0191-9067/85 $3.00 + .00 Copyright ® 1985 SUNSAT Energy Council SPACE SEMICONDUCTOR PROCESSING FACTORY ISAO KUDO and HIROYUKI FUJISADA Electrotechnical Laboratory Agency of Industrial Science and Technology 1-1-4 Umezono, Sakura Ibaraki, Japan Abstract — Consistent material processing from crystal growth of bulky material to highly integrated devices is proposed. Especially, feasibility of automated dry microfabrication process in space is discussed. Space has some beneficial characteristics about high-vacuum environment as well as zero-gravity field. INTRODUCTION It is considered to be quite difficult to explain what product will be economically attractive in a space factory (1). For example, no one can answer “yes” to the question of whether crystal growth of silicon or galium arsenide in space is profitable or not. To add another value to the raw material processed in space is quite important. Our concepts originate in this point. Consistent processing from crystal growth of semiconductor material to VLSI device in addition to silicon crystallization is considered to be promising for a space factory. As for terrestrial device fabrication, wet processing using large amounts of water and acid has been replaced by dry processing using low temperature plasma. Microfabrication technology requires more stringent dust free environment — super clean room. It is said that the largest part of yield degradation originates from dust. In situ measurement methods and automatic wafer loader reduces dust from workers considerably. Technology movement to dry processing and advent of dexterous manipulator will make space processing possible. Production in the semiconductor industry in Japan amounts to 4 billion dollars in 1983 as shown in Fig. 1 and the ratio of LSI to total semiconductor devices is about 60% (2). The amount of LSI will increase to 10 billion dollars in 1985. Future key technology of LSI is considered to be in large scale chips with highly integrated circuits, electron beam and X-ray lithography for microfabrication, and large wafer for improved cost achievement. Wet process using large quantity of pure water and acid began to be replaced with dry process using plasma discharge. VLSI fabrication by all dry process will come soon. Dust attachment on a wafer during processing is another big problem. An idea in which all of the process should be carried out in vacuum has been proposed. Although the ratio of vacuum subsystem to total process system is not high so far the amount required for vacuum apparatuses and their operation cost will increase. The physical limit of LSI is considered to be 0.01 /xm. Since the 2 /mi rule governs the fabrication at present, the tendency of microfabrication will continue further. Figure 2 shows the change of device integration (3). The 1 Mbit memory will appear soon. The processing limit in photolithography will be overcome by electron beam or X-ray lithography. The space environment is considered to be suitable for such a microfabrication.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==