WO2006027612A2

WO2006027612A2 - Improved electro-deoxidation method, apparatus and product

Info

Publication number: WO2006027612A2
Application number: PCT/GB2005/003497
Authority: WO
Inventors: Derek John Fray; Carsten Schwandt; Greg Doughty
Original assignee: Cambridge Enterprise Limited; British Titanium Plc
Priority date: 2004-09-09
Filing date: 2005-09-09
Publication date: 2006-03-16
Also published as: US20140209477A1; US20080190777A1; WO2006027612A3

Abstract

In an electro-decomposition process a cathode (8) comprising a metal compound is contacted with a fused salt electrolyte (6) in an electrochemical cell. The metal compound is a compound between a metal and another substance, and a voltage is applied between the cathode and an anode such that the substance is removed from the metal compound. In the improved method, the applied voltage increases with time, either continuously or stepwise, up to a predetermined maximum voltage. In addition, or in the alternative, the fused salt composition is selected so as to maintain a predetermined concentration of the substance in the fused salt during electro-decomposition.

Description

Improved Electrolytic Method, Apparatus and Product

The invention relates to a method and an apparatus for removing a substance from a solid compound by an electrolytic process, and in particular to improvements to the FFC, or electro-decomposition, process.

Background of the Invention

The FFC process, also termed electro-decomposition or electro-reduction, is described in International Patent Application No. WO/99/01781 and in subsequent publications by the inventors named in the International patent application (D. Fray, G. Chen and T. Farthing) , such as "Nature", vol. 407, 361-364 (21 September 2000) . These documents are incorporated herein by reference in their entirety.

As is known from the prior art, the FFC process is able to remove substances from solid compounds between those substances and metals or semi-metals. For example, as is known from the published literature, it is possible to apply electro-decomposition to remove oxygen from a metal compound such as titanium dioxide or from a semi-metal compound such as silica. For the sake of clarity, the term "metal" will be used throughout this document to encompass both metals and semi-metals, and should be construed as such.

Electro-decomposition is an electrolytic process carried out in a fused-salt electrolyte. In order to remove a substance from a solid compound comprising the substance and a metal, a cathode comprising the solid compound is contacted with the fused salt. An anode is also contacted with the fused-salt and a voltage applied between the cathode and the anode such that the substance is transferred into, or dissolves in, the electrolyte. Electro-decomposition may thus advantageously enable extraction of the metal from the metal compound. Electro- decomposition of a mixture of solid compounds of two or more metals, or of a mixture of one or more metals and one or more metal compounds, may advantageously enable fabrication of an alloy comprising the metals.

The present inventors have refined this process and the present invention relates to the resulting improvements to the process.

Summary of the Invention

In its various aspects the invention provides methods, apparatus, metals and alloys as defined in the appended independent claims. Preferred or advantageous features of the invention are defined in dependent subclaims.

In a first aspect, the invention provides a method for removing a substance from a solid compound comprising the substance and a metal, in which a fused-salt electrolyte is contacted with a cathode comprising the solid compound and with an anode, and a voltage is applied between the cathode and the anode that increases with time, or is ramped. The voltage may increase substantially continuously (either linearly or non-linearly) or stepwise or in any combination of these.

In an alternative embodiment, a voltage is applied between the cathode and the anode such that the potential at the cathode increases with time, or is ramped. (A cathodic potential is a negative quantity, and so reference to the potential at the cathode increasing means that the numerical value of the potential increases, the cathodic potential becoming more negative) . The cathode potential may increase substantially continuously (either linearly or non-linearly) or stepwise or in any combination of these.

If it is desired that the increasing potential should start from an initial level at which no electro- decomposition occurs, then the increasing potential may start at 0 V or may start from some other potential. For example, the thermodynamics of a reaction involved in the removal of a substance from a metal compound may determine a minimum cathode potential below which an electro- decomposition reaction cannot proceed. The increasing potential may therefore, in a preferred embodiment, start from a cathode potential less than or equal to this minimum electro-decomposition potential. An increasing cell voltage applied between the cathode and the anode may advantageously rise from a corresponding voltage level.

It is also preferable that the increasing potential should not exceed a maximum level. In an electro-decomposition process, it is desirable that cations from the fused-salt electrolyte should not be continuously discharged or deposited at the cathode, which may eventually consume or disadvantageously change the composition of the electrolyte and risk contaminating the desired metal product at the cathode. Consequently, it is preferable that the increasing potential does not exceed a potential for continuous discharge or deposition of a cation from the electrolyte, namely the cation deposition potential. An increasing applied voltage should therefore preferably not exceed a level corresponding to this potential at the cathode. In order to achieve the desired aim of not consuming or changing the composition of the electrolyte unnecessarily or excessively, in a further embodiment it is preferable that the increasing potential does not cause the cathode potential to rise to a level which is sufficient to cause continuous decomposition of the electrolyte. An increasing voltage applied to the cell should therefore preferably not exceed a corresponding level.

In a preferred embodiment, when the increasing potential or voltage has risen to its respective maximum level, the potential or voltage may be maintained at that level for a further period of time in order to allow the electro- decomposition to proceed.

The potential at the cathode may be controlled, monitored or measured by, for example, the use of a reference electrode which contacts the fused salt. In a preferred embodiment, the reference electrode may be implemented as a so-called pseudo-reference electrode in which an electrode of a suitable material is contacted with the electrolyte and the potential between the cathode and the electrode monitored. In this document, the term "reference electrode" has been used to encompass both reference and pseudo-reference electrodes, and should be construed accordingly. Suitable materials for a pseudo- reference electrode in a fused-salt electrolyte may include carbon, refractory metals or tin oxide.

Advantageously, little or no current may flow between the cathode and the reference electrode, so that little or no polarisation occurs at the reference electrode.

There may be an advantage in using the same material for the anode and for the reference electrode in a cell. In a preferred embodiment, measurement of the voltage between the cathode and the reference electrode can then be subtracted from the voltage between the cathode and the anode in order to give an indication of polarisation at the anode.

Although, in the embodiments described above, the increasing potential or voltage may be followed by the application of a potential or voltage at the maximum level of the increasing potential or voltage, the increasing potential or voltage may in an alternative embodiment be followed by the application of a different potential or voltage.

In a further alternative embodiment, the potential or voltage applied after the increasing potential or voltage may be variable, rather than constant. However, as described below there may in general be a preferred range for the potential to achieve a maximum electro- decomposition effect and so in a preferred embodiment a substantially constant cathode potential within this range, or a cell voltage for generating a cathode potential in this range, should be applied.

During the application of the increasing potential or voltage, the following advantages may preferably be achieved.

First, if the fused-salt electrolyte comprises a mixture of salts, such as a second salt in solution in a first salt, the salts containing different anion species, the application of an increasing potential or voltage may enable control of the anion deposited or evolved at the anode. For example, if the first salt is a chloride and the second salt is an oxide, for environmental or other reasons it may be preferable for the reaction at the anode to involve oxygen discharge rather than chlorine discharge. Chlorine discharge generally requires a higher potential at the anode and so oxygen discharge would normally be favoured, as long as the activity or concentration of oxygen in the region of the anode is sufficient to avoid excessive polarisation of the oxygen discharge reaction. Applying an increasing potential or voltage at the initial stage of an electro-decomposition reaction may advantageously prevent excessive reduction in the oxygen concentration or activity in the region of the anode by controlling the initial rate of oxygen discharge, and thereby encourage oxygen discharge rather than chlorine discharge during electro-decomposition. Similar considerations may apply to other mixtures of salts, as the skilled person would readily appreciate.

Consequently, in a preferred embodiment the increasing potential or voltage increases at a rate sufficiently slow to favour deposition or discharge of a predetermined or desired species of anion at the anode.

