Patent Application
20110100834
This invention relates to the use in a molten electrolyte for the electrowinning of aluminium of a non-carbon anode having a flow-through active structure with an enhanced stability.
Using non-carbon anodes—i.e. anodes which are not made of carbon as such, e.g. graphite, coke, etc. . . . , but possibly contain carbon in a compound—for the production of aluminium in electrolytic cells should drastically improve the aluminium production process by reducing pollution and the cost of aluminium production.
The developments of non-carbon anode materials, in particular metals, led to the design of new anode shapes that are better adapted to the cell's fluid mechanisms and electromagnetic effects than the conventional anodic solid carbon blocks.
Several designs for oxygen-evolving anodes for aluminium electrowinning cells were proposed in the following documents. U.S. Pat. No. 4,681,671 (Duruz) discloses vertical anode plates or blades operated in low temperature aluminium electrowinning cells. U.S. Pat. No. 5,310,476 (Sekhar/de Nora) discloses oxygen-evolving anodes consisting of roof-like assembled pairs of anode plates. U.S. Pat. No. 5,362,366 (de Nora/Sekhar) describes non-consumable anode shapes including roof-like assembled pairs of anode plates. U.S. Pat. No. 5,368,702 (de Nora) discloses vertical tubular or frustoconical oxygen-evolving anodes for multimonopolar aluminium cells. U.S. Pat. No. 5,683,559 (de Nora) describes an aluminium electrowinning cell with oxygen-evolving bent anode plates which are aligned in a roof-like configuration facing correspondingly shaped cathodes. U.S. Pat. No. 5,725,744 (de Nora/Duruz) discloses vertical oxygen-evolving anode plates, preferably porous or reticulated, in a multimonopolar cell arrangement for aluminium electrowinning cells operating at reduced temperature. WO00/40781, WO00/40782 and WO03/006716 (all de Nora) both disclose aluminium production anodes with a series of parallel spaced-apart elongated anode members which are electrochemically active for the oxidation of oxygen.
For the dissolution of the raw material alumina, a highly aggressive fluoride-based electrolyte, such as cryolite, is required. Various modified electrolytes have been proposed to improve cell operation and reduce wear of non-carbon metal-based anode, particularly caused by corrosion by the electrolyte.
WO00/06804 (Crottaz/Duruz) teaches that a nickel-iron anode can be used in an electrolyte at a temperature of 820° to 870° C. containing 23 to 26.5 weight % AlF3, 3 to 5 weight % Al2O3, 1 to 2 weight % LiF and 1 to 2 weight % MgF2. U.S. Pat. Nos. 5,006,209 and 5,284,562 (both Beck/Brooks), U.S. Pat. Nos. 6,258,247 and 6,379,512 (both Brown/Brooks/Frizzle/Juric), U.S. Pat. No. 6,419,813 (Brown/Brooks/Frizzle) and U.S. Pat. No. 6,436,272 (Brown/Frizzle) all disclose the use of nickel-copper-iron anodes in an aluminium production electrolyte at 660°-800° C. containing 6-26 weigh % NaF, 7-33 weight % KF, 1-6 weight % LiF and 60-65 weight % AlF3. The electrolyte may contain Al2O3 in an amount of up to 30 weight %, in particular 5 to 10 or 15 weight %, most of which is in the form of suspended particles and some of which is dissolved in the electrolyte, i.e. typically 1 to 4 weight % dissolved Al2O3. In U.S. Pat. Nos. 6,258,247, 6,379,512, 6,419,813 and 6,436,272 such an electrolyte is said to be useable at temperatures up to 900° C. In U.S. Pat. Nos. 6,258,247 and 6,379,512 the electrolyte further contains 0.004 to 0.2 weight % transition metal additives to facilitate alumina dissolution and improve cathodic operation. U.S. Pat. No. 5,725,744 (de Nora/Duruz) discloses an aluminium production cell having anodes made of nickel, iron and/or copper in a electrolyte at a temperature from 680° to 880° C. containing 42-63 weight % AlF3, up to 48 weight % NaF, up to 48 weight % LiF and 1 to 5 weight % Al2O3. MgF2, KF and CaF2 are also mentioned as possible bath constituents. WO2004/035871 (de Nora/Nguyen/Duruz) discloses a metal-based anode containing at least one of nickel, cobalt and iron. The anode is used for electrowinning aluminium in a fluoride-containing molten electrolyte consisting of: 5 to 14 wt % dissolved alumina; 35 to 45 wt % aluminium fluoride; 30 to 45 wt % sodium fluoride; 5 to 20 wt % potassium fluoride; 0 to 5 wt % calcium fluoride; and 0 to 5 wt % of further constituents.
The materials having the greatest resistance to oxidation are metal oxides which are all to some extent soluble in cryolite. Oxides are also poorly electrically conductive, therefore, to avoid substantial ohmic losses and high cell voltages, the use of non-conductive or poorly conductive oxides should be minimal in the manufacture of anodes. Whenever possible, a good conductive material should be utilised for the anode core, whereas the surface of the anode is preferably made of an oxide having a high electrocatalytic activity. Several attempts have been made in order to develop non-carbon anodes for aluminium electrowinning cells, resistant to chemical attacks of the bath and by the cell environment, and with an electrochemical active surface for the oxidation of oxygen ions to atomic and molecular gaseous oxygen and having a low dissolution rate. Many patents have been filed on non-carbon anodes but none has found commercial acceptance yet, also because of economical reasons.
U.S. Pat. No. 4,614,569 (Duruz/Derivaz/Debely/Adorian) describes metal anodes for aluminium electrowinning coated with a protective coating of cerium oxyfluoride, formed in-situ in the cell or pre-applied, this coating being maintained during electrolysis by the addition of small amounts of a cerium compound to the molten cryolite electrolyte so as to protect the surface of the anode from the electrolyte attack. Several patents disclose the use of an electrically conductive metal anode core with an oxide-based active outer part, in particular U.S. Pat. Nos. 4,956,069, 4,960,494, 5,069,771 (all Nguyen/Lazouni/Doan), U.S. Pat. No. 6,077,415 (Duruz/de Nora), U.S. Pat. No. 6,103,090 (de Nora), U.S. Pat. No. 6,113,758 (de Nora/Duruz) and U.S. Pat. No. 6,248,227 (de Nora/Duruz), U.S. Pat. No. 6,361,681 (de Nora/Duruz), U.S. Pat. No. 6,365,018 (de Nora), U.S. Pat. No. 6,372,099 (Duruz/de Nora), U.S. Pat. No. 6,379,526 (Duruz/de Nora), U.S. Pat. No. 6,413,406 (de Nora), U.S. Pat. No. 6,425,992 (de Nora), U.S. Pat. No. 6,436,274 (de Nora/Duruz), U.S. Pat. No. 6,521,116 (Duruz/de Nora/Crottaz), U.S. Pat. No. 6,521,115 (Duruz/de Nora/Crottaz), U.S. Pat. No. 6,533,909 (Duruz/de Nora), U.S. Pat. No. 6,562,224 (Crottaz/Duruz) as well as PCT publications WO00/40783 (de Nora/Duruz), WO01/42534 (de Nora/Duruz), WO01/42535 (Duruz/de Nora), WO01/42536 (Nguyen/Duruz/ de Nora), WO02/070786 (Nguyen/de Nora), WO02/083990 (de Nora/Nguyen), WO02/083991 (Nguyen/de Nora), WO03/014420 (Nguyen/Duruz/de Nora), WO03/078695(Nguyen/de Nora), WO03/087435 (Nguyen/de Nora).
