratio, R c (%) After experiment: C Fe : 51.3% 3.2% Fe was removed from Ti ore successfully. 76

77 Results of selective chlorination (3): XRD Intensity, I (a. u.) (a) : FeTiO 3 (JCPDS # ) : CaTiO 3 (JCPDS # ) Intensity, I (a. u.) (b) : FeCl 2 (H 2 O) 4 (JCPDS # ) Angle, 2θ (deg.) XRD patterns of (a) the residue in crucible and (b) the deposit inside the quartz tube. (Exp. SCS) The generation of CaTiO 3 was considered to be beneficial for removing iron from Ti ore. (a) (b) The generation of FeCl 2 indicates that iron was removed from Ti ore as the form of FeCl 2. 77

78 Discussion: Mechanism of selective chlorination Generation of CaTiO 3 and FeCl 2 The reactions were considered as follows: I: H 2 O (g) + CaCl 2 (l) 2 HCl (g) + CaO (CaTiO x, s) a CaO <<1 FeO x (FeTiO x, s) + HCl (g) FeCl 2 (l, g) + H 2 O (g) or II: FeO x (FeTiO x, s) + CaCl 2 (l) FeCl 2 (l, g) + CaO (CaTiO x, s) a CaO <<1 78

79 Summary on iron removal from low-grade Ti ore by selective chlorination 1. Iron removal from Ti ore was carried out successfully. In a certain experimental condition: C Fe : 51.3% 3.2% 2. The mechanism of the selective chlorination reactions was considered as follows: FeO x (FeTiO x, s) + HCl (g) FeCl 2 (l, g) + H 2 O (g) or FeO x (FeTiO x, s) + CaCl 2 (l) FeCl 2 (l, g) + CaO (CaTiO x, s) 79

80 Table Analytical results of titanium ores used in this study Sample name Concentration of element i, C i (mass%) a XRF file name Al Si Ca Ti V Cr Mn Fe Ni Nb ilmenaite b 1.60_ 2.30_ 0.10_ 47.62_ 0.56 n.d. 1.78_ 45.49_ 0.03_ n.d ilmenaite c 0.75_ 2.19_ n.d._ 44.46_ 0.36 n.d. 3.43_ 48.38_ n.d._ UGI d 0.05_ 0.42_ 0.03_ 94.68_ 1.02 n.d. 1.30_ 1.96_ n.d._ UGI e 0.04_ 0.36_ n.d._ 95.84_ 1.04 n.d. 0.05_ 2.03_ n.d._ a: Value determined by XRF analysis, n.d.notes not detected (below 0.01 mass%). b: Natural ilmenite ore produced in Australia. c: Natural ilmenite ore produced in Viet Nam. d: Up-graded ilmenite by the Beacher process (see Fig. 1-7). The ore was produced in Australia. e: Up-graded ilmenite by the Benilite process (see Fig. 1-6). The ore was produced in India. 80

81 Table Experimental conditions of selective chlorination (chlorine source: CaCl 2 ) Exp. # Mass of feed materials, w / g Suscepor Version Ilmenite a UGI b CaCl 2 C H 2 O SCA Ni Mixture N 2 + H 2 O full flow SCB Ni Mixture N 2 + H 2 O full flow SCC Ni layer (CaCl 2, C, ore) Ar + H 2 O SCD Ni Mixture N 2 + H 2 O full flow SCE Ni Mixture N 2 + H 2 O SCF Ni Mixture N 2 + H 2 O SCG Ni Mixture N 2 + H 2 O SCH Ni Mixture N 2 + H 2 O SCI Ni Mixture N 2 + H 2 O SCJ Ni Mixture N 2 + H 2 O SCK Ni Mixture N 2 + H 2 O SCL Ni Mixture N 2 + H 2 O SCM Ni Mixture N 2 + H 2 O SCN Ni Mixture N 2 + H 2 O SCO Ni Mixture N 2 + H 2 O SCP Ni Mixture N 2 + H 2 O SCQ Ni Mixture N 2 + H 2 O SCR Ni Mixture N 2 + H 2 O SCS Ni Mixture N 2 + H 2 O SCT Ni Mixture N 2 + H 2 O a: Natural ilmenite ore produced in Vietnam after pulverization. b: Up-graded ilmenite by the Benilite process was produced in India. c: Experiment date. Foil Mixture or layer sample (from bottom to top ) Reaction temp., T / K Reaction time, t / hr Atmosphere Flow ratio, cc / min. Note c 81

