Japan's Josephson Computer Project was a distributed style of collaborative activity. Funding was provided by the Ministry of International Trade and Industry (MITI) to research groups at the Electrotechnical Laboratory (ETL), Fujitsu, Hitachi, and NEC (see Chapter 5) for each to pursue its own independent program to develop the technology of Josephson computers operating at 4.2 K. As described in the 1989 JTEC superconductivity report (Dresselhaus 1989), this project clearly established Japan as the leader in the design and fabrication of Josephson microprocessor and memory circuits. The Nb trilayer process, which was invented at Sperry and Bell Laboratories in the early 1980s, was developed by the Japanese groups into the highly reliable technology that is now used exclusively for LTS digital circuits and superconducting quantum interference devices (SQUIDs). These groups published their work in detail, to the great benefit of scientists worldwide.
Interestingly, the WTEC panel noted different views of this project in the United States and Japan. In the United States, the research of the four groups in Japan is regarded as one of the outstanding achievements of SCE, in that a new technology was established which made all previous technologies obsolete (e.g., the Pb alloy process used in the earlier IBM project) and which has become the worldwide standard. In Japan, at least in MITI, the project is viewed as having failed to build a Josephson computer! It is interesting to speculate whether the style of the activity to some extent determined the achievements. A distributed style of activity, as used in this project, seems ideally suited to development of device technology, in that relatively small groups can effectively work independently to solve a wide variety of materials and process issues. However, if the objective is to build a system, a larger team of people working jointly in one location, as at the Superconducting Sensor Lab (SSL) later, might be a more suitable style. This assumes, of course, that some level of component technology is available. It is probably difficult to achieve materials, device, and process technology development, as well as a prototype system demonstration, within one project, as the skills of the staff required for each task are quite different.
By far the largest collaborative activity in either country (5 to 10 times larger than a typical project), ISTEC adopted a centralized style. Soon after it was formed in 1988, a new laboratory was built in Tokyo, and other rented space has been obtained as the project has continued to grow. The research staff, drawn from a very large number of supporting companies, plus postdocs who are hired by ISTEC and the government, and students, work in the central facilities with technical management and administrative staff. In the first phase of ISTEC, the first 10 years, the emphasis has been on materials research. The second phase will have three thrusts of a more applied nature: applications of bulk material, wires, and thin films. The funding will be about 4 billion yen (~$33 million2) per year, which does not include the salaries of the research staff on loan from industry. More detailed information concerning ISTEC is given in Chapter 9.
After the discovery of the oxide superconductors in 1986, Japan funded research in HTS materials science at the centralized project at ISTEC, but also pursued research into electronic devices based on HTS through a distributed-style project funded by the Research and Development Association for Future Electron Devices (FED). FED is a foundation established by the Japanese electronics industry in 1981 under license from MITI. Its R&D projects are conducted under contract with the New Energy and Industrial Technology Development Organization (NEDO) in accordance with MITI's Industrial Science and Technology Frontier (ISTF) program. Overall research management is carried out by FED, with the R&D subcontracted to member companies, universities, and national institutes. The superconductivity project, described more fully in the site visit reports (Appendix B), funded research carried out independently at 8 companies. The initial objective was to develop three-terminal HTS devices; later, the project also included HTS SNS junctions and single flux quantum (SFQ) circuits, and recently, two LTS hybrid system programs began at Hitachi and NEC. These programs to develop high speed switching systems use technology from the earlier Josephson Computer Project.
The Superconducting Sensor Laboratory is described in detail in Chapter 6. It was a centralized collaborative project funded by MITI from 1990-1996 whose primary mission was to develop both sensor- and system-level SQUID technology. The research staff was contributed by the 10 supporting companies, and work was carried out in a new laboratory. Technical and administrative management staff was hired by MITI. The success of this activity might suggest that the centralized style is well suited to defined and ambitious system-level objectives.
AMTEL and the Western Alliance, the two most recent collaborative activities in Japan, are best described as joint ventures, even though they differ in some details. At the Advanced Mobile Telecommunications Technology Laboratory (AMTEL) (see site visit report in Appendix B), development of the technologies required for HTS wireless systems is carried out at both of the member companies, Alps Electric and Denso Corporation, in existing laboratories. Staff members in some cases have been relocated to the facility of the other partner.
In the joint venture known as the Western Alliance, between Kyocera, Matsushita, and Sumitomo,3 research is carried on within the 3 partners' laboratories, and staff has not been relocated. However, in contrast to the distributed style of the Josephson Computer or FED projects, the work at the 3 laboratories is coordinated, with the common objective being to develop a high power filter system. Films are made at Sumitomo, filter design and device process is carried out at Matsushita, and the refrigerator will be purchased and the cryogenic package will be developed at Kyocera. Under a 4-year project, 50% of the work is supported by MITI.
2The rate of exchange used in this report $1 = ¥120.
3The WTEC panel visited both Matsushita and Sumitomo: see site visit reports in Appendix B.