U.S. Pat. No. 4,374,050 (Ray) discloses numerous multiple oxide compositions for electrodes. Such compositions inter-alia include oxides of iron and cobalt. The oxide compositions can be used as a cladding on a metal layer of nickel, nickel-chromium, steel, copper, cobalt or molybdenum. U.S. Pat. No. 4,142,005 (Cadwell/Hazelrigg) discloses an anode having a substrate made of titanium, tantalum, tungsten, zirconium, molybdenum, niobium, hafnium or vanadium. The substrate is coated with cobalt oxide Co3O4.
U.S. Pat. No. 6,103,090 (de Nora), U.S. Pat. No. 6,361,681 (de Nora/Duruz), U.S. Pat. No. 6,365,018 (de Nora), U.S. Pat. No. 6,379,526 (de Nora/Duruz), U.S. Pat. No. 6,413,406 (de Nora) and U.S. Pat. No. 6,425,992 (de Nora), and WO04/018731 (Nguyen/de Nora) disclose anode substrates that contain at least one of chromium, cobalt, hafnium, iron, molybdenum, nickel, copper, niobium, platinum, silicon, tantalum, titanium, tungsten, vanadium, yttrium and zirconium and that are coated with at least one of ferrites of cobalt, copper, chromium, manganese, nickel and zinc. WO01/42535 (Duruz/de Nora) and WO02/097167 (Nguyen/de Nora), disclose aluminium electrowinning anodes made of surface oxidised iron alloys that contain at least one of nickel and cobalt. U.S. Pat. No. 6,638,412 (de Nora/Duruz) discloses the use of anodes made of a transition metal-containing alloy having an integral oxide layer, the alloy comprising at least one of iron, nickel and cobalt.
Non-carbon anodes have not as yet been commercially and industrially applied and there is still a need for a metal-based anodic material and an appropriate anode shape that can be used for electrowinning aluminium.
The present invention generally relates to aluminium electrowinning with metal-based anodes having a shape for promoting an electrolyte circulation and having an electrochemically active outer part that has an enhanced stability against corrosion by the highly aggressive circulating electrolyte and/or against oxidation by anodically evolved oxygen, the enhanced stability being provided by a layer that contains predominantly cobalt oxide CoO.
In particular, the invention relates to a cell for the electrowinning of aluminium from alumina dissolved in a fluoride-containing molten electrolyte. The cell comprises at least one non-carbon metal-based anode having an electrically conductive metallic structure that comprises an outer part with an electrochemically active anode surface on which, during electrolysis, oxygen is anodically evolved, and which is suspended in the electrolyte substantially parallel to a facing cathode. This metallic structure has one or more flow-through openings extending from the active anode surface through the metallic structure, the flow-through opening(s) being arranged for guiding a circulation of electrolyte driven by the fast escape of anodically evolved oxygen. The outer part of the anode comprises the abovementioned layer that contains predominantly cobalt oxide CoO to enhance the stability of the anode.
In other words, the invention concerns a cell having an anode that has a shape that promotes electrolyte circulation and that has an electrochemically active outer part that is resistant to the circulating electrolyte and/or to anodically evolved oxygen by the presence of a layer made predominantly of a special form of cobalt oxide, i.e. CoO.
There are several forms of stoichiometric and non-stoichiometric cobalt oxides which are based on:
It has been observed that—unlike Co2O3 that is unstable and Co3O4 that does not significantly inhibit oxygen diffusion—CoO forms a well conductive electrochemically active material for the oxidation of oxygen ions and for inhibiting diffusion of oxygen. Thus this material forms a limited barrier against oxidation of the metallic cobalt body underneath.
The anode's CoO-containing layer can be a layer made of sintered particles, especially sintered CoO particles. Alternatively, the CoO-containing layer may be an integral oxide layer on a Co-containing metallic layer or anode core. Tests have shown that integral oxide layers have a higher density than sintered layers and are thus preferred to inhibit oxygen diffusion.
When CoO is to be formed by oxidising metallic cobalt, care should be taken to carry out a treatment that will indeed result in the formation of CoO. It was found that using Co2O3 or Co3O4 in a known aluminium electrowinning electrolyte does not lead to an appropriate conversion of these forms of cobalt oxide into CoO. Therefore, it is important to provide an anode with the CoO layer before the anode is used in an aluminium electrowinning electrolyte.
The formation of CoO on the metallic cobalt is preferably controlled so as to produce a coherent and substantially crack-free oxide layer. However, not any treatment of metallic cobalt at a temperature above 895° C. or 900° C. in an oxygen-containing atmosphere will result in optimal coherent and substantially crack-free CoO layer that offers better electrochemical properties than a CO2O3/Co3O4.
For instance, if the temperature for treating the metallic cobalt to form Coo by air oxidation of metallic cobalt is increased at an insufficient rate, e.g. less than 200° C./hour, a thick oxide layer rich in Co3O4 and in glassy Co2O3 is formed at the surface of the metallic cobalt. Such a layer does not permit optimal formation of the CoO layer by conversion at a temperature above 895° C. of Co2O3 and Co3O4 into CoO. In fact, a layer of CoO resulting from such conversion is not preferred but still useful despite an increased porosity and may be cracked. Therefore, the required temperature for air oxidation, i.e. above 900° C., usually at least 920° C. or preferably above 940° C. should be attained sufficiently quickly, e.g. at a rate of increase of the temperature of at least 300° C. or 600° C. per hour to obtain an optimal CoO layer. The metallic cobalt may also be placed into an oven that is pre-heated at the desired temperature above 900° C.
Likewise, if the anode is not immediately used for the electrowinning of aluminium after formation of the CoO layer but allowed to cool down, the cooling down should be carried out sufficiently fast, for example by placing the anode in air at room temperature, to avoid significant formation of Co3O4 that could occur during the cooling, for instance in an oven that is switched off.
