The oxidation and protection of heterogeneous carbon anodes used for aluminium smelting

Show simple item record Fitchett, Anne Marie 2021-08-26T06:16:40Z 2021-08-26T06:16:40Z 1988
dc.description Full text is available to authenticated members of The University of Auckland only.
dc.description.abstract Pre-baked carbon anodes are commonly employed in the production of aluminium by electrolysis (in the Hall-Heroult process). They are consumed in the process, both by the electrochemical reaction, and as "excess" carbon consumption. All carbon consumed in reactions other than those resulting in the deposition of aluminium metal is classified as excess. Aluminium smelter technologists tend to refer to the total carbon consumption relative to the amount of metal produced. Values reported vary between 0.40 and 0.48 kilograms of carbon per kilogram of aluminium, and between 0.06 and 0.14kgC/kgAl is typically considered to be excess. Oxidation by air ("airburn") of the anode surface area protruding above the melt accounts for between 0.03 and 0.05kgC/kgAl, in spite of attempts to protect the anode carbon. A significant proportion may be saved by improved carbon quality, technology, and operating practice: small improvements often implying considerable financial savings in raw materials costs. The overall objective of the investigations undertaken was to quantify the importance of airburn and airburn protection media for various carbon qualities. Apparatus was developed for the measurement of carbon oxidation rates at temperature. Temperature dependence studies of unprotected carbon core specimens from plant anodes showed that below 400ºC, the oxidation rates were always extremely low. This temperature probably represents the ignition temperature limit. They then increased rapidly up to approximately 700ºC, above which the airburn rates increased only gradually as the temperature was raised. This was probably due to mass transfer effects becoming rate-controlling. At a given temperature, for unprotected carbons, a significant difference was observed in airburn rates for different formulations this difference being up to 60%. Whilst differences were apparent for different pitching levels and formulations, local variations in other properties made it difficult to ascribe accurate trends. This was consistent with work done by other researchers, and is a consequence of the heterogeneous nature of the material. Attempts were made to relate the measured oxidation rates to carbon quality, as quantified using X-ray diffraction and porosimetry techniques. Various forms of anode protection were investigated, including impregnation using a solution of sodium tetraborate ("Borax"), coating with aluminium spray, and coverings of powdered alumina ore and crushed bath. Aluminium spray was shown to reduce the airburn rate by up to 60% in the temperature range considered, while "Borax" was shown to be of limited benefit. A marked reduction in airburn rate was achieved by using alumina or crushed bath at a given temperature. For 15mm coverage the reduction was over fourfold. Above 25 mm alumina or crushed bath cover the rate of decrease became much smaller, but the differences due to carbon quality were also very small. There was little difference in the use of crushed bath as opposed to alumina for airburn protection below 650ºC. In the temperature region 650ºC to 750ºC, it appeared that a decrease in airburn rate occurred when crushed bath protection was used, and this may have been associated with a sintering/blocking effect. Above 750ºC shrinkage allowed ingress of air, and the rate again became similar to that of alumina. A plant trial in an aluminium smelter was initiated particularly to evaluate the protection afforded by powdered alumina and aluminium spray, and the performance of anodes of different heights. An important part of the programme was also the collection of data to assist in the refinement of the airburn model being developed. This model allowed a reasonable prediction of the expected carbon consumption by oxidation for specific anode conditions. Based on the assumption that airburn in the pot is a function of temperature and the availability of oxidizing gas, anode temperature and cover data were recorded. These were then related to the depth of carbon consumed at each face during anode life. Average timetemperature profiles obtained showed a rapid increase in top temperature up to about 500ºC in about 24 hours. This was followed by a linear rate of increase up to a peak temperature which is dependent on anode height, cover depth and stall location. Carbon ratios obtained for the different anode conditions showed that most of the difference in carbon consumption could be accounted for by airburn. Values for depth of carbon consumed at the anode tops were very similar, varying slightly with the depth of ore cover measured. The main differences were found in the mass of carbon consumed at the anode sides, particularly at the centre channel. Aluminium spray provided significant protection on all sides. However, at the centre channel some of the spray was lost by burning, or in some cases, flux wash. In the light of the results obtained, recommendations for improvements in pot operating practices for better anode performance can be made. These include improvements both in the quality and distribution of anode protection. The minimisation of cell disturbances and exposure of the anodes to flame and bath are also recommended, together with the minimisation of unnecessary heat generation.
dc.publisher ResearchSpace@Auckland
dc.relation.isreferencedby UoA9974689814002091
dc.rights Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
dc.rights Restricted Item. Full text is available to authenticated members of The University of Auckland only.
dc.title The oxidation and protection of heterogeneous carbon anodes used for aluminium smelting
dc.type Thesis Engineering The University of Auckland PhD
dc.rights.holder Copyright: The author

Files in this item

Find Full text

This item appears in the following Collection(s)

Show simple item record


Search ResearchSpace