82 Table Analytical results of the sample after heating (position: bottom). Note e XRF Concentration of element i, C i (mass%) a Fe / Ti ratio, Iron removal Exp. # C Fe / C Ti ratio, file name Al Si Ti V Cr Mn Fe Ni Nb Before Exp. c After Exp. R d (%) SCA n.d._ SCB n.d._ SCC n.d n.d._ SCD n.d n.d._ SCE n.d n.d._ SCF n.d n.d._ SCG n.d._ SCH n.d n.d._ SCI n.d n.d._ SCJ n.d n.d._ SCK n.d n.d._ SCL n.d n.d._ SCM n.d n.d._ SCN n.d n.d._ SCO n.d n.d._ SCP n.d SCQ n.d SCR n.d n.d._ SCS n.d SCT n.d a: Value determined by XRF analysis, the value excludes carbon and gaseour elements, and n.d. notes not detected (below 0.01 mass%). b: This data isthe same data shown in Table 3-7, because this sample is powder. c: See Table 3-2 for reference. d: R = 100 {1- (C Fe, after / C Ti, after ) / (C Fe, before / C Ti, before )}, This iron removal ratio includes large uncertanties, and listed for reference. e: Experimental date. 82

83 Table Experimental conditions of selective chlorination (chlorine source: MgCl 2 ) Exp. # Mass of feed materials, w / g Suscepor Version Foil Ilmenite a UGI b MgCl 2 C SMA Ni Mixture N 2 + H 2 O SMB Ni Mixture 1300 Failure N 2 + H 2 O SMC Ni Mixture N 2 + H 2 O a: Natural ilmenite ore produced in Vietnam after pulverization. b: Up-graded ilmenite by the Benilite process was produced in India. c: Experiment date. Mixture or layer sample (from bottom to top ) Reaction temp., T / K Reaction time, t / hr Atmosphere Flow ratio, cc / min. Note c 83

84 XRD Intensity, I (a. u.) _SCT_ilm_dep : FeCl 2 (H 2 O) 4 (JCPDS # ) Intensity, I (a. u.) _Tape_bulk Angle, 2θ (deg.) Blank 84

85 5 μm Scanning electron microscopic image of titanium powder obtained after calciothermic reduction of iron removed titanium ore (Exp. E-red.). 85

86 Materials Ti Ti sponge FeCl 2 Form Powder Granule Powder Table Initial materials used in this study. Purity (%) 98.0 up 99.2 * 99.0 Note / Supplier Toho Titanium Co., Ltd. Toho Titanium Co., Ltd. *: Determined by X-ray fluorescence analysis (XRF). Kojyundo Chemical Laboratory Co., Ltd. 86

87 Table Experimental conditions of chlorine recovery Exp. # Mass of feed materials, w i / g Reaction Reaction Pressure Mass Ratio temp., time, Atmosphere p / atm w Ti / w FeCl2 Ti scraps FeCl 2 (Powder) NaOH T / K t / h Before Max. Note a CA a Ar CB a Ar CC a Ar CD b Ar CE b Ar a: Experiment date. b: Ti powder was used in this experiment. c: Ti granules was used in this experiment. d: Ti turing was used in this experiment. 87

88 Experiment apparatus (3) Glass wool (Deposits after the experiment) Quartz tube Graphite crucible Vacuum pump Ar gas Silicone rubber plug NaOH gas trap Heating element Sample mixture e.g., FeCl 2 + Ti powder Exp. # CE, CF 88

89 Observation Assembled quartz tube after experiment. 515 K 990 K 1100 K Raw materials: Ti powder FeCl 2 powder 25~28cm The form of the obtained residue and deposit after experiment. 5 cm Silicone plug Solid (White) Flake (Brown) Residue (Black) Raw materials: Ti granule FeCl 2 powder Raw materials: Ti turning FeCl 2 powder 5 cm Silicone plug after experiment 5 cm 2 cm 1 cm Deposit on the surface of the NaOH gas trap Deposit inside the quartz tube Residue in the graphite crucible 1 cm Melting point of FeCl 2 : atm Melting point of TiCl 4 : atm 89

90 Ti granule: XRD Intensity, I (a. u.) : α-ti (JCPDS # ) : α-fe (JCPDS # ) Angle, 2θ (deg.) Fe phase appeared in the obtained residue. Ti (s) + FeCl 2 (l. g) TiCl 4 (g) + Fe (s) 90