An anode with a CoO layer obtained by slow heating of the metallic cobalt in an oxidising environment will not have optimal properties but still provides better results during cell operation than an anode having a Co2O3—Co3O4 layer and therefore also constitutes an improved aluminium electrowinning anode according to the invention.
The anode structure can be foraminate. For instance, the anode structure can have any of the shapes disclosed in the abovementioned WO00/40781, WO00/40782 and WO03/006716. For example, the anode structure comprises a series of parallel anode members, in particular horizontal anode members having electrochemically active surfaces in a generally coplanar arrangement to form said active anode surface, the anode members being spaced apart to form longitudinal flow-through openings for the circulation of electrolyte driven by the fast escape of anodically evolved oxygen. Typically these anode members are blades, bars, rods or wires. The active anode surface can be substantially horizontal, substantially vertical or inclined to the horizontal, for example as disclosed in WO00/40782 or WO03/023092 (both de Nora).
In one embodiment, the molten electrolyte is at a temperature below 950° C., in particular in the range from 910° to 940° C., and consists of:
The presence in the electrolyte of potassium fluoride in the above amount has two effects. On the one hand, it leads to a reduction of the operating temperature by up to several tens of degrees without increase of the electrolyte's aluminium fluoride content or even a reduction thereof compared to standard electrolytes operating at about 950° C. with an aluminium fluoride content of about 45 weight %. On the other hand, it maintains a high solubility of alumina, i.e. up to above about 8 or 9 weight %, in the electrolyte even though the temperature of the electrolyte is reduced compared to conventional temperature.
Hence, in contrast to prior art low temperature electrolytes which carry large amounts of undissolved alumina in particulate form, a large amount of alumina is in a dissolved form in the above electrolyte.
Without being bound to any theory, it is believed that combining a high concentration of dissolved alumina in the electrolyte and a limited concentration of aluminium fluoride leads predominantly to the formation of (basic) fluorine-poor aluminium oxyfluoride ions ([Al2O2F4]2−) instead of (acid) fluorine-rich aluminium oxyfluoride ions ([Al2OF6]2−) near the anode. As opposed to acid fluorine-rich aluminium oxyfluoride ions, basic fluorine-poor aluminium oxyfluoride ions do not significantly dissolve the anode's CoO and do not noticeably passivate or corrode metallic cobalt. The weight ratio of dissolved alumina/aluminium fluoride in the electrolyte should be above 1/7, and often above 1/6 or even above 1/5, to obtain a favourable ratio of the fluorine-poor aluminium oxyfluoride ions and the fluorine-rich aluminium oxyfluoride ions.
It follows that the use of the above described electrolyte with metal-based anodes that contains CoO inhibits its dissolution, passivation and corrosion. Moreover, a high concentration of alumina dissolved in the electrolyte further reduces dissolution of oxides of the anode, in particular CoO.
The electrolyte may consist of: 7 to 10 weight % dissolved alumina; 36 to 42 weight % aluminium fluoride, in particular 36 to 38 weight %; 39 to 43 weight % sodium fluoride; 3 to 10 weight % potassium fluoride, such as 5 to 7 weight %; 2 to 4 weight % calcium fluoride; and 0 to 3 weight % in total of one or more further constituents. This corresponds to a cryolite-based (Na3AlF6) molten electrolyte containing an excess of aluminium fluoride (AlF3) that is in the range of about 8 to 15 weight % of the electrolyte, in particular about 8 to 10 weight %, and additives that can include potassium fluoride and calcium fluoride in the abovementioned amounts.
The electrolyte can contain as further constituent(s) at least one fluoride selected from magnesium fluoride, lithium fluoride, cesium fluoride, rubidium fluoride, strontium fluoride, barium fluoride and cerium fluoride.
Advantageously, The electrolyte contains alumina at a concentration near saturation on the active anode surface.
In order to maintain the alumina concentration above a given threshold in the abovementioned range during normal electrolysis, the cell is preferably fitted with means to monitor and adjust the electrolyte's alumina content.
The CoO-containing anode layer can be integral with a core made of cobalt or a cobalt alloy. Such an anode core can be made of the same materials as the Co-containing alloys described below. The cobalt-containing anode core can advantageously be cast.
Alternatively, the anode comprises an electrically conductive substrate that is covered with an applied electrochemically active coating that comprises the CoO-containing layer.
The CoO-containing layer can be a layer of sintered particles. In particular, the CoO-containing layer can be formed by applying a layer of particulate CoO to the anode and sintering. For instance, the CoO-containing layer is applied as a slurry, in particular a colloidal and/or polymeric slurry, and then heat treated. Good results have been obtained by slurring particulate metallic cobalt or CoO, optionally with additives such as Ta, in an aqueous solution containing at least one of ethylene glycol, hexanol, polyvinyl alcohol, polyvinyl acetate, polyacrylic acid, hydroxy propyl methyl cellulose and ammonium polymethacrylate and mixtures thereof, followed by application to the anode, e.g. painting or dipping, and heat treating.
The CoO-containing layer can be an integral oxide layer on an applied Co-containing metallic layer of the coating.
The CoO-containing layer can be formed by applying a Co-containing metallic layer to the anode and subjecting the metallic layer to an oxidation treatment to form the CoO-containing layer on the metallic layer, the CoO-containing layer being integral with the metallic layer.
Conveniently, the oxidation treatment can be carried out in an oxygen containing atmosphere, such as air. The treatment can also be carried out in an atmosphere that is oxygen rich or consists essentially of pure oxygen.
It is also contemplated to carry out this oxidation treatment by other means, for instance electrolytically. However, it was found that full formation of the CoO integral layer cannot be achieved in-situ during aluminium electrowinning under normal cell operating conditions. In other words, when the anode is intended for use in a non-carbon anode aluminium electrowinning cell operating under the usual conditions, the anode should always be placed into the cell with a preformed integral oxide layer containing predominantly CoO.
As the conversion of Co(III) into Co(II) occurs at a temperature of about 895° C., the oxidation treatment should be carried out above this temperature. Usually, the oxidation treatment is carried out at a treatment temperature above 895° C. or 920° C., preferably above 940° C., in particular within the range of 950° C. to 1050° C. The Co-containing metallic layer can be heated from room temperature to this treatment temperature at a rate of at least 300° C./hour, in particular at least 450° C./hour, or is placed in an environment, in particular in an oven, that is preheated to said temperature. The oxidation treatment at this treatment temperature can be carried out for more than 8 or 12 hours, in particular from 16 to 48 hours. Especially when the oxygen-content of the oxidising atmosphere is increased, the duration of the treatment can be reduced below 8 hours, for example down to 4 hours.
The Co-containing metallic layer can be further oxidised during use. However, the main formation of CoO is preferably achieved before use and in a controlled manner for the reasons explained above.