Second, electro-decomposition of a solid metal compound involves the transfer of the substance removed from the compound into the fused-salt electrolyte. If the substance is removed from the solid compound at a rate faster than the substance can dissolve or disperse in the electrolyte, then the presence of the substance, or of a material comprising the substance, in the electrolyte in the region of the cathode may reduce the rate of the electro-decomposition reaction, or even prevent further reaction. For example, if the local concentration of the substance or a material comprising the substance at or in the region of the cathode reaches saturation, it may precipitate or solidify in the region of the cathode, and prevent or restrict further access of the fused-salt electrolyte to the solid metal compound and so prevent or restrict further electro-decomposition. Even if the local concentration does not reach saturation, an increase in concentration of the substance in the region of the cathode may still slow electro-decomposition by restricting access of the electrolyte to the solid compound. For example, the solid compound is commonly in the form of a porous pellet or other precursor and so the rate of transport or diffusion of the electrolyte into the pores and the substance out of the pores may be limited. The concentration of the substance in the metal compound is generally highest at the initial stages of electro- decomposition and therefore the rate of its removal from the metal compound is likely to be highest in the initial stages. Consequently, the application of an increasing potential or voltage starting from a low level may advantageously control or limit the rate of electro- decomposition in the early stages of the reaction and reduce any tendency for the concentration of the removed substance or for undissolved material to build up at the cathode.

Consequently, in a preferred embodiment the increasing potential or voltage increases at a rate sufficiently slow to control or limit the local concentration of the substance or of a material comprising the substance, or to suppress the precipitation of the substance or of the material containing the substance, in the region of the cathode.

Third, it has been found that in many cases, application of the electro-decomposition process to remove a substance from a solid compound to produce metal does not proceed at the same rate at all parts of the solid compound at the cathode. Commonly, for example, such uneven reaction may cause the formation of metallic layer encasing a partially-reduced core of the solid compound at the cathode, which may disadvantageously reduce the rate of completion of electro-decomposition of the core. In such cases, it may be advantageous to control or limit the rate of electro-decomposition in order to allow a more even reduction of the solid metal compound throughout the cathode structure.

Consequently, in a preferred embodiment the increasing potential or voltage increases at a rate sufficiently slow to favour an even progression of the reduction throughout the solid metal compound at the cathode.

Fourth, it has been found that in many cases electro- decomposition involves the formation of one or more intermediate compounds comprising the substance and the metal. In such cases, it may be advantageous to control the progress of the electro-decomposition reaction through the formation of any intermediate compounds, for example to avoid the formation of fully-reduced metal at one portion of the cathode while reduction has only proceeded as far as an intermediate compound at another portion of the cathode.

Consequently, in a preferred embodiment, the increasing potential or voltage increases at a rate sufficiently slow to control the formation of intermediate compounds, for example to allow formation of an intermediate compound throughout a predetermined proportion of the bulk of the solid compound at the cathode before further reduction proceeds.

In order to achieve these effects, the applied potential or voltage may be increased at a rate of less than 1 V/s or 100 mV/s, preferably at a rate of less than 50 mV/s or 10 mV/s, particularly preferably at a rate of less than 1 iαV/s and in a preferred embodiment at a rate of less than 0.5 mV/s or 0.2 mV/s. If the applied potential or voltage is not increased linearly, the average rate of increase may advantageously not exceed these preferred ramp rates.

A second aspect of the invention provides a method for electro-decomposition of a solid compound in a fused-salt electrolyte, in which the electrolyte comprises a second salt in solution in a first salt, the second salt but not the first salt comprising an anion which is the same as the substance to be removed from the solid compound. The concentration of the second salt in solution in the first salt is advantageously greater than zero during electro- decomposition, and is preferably sufficient to enable transport of the substance through the electrolyte from the cathode to the anode throughout electro-decomposition.

Advantageously, the concentration of the second salt in solution in the first salt is greater than 0.1 mol%, preferably greater than 0.5 mol% and particularly preferably greater than 1 mol% or greater than 2 mol%, during the application of the electro-decomposition voltage.

It is believed that it is advantageous to maintain a predetermined minimum concentration of the second salt in the first salt in order to maintain a pre-determined minimum concentration of the substance (which is the same as the anion in the second salt) in the electrolyte throughout the electro-decomposition process. Although it is not an essential feature of electro-decomposition, it is preferred that the anion discharged at the anode should be the same as or should comprise the substance removed from the solid metal compound. Under preferred conditions, this enables electro-decomposition to be carried out without excessive or unnecessary consumption of the fused-salt electrolyte and preferably with minimum, or limited, variation of the composition of the fused-salt electrolyte. Clearly, if the substance dissolved at the cathode is different from the anion discharged at the anode at any stage during electro-decomposition, the composition of the fused-salt electrolyte must change, and it may be desirable to limit or control any such effects, for example if it is desired to recycle or reuse the electrolyte.

In order to ensure as far as possible that the anion discharged at the anode is the same as or comprises the substance dissolved at the cathode, it is desirable to set up a flux of the substance or anion from the cathode to the anode, which requires that the electrolyte between the cathode and the anode preferably always comprises a pre¬ determined minimum concentration of the substance or anion. Consequently, it is preferred that when a fused- salt electrolyte comprises a second salt in solution in a first salt, or in other words a mixture of first and second salts, and only one of the salts contains an anion corresponding to the substance being dissolved at the cathode, then the concentration of that salt should be maintained above a pre-determined minimum level.

A fused-salt electrolyte comprising first and second salts may comprise further components, such as other salts. For example an electrolyte for electro-decomposition of a metal oxide may comprise a salt such as calcium oxide as the second salt for providing a predetermined oxide concentration in the melt, in solution in, or mixed with, a first salt comprising a eutectic mixture of calcium chloride and lithium chloride. Similarly, the electrolyte may comprise more than one component for the second salt; for example an electrolyte for electro-decomposition of an oxide may comprise more than one oxide in the melt, such as calcium oxide and barium oxide.

In some cases, an electro-decomposition process may involve a reaction between a component of the electrolyte and the solid metal compound, such as a chemical or electrochemical reaction. A proportion of the second salt or its constituent ions may then be consumed at the cathode by this reaction. For example, such a reaction may involve the formation or reaction of an intermediate compound or compounds, formed as an intermediate stage in the electro-decomposition process. In such a case, it is preferable that a sufficient quantity of the second salt should initially be present in the electrolyte to sustain a pre-determined minimum concentration of the second salt in solution in the first salt even when a portion of the second salt has been consumed or reacted at the cathode. Consumption or reaction of the second salt at the cathode is generally temporary; this can be seen by considering the case where electro-decomposition achieves complete reduction of the solid compound to the metal, in which case any of the second salt which was consumed or reacted at the cathode during electro-decomposition must have been released back into the fused salt later in the electro- decomposition process.

The pre-determined minimum concentration of the second salt, or of anions corresponding to the dissolved substance, to be maintained in solution may vary depending on the metal compound being reduced and the composition of the fused-salt electrolyte, but in the case of a fused- salt electrolyte comprising calcium chloride as the first salt and calcium oxide as the second salt being used to reduce a metal oxide, it is understood that a calcium oxide concentration of 5 mol% is generally sufficient. In preferred embodiments, lower concentrations such as concentrations of less than 2 mol% or less than 1 mol% may suffice.

The minimum pre-determined concentration may also depend on the rate of electro-decomposition and the rate at which the substance is dissolved at the cathode. It may therefore be advantageous to control the electro- decomposition voltage, and therefore the driving force for electro-decomposition, as described above as well as controlling the electrolyte composition.

In the embodiment discussed above in which a portion of one component of the fused-salt electrolyte (usually the second salt) is consumed or reacts at the cathode, the quantity of the second salt in the electrolyte may reduce during electro-decomposition. The quantity of the second salt removed from the electrolyte will generally be related to the quantity of the solid metal compound present at the cathode. In such a case, it is advantageous if the initial quantity of the second salt in the electrolyte is sufficient, taking into account the quantity of the solid metal compound at the cathode, to maintain a pre-determined minimum concentration of the second salt in the electrolyte throughout electro- decomposition.

