ELECTROCHEMICAL METHOD AND APPARATUS
Field of Invention
The invention relates to an electrochemical method and apparatus, specifically for the reduction of electronic conductivity of molten salts.
Background to Invention
At the present time there are two ways of electrochemically extracting metals from molten salts. In the first, a halide salt, frequently a chloride, is electrolysed to produce the metal and chlorine. This applies to the production of such metals as magnesium and sodium. The second method, which is practised on an immense scale in the aluminium industry, involves the dissolution of an oxide into a molten fluoride and the electrolysis of the oxide. There are many other metals that can be deposited from molten salts but unless the metal is deposited in the molten state, the deposit is very dendritic and it is difficult to prevent oxidation, especially of the reactive metals. As a result, such metals are prepared by metallothermic reduction, using elements such calcium, aluminium and magnesium as reductants .
Another problem is that with titanium and other reactive metals, there is considerable solubility of oxygen in the metal and this can adversely affect the performance of the product. The present extraction route for titanium and zirconium excludes oxygen at an early stage in the process by carbochlorinating the oxide to the tetrachloride which is then purified and subsequently reduced by magnesium (Kroll process) or sodium (Hunter process) . This takes place in a relatively small reactor (5-10 tonnes) and
takes 2-3 weeks. The reduced titanium is separated from the excess reductant and the by-product chloride by leaching, or in the Kroll process, often, by vacuum distillation. Thereafter, the titanium "sponge" product is crushed to about 30 mm and smaller and is inspected visually for the presence of pieces displaying oxide and/or nitride contamination. Such pieces are removed manually.
Many of the reactive metals have very high melting points and this can cause problems with oxidation and the addition of alloying elements. There are many interesting alloying systems which are precluded from use by difficulties in alloy preparation on a large scale.
In 1961, Ward and Hoar described a process whereby molten copper containing oxygen, sulphur, selenium and tellurium was made the cathode in a bath of molten BaCl2 at temperatures ranging from 1353 K to 1413 K. It was found that the metalloids were preferentially ionised in comparison with the deposition of barium metal.
More recently, Okabe and co-workers have investigated the removal of oxygen dissolved in metals, such as titanium and yttrium, by making the metal the cathode in a bath of molten CaCl2 at 1223 K. The parts per million of oxygen in the samples were reduced from several thousands down to about fifty. The reaction was assumed to be the deposition of calcium that reacted with the oxygen dissolved in the cathode metal, to give calcium oxide that dissolved in the calcium chloride. This can be summarised as follows:
Ca2+ + 2e" = Ca (1)
O-Ti + Ca = Ti 4- CaO (2)
Fray, Farthing and Chen found that by making an oxide the cathode in a molten bath of calcium chloride the favoured cathodic reaction was the ionisation of oxygen rather than the deposition of calcium.
Ti02 + 4e~ = Ti + 202_ (3)
It was found that the current efficiency of this reaction is very high but that the current efficiency tends to fall over a period of hours, and especially in the later stages of the electro-deoxidation process.
Summary of Invention
The invention provides a method and an apparatus as defined in the appended independent claims, to which reference should now be made. Preferred or advantageous features of the invention are set out in dependent subclaims .
The invention is thus based on the realisation by the inventors that during the electro-deoxidation process of Fray, Farthing and Chen, or during other electrolytic processes, the observed decrease in current efficiency over time may be due, at least in part, to a gradual build-up of dissolved metals in the electrolyte. Thus, in the electro-deoxidation of titanium dioxide in calcium chloride, due to the solubility of calcium in calcium chloride there may be a gradual build-up of calcium over a period of hours, in the calcium chloride melt. The presence of calcium may cause the electrolyte to become partially electronically conducting and this greatly reduces the current efficiency of the electro-deoxidation process. The same phenomenon may similarly affect other electrolytic processes.
The invention may advantageously solve this problem by adding or introducing to the melt, or electrolyte, a reactant which reacts with the dissolved metal, which.may be present in the form of a dissolved metallic species such as metal ions, and removes it from solution. Thus, in the example described above, the reactant may decrease the electronic conductivity of the melt by reacting with the dissolved calcium or Ca+ and removing it from the solution.
In a preferred embodiment, the reactant may correspond to an anion already present in the melt, so that the reaction generates the molten salt as its product. For example, chlorine may be added to a calcium chloride melt containing dissolved calcium, to generate calcium chloride as the reaction product.
In an alternative embodiment, the reactant may react with the dissolved metal to form a product which is insoluble in the melt and precipitates out of solution.
In a further alternative, a dissolved metal may be removed by the addition of a suitable salt. For example dissolved calcium may be removed from a calcium chloride melt by the addition of a salt of a less stable chloride, such as SnCl2, where the reaction would be:
Ca + SnCl2 = CaCl2 + Sn.
This reaction, like the other reactions within the scope of the invention, may optionally be carried out in an external reactor coupled to an electrolysis cell containing the melt.
