results (Figure IV.B.l.a.22) indicate that a very high concentration ratio is desirable, and a substrate other than GaAs is needed. Also, further research is needed in the recovery of gallium. The technology of gallium arsenide is not as well developed as silicon. Some of the immediate areas which must be explored are (1) methods of extracting large quantities of gallium from bauxite, (2) reducing the cost of gallium, (3) elimination of the gallium arsenide substrate, and (4) the development of concentrators for the use in space with ratios greater than 10. Arsenic is a by-product from processing metal based ores such as copper, lead, and iron. Roasting the ores produces volatile oxides which are collected as flue dusts. The flue dusts are purified to obtain white arsenic (As^On) or reduced with carbon to yield arsenic metal. In general, these processes are well developed and, for the most part, arsenic is considered a troublesome by-product. Arsenic production consists of approximately 97 percent white arsenic and 3 percent metallic oxide. The U.S. imports approximately 16,000 tons of (Figure IV.B.l.a.23) white arsenic and 600 tons of metallic arsenic. The price of white arsenic fluctuates from $2 to $4 per kg (1972). The availability of arsenic or refining it does not appear as a significant problem for the SPS. At most, facilities will have to be increased to reduce white arsenic to arsenic metal. Gallium will be the key element in fabricating GaAs solar cells. The availability of gallium arsenide solar cells for the SPS arrays is as dependent on the availability of large scale raw material processing facilities and techniques as much as it is on the basic elements to make the compound. At present, vast sums are being spent each year by the ERDA to improve the supply of device-grade silicon and to lower its price. These improvements are being made in a material technology that is already mature and has more than two decades of manufacturing experience behind it. It also has a number of vendors who are producing it in large quantities now, and have a capability to expand their operations in the future. With respect to the control of surface states and bulk impurities, gallium arsenide has not been developed to the level of silicon technology. Many researchers believe that the recent developments in epitaxial methods of growing heterojunctions in semiconductors will lead to gallium arsenide devices which are superior to silicon devices. The SPS array requirements could be enough of a special need to justify such an independent effort because of the weight
RkJQdWJsaXNoZXIy MTU5NjU0Mg==