91 Ti turning: XRF Analytical results of the samples before and after heating, and the deposits obtained within the quartz tube and on the surface of the NaOH gas trap after heating. Exp. CF Ti Concentration of element i, C i (mass%) a Fe Cl Initial sample in the graphite crucible before heating 13.6 a 38.1 a 48.3 a Residue in the graphite crucible 29.1 b 64.7 b 6.18 b Deposit inside the quartz tube 0.06 b 50.3 b 49.6 b Deposit on the surface of the NaOH gas trap 0.04 b 1.29 b 98.7 b a: Calculated. b: Determined by X-ray fluorescence analysis (XRF). Black coat was obtained on the surface of the residue. The residue is of magnetic character. Fe element exists in the residue after heating. 91

92 4. Experimental results Discussion (3) Although Ti in Ti granule or turning was extracted by FeCl 2, the efficiency of the recovery of Ti scraps is low and the reaction speed in the system decreased when Ti granule or turning was used as the feed material. The reaction speed was affected by the morphology of the Ti scraps. 92

93 The purpose of this study Development of a new process for producing high-purity Ti powder High-productivity, low-cost process has to be developed for producing high-purity Ti. Preform reduction process (PRP) Feed preform (TiO 2 feed + flux) Reductant vapor Reductant (R = Ca, or Ca-X alloy) 93

94 Preform Reduction Process (PRP)? Feed preform (TiO 2 feed + flux) Reductant vapor Reductant (R = Ca, or Ca-X alloy) TiO 2 ore or UGI TiO 2 preform Ti powder Starting material Preform Powder TiO 2 (s, in feed preform) + Ca (g) Ti (s, powder) + CaO (s, flux) 94

95 Features of PRP Advantages of PRP: simple and low-cost process Suitable for uniform reduction Flexible scalability Possible to control the morphology of powder by varying the flux content in the preform Possible to prevent the contamination from reaction container Amount of waste solution is minimized Molten salt as a flux can be reduced compared to other direct reduction process Disadvantages of PRP: Leaching process is required Calcium production and handling of calcium vapor is difficult 95

96 PRP in this study Previous study: Artificial feed materials TiO 2 powder Upgraded ilmenite (India) Titanium Powder Or De-ionized ilmenite ore 99 % up matellic titanium powder was obtained by using titanium oxide (TiO 2 ) or upgraded ilmenite (UGI) as the staring materials. This study: Natural titanium ore (Rutile, South Africa) used as feed material Rutile ore (South Africa) XRF analysis (mass %) Ti Fe Si V Al So far, it was difficult to produce high-purity Ti directly from natural Ti ore! 96

97 Titanium ore used in this study. Sample Ti Si Nb Concentration of element i, C i (mass%) a Al Fe V Cr Mn Ca Mg Ni Note UGI b ND d ND d ND d ND d 0.07 ND d ND d ND d Rutile c ND d ND d ND d 0.03 ND d a : Determined by X-ray fluorescence analysis (XRF). b : Up-graded ilmenite produced in India by the Beacher process. c : Natural rutile ore produced in South Africa. d : Not detected. Below detection limit of XRF (<0.01%). 97

98 Table Starting materials used in this study. Materials Form Purity or conc. (%) Note / Supplier Rutile a Powder 93.1 c Produced in South Africa. CaCl 2 Powder 95.0 up Kanto Chemicals., Inc. Collodion b Aqueous 5.0 d Wako Pure Chemical Industries, Ltd. Ca Chip 98.0 up Mintech Japan K. K. Ti Sponge 98.0 up Toho Titanium Co., Ltd. CH 3 COOH Aqueous 99.7 up Kanto Chemicals., Inc. HCl Aqueous 35.0 d Kanto Chemicals., Inc. 2-Propanol Liquid 99.5 up Wako Pure Chemical Industries, Ltd. Acetone Liquid 99.0 up Wako Pure Chemical Industries, Ltd. a Natural rutile ore produced in South Africa. b 5 mass% nitro cellulose, mass% ethanol, mass% diethylether. c Purity of TiO 2 in the ore. d Concentration of the solution. Ito-san 98

99 Table Experimental condition of preform reduction process. Mass of sample, w i /g Cationic Calcination Mass of Reduction Exp. # PCD molar., Reductant, b Feed Flux C powder Binder R Cat. / Ti temp., time, w i /g temp., time, Ti ore a CaCl 2 Reductant? Collodion T cal. / K t' cal. / hr T red. / K t' red. / hr _ _ _ _ a : Natural rutile ore produced in South Africa after pulverization. b : Cationic molar ratio, R Cat. / Ti = N Cat. / N Ti, where N Cat. and N Ti are mole amount of cation in flux and that of titanium, respectively. 99

