Site: Sumitomo Electric Industries, Ltd.
1-1-1, Koya-kita, Itami
Hyogo 664-0016, Japan
Date Visited: 13 October 1998
WTEC Attendees: R. Harris, H. Morishita
Present work at Sumitomo Electric Industries is a partial result of JKTC support of the Superconducting Sensor Laboratory (SSL). The SSL project ran from March 1990 to March 1996, at a total cost of ¥5,700 million, of which ¥3,990 million was from JKTC. JKTC officials describe the project as "advanced biomagnetic field measurement system (SQUID) research and development" which included "studies and development of fundamental technologies related to super high-sensitive magnetic field measurement of living body (brain) activity non-invasively, and analysis and image-processing of brain information." The project was a joint effort amongst Hitachi, Yokogawa Electric, Sumitomo Electric Industries and seven other companies.
Mr. Itozaki gave a complete explanation of the history of Japanese funding for superconducting electronics since the discovery of high Tc superconductivity. Much of this history is in the previous WTEC report on electronic applications of superconductivity in Japan (Rowell et al. 1998) and is not reported here.
At the end of the SSL project the equipment, primarily the large multichannel SQUID system, was donated to Tokyo Denki University for a project under Professor Kotani, who has since become President of the University. The fabrication line is still in the buildings constructed for the SSL. Tokyo Denki University has loaned it to ISTEC-SRL. It is supported by some new NEDO funding.
After the SSL project ended, the researchers returned to their home companies, which are now working separately and competitively to make commercial SQUID systems. Tokogawa, Hitachi, Daikin, Shimadzu and the Seiko Institute are working on low Tc SQUID systems. The latter two are developing applications in non-destructive evaluation. Commercial systems have been announced by Yokogawa and Shimadzu in 150 and 20 channel varieties. The Yokogawa work takes advantage of a joint research project with the Kanazawa Technical Institute (KIT).
The SSL still exists as a patent controlling company using 0.1 staff-years of administrative support. For the first two years, administration was handled by Hitachi. For the next two years Sumitomo will be responsible.
Originally SSL was funded 60% from JKTC and 40% from member companies. Most of SSL's assets were in its buildings, which have been donated to KIT. The remaining SSL funds came from its sale of some equipment to participating companies and from license royalties. Since there is a small royalty stream to SSL, JKTC has requested that all its original funding be returned, even though there was no original expectation in the view of the participating companies. Some companies are very upset with this request.
A projection of SSL's continuing income has been made which shows royalty funding asymptotically approaching ¥100 million per year and a payback period of 15 years. At present the royalty stream is ¥1.2 million per year. This is highly unusual among the completed JKTC projects. Only a few of the 50 or so projects exhibit any royalty stream. Nevertheless, there is no guarantee that the projected royalties for SSL will materialize.
During the SSL funding period only Sumitomo was interested in HTS SQUIDs, and it performed only about 10% of the total work. The HTS work was physically located at Sumitomo. At the end of the SSL project, Sumitomo bought all the equipment at its site.
The SSL did go through the mid-term and final evaluations. Each evaluation team was made up of four to five professors, and they were different for the two evaluations. The researchers did not know the members of the team until the actual "hearing." There was quite a buzz as the staff recognized the team members. WTEC did not ask the identity of any team members, and the information was not volunteered.
SSL passed the evaluations easily. It was very successful. Its goals were not just research but to construct a building and to demonstrate equipment. Both the building and the working 200 channel instrument, as projected, were tangible signs of success. Subsequently, even a royalty stream has resulted, although small.
Supporting the royalty stream, Sumitomo already has modest sales. Moreover, about ¥40 million comes from Daikin, Hitachi, and Shimadzu, the latter two of which have sold improvements of the original system to Tokyo Denki University.
This year or next Sumitomo may sell some systems to hospitals.
Two and a half years after the end of the project, the companies still feel pressure to contribute to the royalty stream. But the biggest pressure is on the people who are trying to market equipment. The WTEC panelists presume this means that if you don't sell anything you aren't providing any royalties at all, so there is no pressure.