It is further preferred that the electrolyte contains a sufficient quantity of the first salt to keep the sufficient quantity of the second salt in solution in the electrolyte at all times, bearing in mind that the second salt may have a solubility limit for mixing with the first salt. In addition, if the transfer of the substance from the solid compound into the electrolyte may affect the solubility of the second salt in the first salt, then this may also affect the quantity of the first salt desired in the electrolyte. For example, if the substance is the same as the anion in the second salt, then the electrolyte should preferably contain a sufficient quantity of the first salt to accommodate transfer of the substance into the melt without the solubility limit of the second salt in the first salt being approached or exceeded.

These considerations can readily be extended by the skilled person to cases in which the electrolyte contains more complex mixtures of salts, such as the example described above in which the first salt comprises a eutectic or other mixture of salts.

As a particular example, during electro-decomposition of titanium dioxide in an electrolyte comprising calcium chloride as the first salt and calcium oxide as the second salt, it is found that for each mole of titanium dioxide at the cathode, approximately 0.67 moles of calcium oxide is consumed in the formation of intermediate compounds such as calcium titanates at the cathode. It is therefore preferred that in this electro-decomposition process, for each mole of titanium dioxide at the cathode, the electrolyte should contain more than 0.67 moles of calcium oxide. For example, in a preferred embodiment, for each mole of titanium dioxide at the cathode, one mole of calcium oxide may be present in the electrolyte.

In this preferred embodiment, electro-decomposition may typically be carried out at about 900C, at which temperature the solubility limit for calcium oxide in calcium chloride is about 20 mol% . Consequently, for each mole of titanium dioxide present, and therefore for each mole of calcium oxide present at the start of the process, the electrolyte must contain at least about 4 moles of calcium chloride to ensure that the calcium oxide remains in solution at all times. It may be preferred, however, for the electrolyte to contain more calcium chloride than this in order to allow for any increase in the oxygen content of the melt caused by the removal of oxygen from the titanium dioxide, including any local increase in oxygen content in the region of the cathode caused by the removal of oxygen from the titanium dioxide, which might otherwise encourage calcium oxide precipitation in the region of the cathode.

In a further aspect, the invention provides a method for electro-decomposition in which a voltage is applied within a pre-determined voltage range in order to enhance the effectiveness of the electro-decomposition process. As described above, the thermodynamics of the reaction in which the substance is removed from the metal compound to form the metal determines a minimum cathode potential. A preferred upper limit arises because it is advantageous for the cathode potential to be less than a potential for continuous discharge or deposition of cations from the electrolyte at the cathode. However, it has been found that within this range of cathode potential there is a range for which electro-decomposition is enhanced.

For example, when the solid metal compound for electro- decomposition is an oxide it has been found that a fused- salt electrolyte comprising calcium oxide may advantageously be used. In this embodiment the potential at the cathode, measured with respect to a carbon reference electrode, is advantageously less than 1.7 V, preferably less than 1.6 V and is particularly preferably less than about (1541-116 log C) mV, where C is the normalised concentration (i.e. the actual concentration divided by the saturation concentration) of calcium oxide in the fused-salt electrolyte.

In an embodiment in which the metal oxide comprises a titanium oxide, a voltage applied between the cathode and the anode is preferably such that the potential at the cathode, measured with respect to a carbon reference electrode, is more than 1.1 V, and preferably more than 1.3 V. Thus, the preferred cathode potential range for reduction of a titanium oxide in a fused-salt electrolyte comprising calcium oxide is between 1.3 V and 1.7 V, and particularly preferably is between 1.4 V and 1.6 V, or the potential is about 1.5 V, measured with respect to a carbon reference electrode.

In all of these embodiments, if a different reference electrode is used, then the same principles apply to the enhancement of the electro-decomposition process, except that a voltage offset may need to be introduced in order to account for the difference in potential measured using the reference electrode rather than a carbon reference electrode as described above. Determining or evaluating the necessary voltage offset in each case would be well within the competency of the skilled person.

Thus, in order for a method, apparatus or product to fall within this aspect of the invention, it should be noted that a carbon reference electrode does not have to be used in the method or apparatus or in the manufacture of the product. The reference to the carbon reference electrode in relation to this aspect of the invention is only to define precisely the cathode potentials involved. Any appropriate method may be used in practice to measure or to generate the cathode potential, as the skilled person would appreciate.

In a further aspect, the invention provides a method and a corresponding apparatus in which a carbon or refractory metal or tin oxide reference electrode is used to monitor the cathode potential. In this method, the cathode, an anode and the reference electrode are contacted with the fused-salt electrolyte and an electro-decomposition voltage applied between the cathode and the anode.

Measurement of the voltage between the cathode and the reference electrode can then be used to monitor the cathode potential and, in turn, used to control the electro-decomposition voltage in order to achieve a pre- determined desired cathode potential at each stage of the electro-decomposition process.

The foregoing discussion describes various aspects of the electro-decomposition process. These aspects may advantageously be used in combination. For example, an increasing voltage may initially be applied between the cathode and the anode, for example so as to generate an increasing cathode potential, terminating at a voltage within the preferred range for continuous electro- decomposition. A substantially-constant voltage may then be applied at that level, or a voltage applied so as to maintain a substantially-constant cathode potential at the preferred level. The fused-salt-electrolyte composition during this procedure may advantageously be predetermined such that a sufficient concentration of anions corresponding to the substance to be removed from the solid compound is maintained in the electrolyte at all times, so as to ensure a sufficient flux of the substance from the cathode to the anode. A carbon or other reference electrode may advantageously be contacted with the electrolyte to monitor the cathode potential during the process.

In alternative embodiments, other combinations of the various aspects of the invention may advantageously be used, as the skilled person would appreciate.

The various aspects of the invention may advantageously be applied to the electro-decomposition of a wide range of solid metal compounds, including compounds of titanium, silicon, germanium, zirconium, hafnium, samarium, uranium, aluminium, magnesium, neodymiuτn, molybdenum, chromium, niobium, boron, scandium, vanadium, manganese, iron, cobalt, nickel, copper, gallium, yttrium, tantalum, tungsten, rhenium, lead, cerium and plutonium. Aspects of the invention may find particular application in reducing oxides of these metals, for forming the metals and for forming alloys comprising these metals.

In general, the fused salt electrolyte for electro- decomposition may advantageously contain a chloride of calcium, strontium, barium, lithium or a rare earth metal. These salts are particularly efficacious for electro- decomposition of oxides as these salts dissolve their own oxides,- that is, for example, calcium oxide is soluble in calcium chloride, and so on. Thus the first salt in the foregoing description may comprise one of the chlorides mentioned above and the second salt may comprise the corresponding oxide. Alternatively the" second salt may comprise a different cation from the first salt, such as barium oxide mixed with calcium chloride. In further embodiments, more complex mixtures of salts may be used.

For example an electrolyte containing calcium chloride may also contain sodium chloride or potassium chloride, for instance to modify the melting point of the electrolyte. A suitable oxide may then be included in the salt to enable oxygen transfer to the anode, where an oxide is to be electro-decomposed.

Specific Embodiments and Best Mode of the 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 apparatus for carrying out an electro-decomposition process according to an embodiment of the invention;

Figure 2 is a diagram of a cathode structure comprising a titanium dioxide pellet suspended on a nickel wire, for use with the apparatus of Figure 1;

Figure 3 is a diagram of an electrode arrangement for carrying out pre-electrolysis of a fused-salt electrolyte, according to a further embodiment of the invention; and

Figure 4 is a diagram of an electrode arrangement for use with the apparatus of Figure 1, and incorporating a reference electrode.

