The University of Tokyo|
Research Center for Advanced Science and Technology (RCAST)
4-6-1 Komaba, Meguro-ku
Tokyo 153, Japan
|Date Visited:||January 29, 1997|
|WTEC Attendees:||G. Gamota (report co-author), J. Rowell (report co-author), H. Morishita, F. Patten|
|Hosts:||Prof. Yoichi Okabe|
Professor Okabe reviewed for the panel members the current state of superconductivity research activities in Japan. He made many interesting comments about both LTS and HTS technologies and the changes in funding responsibilities of the agencies. A recent important event was the convening by MITI of a Special Group whose mission was to review the projects funded by MITI in this area. The group membership included Professor Sekine (Chair), Professor Kitazawa, Professor Masada (the University of Tokyo) and Professor Okabe. It also had representatives from Hitachi, Sumitomo, Tokyo Power, and Toshiba, with representatives from MITI completing the membership.
Projects under review included those funded at ISTEC and through the New Energy and Industrial Technology Development Organization (NEDO) and the Research and Development Association for Future Electron Devices (FED).
Besides MITI, other Japanese government organizations that fund superconducting projects include STA (Science & Technology Agency), Monbusho (Ministry of Education, Science, Sports and Culture), and MPT (Ministry of Posts and Telecommunications).
MITI has had one project (funded through FED) in high temperature superconducting electronics. The focus was on 3-terminal devices, and it was slated to end in March 1998. Its second project is Hybrid Systems. A decision will be made in 1998 how to continue the project (not necessarily at FED, we learned later -- see the FED site visit report) but almost certainly it will be HTS electronics with emphasis on single flux quantum (SFQ) logic.
There is an ongoing and serious discussion of the styles to be used in these collaborative projects. The government seems to want to centralize the new electronics project (similar to the ISTEC style of the past 10 years for materials research), but industry wants it distributed, as in the previous FED project. There is concern that should it be centralized, industry might stop all superconductivity work.
MITI considers low temperature superconducting electronics a failure since no prototype came out of the Supercomputer for Science and Technology (Josephson Computer) project that was carried out at ETL, Fujitsu, Hitachi, and NEC from 1981 to 1990. (That project is very highly regarded in the United States, as it established the LTS Nb trilayer process as a highly reliable technology.) Therefore, MITI will focus only on HTS and let researchers in LTS, who are largely still in industry, obtain other sources of funding. A possible source of this funding is STA, but traditionally STA has only funded research in STA national laboratories and universities.
This discussion illustrated that there now appears to be a mixing of roles and functions in Japanese government R&D funding organizations. STA, normally a supporter of basic research in STA national laboratories and universities, is moving toward more applied research and might fund work in the laboratories of industry, while MITI is moving toward supporting the longer-range research in a central superconductivity laboratory, ISTEC. Almost all MITI-funded work in digital will be SFQ, thus agreeing with IBM in its 1983 decision to end work on latching logic.
MITI, however, has not been able to define a system orientation for its activities. SFQ appears to be the only technology that might be useful in satellite telecommunications. There is also talk about a "balloon" solution, in which a set of semistationary balloons would be launched around the globe.
In keeping with comments the WTEC panelists heard throughout their week in Japan, Prof. Okabe suggested that NTT is depressing interest in passive applications in Japan. Although there are about 5 providers of cellular service in Japan, NTT still determines technical directions and policy. NTT takes the position that cryogenic systems will not be reliable enough for long-term operation and has pulled back its R&D activities. (See the Saitama University site visit report.)
There are 3 companies in Japan engaged in SQUID work: Shimadzu, Sumitomo, and Seiko. Professor Ogawa feels that technology transfer is very poor in Japan, and that the high level of technology developed within the Superconducting Sensor Laboratory (SSL) has not been fully utilized by industry. There are no start-up companies as in the United States to pick up the technology. (Prof. Okabe was not fully aware of the activities at the Kanazawa Institute of Technology, as described in Chapter 6 of this report on SQUIDs)
Prof. Okabe believes there are strong superconductivity science programs in Japan. Many are focused on epitaxial growth, doping, and substitution in the wide variety of HTS materials. He also feels that ISTEC does not have the ability to distribute enough single crystal materials to satisfy the needs of university projects.
There is another change, in that a few years ago Japanese could not order "chips" from outside Japan (say from Hypres), whereas now they can.
Prof. Okabe summarized the current funding situation for superconductivity as follows. The support by MITI for all of their superconductivity work is ¥11 billion, but of this, ¥3 billion is for microgravity experiments (which includes some testing of superconducting materials). The allocation is ¥9 billion for basic work, including ¥3 billion for electronics. The electronics work is mostly in ISTEC (85%); FED accounts for 15%.
ISTEC is losing industrial (management) support, since industry researchers often go on assignment to ISTEC and are then reluctant to go back to their companies. So industry has lost good people to ISTEC. Company managers feel that ISTEC is not doing enough training, as there is no formal education in materials science and superconductivity at ISTEC. Also, due to Japan's economic problems, there is now less interest in industry in supporting basic research, whether in their own laboratories, at ISTEC, or in universities. This is causing some major shifts in the traditional ways Japan has done business. Industry is shifting even more to applied research. The Japanese government is picking up support of basic research, and that research is moving from industry to the campuses. University professors will be able to consult for industry as of April 1997. This is a major shift from past practices.
The number of graduate students is decreasing, because they prefer software rather than hardware. The previous maximum number of graduate students was 7. There are now 4 graduate students including 3 PhD students: one for film and device work, another for circuit design, and the third for the MEG inverse problem.
Low Tc: New concepts of SFQ logic, and SQUID applications.
HTS: Also SFQ, and some trials to make reproducible Josephson junctions.
Monbusho grants of about $100,000/year stopped a few years ago. Monbusho grants of about $20,000/year are continuing. Support from FED, which originates from MITI, is about $40,000/year. Support from SRL is about $30,000/year; to this Monbusho adds $15,000/year.
Support from STA of about $100,000/year was to be started in the next fiscal year. This program is a new initiative at STA, which will support an LTS SFQ project being organized by NEC, Hitachi, Toshiba, Fujitsu, and several universities.
MITI is planning another new funding initiative: a project on HTS SFQ will start in FY 1998 that will sustain superconducting activities. Prof. Okabe expressed the view that this new initiative will excite the Japanese superconductive electronics R&D community.
Total support from industry for Prof. Okabe's lab is estimated at about $10,000/year. This comes in the form of "donations," and the purposes are not well specified. The amount for each company is less than $5,000/year. Donations from NEC and Fujitsu will be related to superconductivity.
Prof. Okabe reported that most of his graduate students end up in the LSI field because companies recognize the skills they have developed in a wide variety of areas: process, device, design, circuits, etc. Some who graduated more than ten years ago joined superconductivity groups, but most of them are working in the semiconductor field now.