100 Table Experimental condition of preform reduction process. Mass of sample, w i /g Cationic molar., R Cat. / Ti c Calcination Mass of Reductant, w i /g Reduction Exp. # PE Feed Flux Binder temp., time, temp., Rutile a Ilmenite b CaCl 2 Collodion T cal. / K t' cal. / hr T red. / K _ _ _ _ _ _ a : Natural rutile ore produced in South Africa. b: Natural ilmenite ore produced in Vietnam after pulverization. c : Cationic molar ratio, R Cat. / Ti = N Cat. / (N Ca +N Ti ), where N Cat. and N Ti are mole amount of cation in flux and that time, t' red. / hr 100

101 PRP Ti ore Flux Binder Mixing Ti ore: Rutile Flux: CaCl 2 Binder: Collodion Rutile+CaCl 2 +Binder T: Room temp., t : 6 h e.g. 40mm 20mm 8mm Slurry Preform fabrication Feed preform Ti ore + flux T: 1273 K, t : 1 ~ 2 h T: 1273 K, t : 6 ~10 h Calcium vapor T: Room temp. 50% CH 3 COOH aq., t : 6 h 20% HCl aq., t : 1 h Calcination/iron removal Sintered feed preform Reduction Reduced preform Leaching Vacuum drying Powder FeCl x TiO 2 feed in flux Ti + CaO + Ca Waste solution Ti powder 101

102 Reduction experiment in PRP TIG welding Stainless steel reaction vessel Stainless steel cover Feed preform after Fe removal Stainless steel net Stainless steel holder Reductant (Ca granules) Ti sponge getter Schematic illustration of the experimental apparatus for the reduction experiment. TiO 2 (s, in feed preform) + Ca (g) Ti (s) + CaO (s, in flux) 102

103 Experimental apparatus Stainless steel reaction vessel Feed preform Ca shot Ti sponge Stainless steel holder for Ca Stainless steel net Stainless steel plate 1 cm Left: Arrangement of stainless steel net and holder tentatively installed in transparent container; Right: Apparatus parts and materials used in reduction experiment before assembly. 103

104 Experiment conditions Exp. a b A B C Feed Ti ore a Mass of sample, w i / g Flux CaCl Binder Collodion Additive Carbon powder Natural rutile ore produced in South Africa after pulverization. Cationic molar ratio, R Cat. / Ti = N Cat. / N Ti, where N Cat. and N Ti are mole amount of cation in flux and that of titanium, respectively. Cationic molar ratio., R Cat./Ti b Calcination Temp., Time, T cal. / K t' cal. / h Fe removal Reduction Temp., Time, T red. / K t' red. / h Ti powder production Fe removal: Reduction: FeO x (s, in Ti ore) + MCl y (g, l) FeCl 2 (g) + MO z TiO 2 (s) + Ca (g) Ti (s) + CaO (s) in flux 104

105 Results of PRP (1): Images Exp. A, Cationic molar ratio, R Cat./Ti = 0.2 (a) Fabricated feed preform (b) After calcination Ti ore + flux (c) After reduction TiO 2 feed in flux (d) After leaching Ti + CaO + Ca Ti powder A(C-2) 105

106 Results of PRP (2): XRD-1 Exp. A, Cationic molar ratio, R Cat./Ti = 0.2 : TiO 2 : CaCl 2 : CaCl 2 (H 2 O) 4 JCPDS # JCPDS # JCPDS # After fabrication: TiO 2 + CaCl 2 + CaCl 2 (H 2 O) 4 Ti ore + flux Intensity, I (a.u.) : TiO 2 : CaCl 2 : CaCl 2 (H 2 O) 4 : Ti : Ca : CaO JCPDS # JCPDS # JCPDS # JCPDS # JCPDS # JCPDS # After Calcination: TiO 2 + CaCl 2 + CaCl 2 (H 2 O) 4 After reduction: Ti + CaO + Ca TiO 2 feed in flux Ti + CaO + Ca : Ti JCPDS # After leaching: Ti Ti powder Angle, 2θ (deg.) Metallic Ti powder was successfully produced directly from natural titanium ore by the PRP. 106