The method for forming the CoO-containing layer on the Co-containing metallic layer can be used to form the CoO-containing layer on the previously mentioned Co-containing anode core.
The Co-containing metallic layer can contain alloying metals for further reducing oxygen diffusion and/or corrosion through the metallic layer.
In one embodiment, the anode comprises an oxygen barrier layer between the CoO-containing layer and the electrically conductive substrate. The oxygen barrier layer can contain at least one metal selected from nickel, copper, tungsten, molybdenum, tantalum, niobium and chromium, or an oxide thereof, for example alloyed with cobalt, such as a cobalt alloy containing tungsten, molybdenum, tantalum and/or niobium, in particular an alloy containing: at least one of nickel, tungsten, molybdenum, tantalum and niobium in a total amount of 5 to 30 wt %, such as 10 to 20 wt %; and one or more further elements and compounds in a total amount of up to 5 wt % such as 0.01 to 4 weight %, the balance being cobalt. These further elements may contain at least one of aluminium, silicon and manganese.
Typically, the oxygen barrier layer and the CoO-containing layer are formed by oxidising the surface of an applied layer of the abovementioned cobalt alloy that contains nickel, tungsten, molybdenum, tantalum and/or niobium. The resulting CoO-containing layer is predominantly made of CoO and is integral with the unoxidised part of the metallic cobalt alloy that forms the oxygen barrier layer.
When the CoO layer is integral with the cobalt alloy, the nickel, when present, should be contained in the alloy in an amount of up to 20 weight %, in particular 5 to 15 weight %. Such an amount of nickel in the alloy leads to the formation of a small amount of nickel oxide NiO in the integral oxide layer, in about the same proportions to cobalt as in the metallic part, i.e. 5 to 15 or 20 weight %. It has been observed that the presence of a small amount of nickel oxide stabilises the cobalt oxide CoO and durably inhibits the formation of Co2O3 or Co3O4. However, when the weight ratio nickel/cobalt exceeds 0.15 or 0.2, the advantageous chemical and electrochemical properties of cobalt oxide CoO tend to disappear. Therefore, the nickel content should not exceed this limit.
Alternatively, an oxygen barrier layer, for example made of the above cobalt alloy that contains nickel, tungsten, molybdenum, tantalum and/or niobium, can be covered with an applied layer of CoO or a precursor thereof, as discussed above. In this case the oxygen barrier layer can be an applied layer or it can be integral with the electrically conductive substrate.
In another embodiment, the Co-containing metallic layer consists essentially of cobalt, typically containing cobalt in an amount of at least 95 wt %, in particular more than 97 wt % or 99 wt %.
Optionally the Co-containing metallic layer contains at least one additive selected from silicon, nickel, manganese, niobium, tantalum and aluminium in a total amount of 0.1 to 2 wt %.
Such a Co-containing layer can be applied to an oxygen barrier layer which is integral with the electrically conductive substrate of the flow-through anode structure or applied thereto.
The electrically conductive substrate can comprise at least one metal selected from chromium, cobalt, hafnium, iron, molybdenum, nickel, copper, platinum, silicon, titanium, tungsten, molybdenum, tantalum, niobium, vanadium, yttrium and zirconium, or a compound thereof, in particular an oxide, or a combination thereof. For instance, the electrically conductive substrate may have an outer part made of cobalt or an alloy containing predominantly cobalt to which the coating is applied. For instance, this cobalt alloy contains nickel, tungsten, molybdenum, tantalum and/or niobium, in particular it contains: nickel tungsten, molybdenum, tantalum and/or niobium in a total amount of 5 to 30 wt %, e.g. 10 to 20 wt %; and one or more further elements and/or compounds in a total amount of up to 5 wt %, the balance being cobalt. These further elements may contain at least one of aluminium, silicon and manganese. The electrically conductive substrate, or an outer part thereof, may contain or consist essentially of at least one oxidation-resistant metal, in particular one or more metals selected from nickel, tungsten, molybdenum, cobalt, chromium and niobium, and for example contains less than 1, 5 or 10 wt % in total of other metals and metal compounds, in particular oxides. Alternatively, the electrically conductive substrate can be made of an alloy of nickel, iron and copper, in particular an alloy containing: 65 to 85 weight % nickel; 5 to 25 weight % iron; 1 to 20 weight % copper; and 0 to 10 weight % further constituents. For example, the alloy contains about: 75 weight % nickel; 15 weight % iron; and 10 weight % copper.
Advantageously, the anode's CoO-containing layer, in particular when the CoO layer is integral with the applied Co-containing metallic layer or the anode body, has an open porosity of below 12%, such as below 7%.
The anode's CoO-containing layer can have a porosity with an average pore size below 7 micron, in particular below 4 micron. It is preferred to provide a substantially crack-free CoO-containing layer so as to protect efficiently the anode's metallic outer part which is covered by this CoO-containing layer.
Usually, the CoO-containing layer contains cobalt oxide CoO in an amount of at least 80 wt %, in particular more than 90 wt % or 95 wt % or 98 wt %.
Advantageously, the CoO-containing layer is substantially free of cobalt oxide Co2O3 and substantially free of Co3O4, and contains preferably below 3 or 1.5% of these forms of cobalt oxide.
The CoO-containing layer may be electrochemically active for the oxidation of oxygen ions during use, in which case this layer is uncovered or is covered with an electrolyte-pervious layer.
Alternatively, the CoO-containing layer can be covered with an applied protective layer, in particular an applied oxide layer such as a layer containing cobalt and/or iron oxide, e.g. cobalt ferrite. The applied protective layer may contain a pre-formed and/or in-situ deposited cerium compound, in particular cerium oxyfluoride, as for example disclosed in the abovementioned U.S. Pat. Nos. 4,956,069, 4,960,494 and 5,069,771. Such an applied protective layer is usually electrochemically active for the oxidation of oxygen ions and is uncovered, or covered in turn with an electrolyte pervious-layer.
The anode's electrochemically active surface can contain at least one dopant, in particular at least one dopant selected from iridium, palladium, platinum, rhodium, ruthenium, silicon, tungsten, molybdenum, tantalum, niobium, tin or zinc metals, Mischmetal and metals of the Lanthanide series, as metals and compounds, in particular oxides, and mixtures thereof. The dopant(s) can be present at the anode's surface in a total amount of 0.1 to 5 wt %, in particular 1 to 4 wt %.
Such a dopant can be an electrocatalyst for fostering the oxidation of oxygen ions on the anode's electrochemically active surface and/or can contribute to inhibit diffusion of oxygen ions into the anode.
The dopant may be added to the precursor material that is applied to form the active surface or it can be applied to the active surface as a thin film, for example by plasma spraying or slurry application, and incorporated into the surface by heat treatment.
