Metallic aluminum is not found in nature, but the oxides, hydroxides, and especially the silicates are plentiful. The estimated percentage of aluminum in the crust of the earth is about 8 pct while that of iron is about 5 pct. By far the larger portion of this is combined with silica in the form of various clay minerals and igneous silicate rocks. From the point of view of extractive metallurgy of aluminum, these are low grade ores while the better qualities of bauxite are the high grade ores. There have been various definitions of bauxite but perhaps the best definition at the present time is that bauxite is a rock or earth commonly used as an ore of aluminum or its salts in which the aluminum is present predominantly as a hydrate or a mixture of hydrates and hydrous oxides. It contains varying amounts of oxides of silicon, iron, and titanium and traces of compounds of some of the less common elements. The silica is mainly combined with alumina as clay or clay minerals which are hydrous aluminum silicates, although a part of it may be present as quartz sand. On the American continents, the alumina is mainly present as gibbsite, Al2O3 · 3H2O, and the same may be said of the best known deposits of the Dutch East Indies and some of the deposits in India. In France and other countries in Europe as well as in Africa, the alumina is present mainly as boehmite, Al2O3 · H2O, but in some of these deposits it is mixed with minor amounts of gibbsite. Some other deposits, such as those in the islands of Haiti and Jamaica, evidently contain two or more hydrates or hydrous oxides of alumina in varying proportions. Perhaps the main portion of the alumina may be present as gibbsite and boehmite with the proportion between the two varying rather widely.
The commercial production of high grade aluminum from its ores is divided into two separate and distinct processes. The first process produces pure alumina from the ores, and the second process reduces the pure aluminum oxide to metallic aluminum by electrolysis of a solution of alumina in molten cryolite using carbon electrodes.
Production of Alumina
While many processes have been devised for the production of pure alumina from bauxite, clay, and other minerals, the one which is used predominantly and almost to the exclusion of the others is the Bayer process invented by Carl Joseph Bayer some sixty years ago. Briefly, this process consists of dis¬solving alumina from bauxite in caustic soda solution to form a sodium aluminate solution, settling or filtering to remove the insoluble residue, precipitating the alumina as hydrate from the solution, and calcining the hydrated alumina to aluminum oxide for use in the electrolytic reduction process.
Among the processes which have been tried was the production of molten alumina in an electric furnace with carbon to reduce the iron, silicon, and titanium to form a heavy alloy in the bottom of the furnace and a molten alumina slag which could be separated, cooled, and used for the production of commercial aluminum by electrolytic reduction. A limited market was found for the iron-silicon-titanium alloy which also contained some aluminum, and the alumina slag was used in the commercial reduction of aluminum. There was some difficulty in obtaining aluminum of the purity desired in the market, and the cost of the process was high so that it was abandoned.
Among the acid salt processes, the potassium alum process of the Kalunite Co. dissolved the alumina in a solution of sulphuric acid and potassium sulphate to form potash alum which was crystallized, redissolved in a minimum amount of water, heated to precipitate a basic potash alum which was then washed and calcined to form a mixture of potassium sulphate and anhydrous alumina. After removing the potassium sulphate by solution, the alumina was dried for use in the reduction process. Another acid salt process was the ammonium alum method in which clay was treated with ammonium bisulphate to produce an ammonium alum solution. This solution was then treated with ammonia to precipitate the alumina and convert the ammonium bisulphate to ammonium sulphate. The alumina was calcined for use in the aluminum industry, and the ammonium sulphate solution was evaporated and re-converted to ammonium bisulphate by heating in an electric furnace. This made it possible to recycle both the ammonium sulphate and the ammonia in the process. Both of these acid salt processes were tried during World War II but are not currently in use.
The only alkaline furnace process for the production of alumina which is still in operation in this country is a modification of an old process for the treatment of bauxite and other materials containing alumina which had been tried many times. In this process, bauxite containing approximately 50 pct alumina and up to 15 pct silica is first treated by the American modification of the Bayer process, and the insoluble material, usually called red mud, is partially dewatered, mixed with limestone and soda ash, and sintered in rotary kilns.
