TREATMENT OF TITANIUM-CONTAINING MATERIAL
The present invention relates to the removal of metallic iron from a titanium-containing material. In particular the invention concerns the oxidation and removal of metallic iron from a titanium-containing material.
Titanium metal and certain titanium compounds possess properties which make them useful in certain applications. For example, titanium dioxide, Ti02, which accounts for over 95% of world-wide use of titanium [1], is a compound which is non-toxic, has a high refractive index, and is white when pure, which makes it an ideal filler and pigment for many paint, paper, plastic, and rubber products. Also, titanium metal and alloys of titanium are commonly used in the aerospace and related industries because they are light, strong and very corrosion resistant. Other titanium compounds are used as fluxes and in the manufacture of ceramics. A significant quantity of the world's titanium products is obtained from synthetic rutile (TiO2), a material which is manufactured from the mineral ilmenite (FeTiO3) via the Becher process. This process is based on a two-step procedure developed by the Western Australian Government Chemical Laboratories in 1961 [2,3], see Fig. 1 , and is currently used in commercial operations in Western Australia.
In the first step ilmenite in granular form and bituminous coal and/or coal char are fed into a kiln which is heated to about 1100°C. Many complex reactions occur during this stage [4,5,6], but, ultimately, a composite material called reduced ilmenite (Fe,Ti02) is formed which mainly comprises fine-grained metallic iron dispersed within fine particles of titanium dioxide. Formally, the reaction can be represented by the equation
FeTiO3 + C → Fe,TiO2 + CO (1 )
In the second step, the aqueous oxidation step, reduced ilmenite particles are suspended in a dilute solution of ammonium chloride which is then stirred and aerated for circa 14 to 16 hours. Under these conditions, the metallic iron is oxidised to soluble Fe2+ which diffuses out of the reduced ilmenite particles,
eventually precipitating as various iron oxides. Thus, at the completion of this step, the ammonium chloride solution contains a mixture of synthetic rutile particles and even finer particles of iron oxide. Because the particles of iron oxide are generally between 1 μm and 10 μm in diameter they are easily separated by washing from the larger particles of synthetic rutile, which are generally between 75 μm and 850 μm in diameter. Ultimately, the electrochemical half-reaction driving the oxidation of the metallic iron is the reduction of oxygen which is supplied by the air bubbling through the solution.
1/2O2 + 2H+ + 2e- → H2O (2)
Fe → Fe2+ + 2e~ → Iron Oxides ^
Reactions (2) and (3) present a greatly simplified description of the oxidation of metallic iron. In fact, the process is an extremely complex one and a number of variables affect the rate at which metallic iron is removed and the identity of the iron oxides which are formed. The iron oxide which plant operators prefer is magnetite, Fe3O4, a black compound which is easily separated from the synthetic rutile. The principal product, synthetic rutile, can then be further purified and used as a feedstock for the titanium industry.
Accordingly, it is an object of the present invention to provide a method of oxidising metallic iron in a titanium-containing material in which the rate of oxidation of the metallic iron is accelerated.
Accordingly, the present invention provides a method of oxidising and removing metallic iron from a titanium-containing material which method includes oxidising the metallic iron in the presence of an additive that accelerates the rate of oxidation of the metallic iron.
The titanium-containing material may be selected from ilmenite, reduced ilmenite, rutile, pseudorutile, leucoxene, pseudobrookite and other iron titanate phases or combinations thereof. In a preferred embodiment the titanium- containing material is reduced ilmenite. The reduced ilmenite may be for example, that produced in the first step of the Becher process but is not limited to this source of reduced ilmenite.
The metallic iron may be oxidised to soluble Fe + which may diffuse out of the reduced titanium-containing material and precipitate as iron oxide leaving behind a titanium product with diminished metallic iron content. In a preferred embodiment the precipitated iron oxide is magnetite. Preferably, the method includes a further step of separating the iron oxide from the titanium product. Optionally the titanium product is subject to acid leaching.