107 (a) Results of PRP (3): XRD-2 : TiO 2 : JCPDS # Exp. B, Cationic molar ratio, R Cat./Ti = 0.2, Carbon powder: 0.2 g After fabrication: TiO 2 (+ CaCl 2 + CaCl 2 (H 2 O) 4 ) (b) : TiO 2 JCPDS # JCPDS # : CaTiO 3 After Calcination: Intensity, I (a.u.) (c) : Ti : Ca : CaO JCPDS # JCPDS # JCPDS # TiO 2 + CaTiO 3 (+ CaCl 2 + CaCl 2 (H 2 O) 4 ) After reduction: Ti + CaO + Ca (d) : Ti JCPDS # After leaching: Ti Angle, 2θ (deg.) The generation of CaTiO 3 was considered to be benificial for the iron removal. 107

108 Results of PRP (4): XRF Exp. B, Cationic molar ratio, R Cat./Ti = 0.2, Carbon powder: 0.2 g Sample Ti ore a Preform After calcination After reduction After leaching Analytical results of the samples in this study. Concentration of element i, C i (mass%) b Al Cl Ca Ti 1.01 < < Fe a : Natural rutile ore produced in South Africa after pulverization. b : Determined by X-ray fluorescence analysis, and the value excludes carbon and gaseous elements. Iron removal ratio: 90% The ratio is higher than the iron removal ratio of 56% in the previous experiment, in which carbon powder was not introduced in the preform. Purity of Ti powder: 98.23% Iron removal ratio: (C Fe, Bef. /C Ti, Bef. C Fe, Aft. /C Ti, Aft. ) / (C Fe, Bef. /C Ti,Bef. ) 108

109 Results of PRP (5): SEM images Exp. C, Cationic molar ratio, R Cat./Ti = 0.3 (a) Fabricated feed preform (b) After calcination (c) After reduction 100 μm (d) After leaching 100 μm 100 μm 5 μm 109

110 Results of PRP (6): Comparison of the SEM images Exp. B, Cationic molar ratio, R Cat./Ti = 0.2, Carbon powder: 0.2 g Exp. C, Cationic molar ratio, R Cat./Ti = 0.3, Carbon powder: 0.2 g 5μm 5 μm Cationic molar ratio, R Cat. / Ti = N Cat. / N Ti,where N Cat. and N Ti are mole amount of cation in flux and that of titanium, respectively. 110

111 Results of PRP (7): Composition and yields of the obtained Ti product Exp. Cationic molar ratio, R Cat/Ti. Ti Fe Exp. B and C Analytical results of the titanium samples obtained after leaching. Concentration of element i in obtained Ti powder, C i (mass %) Al Ca Cl Yield (%) B (0.00) C (0.00) a b Natural rutile ore produced in South Africa after pulverization. Cationic molar ratio, R Cat/Ti. = N Cat. / N Ti, where N Cat. and N Ti are mole amount of cation in flux and that of titanium, respectively. Still high for practical application, but will be improved. Loss occurred mainly at leaching process. 111

112 Results of PRP: SEM images Exp. C-7, Cationic molar ratio, R Cat./Ti = 0.3 (UGI) (a) Fabricated feed preform (b) After calcination 5μm 5μm (c) After reduction (d) After leaching 5μm 5μm 112

113 Chapter 5 After leaching XRD pattern Experimental results: XRD, SEM images, and XRF Exp. B, Cationic molar ratio, R Cat / Ti. = 0.2, Carbon powder: 0.2 g Step XRF analysis (mass %) Ti Fe Al Ca Cl : Ti JCPDS # SEM image (a) After fabrication (b) After calcination Iron removal ratio is 90 %. (c) After reduction (0.00) μm (d) After leaching (0.00) Iron removal ratio: (C Fe, Bef. /C Ti, Bef. C Fe, Aft. /C Ti, Aft. ) / (C Fe, Bef. /C Ti,Bef. ) C(PCD-2) 113

114 After leaching XRD pattern XRF analysis (mass %) Ti Fe Al Ca Experimental results: XRD, SEM images, and XRF Exp. D, Cationic molar ratio, R Cat. = 0.3, Carbon powder: 0.2 g Step Cl : Ti JCPDS # SEM image (a) After fabrication (b) After calcination Iron removal ratio is 65 %. (c) After reduction (0.00) μm (d) After leaching (0.00) Iron removal ratio: (C Fe, Bef. /C Ti, Bef. C Fe, Aft. /C Ti, Aft. ) / (C Fe, Bef. /C Ti,Bef. ) D(PCD-4) 114