The combined Bayer and sinter process seems to be commercially profitable on high silica bauxite where the high silica bauxite is cheap enough to be competitive with low silica imported bauxite and where fuel is cheap and where the sinter plant already exists. It might not be profitable to build a new sinter plant for the purpose, especially when imported bauxite is plentiful.
Commercial Production of Alumina
While many processes have been devised for the production of pure aluminum oxide, the Bayer process is used predominantly and almost to the exclusion of the others. This process was originally developed in Europe for application to European ores in which the alumina is mainly present as the mono- hydrate, boehmite. The ground bauxite, either with or without partial calcination, is treated with caustic soda solution containing approximately 350 g NaOH per liter at a temperature approximating 170 °C (338°F) for 2 to 4 hr with a total heating time of 4 to 8 hr so as to produce a solution of sodium aluminate with approximately 1.6 mol Na2O per mol Al2O3.
In the European Bayer process, the practical limit of caustic soda concentration is just below that concentration which, when nearly saturated with alumina, would be sufficient to salt out some of the alumina and some of the soda as a hydrated sodium aluminate. The caustic soda concentration and the temperature vary somewhat in different alumina plants, but in general the limit of concentration is a little short of that concentration which might salt out sodium aluminates at the lowest temperatures to which strong solutions may be subjected in certain parts of the plant.
In the American Bayer process, the practical limit of caustic soda concentration is that which gives fairly efficient use of the lime in causticizing without precipitating appreciable quantities of alumina as calcium aluminate. The limiting temperature of digestion is purely economic and varies somewhat in different plants. In general, the temperature and time of digestion should be just sufficient to get the proper amount of alumina into the solution and to precipitate any silica which may have gone into solution as a hydrated sodium aluminum silicate.
The loss of soda and alumina is due to the formation of a somewhat indefinite compound or group of compounds which are hydrous sodium aluminum silicates. There is some difference of opinion as to whether there is a definite compound or whether a part of the loss of soda and alumina is due to adsorption on some of these hydrous sodium aluminum silicates. In any case, the chief loss of soda is due to silica in the bauxite, and as the silica in the bauxite increases, the amount of required make-up soda in the process increases rapidly.
There are no definite specifications for analysis or particle size of alumina for the industry as a whole. Each producing company strives to keep the impurities as low as possible consistent with low costs. Off grade alumina is sometimes segregated for use in production of certain alloys which may require the addition of Si, Fe, Mn, Cu, etc. As far as any definite specifications can be given for the industry in either this country or Europe, we might give the following:
Each cell may take anywhere from 8000 to 60,000 amp or in special cases 100,000 amp, according to the size of the cell, at a voltage per cell, including bus bar losses of 4½ to 6½ v, depending on the design of the circuit and the cell, age of cell, anode-cathode distance, and other factors. Since it is not economical to generate electricity at such low voltages, a number of cells anywhere from 30 to 100 or more are connected in a line or series. The line current may be supplied by direct current generators, rotary converters, or rectifiers. The tendency in the last few years has been to go to rectifiers. The power required per pound of aluminum varies with various factors, including the anode-cathode distance and the care with which the cell is operated. Possibly 10 kw-hr per lb of metal may be the average overall consumption of power, including line losses and accessories, although the power actually used in the cell may be 8 kw-hr per lb or slightly less under the best conditions. The consumption of carbon anodes varies a great deal with the quality of the anodes and the other conditions but in general may be from 6/10 to 7/10 of a lb per lb of aluminum.
The electrolytic cell works more efficiently when the carbon bottom is covered with a layer of aluminum. The reasons for this are somewhat controversial, but it appears that a bare carbon cathode causes the production of somewhat more sodium by electrolysis than a cathode covered with molten aluminum. In practice, the current efficiency may vary between 75 and 95 pct. Under good operating conditions, it usually runs in the high 80’s.