Preferably the titanium product is synthetic rutile. Preferably the synthetic rutile is produced by the Becher process. The additive may consist of, or include, any redox couple, or mixture of redox couples. Both the reduced and oxidised forms may be soluble in the oxidation solution. The reduced form or forms may be capable of being oxidised by oxygen, and the oxidised form or forms may be capable of being reduced by metallic iron. The redox couple may be selected from chemical compounds of the quinone family, quinonimines, diimines, indigo derivatives, diketones, and thiazines, or Diels-Alder adducts of any of the foregoing.
The additive may consist of, or include, a substituted or unsubstituted anthraquinone sulfonic acid or salt thereof. The anthraquinone sulfonic acid may be a substituted or unsubstituted anthraquinone disulfonic or monosulfonic acid or a salt thereof. The disodium salt of 9,10-anthraquinone-2,6-disulfonic acid (C14H6O8S2Na2) or derivatives thereof is particularly preferred, although other anthraquinone sulfonic acids or salts such as those substituted at the (1), (2), and (2,7) positions, or their corresponding sulfonamides, may be used.
The additive may be added such that the concentration of the redox couple is in the range of about 0.001% to 10% weight-for-volume of the oxidation solution, and preferably in the range of about 0.01% to 1.0%. An additive concentration of about 0.2% is particularly preferred.
Preferably, the oxidation is carried out in the presence of a halide salt. Preferably the halide salt is a chloride. More preferably the halide salt is ammonium chloride. Preferably the halide salt is present in an amount in the range of about .01 to 10%.
Preferably the aqueous oxidation of metallic iron is carried out in the presence of a source of oxygen. Preferably the source of oxygen is air.
Preferably the oxidation of metallic iron is carried out at a temperature in the range of about 20 to 95°C. It will be understood that the present invention provides a method whereby the rate of removal of metallic iron from titanium-containing material is accelerated. The rate of removal of metallic iron from titanium-containing material may be accelerated by a factor of approximately three.
Brief Description of the Figures The following notation is used:
AQ-1 , Anthraquinone 1 -sulfonic acid*
AQ-2, Anthraquinone 2-sulfonic acid* AQ-2,6 Anthraquinone 2,6 disulfonic acid*
AQ-2,7, Anthraquinone 2,7 disulfonic acid* *added as the sodium salts.
Fig. 1. Schematic representation of the Becher process. REDUCTION STEP: Ilmenite is mixed with coal and/or coal char and is reduced in a kiln to form Reduced ilmenite. OXIDATION STEP: Reduced ilmenite is added to a solution of ammonium chloride and the metallic iron is dissolved away by an accelerated corrosion process. The product is Synthetic Rutile.
Fig. 2. Metallic iron content of samples of standard reduced ilmenite withdrawn at intervals from the aeration reactor. Top curve: 250 grams reduced ilmenite in 500 mis 2% (w/v) NH4CI. Lower curves: various concentrations of AQ-2,6.
Fig. 3. Metallic iron content of samples of standard reduced ilmenite withdrawn at intervals from the aeration reactor. Top curve: 250 grams reduced
ilmenite in 500 mis 2% (w/v) NH4CI. Lower curves: 0.2% AQ-2,6 and 0.2% AQ- 2,7.
Fig. 4. Metallic iron content of samples of standard reduced ilmenite withdrawn at intervals from the aeration reactor. Top curve: 250 grams reduced ilmenite in 500 mis 2% (w/v) NH4CI. Lower curves: 0.2% AQ-1 and AQ-2.
The invention will now be described with reference to the following examples. These examples are useful for illustrating the scope of the invention and the advantages thereof but should not be seen as limiting in any way.
Example
A stockpile of approximately 12 kg of a standard reduced ilmenite formed the basis of our aeration tests. CHARACTERISATION
The standard reduced ilmenite was characterised by the following measures.
(1) Particle size distribution by dry screening.
(2) Specific surface area versus particle size by the BET method. (3) Pore volume distribution versus particle size by mercury porosimetry.
(4) Metallic iron content versus particle size using a Saturation Magnetisation Analyser.
(5) Elemental analysis by X-ray fluorescence.
(6) Crystal phase analysis by X-ray diffraction.
Particle Size Distribution
The particle size distribution of the standard reduced ilmenite was determined by dry screening. The results are given in Table 1.