The cell can have a cathode that has an aluminium-wettable surface, in particular a horizontal or inclined drained surface. This surface can be formed by an aluminium-wettable material that comprises a refractory boride and/or an aluminium-wetting oxide. Examples of such materials are disclosed in WO01/42168, WO01/42531, WO02/070783, WO02/096830 and WO02/096831 (all in the name of MOLTECH).
The anode can be suspended in the electrolyte by a stem, in particular a stem having an outer part comprising a layer that contains predominantly cobalt oxide CoO.
Another aspect of the invention relates to a method of electrowinning aluminium in a cell as described above The method comprises electrolysing the dissolved alumina to produce oxygen on the anode and aluminium cathodically, and supplying alumina to the electrolyte to maintain therein a concentration of dissolved alumina of 6.5 to 11 weight %, in particular 7 to 10 weight %.
Oxygen ions may be oxidised on the anode's CoO-containing layer that contains predominantly cobalt oxide CoO and/or, when present, on an active layer applied to the anode's CoO layer, the CoO layer inhibiting oxidation and/or corrosion of the anode's metallic outer part.
The invention also relates to a non-carbon metal-based anode for the electrowinning of aluminium from alumina dissolved in a fluoride-containing molten electrolyte. This anode comprises an electrically conductive metallic structure that comprises an outer part with an electrochemically active anode surface on which oxygen is anodically evolved and which is suspended in the electrolyte substantially parallel to a facing cathode during use. This metallic structure has one or more flow-through openings extending from the active anode surface through the metallic structure. These flow-through opening(s) are arranged for guiding during use a circulation of electrolyte driven by the fast escape of anodically evolved oxygen. The outer part of the anode comprises the abovementioned layer that contains predominantly cobalt oxide CoO to enhance the stability of the anode. This anode can include any of the above described anode features or a combination thereof.
The invention will now be described with reference to the schematic drawings, wherein:
a and 1b show respectively a side elevation and a plan view of an anode according to the invention;
a and 2b show respectively a side elevation and a plan view of another anode according to the invention;
a and 1b schematically show an anode 10 of a cell for the electrowinning of aluminium according to the invention.
The anode 10 comprises a vertical current feeder 11 for connecting the anode to a positive bus bar, a cross member 12 and a pair of transverse connecting members 13 for connecting a series of anode members 15.
The anode members 15 have an electrochemically active lower surface 16 where oxygen is anodically evolved during cell operation. The anode members 15 are in the form of parallel rods in a coplanar arrangement, laterally spaced apart from one another by inter-member gaps 17. The inter-member gaps 17 constitute flow-through openings for the circulation of electrolyte and the escape of anodically-evolved gas released at the electrochemically active surfaces 16.
The anode members 15 are transversally connected by the pair of transverse connecting members 13 which are in turn connected together by the cross member 12 on which the vertical current feeder 11 is mounted. The current feeder 11, the cross member 12, the transverse connecting members 13 and the anode members 15 are mechanically secured together by welding, rivets or other means.
In accordance with the invention, the electrochemically active surface 16 of the anode members is formed by an outer part that comprises a layer containing predominantly CoO. This CoO layer can form the electrochemically active surface 16 and be directly exposed to the electrolyte during use or the CoO layer can be covered with a further layer, for instance a layer containing predominantly a cerium compound such as cerium oxyfluoride.
The cross-member 12 and the transverse connecting members 13 are so designed and positioned over the anode members 15 to provide a substantially even current distribution through the anode members 15 to their electrochemically active surfaces 16. The current feeder 11, the cross-member 12 and the transverse connecting members 13 do not need to be electrochemically active and their surface may passivate when exposed to electrolyte. However they should be electrically well conductive to avoid unnecessary voltage drops and should not substantially dissolve in electrolyte. The electrochemically-inactive current-carrying elements (11,12,13) can have an outer part with a protective layer containing predominantly CoO.
a and 2b schematically show a variation of the anode 10 shown in
Instead of having transverse connecting members 13, a cross-member 12 and a current feeder 11 for mechanically and electrically connecting the anode members 15 to a positive bus bar as illustrated in
The anode members 15 may be secured by force-fitting or welding in the horizontal foot 14a. As an alternative, the shape of the anode members 15 and corresponding receiving slots in the foot 14a may be such as to allow only longitudinal movements of the anode members. For instance the anode members 15 and the foot 14a may be connected by dovetail joints.
The anode members 15 of the anode shown in
The anode members 15 shown in
The intermediate connecting member 15c shown in
In accordance with the invention, the electrochemically active member 15a shown in
The support member 15b shown in
Pairs of anodes 10 are connected to a positive bus bar through a primary vertical current feeder 11′ and a horizontal current distributor 11″ connected at both of its ends to a foraminate anode 10 through a secondary vertical current distributor 11′″.
The secondary vertical current distributor 11′″ is mounted on the anode structure 12,13,15, on a cross member 12 which is in turn connected to a pair of transverse connecting members 13 for connecting a series of anode members 15. The current feeders 11′,11″,11′″, the cross member 12, the transverse connecting members 13 and the anode members 15 are mechanically secured together by welding, rivets or other means.
The anode members 15 have an electrochemically active lower surface 16 on which during cell operation oxygen is anodically evolved. The anode members 15 are in the form of parallel rods in a foraminate coplanar arrangement, laterally spaced apart from one another by inter-member gaps 17. The inter-member gaps 17 constitute flow-through openings for the circulation of electrolyte and the escape of anodically-evolved gas from the electrochemically active surfaces 16.
The cross-member 12 and the transverse connecting members 13 provide a substantially even current distribution through the anode members 15 to their electrochemically active surfaces 16. The current feeder 11, the cross-member 12 and the transverse connecting members 13 do not need to be electrochemically active and their surface may passivate when exposed to electrolyte. However they should be electrically well conductive to avoid unnecessary voltage drops and should not substantially dissolve in the molten electrolyte.
In accordance with the invention, the active surface 16 of the anode members 15 can be CoO-based.
The CoO-based surface may extend over all immersed parts 11′″,12,13,15 of the anode 10, in particular over the immersed part of the secondary vertical current distributor 11′″ which is preferably covered with a CoO-based layer at least up to 10 cm above the surface of the electrolyte 30.
The anodes 10 are further fitted with means for enhancing dissolution of fed alumina in the form of electrolyte guide members 5 formed of parallel spaced-apart inclined baffles 5 located above and adjacent to the foraminate anode structure 12,13,15. The baffles 5 provide upper downwardly converging surfaces 6 and lower upwardly converging surfaces 7 that deflect gaseous oxygen which is anodically produced below the electrochemically active surface 16 of the anode members 15 and which escapes between the inter-member gaps 17 through the foraminate anode structure 12,13,15. The oxygen released above the baffles 5 promotes dissolution of alumina fed into the electrolyte 30 above the downwardly converging surfaces 6. Further details of such baffles are disclosed in the abovementioned WO00/40781 and WO00/40782.
