In the Alcoa smelting process, anhydrous aluminum chloride is electrolytically converted to aluminum metal. This process could have a major impact on the aluminum industry if an efficient way of making the anhydrous aluminum chloride from alumina were found (Othmer et al., 1978). To make aluminum chloride, carbon and chlorine must be added to alumina. The carbon and chlorine reactants can take many forms, but carbon tetrachloride is convenient in that the carbon and chlorine are together in a single molecule. The net reaction is written as:The kinetics of this reaction have received considerable attention but the rates of the reaction have been too slow to justify industrial development.In this study, means of increasing the reaction rate by operating at supercritical conditions are experimentally investigated. The reaction product, AlCl,, is soluble in the supercritical reactant, and the solubility enhancement that supercritical fluids display is a reasonable explanation for the increases in the observed rate, which are reported.
Experimental ProcedureAlumina, 5-10 g, solid, and excess carbon tetrachloride, 30-60 g, liquid, were combined in a batch reactor and heated rapidly to reaction conditions. Pressures of 4 to 10 MPa and temperatures of 563 to 643 K were supercritical with respect to carbon tetrachloride (T, = 283"C, P, = 4.5 MPa) and were monitored continuously. Conversions were determined by extracting the reaction products with water and measuring the amount of unconverted alumina. Alumina is the only water-insoluble solid in the reactor. At the conditions studied, particle size (mass average dia. 60 pm, number average dia. 6 pm N, BET surface area 60 m'/g) and the total mass of alumina charged had no effect on the reaction rate per unit mass and hence fluid-solid mass transfer is considered to be unimportant in the overall reaction rate process. Runs with smaller particles produced no increase in the reaction rate, which indicates that internal mass transfer limitations are not important.Reaction rates were measured by dividing the amount of alumina converted by the isothermal reaction time. A correction was made for the conversion during the heat-up period using the measured activation energy. This correction was verified experimentally by stopping several runs as soon as run temperature was reached. These runs yielded 1 to 5% conversion during heatup. Conversions ranged from 6 to 48% except for one high-temperature run where the conversion reached 73%. In addition, the assumption of irreversibility of the reaction for these conversions was verified by two runs showing very high conversion (88%) as run time was lengthened to 4 h.
Results and DiscussionThe results obtained in this study are summarized in Figure 1, which also shows, under noncritical conditions, the results of previous investigators (Bertoti et al., 1980(Bertoti et al., , 1982. The primary difference between the present results and those of previous investigators is that the present experiments were conducted at much higher pressures (d...