Figure 1 is a schematic diagram of an apparatus 2 for carrying out an electro-decomposition process. The apparatus comprises an alumina crucible 4 containing a fused-salt electrolyte 6. A cathode 8, an anode 10 and a shielded thermocouple 12 are immersed in the electrolyte. The anode and the cathode are connected to a potentiostatically-controlled voltage source 14. The crucible is positioned on a ceramic insulator 16 within an Inconel (RTM) reactor 18 of height 65 cm and inner diameter 13.4 cm. An upper end of the reactor is provided with water cooling 20 and closed by a stainless-steel cover 22 sealed with an O-ring 24. The cover is provided with electrical feedthroughs for the power supply and for a lead for the thermocouple, as well as with a gas inlet 26 and gas outlet 28 which enable control of the atmosphere within the reactor. The reactor is externally- heated inside a vertical-tube furnace (not shown) ; in an alternative embodiment, the reactor may be internally heated.

In the apparatus of Figure 1, the anode is a graphite rod connected by a nickel wire to the power supply, or voltage source; the graphite rod and the nickel wire are each threaded into blind bores at opposite ends of a stainless- steel connector.

Example 1: TiO₂

In a first Example, in the apparatus of Figure 1 the cathode comprises a disk or pellet 32 of titanium dioxide, prepared as described below, connected to the power source by means of a nickel wire 30. The titania disk has a hole drilled through its centre; the nickel wire is then threaded through the hole and twisted back on itself to secure the disk at the end of the wire, as illustrated in Figure 2.

The titania disk is formed from a commercial powder (Alfa Aesar) specified as rutile, 99.5% pure, average particle size between 1 and 2 μm. The powder is dried in an oven for several days at temperatures around 100⁰C. Then 1.0% by mass of a PVB/PVA mixture (poly vinyl butyral-co-vinyl alcohol-co-vinyl acetate, approximately 80 wt % butyral) and 0.5% by mass of PEG (poly (ethylene glycol), average molecular mass 200) are added. The components are mixed by wet milling using alumina balls in iso-propanol for 24 hours. The powder is dried at around 100⁰C and passed through a vibrating 53 μm stainless steel sieve. The treated titanium dioxide powder is then pressed into disks weighing between 1 g and 8 g which are prepared by, firstly, uniaxial pressing at about 50 MPa to form a green body and, second, isostatic pressing at 175 MPa in order to further densify the green body. Disks are then sintered at 1050C for 150 iriin in air. Sintering temperatures of between IOOOC and 1250C, or preferably between IOOOC and 1100C, are found to be particularly effective in producing disks or other electro- decomposition precursors of titania that have sufficient mechanical strength to be used in electro-decomposition without damage, and open porosity of about 30%, which advantageously enables intimate access of the fused-salt electrolyte to the interior of the disk or other precursor during electro-decomposition.

It is particularly important that the sintering temperature for the pellets is at least as high, and preferably higher than, the electro-decomposition temperature (electrolyte temperature) to be used. This advantageously avoids further sintering of the pellet on immersion into the fused-salt electrolyte and thus enhances control of the pellet properties. Consequently, the sintering temperatures and ranges described above are particularly effective for electro-decomposition at 900C.

Clearly, the preparation procedure may be modified by the skilled person in order to prepare electro-decomposition precursor materials of compounds other than titanium dioxide, for example if metals other than titanium are to be extracted, or to incorporate mixtures of compounds if alloys are to be prepared.

The fused-salt electrolyte in the embodiment is as follows. 1.95 mol of anhydrous calcium chloride (Fluka, >97% purity) (corresponding to 216 g) and 0.05 mol of anhydrous calcium oxide (formed by calcining CaCO₃, Aldrich, >99% purity) (corresponding to 2.8 g) are melted to form the electrolyte, giving a nominal calcium oxide concentration of 2.5 mol%. This gives a molten-salt depth of between 4 and 5 cm in the apparatus of Figure 1.

Alternative embodiments may vary from the apparatus of Figure 1 in a number of ways in accordance with known electro-decomposition processes as published in patent application WO 99/01781 and elsewhere. For example the cathode may comprise the solid compound in any form of electro-decomposition precursor, such as pellets or other artefacts contained in a metal basket or other cathode assembly, or a precursor of a predetermined shape for forming an electro-decomposition product of a predetermined shape. The crucible may be of a conducting material and used as the anode or as the cathode, in the latter case contacting the solid metal compound. The composition of the fused-salt electrolyte may be selected according to, for example, the metal compound to be reduced. The electrolyte may comprise a single salt or a mixture of any number of salts. For example it may comprise a eutectic mixture of salts if it is desired to obtain a low-melting-point electrolyte. Advantageously, salts or mixtures of salts may comprise salts of the Group 1 or Group 2 metals, including halides and oxides of these metals. The electrolyte temperature, anode materials, and other parameters of the electro- decomposition process may also be varied. For example, higher temperatures might be used to accelerate diffusion rates and so accelerate electro-decomposition, but may also disadvantageously affect corrosion of the apparatus.

In the embodiment described above, before electro- decomposition begins a thermal drying and pre-electrolysis procedure is performed as follows, in order to remove any- remaining water from the electrolyte. The thermal drying procedure comprises the following heating sequence; ramp at 2C/min to 150C and hold for at least 5 hours, ramp at 2C/min to 300C and hold for 5 hours, and ramp to target temperature of 900C. Pre-electrolysis is then performed as follows. Three graphite electrodes are immersed into the molten-salt electrolyte and are applied, as illustrated in Figure 3, as working, counter and pseudo- reference electrodes 34, 36, 38. A voltage source 40 is applied between the working electrode (cathode) and the counter electrode (anode) and the cathodic polarisation of the working electrode against the reference electrode monitored and used to control the voltage applied by the voltage source. During pre-electrolysis, the applied voltage is controlled so that the cathodic polarisation of the working electrode versus the reference electrode is first ramped from 0 V to 1 V at a ramp rate of 0.087 mV/s and then maintained at a constant potential of 1 V. Pre- electrolysis is continued until a small and constant background current is encountered.

In a preferred embodiment, electro-deoxidation is then performed using the apparatus illustrated in Figure 1 but additionally incorporating a graphite pseudo-reference electrode 42, immersed in the electrolyte and connected as illustrated in Figure 4 so that the voltage applied by the voltage source can be controlled in response to the cathodic polarisation of the cathode, measured with respect to the reference electrode.

In each of a set of experiments using the apparatus of Figure 1, a titanium dioxide pellet of approximately 4 g weight, 4 mm thickness and 30% open porosity was connected to a nickel wire to form the cathode. A graphite rod was employed as the anode and the thermally dried and pre- electrolysed calcium chloride/ calcium oxide melt described above was used as the electrolyte. The anode- cathode separation was about 4 cm. During reduction the voltage source 14 was controlled so that the cathodic polarisation of the cathode against the pseudo-reference electrode was first ramped from 0 V to a maximum voltage level, and then the maximum voltage level maintained in a constant manner for an additional period of 16 hours.

A number of experiments were carried out in which the ramp rate was 0.087 mV/s and the maximum voltage levels were 0.9, 1.0, 1.1, 1.3, 1.4, 1.5, 1.6, 1.7 and 1.9 V respectively, measured between the cathode and the reference electrode. The durations of the individual experiments were thus between about 18.9 and 22.1 hours.

In each experiment, an atmosphere of dry argon and a reaction temperature of 900C were applied. After each experiment, the electro-reduced pellet was removed from the electrolyte, rinsed with water, acid-leached and dried, before being broken up and inspected.