The aluminium-wettable cathodic coating 22 of the cell shown in
During cell operation, alumina is fed to the electrolyte 30 all over the baffles 5 and the metallic anode structure 12,13,15. The fed alumina is dissolved and distributed from the bottom end of the converging surfaces 6 into the inter-electrode gap through the inter-member gaps 17 and around edges of the metallic anode structure 12,13,15, i.e. between neighbouring pairs of anodes 10 or between peripheral anodes 10 and sidewalls 25. By passing an electric current between anodes 10 and facing cathode cell bottom 20 oxygen is evolved on the electrochemically active anode surfaces 16 and aluminium is produced which is incorporated into the cathodic molten aluminium 35. The oxygen evolved from the active surfaces 16 escapes through the inter-member gaps 17 and is deflected by the upwardly converging surfaces 7 of baffles 5. The oxygen escapes from the uppermost ends of the upwardly converging surfaces 7 enhancing dissolution of the alumina fed over the downwardly converging surfaces 6.
The aluminium electrowinning cells partly shown in
In
Also shown in
By guiding and confining anodically-evolved oxygen towards the surface of the electrolyte 30 with baffles or other confinement means as shown in
It is understood that the electrolyte confinement members 5 shown in
The anode 10′ shown in
As shown in
Each electrolyte guide member 5′ is in the general shape of a funnel having a wide bottom opening 9 for receiving anodically produced oxygen and a narrow top opening 8 where the oxygen is released to promote dissolution of alumina fed above the electrolyte guide member 5′. The inner surface 7 of the electrolyte guide member 5′ is arranged to canalise and promote an upward electrolyte flow driven by anodically produced oxygen. The outer surface 6 of the electrolyte guide member 5′ is arranged to promote dissolution of alumina fed thereabove and guide alumina-rich electrolyte down to the inter-electrode gap, the electrolyte flowing mainly around the foraminate structure.
As shown in
The arrangement of the electrolyte guide members 5′ and the anode 10′ can be moulded as units. This offers the advantage of avoiding mechanical joints and the risk of altering the properties of the materials of the electrolyte guide members 5′ or the anode 10′ by welding.
The anodes 10′ and electrolyte guide members 5′ can be made of the same materials, in particular they can be made of a metallic body having an outer part with a layer containing predominantly CoO.
The anodes 10′ of
Useful variations of this anode structure are disclosed in the abovementioned WO00/40782. Further suitable anode designs are disclosed WO99/027064 (de Nora/Duruz), WO01/31088, WO03/006716 and WO03/023092 and U.S. Pat. No. 5,368,702 (all de Nora)
The manufacturing and behaviour in an aluminium electrowinning cell of the cobalt-oxide containing material used for the anode of the present invention will be further described in the following examples:
A cylindrical metallic cobalt sample was oxidised to form an integral cobalt oxide layer that did not predominantly contain CoO. The cobalt samples contained no more than a total of 1 wt % additives and impurities and had a diameter of 1.94 cm and a height of 3 cm.
Oxidation was carried out by placing the cobalt sample into an oven in air and increasing the temperature from room temperature to 850° C. at a rate of 120° C./hour.
After 24 hours at 850° C., the oxidised cobalt sample was allowed to cool down to room temperature and examined.
The cobalt sample was covered with a greyish oxide scale having a thickness of about 300 micron. This oxide scale was made of: a 80 micron thick inner layer that had a porosity of 5% with pores that had a size of 2-5 micron; and a 220 micron thick outer layer having an open porosity of 20% with pores that had a size of 10-20 micron. The outer oxide layer was made of a mixture of essentially Co2O3 and Co3O4. The denser inner oxide layer was made of CoO.
As shown in Comparative Examples 2 and 3, such oxidised cobalt provides poor results when used as an anode material in an aluminium electrowinning cell.
A cobalt sample which can be used to manufacture for an anode according to the invention was prepared as in Comparative Example 1 except that the sample was oxidised in an oven heated from room temperature to a temperature of 950° C. (instead of 850° C.) at the same rate (120° C./hour).
After 24 hours at 950° C., the oxidised cobalt sample was allowed to cool down to room temperature and examined.
The cobalt sample was covered with a black glassy oxide scale having a thickness of about 350 micron (instead of 300 micron). This oxide scale had a continuous structure (instead of a layered structure) with an open porosity of 10% (instead of 20%) and pores that had a size of 5 micron. The outer oxide layer was made of CoO produced above 895° C. from the conversion into CoO of Co3O4 and glassy Co2O3 formed below this temperature and by oxidising the metallic outer part of the sample (underneath the cobalt oxide) directly into CoO. The porosity was due to the change of phase during the conversion of Co2O3 and Co3O4 to CoO.
Such a material can be used for making an aluminium electrowinning anode according to the invention. However, the density of the CoO layer and the performances of this material can be further improved as shown in Examples 1c and 1d.
In general, to allow appropriate conversion of the cobalt oxide and growth of CoO from the metallic outer part of the substrate, it is important to leave the sample sufficiently long at a temperature above 895° C. The length of the heat treatment will depend on the oxygen content of the oxidising atmosphere, the temperature of the heat treatment, the desired amount of CoO and the amount of Co2O3 and Co3O4 to convert into CoO.
Example 1a was repeated with a similar cylindrical metallic cobalt sample. The oven in which the sample was oxidised was heated to a temperature of 1050° C. (instead of 950° C.) at the same rate (120° C./hour).
After 24 hours at 1050° C., the oxidised cobalt sample was allowed to cool down to room temperature and examined.
The cobalt sample was covered with a black crystallised oxide scale having a thickness of about 400 micron (instead of 350 micron). This oxide scale had a continuous structure with an open porosity of 20% (instead of 10%) and pores that had a size of 5 micron. The outer oxide layer was made of CoO produced above 895° C. like in Example 1a.
Such a oxidised cobalt is comparable to the oxidised cobalt of Example 1a and can likewise be used as an anode material to produce aluminium according to the present invention.
In general, to allow appropriate conversion of the cobalt oxide and growth of CoO from the metallic outer part of the substrate, it is important to leave the sample sufficiently long at a temperature above 895° C. The length of the heat treatment above 895° C. will depend on the oxygen content of the oxidising atmosphere, the temperature of the heat treatment, the desired amount of CoO and the amount of Co2O3 and Co3O4 (produced below 895° C.) which needs to be converted into CoO.
