-
4. MOS in a Bipolar Company: FAIRCHILD AND THE MOS TRANSISTOR, 1963 – 1968
- Johns Hopkins University Press
- Chapter
- Additional Information
107 FOUR MOS in a Bipolar Company F A I R C H I L D A N D T H E MOS T R A N S I S T O R , 1 9 6 3 – 1 9 6 8 From 1963 to 1968 Fairchild succeeded in developing a scientific understanding of MOS structures but utterly failed in developing an MOS business. Fairchild ’s MOS program faced difficulties common throughout the industry, such as the challenge of finding a role for the new technology and the difficulties of making stable MOS transistors. But the crucial problem in Fairchild’s MOS program was organizational. Initially, Fairchild R&D’s MOS program suffered from being decentralized, spread out across departments where bipolar technology represented a satisfactory (and simpler) solution for a given application. As R&D centralized its MOS work, a competing site for MOS work arose at the company’s Mountain View plant. The Mountain View work was not only not under the management control of R&D, it was under no management control whatsoever. The R&D and Mountain View groups had very different conceptions of MOS technology, particularly how it should be made and what it was best suited to do. The two sides battled to a stalemate over a period of years, to the detriment of Fairchild. In the industry at large, despite the publicity given to it, the MOS technology remained in an inchoate state, with no firm able to establish an unequivocally successful MOS product. Thus, despite the frustration both sides at Fairchild felt, the situation was not clearly unsatisfactory 108 T O T H E D I G I TA L A G E and persisted until the calamitous summer of 1968. Ultimately, Fairchild had enough MOS talent for two successful companies, but too much for one. Studying the Silicon–Silicon Dioxide System In early 1963, after Frank Wanlass’s initial MOS transistor work, Gordon Moore, the director of Fairchild R&D, began putting together a team to understand the MOS structure and the silicon–silicon dioxide system in a systematic way. Moore’s main reason for this study was to produce better bipolar transistors —Fairchild’s main area of business. But this work would also be expected to address the problems of stability of MOS transistors. Up to this time the problems of MOS stability were so great—an MOS transistor’s characteristics might vary by over one hundred volts with changes in time or temperature— that they made MOS transistors useless as a product. If these problems were solved, MOS transistors would be technically viable. In 1963 Moore and the manager of the Solid State Physics Department, C. T. Sah, hired three people who would play key roles in developing a systematic understanding of MOS structures.1 The first member of the group was Bruce Deal, who joined Fairchild in March 1963. Deal had earned his Ph.D. in chemistry from Iowa State University in 1955, working on the oxidation of uranium and thorium. He then worked for Kaiser Aluminum on the oxidation of aluminum, and in 1959 he joined Rheem Semiconductor, the first of the start-up companies formed from Fairchild. In 1962 Rheem was acquired by Raytheon, a Massachusetts-based electronics firm, which sent a group of its Massachusetts semiconductor team to California to direct Rheem. In the ensuing turmoil, large numbers of pre-Raytheon Rheem employees left, including Deal. Deal came to Fairchild with four years’ experience in the semiconductor industry and many more years’ experience with the processes of oxidation, a key part of the MOS structure. Deal was an extremely talented and methodical experimentalist.2 Andrew Grove joined the group later that spring after earning his Ph.D. in chemical engineering from Berkeley. Grove, a Hungarian, had emigrated to the United States following the Soviet suppression of the 1956 uprising. He earned a bachelor’s degree in engineering from the City College of New York and then [3.239.214.173] Project MUSE (2024-03-19 10:36 GMT) 109 MOS in a Bipolar Company went to Berkeley, where he received his Ph.D. in chemical engineering, concentrating on fluid mechanics. Grove was to be the chief theoretician of the group.3 The final member of the group, Ed Snow, joined later that year, after receiving a Ph.D. in solid-state physics from the University of Utah. Although he had no explicit background in semiconductors as such, his dissertation had been on the migration of ions in quartz. Snow had initially accepted a job with one...