Particle Size Fraction Recovered Fraction/μm of Total/%
-75 +53 0.03
+75 -106 1.21
+106 -150 21.23
+150 -212 37.60
+212 -250 19.15
+250 -300 12.17
+300 -425 7.47
+425 -600 HO
+600 -850 0.04
Table 1. Particle size distribution of the standard reduced ilmenite determined by dry screening.
These results show the particle size distribution of the standard reduced ilmenite had a maximum in the (-212 μm +150 μm) fraction, and a "tail" towards larger particles. The upper and lower limits of the distribution occur at 850 μm and 75 μm.
On the basis of these results the standard reduced ilmenite material was dry screened and the -425 μm +106 μm size fraction was used for all further work.
Specific Surface Areas
The specific surface areas of each of the fractions of the standard reduced ilmenite recovered by dry screening were determined using the Brunauer- Emmett-Teller (B.E.T.) method for analysis. The results are presented in Table 2.
Pore Volume Distribution
The pore volumes (internal volumes) of the fractions of the standard reduced ilmenite recovered by dry screening were measured using mercury porosimetry. The results are presented in Table 2.
Metallic Iron Content
The metallic iron content of the fractions of the standard reduced ilmenite recovered by dry screening were measured using a Saturation Magnetisation Analyser, and the results are given in Table 2. The metallic iron content is not uniform with average particle size but increases from 26% to just over 30% as the particle size increases, which is believed to reflect a variation in the composition of the native ore, not variations introduced by processing.
Particle Size Specific Pore Metallic Iron Fraction Surface Volume/cc g~ Content/%
Area/m 2 g- 1 (w/w)
+75 -106 not determined 0.24 not determined
+106 -150 1.4 0.32 26.2
+150 -212 1.3 0.25 26.0
+212 -250 2.5 0.21 27.4
+250 -300 3.4 0.16 27.8
+300 -425 4.8 0.13 30.0
+425 -600 5.0 0.12 30.4
Table 2. Specific surface areas, pore volumes, and metallic iron content for the particle size fractions obtained from the standard reduced ilmenite by dry screening.
X-Rav Fluorescence The results for elemental analysis by X-ray fluorescence are presented in
Table 3. The selected elements are all reported as oxides because the samples are oxidized at high temperature prior to analysis.
Element (as Concentration/% Element (as Concentration/% oxide) w/w oxide) w/w
TiO2 66.03 Fe2O3 41.02
Mn304 1.49 Si02 1.02
Al203 0.97 MgO 0.48
Cr2O3 0.11 ZrO2 0.075
Table 3. Results for the elemental analysis of the standard reduced ilmenite by X-ray fluorescence.
X-Rav Diffraction
The crystal phase analysis revealed that the standard reduced ilmenite contained three types of crystal phases: (i) a solid solution of the general formula M3O5 (for example, FeTi2O5, Ti3O5, (Mn,Mg)Ti2O5), (ii) reduced rutile phases of the general formula Ti02-χ, and (iii) metallic iron. This mixture is typical of reduced ilmenite, and no unexpected crystal phases were observed.
METHOD
A two-litre round bottom flange flask with detachable five-port top formed the aeration reactor, into which were introduced two air bubblers and an open- ended glass sheath for a temperature sensor. The two remaining ports were used to insert a stirrer blade and a high-efficiency condenser which trapped any vapours produced during the reaction. The reactor was thermostatted at 70°C using a heating mantle which was controlled using a temperature setpoint controller. Feedback for each controller was provided by a teflon-coated 100 Ω Resistance Temperature Detector. A titanium stirrer blade immersed as low as possible in the reactor was rotated using a stirrer motor such that the mixture was forced to flow towards the bottom of the reactor during operation. The stirring rate was 750-770 rpm. Air flow into the reactor was measured by a precision bore flow meter and the flow rate was accurately controlled using a fine- adjustment teflon valve. The air flow rate was 2.3-2.6 litres per minute.