Example 1a was repeated with a similar cylindrical metallic cobalt sample. The oven in which the sample was oxidised was heated to the same temperature (950° C.) at a rate of 360° C./hour (instead of 120° C./hour).
After 24 hours at 950° C., the oxidised cobalt sample was allowed to cool down to room temperature and examined.
The cobalt sample was covered with a dark grey substantially non-glassy oxide scale having a thickness of about 350 micron. This oxide scale had a continuous structure with an open porosity of less than 5% (instead of 10%) and pores that had a size of 5 micron.
The outer oxide layer was made of CoO that was formed directly from metallic cobalt above 895° C. which was reached after about 2.5 hours and to a limited extent from the conversion of previously formed Co2O3 and Co3O4. It followed that there was less porosity caused by the conversion of Co2O3 and Co3O4 to CoO than in Example 1a.
Such an oxidised cobalt sample has a significantly higher density than the samples of Examples 1a and 1b, and is substantially crack-free. This oxidised cobalt constitutes a preferred material for making an improved aluminium electrowinning anode for use in a cell according to the invention.
Example 1c was repeated with a similar cylindrical metallic cobalt sample. The oven in which the sample was oxidised was heated to the same temperature (1050° C.) at a rate of 600° C./hour (instead of 120° C./hour in Example 1a and 1b and 360° C./hour in Example 1c).
After 18 hours at 1050° C., the oxidised cobalt sample was allowed to cool down to room temperature and examined.
The cobalt sample was covered with a dark grey substantially non-glassy oxide scale having a thickness of about 300 micron (instead of 400 micron in Example 1b and 350 micron in Example 1c). This oxide scale had a continuous structure with a crack-free open porosity of less than 5% (instead of 20% in Example 1b) and pores that had a size of less than 2 micron (instead of 5 micron in Example 1b and in Example 1c).
The outer oxide layer was made of CoO that was formed directly from metallic cobalt above 895° C. which was reached after about 1.5 hours and to a marginal extent from the conversion of previously formed Co2O3 and Co3O4. It followed that there was significantly less porosity caused by the conversion of Co2O3 and Co3O4 to CoO than in Example 1b and in Example 1c.
Such an oxidised cobalt sample has a significantly higher density than the samples of Examples 1a and 1b, and is substantially crack-free. This oxidised cobalt constitutes a preferred material for making an improved aluminium electrowinning anode according to the invention.
An anode made of metallic cobalt oxidised under the conditions of Comparative Example 1 was tested in an aluminium electrowinning cell.
The cell's electrolyte was at a temperature of 925° C. and made of 11 wt % AlF3, 4 wt % CaF2, 7 wt % KF and 9.6 wt % Al2O3, the balance being cryolite Na3AlF6.
The anode was placed in the cell's electrolyte at a distance of 4 cm from a facing cathode. An electrolysis current of 7.3 A was passed from the anode to the cathode at an anodic current density of 0.8 A/cm2.
The electrolysis current was varied between 4 and 10 A and the corresponding cell voltage measured to estimate the oxygen overpotential at the anode.
By extrapolating the cell's potential at a zero electrolysis current, it was found that the oxygen overpotential at the anode was of 0.88 V.
A test was carried out under the conditions of Comparative Example 2 with two anodes made of metallic cobalt oxidised under the conditions of Example 1c and 1d, respectively, in cells according to the invention using the same electrolyte as in Comparative Example 2. The estimated oxygen overpotential for these anodes were at 0.22 V and 0.21 V, respectively, i.e. about 75% lower than in Comparative Example 2.
It follows that the use of metallic cobalt covered with an integral layer of CoO instead of Co2O3 and Co3O4 as an aluminium electrowinning anode material in a cell according to the invention leads to a significant saving of energy.
Another anode made of metallic cobalt oxidised under the conditions of Comparative Example 1, i.e. resulting in a Co2O3 and Co3O4 integral surface layer, was tested in an aluminium electrowinning cell. The cell's electrolyte was at 925° C. and had the same composition as in Comparative Example 2. A nominal electrolysis current of 7.3 A was passed from the anode to the cathode at an anodic current density of 0.8 A/cm2.
The cell voltage at start-up was above 20 V and dropped to 5.6 V after about 30 seconds. During the initial 5 hours, the cell voltage fluctuated about 5.6 V between 4.8 and 6.4 V with short peaks above 8 V. After this initial period, the cell voltage stabilised at 4.0-4.2 V.
Throughout electrolysis, fresh alumina was fed to the electrolyte to compensate for the electrolysed alumina.
After 100 hours electrolysis, the anode was removed from the cell, allowed to cool down to room temperature and examined.
The anode's diameter had increased from 1.94 to 1.97 cm. The anode's metallic part had been heavily oxidised. The thickness of the integral oxide scale had increased from 350 micron to about 1.1-1.5 mm. The oxide scale was made of: a 300-400 micron thick outer layer containing pores having a size of 30-50 micron and having cracks; a 1-1.1 mm thick inner layer that had been formed during electrolysis. The inner layer was porous and contained electrolyte under the cracks of the outer layer.
An anode made of metallic cobalt oxidised under the conditions of Example 1c, i.e. resulting in a CoO integral surface layer was tested in an aluminium electrowinning cell under the conditions of Comparative Example 3. A nominal electrolysis current of 7.3 A was passed from the anode to the cathode at an anodic current density of 0.8 A/cm2.
At start-up the cell voltage was 4.1 V and steadily decreased to 3.7-3.8 V after 30 minutes (instead of 4-4.2 in Comparative Example 3). The cell voltage stabilised at this level throughout the test without noticeable fluctuations, unlike in Comparative Example 3.
After 100 hours electrolysis, the anode was removed from the cell, allowed to cool down to room temperature and examined.
The anode's external diameter did not change during electrolysis and remained at 1.94 cm. The metallic cobalt inner part underneath the oxide scale had slightly decreased from 1.85 to 1.78 cm. The thickness of the cobalt oxide scale had increased from 0.3 to 0.7-0.8 mm (instead of 1-1.1 mm of Comparative Example 3) and was made of: a non-porous 300-400 micron thick external layer; and a porous 400 micron thick internal layer that had been formed during electrolysis. This internal oxide growth (400 micron thickness over 100 hours) was much less than the growth observed in Comparative example 3 (1-1.1 mm thickness over 100 hours).
It follows that the anode's CoO integral surface layer inhibits diffusion of oxygen and oxidation of the underlying metallic cobalt, compared to the Co2O3 and Co3O4 integral surface layer of the anode of Comparative Example 3.
The anode material of Examples 1a to 1d, 2 and 3 can be covered upon formation of the integral CoO layer with a slurry applied layer, in particular containing CoFe2O4 particulate in a iron hydroxide colloid followed by drying at 250° C. to form a protective layer on the CoO integral layer.