The pellets obtained from reductions performed at maximum voltage levels of 0.9 and 1.0 V were violet inside and dark yellow outside, suggesting the presence of CaTi₂O₄ and TiO respectively. This suggests that only partial reduction of the titanium dioxide was achieved. The pellets processed at 1.1, 1.3 and 1.4 V maximum voltage levels exhibited a grey metallic outer scale around a partially-reduced core, the thickness of the metallic scale being greater for the higher voltage levels. The pellets made at 1.5 and 1.6 V maximum voltages were metallic grey throughout their entire volumes, indicating complete conversion into titanium metal. The pellets processed at 1.7 and 1.9 V maximum voltages were reduced to metal only at the surface, over a partially-reduced core.

In other experiments, faster and slower ramp rates

(including 0.174 and 0.044 mV/s) were used, terminating at the same potentials as described above, but a ramp rate of 0.087 mV/s was found to be most effective in accelerating electro-decomposition in the embodiment.

During electro-decomposition, it was observed that the cell voltage applied between the anode and the cathode was approximately 1 V greater than the voltage measured between the cathode and the pseudo-reference electrode. This suggests that polarisation of the reaction at the anode consumes about 1 V of the applied cell voltage.

The fact that the formation of titanium metal is not observed below an effective voltage (measured against the pseudo-reference electrode) of 1.1 V is in good agreement with thermodynamic expectations, as described below. It is believed that the electro-deoxidation of TiO₂ to Ti metal may involve the reactions 1 to 8 below. Reactions 1 to 5 describe the progressive reduction of titanium and the associated formation of calcium titanates. The calcium titanates CaTiO₃ and CaTi₂O₄ are observed in the electro-decomposition of TiO₂ and it is believed that these decompose through reactions 6, 7 and 8. Reaction 6 is a chemical reaction and reaction 7 is an electrochemical reaction, both converting CaTiO₃ to CaTi₂O₄. Reaction 8 is an electrochemical reaction converting CaTi₂O₄ to TiO, which is then believed to reduce according to reaction 5. It is assumed that CO is evolved at the anode; in practice, this may depend on the anode material and reaction kinetics at the anode. It is believed that an equilibrium between C and CO occurs at a surface of the graphite pseudo-reference electrode and therefore that it is appropriate to refer to this reaction to calculate expected cathodic polarisations E relative to the reference electrode, as listed below.

5TiO₂ + CaO + C = Ti₄O₇ + CaTiO₃ + C0 E = +266mV (1)

4Ti₄O₇ + CaO + C = 5Ti₃O₅ + CaTiO₃ + C0 E = +137mV (2)

3Ti₃O₅ + CaO + C = 4Ti₂O₃ + CaTiO₃ + C0 E = +33mV (3)

2Ti₂O₃ + CaO + C = 3TiO + CaTiO₃ + C0 E = -342mV (4)

TiO + C = Ti + CO E = -1114mV (5)

CaTiO₃ + TiO = CaTi₂O₄ (6)

2CaTiO₃ + C = CaTi₂O₄ + CaO + C0 (7)

CaTi₂O₄ + C = 2TiO + CaO + C0 (8)

The potentials E listed above are as calculated using standard software (HSC Chemistry, Version 4.1, Outokumpu Research Oy, Pori, Finland) . The calculations are based on the free-enthalpy change in each reaction, at a temperature of 900C and assuming that all compounds are present at unit activity (clearly this may not be the case but actual activity differences are expected to have only a limited effect on the calculated potentials) . Following common conventions, a negative potential E corresponds to a positive enthalpy change.

If CO₂ rather than CO were to be formed (i.e. to be the species determining the reference electrode potential) at the reference electrode, all of the potentials would become more negative by 90 mV.

It can clearly be seen that the expected potential required to prepare Ti metal, as indicated by equation 5, is approximately 1.1 V, which shows good agreement with the experiment described above.

The preferred potential for reduction of the entire pellet is between 1.4 V and 1.7 V, as evidenced by the successful reduction to Ti in the experiments carried out at 1.5 V and 1.6 V. The requirement for a potential in excess of 1.1 V indicates that there may be some polarisation of the reaction at the cathode. The less-successful results at potentials of 1.7 V and above indicate that these potentials are less preferred.

Example 2: Cr₂O₃

This example illustrates the reduction of Cr₂O₃ to Cr metal. Chromium sesquioxide, Cr₂O₃, disks were formed from a commercial powder of particle size less than 3μm (Elementis Pigments) . The powder was dried, mixed with a binder and sieved in the same way as the titania powder described above. The treated powder was then pressed into disks weighing 2.7g which were prepared by uniaxial pressing at about 100 Mpa to form a green body. The subsequent isostatic pressing ^■ step used for preparing titania disks was found to be dispensable as the uniaxially-pressed disks were sufficiently robust. The Cr₂O₃ disks were then sintered at 1300C in air for 150 rain. Sintering temperatures of between 1100 and 1500C are found to be particular effective in producing disks or other electro-decomposition precursors of chromia that have sufficient mechanical strength to be used in electro- decomposition without damage, while maintaining a substantial degree of open porosity that enables intimate access of the fused-salt electrolyte to the interior of the disk or other precursor during electro-decomposition.

Using the apparatus of Figure 1, a porous chromium sesquioxide pellet (disk) of approximately 2.7g weight and 3mm thickness was connected to nickel wire to form the cathode. A graphite rod was employed as the anode and the thermally dried and pre-electrolysed calcium chloride/calcium oxide melt described above was used as the electrolyte. The anode-cathode separation was 4cm, and a carbon reference electrode was provided. During reduction the voltage source 14 was controlled so that the cathodic polarisation of the cathode against the carbon reference electrode was first ramped from 0 V to a maximum voltage level, and then the maximum voltage maintained in a constant manner. The ramp rate was 0.087 mV/s, the maximum voltage level was 1.0 V measured between the cathode and the reference electrode, and the dwell time at the maximum voltage level was 8 h, rendering the duration of the experiment about 11.2 h.

In the experiment, an atmosphere of dry argon and a reaction temperature of 900C were applied. After the experiment, the electro-reduced pellet was removed from the electrolyte, rinsed with water, leached with semi- concentrated acetic acid and dried, before being broken up and inspected.

The processed specimen was metallic throughout its entire volume. X-ray diffraction analysis proved the exclusive presence of chromium metal, and the quantitative determination of oxygen content yielded a numerical value of approximately 2800 ppm. The fact that the formation of chromium metal is observed at a relatively low effective voltage (measured against the pseudo-reference electrode) of 1.0V is in accordance with the thermodynamics, as illustrated in equations 9 to 11. The electrode potentials calculated for each reaction based on free-enthalpy changes at 900C are given.

Cr₂O₃ + 3C = 2Cr + 3CO E=-322mV (9)

3Cr₂O₃ + 2CaO + 3C = 2Cr + 2CaCr₂O₄ + 3CO E=-117mV (10) CaCr₂O₄ + 3C = 2Cr + CaO + 3CO E=-424mV (11)

The temporary presence of calcium chromites during electro-reduction of chromium sesquioxide is observed on analysis of incompletely-reduced samples.

The preferred potential for reduction of chromium sesquioxide is between 0.5 V (i.e. a little greater than the potential of -424 mV for reaction 11) , as measured with respect to a carbon reference electrode, and the potential at which substantial electronic conduction through the electrolyte occurs. The potential (as measured against a carbon reference electrode) is therefore preferably between 0.5 V and 1.7 V and particularly preferably between 0.6 V and 1.3 V, or between 0.65 V and 1.0 V.

The rapid and successful preparation of Cr through electro-decomposition of Cr₂O₃ clearly demonstrates the strength of the process. Through the choice of a potential which is sufficiently high to reduce the Cr₂O₃ but low enough to avoid unnecessary or excessive damage to or consumption of the electrolyte, decomposition of the CaO dissolved in the CaCl₂ is substantially precluded, and the problems arising from electronic transference in the electrolyte are avoided. This may advantageously improve process control and increase current efficiency in a significant manner.