Before commencing aeration tests 250 g samples of standard reduced ilmenite were rigorously split from the fully-characterised stockpile using a commercial sample divider, and the metallic iron content of the samples was measured. A 500 ml solution of 2% (w/v) AR grade ammonium chloride was made up using deionised water. At the beginning of each run the solution, the standard reduced ilmenite, and the additive were added to the reactor, the air flow was started, and the heating mantle was switched on. The reactor reached temperature within 15 min of switching on. Samples of standard reduced ilmenite were recovered every hour by stopping the stirrer and drawing 5 millilitres of the settled contents of the reactor into a glass tube, transferring them to a beaker and washing away any iron oxides present. The remaining solids (typically 5 grams) were dried in an evacuable oven at 50°C and the metallic iron content was measured using a Saturation Magnetisation Analyser. The solids were then returned to the reactor. At the end of the experiment the synthetic rutile product was washed free of iron oxides and dried at 50°C under vacuum. Samples of the iron oxides were collected, and the crystal phases present were identified by X- ray diffraction. The elemental composition of the synthetic rutile was also confirmed by X-ray fluorescence.
RESULTS
The metallic iron contents of samples of standard reduced ilmenite withdrawn at hourly intervals from the reactor are plotted in Figs. 2, 3 and 4. In Fig. 2, plots shown are for a standard run (top curve) and for runs in which 0.2%, 0.4% and 0.8% (w/v of the aeration solution) of the disodium salt of 9,10- anthraquinone-2,6-disulfonic acid was added (lower curves). In the standard run the metallic iron content of the standard reduced ilmenite decreased to approximately 2% after 12 hours of reaction, whereas with 0.2% of the additive the same level was reached after 5 hours and with 0.4% and 0.6% of the additive the same level was reached in 3 hours.
Element TiO2 Fe2O3 Mn3O4 SiO2 AI2O3 MgO Cr2O3
Standard 90.06 5.83 1.62 1.38 1.27 0.68 0.11 run
With 0.2% 90.04 5.23 1.53 1.45 1.27 0.60 0J 3 additive
Table 4. Compositions of the synthetic rutile products determined by X-ray fluorescence. The figures are weight percentages, and the elements are reported as oxides because the method used oxidises the samples before analysis.
The compositions of the synthetic rutile products were determined by X-ray fluorescence. As is evident from Table 4, no significant differences in composition were observed in the runs with and without the additive.
X-ray diffraction of the iron oxides produced during both runs was also carried out. The results are summarised in Table 5, where it can be seen that only magnetite was detected, the oxide most desired in industrial practice.
D-Spacing /A (relative intensity / %)
Standard run 4.836 2.963 2.528 2.419 2.096 1.712 1.613 1.483 (10) (29) (100) (8) (21) (8) (24) (35)
With 0.2% 4.837 2.963 2.528 2.422 2.098 1.713 1.615 1.484 additive (10) (30) (100) (8) (20) (8) (23) (32)
Pure magnetite 4.852 2.967 2.532 2.424 2.099 1.715 1.616 1.485 (ICDD data) (8) (30) (100) (8) (20) (10) (30) (40)
Table 5. Data from X-ray diffractograms of the iron oxides removed during the aeration process compared with pure magnetite data from the International Centre for Diffraction Data (Swarthmore, USA)
In Fig. 3 the standard run is shown in the upper curve with the plots for runs in which 0.2% AQ-2,6 and AQ-2,7 was added (lower curves).
In Fig. 4 the standard run containing no additive is shown in the upper curve with runs in which 0.2% of AQ-1 and AQ-2 are used (lower curves).
It is clear that the additives are behaving as catalysts for iron oxidation and removal. References
[1] Towner, R. R., Gray, J.M., and Porter, L.M., International Strategic Minerals Inventory Summary Report - Titanium. U.S. Geological Survey Circular 930-G (1988). [2] Becher, R.G., Australian Patent 24710 (1963). [3] Becher, R.G., Canning, R.G., Goodheart, B.A., and Uusna, S., Proceedings Aus. Inst. Min. Metall. 214, 21 (1965).
[4] Ostberg, G., Jemkont. Annin. 144, 46 (1960). [5] Jones, D.G., Trans. Inst. Min. Metall. Sect. C. 82, C186 (1973). [6] Grey, I.E., and Reid, A.F., Trans. Inst. Min. Metall. Sect. C. 83, C39 (1974).