A coated anode for use in a cell according to the invention was made by covering a metallic cobalt substrate with an applied electrochemically active coating comprising an outer CoO layer and an inner layer of tantalum and cobalt oxides.
The coating was formed by applying cobalt and tantalum using electrodeposition. Specifically, tantalum was dispersed in the form of physical inclusions in cobalt electrodeposits.
The electrodeposition bath had a pH of 3.0 to 3.5 and contained:
The tantalum particles had a size below 10 micron and were dispersed in the electrodeposition bath.
Electrodeposition on the cobalt substrate was carried out at a current density of 35 mA/cm2 which led to a cobalt deposit containing Ta inclusions, the deposit growing at a rate of 45 micron per hour on the substrate.
After the deposit had reached a total thickness of 250-300 micron, electrodeposition was interrupted. The deposit contained 9-15 wt % Ta corresponding to a volume fraction of 4-7 v %.
To form a coating, the substrate with its deposit were exposed to an oxidation treatment at a temperature of 950° C. The substrate with its deposit were brought from room temperature to 950° C. at a rate of 450-500° C./hour in an oven to optimise the formation of CoO instead of Co2O3 or Co3O4.
After 8 hours at 950° C., the substrate and the coating that was formed by oxidation of the deposit were taken out of the oven and allowed to cool down to room temperature. The coating had an outer oxide layer CoO on an inner oxide layer of Co—Ta oxides, in particular CoTaO4, that had grown from the deposit. The innermost part of the deposit had remained unoxidised, so that the Co—Ta oxide layer was integral with the remaining metallic Co—Ta deposit. The Co—Ta oxide layer and the CoO layer had a total thickness of about 200 micron on the remaining metallic Co—Ta.
As demonstrated in Example 6, this CoO outer layer can act as an electrochemically active anode surface. The inner Co—Ta oxide layer inhibits oxygen diffusion towards the metallic cobalt substrate.
A coated anode was made of a cobalt substrate covered with a Co—Ta coating as in Example 5 and used in a cell for the electrowinning aluminium according to the invention.
The anode was suspended in the cell's electrolyte at a distance of 4 cm from a facing cathode. The electrolyte contained 11 wt % AlF3, 4 wt % CaF2, 7 wt % KF and 9.6 wt % Al2O3, the balance being Na3AlF6. The electrolyte was at a temperature of 925° C.
An electrolysis current was passed from the anode to the cathode at an anodic current density of 0.8 A/cm2. The cell voltage remained remarkably stable at 3.6 V throughout electrolysis.
After 150 hours electrolysis, the anode was removed from the cell. No significant change of the anode's dimensions was observed by visual examination.
Example 7
Example 5 was repeated by applying a Co-Ta coating onto an anode substrate made of a metallic alloy containing 75 wt % Ni, 15 wt % Fe and 10 wt % Cu.
The anode was tested as in Example 6 at an anodic current density of 0.8 A/cm2. At start-up, the cell voltage was at 4.2 V and decreased within the first 24 hours to 3.7 V and remained stable thereafter.
After 120 hours electrolysis, the anode was removed from the cell. No sign of passivation of the nickel-rich substrate was observed and no significant change of dimensions of the anode was noticed by visual examination of the anode.
Examples 5 to 7 can be repeated by substituting tantalum with niobium.
Another anode for use in a cell according to the invention was made by applying a coating of Co—W onto an anode substrate made of a metallic alloy containing 75 wt % Ni, 15 wt % Fe and 10 wt % Cu.
The coating was formed by applying cobalt and tungsten using electrodeposition. The electrodeposition bath contained:
Moreover, NH4OH had been added to this bath so that the bath had reached a pH of 8.5-8.7.
Electrodeposition on the Ni—Fe—Cu substrate was carried out at a temperature of 82-90° C. and at a current density of 50 mA/cm2 which led to a cobalt-tungsten alloy deposit on the substrate, the deposit growing at a rate of 35-40 micron per hour at a cathodic current efficiency of about 90%.
After the deposit had reached a total thickness of about 250 micron, electrodeposition was interrupted. The deposited cobalt alloy contained 20-25 wt % tungsten.
To form a coating, the substrate with its deposit were exposed to an oxidation treatment at a temperature of 950° C. The substrate with its deposit were brought from room temperature to 950° C. at a rate of 450-500° C./hour in an oven to optimise the formation of CoO instead of Co2O3 or Co3O4.
After 8 hours at 950° C., the substrate and the coating that was formed by oxidation of the deposit were taken out of the oven and allowed to cool down to room temperature. The coating contained at its surface cobalt monoxide and tungsten oxide.
The structure of the coating after oxidation was denser and more coherent than the coating obtained by oxidising an electrodeposited layer of Ta—Co as disclosed in Example 1.
As demonstrated in Example 10, this coating can act as an electrochemically active anode surface. The presence of tungsten inhibits oxygen diffusion towards the metallic cobalt substrate.
An anode was made as in Example 9 and used in a cell for the electrowinning of aluminium according to the invention.
The anode was suspended in the cell's electrolyte at a distance of 4 cm from a facing cathode. The electrolyte contained 11 wt % AlF3, 4 wt % CaF2, 7 wt % KF and 9.6 wt % Al2O3, the balance being Na3AlF6. The electrolyte was at a temperature of 925° C.
An electrolysis current was passed from the anode to the cathode at an anodic current density of 0.8 A/cm2. The cell voltage remained stable at 3.5-3.7 V throughout electrolysis. After 100 hours electrolysis, the anode was removed from the cell. No change of the anode's dimensions was observed by visual examination.
Examples 9 and 10 can be repeated with an anode substrate made of cobalt, nickel or an alloy of 92 wt % nickel and 8 wt % copper.
Comparative tests show that the use in a conventional cryolite-based electrolyte at 960° C. of a metal-based anode having an electrochemically active outer part comprising a layer that contains predominantly cobalt oxide CoO, leads to accelerated oxidation of the anode and dissolution into the electrolyte of oxides of the anode, in particular CoO. Moreover, use of such an anode in an electrolyte at 910°-940° C. without potassium fluoride leads to corrosion or passivation the anode.
These Examples demonstrate that a material having an outer part with a layer that contains predominantly cobalt oxide CoO as described above, provides an enhanced stability during use in an aluminium electrowinning cell and is therefore suitable to protect anodes having a flow-through structure which is exposed to the fluoride-based molten electrolyte that is rendered more aggressive to the anodes by its circulation through the anodes.
| Number | Date | Country | Kind |
|---|---|---|---|
| PCT/IB04/01900 | Jun 2004 | IB | international |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/IB05/51718 | 5/25/2005 | WO | 00 | 12/2/2006 |