Background Understanding

Although the reaction mechanisms involved in the electro- decomposition process are not yet fully understood, the inventors' current understanding of the mechanism of electro-decomposition in the embodiments described above is as follows. Since this understanding is not yet complete or definitive, while the following comments may guide the skilled person they should not be considered to limit the present invention beyond the definitions set out in the claims.

Reactions 1 to 8 illustrate that the reduction of TiO₂ to Ti is believed to involve the formation of a sequence of intermediate compounds, including calcium titanates and titanium oxides containing titanium in various oxidation states, and the corresponding potentials show that increasing potential values are required to reduce the titanium to progressively lower oxidation states. Thus, it can be seen that the ramped potential applied in the experiments above may advantageously encourage the formation of the intermediate compounds in sequence throughout the bulk of the metal compound at the cathode. The ramped voltage may also help to reduce any tendency for full conversion to metal to occur most rapidly at the surface of the metal compound, and so may ensure that the electro-decomposition reaction proceeds more evenly throughout the bulk of the titania pellet, reducing the likelihood of creating a metallic surface surrounding an unreduced or partially-reduced core. The inventors believe that a further factor in increasing the rate of electro-decomposition may concern the formation of calcium oxide at the cathode, as oxygen is removed from the cathode and dissolved in the electrolyte. Calcium oxide has limited solubility in calcium chloride (approximately 20 mol% at 900C) and a limited rate of dissolution, and if oxygen is removed from the cathode too rapidly, these limits may be exceeded and cause precipitation of calcium oxide in the region of the cathode. Since the cathode is a porous body of titanium dioxide, if calcium oxide precipitates in the pores of the body, further reaction may become extremely slow or even stall completely. The inventors term this phenomenon "oxide quenching". It is believed that a further advantage of the application of an increasing potential or voltage starting from a low level at the beginning of electro-decomposition may be to control the rate of reaction at the cathode to a rate at which dissolved oxygen can be transported away from the cathode region before calcium oxide precipitation occurs .

In cases where the solid compound reacts with the electrolyte, as in the reactions of TiO₂ and Cr₂O₃ with CaO described above, it is believed that the compounds formed in such reactions generally decompose as electro-reduction proceeds to produce the desired metal. This may lead to an effective increase in the rate of removal of the substance from the solid compound; in the example of TiO₂ this would involve the decomposition of calcium titanates releasing CaO back into the electrolyte. This process may increase the risk of precipitation of the substance or of a material comprising the substance (e.g. CaO) in the region of the cathode, and it is thought that this may advantageously be controlled by applying an increasing voltage rising from a low level at the start of electro- decomposition, and/or by controlling the electro- decomposition voltage to a suitable level throughout electro-decomposition so as not to drive the process too fast. Suitable design of the solid-compound precursor, for example to increase porosity and reduce precursor size, may also assist in accelerating diffusion and dissolution of the substance.

The specific oxide-ion conductivity of an electrolyte is given by the following expression.

σ ₀2- = 2 F c ₀2- u ₀2- (12)

where σ, F, c_r u are specific conductivity, Faraday constant, molar concentration and electrical mobility, respectively, and the subscript denotes the ion species concerned. Thus, increasing the oxide concentration in the electrolyte by adding calcium oxide to calcium chloride is believed to increase the specific oxide-ion conductivity of the electrolyte. Although it is known that electro-decomposition of metal oxides can proceed at low oxygen-ion concentrations, for example by discharging chlorine from a calcium chloride electrolyte at the anode, it has been observed by the inventors that the presence of calcium oxide dissolved in calcium chloride tends to accelerate the rate of the reduction process for metal oxides. This points to the possible occurrence of a transport limitation in the electrolyte for the transport of oxygen dissolved at the cathode, through the electrolyte, to the anode under conditions when very little or no calcium oxide is present in the electrolyte.

In order to benefit from the accelerating effect of the dissolved calcium oxide, it is believed that the calcium oxide should not be reduced to too low a concentration in the electrolyte during electro-decomposition, through chemical or electrochemical reactions . During electro- decomposition of titanium dioxide, dissolved calcium oxide in the electrolyte may react with the titanium dioxide or other titanium oxides at the cathode to form calcium titanates, as in reactions 1 to 4 for example. For this reason, it is believed to be advantageous to ensure that the electrolyte contains a sufficient amount of calcium oxide to ensure that a sufficient concentration of calcium oxide will still be present in the electrolyte to achieve oxygen transport to the anode after the reaction of calcium oxide with the cathode material has removed some of the calcium oxide from the electrolyte. In this preferred embodiment, maintenance of an adequate calcium oxide concentration in the melt may not only accelerate oxygen transport to the anode but also reduce any polarisation of the oxygen-discharge reaction at the anode and thereby suppress chlorine discharge from the calcium chloride.

It is important to note that reaction of the electrolyte with the metal compound in this way does not involve continuous electrochemical decomposition of the salt or deposition of metal from the salt. As clearly shown in reactions 1 to 4, the reaction is between CaO and the metal compounds at the cathode, and therefore stops when one of these reagents is exhausted. As described above, it is believed to be preferable that a concentration of oxygen ions is maintained in the electrolyte throughout the electro-decomposition and so it is preferred that the amounts of the solid metal compound and CaO present are predetermined so that the metal compound is the reagent exhausted by the reaction between them. It is believed that the concentration of dissolved calcium oxide in the calcium chloride, and hence the concentration of oxide ions, may disadvantageously decrease if the effective potential across the electrolyte, i.e. the cell 5 voltage minus external losses and electrode polarisation, were to reach or exceed the decomposition potential of the dissolved calcium oxide. Assuming that the cathode is essentially inert, this potential as measured against a carbon reference electrode may be calculated. For an 0 electrolyte temperature of 900C, unit activities of the chemical components, and under the assumption that CO is formed at the reference electrode, the calculated decomposition potential is -1541 mV and becomes more negative by 116 mV for each order of magnitude by which s the calcium oxide concentration falls. (As noted above, the inventors' experiments indicate that significant polarisation occurs at a working carbon anode during electro-decomposition and that the effective potential between the cathode and the anode is typically at least 0 about 1 V higher than the potential between the cathode and the reference electrode. Of course, the applied cell voltage required to achieve this effective potential would be higher still, to account for IR losses in the cell.) During electro-decomposition, the inventors believe that S this decomposition potential, as measured between the cathode and the carbon reference electrode, should advantageously not be exceeded because if such a high voltage is applied then calcium oxide decomposition may reduce the oxide concentration in the electrolyte and the 0 rate of electro-deoxidation may disadvantageously be reduced by any resulting oxide-ion transport limitation in the electrolyte. This is expected to be the case even though the driving force for electro-deoxidation may be high (corresponding to the high applied voltage) and even S if there is a considerable amount of oxygen left in the cathode. In effect, it may be proposed that calcium oxide electrolysis (decomposition) and electro-deoxidation are opposing processes.

It is, however, believed that some discharge of calcium ions from the electrolyte may occur at the cathode even at voltages below the decomposition voltage for any of the salts in the electrolyte. Calcium, in the form of Ca⁰ or Ca¹⁺, has some solubility in calcium chloride. Consequently, when a voltage below the calcium oxide decomposition voltage is applied to a cell, the inventors believe that some reaction of calcium ions may occur at the cathode, converting Ca²⁺ to Ca¹⁺ or Ca⁰ in solution in the electrolyte until a corresponding activity for the dissolved calcium species in the melt is reached. It is important to note, however, that this is not a continuous process and therefore does not constitute decomposition of the calcium oxide. It is a self-limiting process which only proceeds until a calcium-activity level corresponding to the applied cathode potential is reached in the salt, and then stops. This process does not involve continuous discharge or deposition of calcium. However, the generation of Ca⁰ or Ca¹⁺ in solution in the melt may have an important impact on electro-decomposition in that these species are understood to increase the electronic conductivity of the electrolyte. Clearly, this is a disadvantageous effect, as any electronic conduction does not contribute to the electro-decomposition process and reduces the electrical efficiency of the process. Consequently, the inventors believe that it may be desirable to operate the electro-decomposition process at as low a voltage as possible in order to reduce or limit the concentration of Ca⁰ or Ca⁺ in the electrolyte. An adequate voltage is required to achieve electro- decomposition, as illustrated by the examples discussed above, but excessive voltages may also advantageously be avoided.

