Journal of Solar Energy Engineering-Transactions of The ASME, Vol.123, No.2, 125-132, 2001
Solar production of aluminum by direct reduction: Preliminary results for two processes
The production of aluminum or silicon by reduction of their oxides with carbon is a technical challenge. The temperature required, in the range 2100-2300 degreesC, is too high for practical process heat addition from a combustion sour ce alone. Wizen an electrothermal process is used, only about a third of the energy contained in the fuel used to generate electricity enters the process. Thus, for materials produced electrolytically or in an electric furnace, the energy cost dominates the cost of the final product. By contrast, highly-concentrated solar energy is capable of supplying large amounts of process heat at very high temperatures, and may have real advantages for metals reduction processes. An are introduces too much energy to the reaction zone, hz the case of aluminum, the metal floats and it short circuits the nl c. Ideally, the heat would enter at the bottom or side of a reactor, which could be achieved with solar process heat. Among industries, the primary aluminum industry is a major consumer of electricity. It uses about 10 percent of the electricity generated globally for industrial purposes, and about half comes from coal-fired generation stations. This represents about 5 percent of the electricity generated for all sectors. A solar-thermal process would drastically reduce the emission of climate-altering gases, reduce the reliance on electricity, and might be a critical factor in making a direct thermal route Si om the ore to metal possible. Two industrially-developed processes appear to be attractive candidates for a solar process. Preliminary tests have been performed using a black-body cavity receiver placed at the focus of the Paul Scherrer Institute's 70kW tracking parabolic concentrator, and though the experiment had to be ended earlier than planned, a small amount of 61/37 weight percent Al/Si alloy was formed, and the partially reacted pellets showed conversion to Al4C3 and SiC. Further qualitative tests have been performed using the facilities at Odeillo in a 2 kW solar furnace, where the onset of production of both aluminum by direct carbothermal reduction, and Al-Si alloy via carbothermal reduction of a mixture of alumina, silica and carbon could be directly observed.