Dr. Itozaki was the sole host for the WTEC visit. He received his PhD from Northwestern University in the early 1980s. There are three professional staff members working on this project, Itozaki with a PhD and two MS people. There are one or two technical support staff members. All funding is from Sumitomo.
The entire focus of the Sumitomo program is on HTS SQUIDs, which provide high-resolution measurement of magnetic fields. The use of liquid nitrogen is a big advantage over LTS SQUIDs. Sumitomo is trying to generate a new field of interest, especially in non-destructive evaluation. It is a big field overall. More generally Sumitomo managers have no specific market in mind and know they must make a market.
The simple SQUID system Sumitomo sells is designed to stimulate buyers to create markets. It is hoped the buyers will use the SQUID, not do research on it. The SQUID demonstration system is similar to Mr. SQUID, produced by Conductus in the United States. For example, there is a geographic survey project coming up. WTEC panelists presume this is something like the magnetotellurics work John Clarke once did. In addition, big companies may want to detect iron particles on the production lines for copper wire, an application that may interest Sumitomo itself. Dr. Itozaki explained that the sale of about 20 of the HTS magnetometers was highly important in establishing and maintaining the credibility of the HTS SQUID project within Sumitomo. He has no funding except from Sumitomo.
The Sumitomo SQUIDs have a noise level of 10 fT/Hz1/2 compared with 8 from the University of California at Berkeley. The best commercial niobium SQUIDs are 5-10 fT/Hz1/2, and the best ever is less than one. On the other hand, it is expensive for Sumitomo to make the best HTS SQUIDs, so its simple system is specified at about 1000 fT/Hz1/2. Extremely high sensitivity is very difficult for end users, so for now low sensitivity is better.
The WTEC team was given a complete lab tour beginning with a clean room, about Class 10,000, with two lasers for laser ablation. The lasers were stacked on top of each other. Another clean room contained a third laser. During the SSL the first two lasers were owned by SSL but located at Sumitomo. The third was owned by Sumitomo and kept separated from the other two. Now they are located near the other labs. The Sumitomo group makes its own targets, providing better opportunity for changing composition and materials compared with buying commercial targets.
In addition, a better clean room for lithography is available solely for the HTS SQUID operation. Its contents include a contact printer, an asher and a resist oven.
Sumitomo SQUIDs are made from HoBCO, primarily for historical reasons.
The WTEC team was shown a cardiography system with a 32 SQUID array detector. It has been in operation for three years. Detectors are individually encapsulated (0.5 cm diameter) and plugged into a PC card. The patient lies in a horizontal, three-layer mu-metal tube of approximately one meter in diameter. The SQUIDs are cooled by N2 flowing through an uninsulated plastic tube. Electronics consisted of a rack of control cards for the SQUIDs, coupled to a computer. Dr. Itozaki explained that by using the array researchers have seen anomalous behavior from one part of the heart but not other parts.
Dr. Itozaki explained that the technique for finding iron inclusions in copper wire has not been further developed. To be useful in production, the system would have had to be further engineered with a refrigerator and software for fully automatic operation. No funds were provided by the Sumitomo production facility. Inclusions only occur once or twice a year. When they do, there is chaos, but apparently it is felt that the diagnostic ability made possible by the SQUID system is not essential.
There was no evidence of the 64 channel HTS SQUID mentioned in the Rowell report.
Detection of fine iron particles in high speed scrolled wire by high-Tc SQUID. 1997. IEICE Trans. Electron. Vol. E80-C, no. 10. October: 1247-1251.
High Tc SQUID. Advertising brochure.
High Tc SQUID system for magnetocardiography. Advertising brochure.
Rowell, J.M., M.R. Beasley, R.W. Ralston. 1998. WTEC Panel Report on Electronic Applications of Superconductivity in Japan. International Technology Research Institute, Loyola College, Baltimore, MD. July. NTIS PB98-150139. http://itri.loyola.edu/scel96/.
Sendai International Center. 1998. 16 ch high Tc SQUID MCG system. Biomag 98. August 28th to September 2nd. Sendai, Japan.