Application of Background Understanding

Following this explanation of the inventors' understanding of the mechanisms involved in electro-decomposition, it is possible to apply these concepts to the various aspects of the invention.

First, the ramp rate, or rate of increase of the voltage initially applied during electro-decomposition, should preferably be selected so as to encourage an even progression of the reaction throughout the bulk of the metal compound at the cathode, and so as to avoid the initial stages of the reaction progressing too quickly to allow dissolution or dispersal of the substance removed from the metal compound in the electrolyte. Otherwise, the reaction may be expected to proceed most rapidly in its early stages, when the maximum amount of the substance is present in the solid compound.

In the embodiments described above, ramp rates of 0.087 mV/s and faster and slower ramp rates (including

0.174 and 0.044 mV/s) were used, although 0.087 mV/s was found to be more effective in accelerating electro- decomposition of TiO₂ under the conditions in the embodiments. However, in view of the discussion above the skilled person would appreciate that the optimum rate of voltage increase is likely to depend on factors such as the rate at which the removed substance can dissolve in the electrolyte and be transported away from the cathode, which is a function of the materials involved and the cathode geometry (in that a small, highly-porous cathode would allow more rapid transport) and would involve consideration of the reactions involved at the cathode, which, may vary for different metal compounds and different electrolytes . The skilled person may thus apply the teaching. in the present patent application to different electro-decomposition processes and parameters, including for example the electro-reduction of different materials or different structures (e.g. using feed materials of different dimensions and shapes, porosity and particle size) , the use of different electrolytes at different temperatures, and the design of different reactor geometries, without inventive input.

The starting level for the increasing voltage may advantageously be selected depending on the thermodynamics of the reaction at the cathode. The increase in voltage may start from 0 V but if no reaction can occur at the cathode until a higher voltage level (because a certain minimum applied voltage, or a corresponding minimum cathode potential, is required to drive the electro- decomposition reaction) then any voltage or potential below that minimum voltage or potential may be selected as the initial level for the increasing voltage or potential.

The maximum voltage level of the increasing voltage may advantageously be selected so as to maximise the driving force for electro-decomposition while not exceeding a level at which excessive consumption of or damage to the electrolyte may occur. In addition, if the substance removed from the solid compound has low solubility in the electrolyte, it may be important not to drive the electro- decomposition reaction too rapidly, as this may cause precipitation of the substance or a material comprising the substance at the cathode. In that case, the maximum voltage level should be selected so as to apply a reduced driving force for the reaction, so as to avoid the rate of reaction being limited by kinetic effects such as the rate of dissolution of the substance.

Second, the voltage level selected for continuous application during electro-decomposition, either from the start of the reaction or following a ramped, increasing voltage, may similarly be selected so as to drive the reaction as rapidly as possible without excessively or unnecessarily consuming or damaging the electrolyte and without introducing kinetic limitations such as may be caused by the precipitation of material at the cathode.

Third, it may be desirable for the electrolyte to contain a sufficient concentration of anions of the same species as the substance to be removed from the solid compound, if it is envisaged that that substance will be transported through the electrolyte from the cathode for discharge at the anode. If the electrolyte comprises a mixture of salts in which the salt containing this anion species is present in relatively low concentration, then it may be desirable to maintain a sufficient concentration of this salt at all times during electro-decomposition in order to avoid introducing anion-transport limitations. If, as in the case of the reduction of titanium dioxide in an electrolyte of calcium chloride and calcium oxide, a reaction between the salt and the solid compound occurs, then this should be taken into account when selecting the quantity of the metal compound and the quantity of the corresponding salt in an electro-decomposition reactor. A sufficient quantity of the electrolyte should then preferably be provided to keep the corresponding salt in solution.

Fourth, it may be important to monitor the cathode potential accurately, particularly in an experimental situation where the full details of the reaction may not be understood, for example by using a reference electrode or pseudo-reference electrode as described above. Otherwise, polarisation effects at the anode, which may be very significant, may not be accurately accounted for.

Although these aspects of the invention and the factors involved in enhancing the electro-decomposition process have been exemplified with reference to the reduction of titanium dioxide in a mixture of calcium chloride and calcium oxide, the various aspects of the invention may equally be applicable to other metal compounds and other electrolytes, and other parameters of the process such as the operating temperature may be varied, as the skilled person would appreciate. For example, other salts may be used as the electrolyte, including mixtures of salts such as eutectic mixtures, and including mixtures of two or more salts which may comprise the same or different cations and/or anions. The use of different electrolytes would typically change the electrolyte or salt decomposition potential or potentials, and so may alter parameters such as the preferred range of the applied voltage during electro-decomposition, but such modifications to the embodiments described herein could easily be made by the skilled person, using his common general knowledge in the field.

Claims

1. A method for removing a substance from a solid compound comprising the substance and a metal, comprising the steps of:

contacting a cathode comprising the solid compound with a fused-salt electrolyte;

contacting an anode with the electrolyte; and,

applying a voltage between the cathode and the anode which increases with time.

2. A method according to claim 1, in which the voltage increases substantially continuously with time.

3. A method according to claim 1 or 2, in which the voltage increases stepwise with time.

4. A method according to claim 1, 2 or 3, in which the electrolyte comprises a cation and the voltage is applied such that the potential at the cathode increases to a potential which is less than a potential for continuous deposition of the cation from the electrolyte at the cathode.

5. A method according to any preceding claim, in which the voltage is applied such that the effective potential between the cathode and the anode increases to a potential which is less than a potential sufficient to cause continuous decomposition of the electrolyte.

6. A method according to any preceding claim, in which the voltage is applied such that the potential at the cathode increases from a potential which is insufficient to cause removal of the substance from the compound.

7. A method according to any preceding claim, in which the voltage is applied such that the potential at the cathode increases from a potential of approximately 0 V.

8. A method according to any preceding claim, in which, after the increasing voltage has been applied, a substantially constant voltage is applied between the cathode and the anode.

9. A method according to claim 8, in which the electrolyte comprises a cation and the substantially- constant voltage is such that the potential at the cathode is less than a potential for continuous deposition of the cation from the electrolyte at the cathode.

10. A method according to claim 8 or 9, in which the substantially-constant voltage is insufficient to cause continuous decomposition of the electrolyte.

11. A method according to any preceding claim, in which, after the increasing voltage has been applied, a voltage is applied between the cathode and the anode such that the potential at the cathode is substantially constant.

12. A method according to claim 11, in which the electrolyte comprises a cation and the substantially- constant potential is less than a potential for continuous deposition of the cation from the electrolyte at the cathode.

13. A method according to claim 11 or 12, in which the substantially-constant potential is insufficient to cause continuous decomposition of the electrolyte.

14. A method according to any preceding claim, in which the potential at the cathode is monitored using a reference electrode.

15. A method according to any preceding claim, in which 5 the reference electrode comprises a carbon electrode.

16. A method according to any preceding claim, in which the fused-salt electrolyte comprises a second salt in solution in a first salt, the first and second salts containing different anions, and in which the increasing 0 voltage increases at a rate sufficiently slow to favour discharge of anions from the second salt rather than anions from the first salt at the anode.

17. A method according to claim 16, in which the substance comprises oxygen, the solid compound comprises 5 an oxide and the electrolyte comprises calcium chloride containing calcium oxide in solution therein, and in which the increasing voltage increases at a rate sufficiently slow to favour discharge of oxygen anions rather than chlorine anions at the anode.