In a further alternative, a metal could be added to the melt which reacts with the dissolved metal. For example,
to remove dissolved calcium from a calcium chloride melt, aluminium may be added:
2A1 + Ca = Al2Ca.
The calcium could then be removed from the aluminium by either oxidation or chlorination.
Description of Specific Embodiments and Best Mode of Invention
Specific embodiments of the invention will now be described by way of example, with reference to the accompanying drawings, in which:
Figure 1 is a diagram of an electrolytic cell according to a first embodiment;
Figure 2 is a diagram of an electrolytic cell according to a second embodiment; and
Figure 3 is a diagram of an electrolytic cell according to a third embodiment.
With reference to the electro-deoxidation process described in the introduction above, it might be expected that if calcium chloride is subjected to a potential below the decomposition potential of calcium chloride, no reaction would take place. However, there is some solubility of calcium in calcium chloride so that even at potentials below the deposition potential for pure calcium, calcium can be formed and dissolve in the molten chloride up to an activity related to the potential at the cathode. It can be appreciated that the rate of calcium liberation at these potentials, which are less negative than the potential for the deposition of calcium, is very slow so that it may take several hours for an equilibrium
mixture of calcium to form in the melt. The amount of calcium and, therefore, the value of the electronic conductivity are related to the potential applied at the cathode .
Calcium is a very reactive metal so that there are many reagents that can remove the calcium from solution. Oxygen can readily react with calcium to form calcium oxide which can dissolve in the calcium chloride. If chlorine is used the chlorine reacts with the calcium to form more calcium chloride.
Example 1
For an electrode area of 1 m2 and 0.1 m3 of salt (calcium chloride) , the application of about 3 V between the anode and cathode during electro-deoxidation of solid TiOs at the cathode has been found to result in a build-up of 13.6 g of calcium in the calcium chloride melt over a 6 h period. This allows about 1000 A to be passed through the electrolyte without decomposition of the salt. This conductivity is entirely due to electrons passing through the electrolyte. To remove this calcium a bleed of chlorine 30 cm3 per second at NTP (normal temperature and pressure) would be sufficient. The chlorine can either be passed into the melt in the electrochemical reactor or the salt circulated to an external reactor where the chlorine is added. This rate of addition of chlorine will ensure that the electronic conductivity of the melt remains negligible allowing the electro-deoxidation to occur in an energy-efficient manner. After treatment the current has fallen to around 10 A.
Figures 1 and 2 illustrate suitable electrolytic cells for introducing chlorine into the electrolysis reactor (Figure 1) or into an external reactor.
Figure 1 shows a cell 2 containing the calcium chloride melt 4. An anode 6, optionally of carbon, and a cathode 8 comprising the titanium dioxide for electrodecomposition are immersed in the melt. A vent or pipe 10 for releasing chlorine into the melt is preferably positioned near the anode. The gas may thus remove calcium from the melt near the anode, creating a depletion zone and reducing the melt conductivity.
Figure 2 shows a cell 2 as in Figure 1, but coupled to an external reactor 12 into which gas is vented to remove dissolved calcium. The molten salt is recirculated 14 between the cell 2 and the reactor 12.
Similar apparatus to that of Figure 1 or Figure 2 may be used to bleed, for example, air, oxygen or steam, or other suitable gas, into the melt.
Example 2
For an electrode area of 1 m2 and 0.1 m3 of salt, as in Example 1, the application of about 3 V between the anode and cathode results in a build-up of 13.6 g of calcium in the molten calcium chloride melt over a 6 h period. This allows about 1000 A to be passed through the electrolyte without decomposition of the salt. This conductivity is entirely due to electrons passing through the electrolyte. To remove this calcium a bleed of oxygen of 15 cm3 per second at NTP or air at 75 cm3 per second would be sufficient. The oxygen or air can either be passed into the melt in the electrochemical reactor or the salt circulated to an external reactor where the oxygen or air is added. This rate of addition of oxygen or air will ensure that the electronic conductivity of the melt remains negligible, allowing the electro-deoxidation to occur in an energy efficient manner. After treatment the current has fallen to around 10 A.
It is clear to anyone of normal skill in the art, after study of the above examples, that removal of calcium build-up in the salt may be effected by use of any of a large number of reagents. The fact that these are not mentioned explicitly is not intended as an intent to restrict the calcium removal method to the use of oxygen, air or chlorine only.
As the skilled person would appreciate, various reagents may be efficacious in any particular case of calcium chloride containing dissolved calcium, the reagent might be or comprise chlorine, oxygen, steam, hydrogen chloride or bromine. Alternatively if the electrolyte were sodium hydroxide containing dissolved sodium, the reagent might be oxygen or steam. The skilled person would also be able to select other reagents appropriate in each case for putting the invention into practice.
If a solid reagent such as Al or SnCl2 is used as described above, an apparatus as illustrated in Figure 3 may be used. This apparatus is the same as in Figure 2, and the same reference numbers are used, but no pipe or vent is required. The solid reagent can be added directly 16 into the external reactor 12.