Q 18. A method according to any preceding claim, in which the increasing voltage increases at a rate sufficiently slow to suppress precipitation of the substance removed from the solid metal compound, or of a material comprising the substance removed from the solid metal compound, in 5 the region of the cathode.

19. A method according to claim 18, in which the substance comprises oxygen, the solid compound comprises an oxide and the electrolyte comprises calcium chloride containing calcium oxide in solution therein, and in which 0 the increasing voltage increases at a rate sufficiently slow to suppress precipitation of calcium oxide at the cathode.

20. A method according to any preceding claim, in which removal of the substance from the solid compound to produce the metal involves the formation of an intermediate compound comprising the substance and the metal, and in which the increasing voltage increases at a rate sufficiently slow to favour formation of the intermediate compound before formation of the metal.

21. A method according to any preceding claim, in which the voltage increases at a rate of less than 0.2mV/s, and preferably at a rate of less than 0.1mV/s.

22. A method according to any preceding claim, in which:

the fused-salt electrolyte comprises a second salt in solution in a first salt, the second salt but not the first salt comprising an anion which is the same as the substance; and

the concentration of the second salt in solution in the first salt is greater than zero, and preferably greater than lmol%, during the application of the voltage.

23. A method according to claim 22, in which the concentration of the second salt in solution in the first salt during the application of the voltage is sufficient to enable transport of the substance through the electrolyte from the cathode to the anode.

24. A method according to claim 22 or 23, in which the removal of the substance from the solid compound to produce the metal involves the formation of an intermediate compound comprising a cation and/or an anion of the second salt, and the fused-salt electrolyte contains a sufficient quantity of the second salt both to enable formation of the intermediate compound and to retain the concentration of the second salt in solution in s the first salt greater than zero, and preferably greater than lmol%, during the application of the voltage.

25. A method according to claim 24, in which the fused- salt electrolyte contains a sufficient quantity of the first salt to retain the sufficient quantity of the second 0 salt in solution.

26. A method according to any of claims 22 to 25, in which the substance comprises oxygen, the compound comprises an oxide, and the anion in the second salt comprises oxygen.

5 27. A method according to claim 26, in which the first salt is calcium chloride and the second salt is calcium oxide.

28. A method according to claim 26 or 27, in which the compound comprises a titanium oxide.

0 29. A method according to any preceding claim, in which the electrolyte comprises calcium oxide; and,

the voltage applied between the cathode and the anode is such that the potential at the cathode is insufficient to cause continuous decomposition of the calcium oxide.

5 30. A method according to claim 29, in which the potential at the cathode, as measured with respect to a carbon reference electrode, is less than about (1541- 116 log C) mV, where C is the normalised concentration of calcium oxide in the fused-salt electrolyte.

31. A method according to claim 29, in which the potential at the cathode, as measured with respect to a carbon reference electrode, is less than 1.7 V, and preferably less than 1.6 V.

32. A method according to any of claims 29 to 31, in which the metal compound comprises a titanium oxide, the anode is a carbon anode, and the voltage applied between the cathode and the anode is such that the potential at the cathode, as measured with respect to a carbon reference electrode, is more than 1.1 V, and preferably more than 1.3 V.

33. A method according to any preceding claim, in which the solid metal compound comprises a titanium oxide, and the voltage applied between the cathode and the anode is such that the potential at the cathode is more than 1.1 V, and preferably more than 1.3 V.

34. A method for removing a substance from a solid compound comprising the substance and a metal, comprising the steps of:

contacting a cathode comprising the solid compound with a fused-salt electrolyte comprising a second salt in solution in a first salt, the second salt but not the first salt comprising an anion which is the same as the substance;

contacting an anode with the electrolyte; and,

applying a voltage between the cathode and the anode such that the potential at the cathode is less than a potential for deposition of cations from the electrolyte; the concentration of the second salt in solution in the first salt being greater than zero, and preferably greater than lmol%, during the application of the voltage.

35. A method according to claim 34, in which the concentration of the second salt in solution in the first salt during the application of the voltage is sufficient to enable transport of the substance through the electrolyte from the cathode to the anode.

36. A method according to claim 34 or 35, in which the removal of the substance from the solid compound to produce the metal involves the formation of an intermediate compound comprising a cation and/or an anion of the second salt, and the fused-salt electrolyte contains a sufficient quantity of the second salt both to enable formation of the intermediate compound and to retain the concentration of the second salt in solution in the first salt greater than zero, and preferably greater than 1 mol%, during the application of the voltage.

37. A method according to claim 36, in which the fused- salt electrolyte comprises a sufficient quantity of the first salt to retain the sufficient quantity of the second salt in solution.

38. A method according to any of claims 34 to 37, in which the substance comprises oxygen, the compound comprises an oxide, and the second salt comprises an oxide.

39. A method according to claim 38, in which the first salt is calcium chloride and the second salt is calcium oxide.

40. A method according to claim 38 or 39, in which the compound comprises a titanium oxide.

41. A method for removing a substance from a solid metal compound comprising the substance and a metal, comprising the steps of:

contacting a cathode comprising the solid metal compound with a fused-salt electrolyte, the electrolyte comprising a corresponding salt which contains the substance as an anion;

contacting an anode with the electrolyte; and,

applying a voltage between the cathode and the anode which is insufficient to cause continuous decomposition of the corresponding salt.

42. A method according to claim 41, in which the solid metal compound comprises a metal oxide and the corresponding salt comprises an oxide.

43. A method according to claim 42, in which the corresponding salt is calcium oxide, the anode is a carbon anode and the potential at the cathode, as measured with respect to a carbon reference electrode, is less than about (1541-116 log C) mV, where C is the normalised concentration of calcium oxide in the fused-salt electrolyte.

44. A method according to claim 42, in which the corresponding salt is calcium oxide, the anode is a carbon anode and the potential at the cathode, as measured with respect to a carbon reference electrode, is less than 1.7 V, and preferably less than 1.6 V.

45. A method according to any of claims 41 to 43, in which the metal oxide comprises a titanium oxide, the anode is a carbon anode, and the voltage applied between the cathode and the anode .is such that the potential at the cathode, as measured with respect to a carbon reference electrode, is more than 1.1 V, and preferably more than 1.3 V.

46. A method for removing oxygen from a solid titanium oxide, comprising the steps of: ,

contacting a cathode comprising the solid titanium oxide with a fused-salt electrolyte;

contacting a carbon anode with the electrolyte; and,

applying a voltage between the cathode and the anode such that the potential at the cathode, as measured with respect to a carbon reference electrode, is more than 1.1 V, and preferably more than 1.3 V.

47. A method for removing a substance from a solid compound comprising the substance and a metal, comprising the steps of:

contacting an anode with the electrolyte; and,

applying a voltage between the cathode and the anode such that the potential at the cathode increases with time.

48. A method according to claim 47, in which the potential at the cathode increases continuously with time.

49. A method according to claim 47 or 48, in which the potential at the cathode increases stepwise with time.

50. A method for removing a substance from a solid metal compound substantially as described herein, with reference to the drawings.

51. A method for forming an alloy comprising two or more metal components, comprising the steps of forming a precursor material containing mixed solid compounds of each of the metal components with a substance or substances, and processing the precursor material to remove the substance or substances as defined in any preceding claim.

52. A metal or alloy formed using the method of any preceding claim.

53. An apparatus for carrying out the method of any preceding claim.

54. An apparatus for removing a substance from a solid compound comprising the substance and a metal, comprising;

a cathode assembly for holding the solid compound;

an anode;

a reference electrode of carbon;

a receptacle for holding a fused-salt electrolyte in contact with the cathode assembly, the anode and the reference electrode; and

a voltage source for applying a voltage between the cathode and the anode; and a means for monitoring the voltage between the cathode and the reference electrode and controlling the voltage source in response to the monitored voltage.