INFRASTRUCTURE

The infrastructure that is used to support and direct MEMS in Japan is different from that in the United States. The most striking is the ten-year commitment by MITI to a very ambitious program to develop enabling technologies. A further goal is to establish a commercial base for the production of MEMS.

While target funding for the total program is 25 billion, only 2.43 billion have been actually allocated to the program in its first three years. The funding is directly linked to tax revenues. It was noted that Japanese companies participating in the MITI program have substantial internally-funded R&D activities in this area. AIST estimated that of the 270 researchers working on micromachines in 23 participating companies, 80 are funded by their respective companies. One company estimated its expenditures to be in the neighborhood of 200-300 million per year.

The structure of the MITI program is shown in Figure 7.8. As indicated in the figure, Three national laboratories are funded directly by AIST under this program: the Electro-Technology Laboratory (ETL), the Mechanical Engineering Laboratory (MEL), and the National Research Laboratory of Metrology (NRLM).


Figure 7.6. Outer duct of the potential microcatheter.


Figure 7.7. Inner duct of the potential microcatheter.


Figure 7.8. Structure of the MITI program.

Industrial participation in the program is managed through the New Energy and Industrial Technology Development Organization (NEDO) under AIST. NEDO in turn contracts with the Micromachine Center (MMC) for management of the individual research projects. Each member of MMC pays a membership fee that is used to pay MMC's overhead and to fund the non-Japanese companies.

There are four types of membership: "Research Supporting," "Group Supporting," "General Supporting," and "Special Supporting." Current membership in each of these categories is shown in the MMC site report (Appendix C, p. 159); research supporting membership is shown in Figure 7.9.

The research supporting members consist of those companies that are funded directly by MMC to perform research under the MITI-funded program. Group supporting members consist of two Japanese industrial associations that are interested in the research performed by the center, and the four non-Japanese institutions that perform research for the MITI project. The general supporting members consist of companies that have access to all information developed through the center, but do not have any intellectual property rights in the research results developed with NEDO funds. Membership in this latter category is open, and new members can be added at any time. Membership in the research and group supporting categories is closed, since the terms of membership were determined by MITI, on an individual basis with each company, at the beginning of the project. The fourth type of member, special supporting member, consists of the banking institutions that handle funds for the center. The fees paid by the members become discretionary funds for MMC.


Figure 7.9. Micromachine Center (MMC) - research supporting membership.

In addition to the national (MITI) project, MMC also administers an independent R&D program using the funds provided by its members. MMC as a private concern is considering supporting university research, which has regulatory constraints for government entities. Most university research is funded through the Ministry of Education.

The MMC is currently staffed by three assignees from industry and one from national laboratories. They serve a term of two years and then return to their institutions. There is little concern over conflict of interest or favoritism by the MMC staff. The objectives of MMC are to: (1) establish the technology of micromachining; (2) disseminate the technology to industry; (3) help industry with the technology; and (4) foster international collaboration in the area of micromachine technology.

MITI and MMC both worked hard to encourage participation by overseas researchers in the national project. However, a number of U.S. firms that applied to be research supporting members (funded to do research) were not able to arrive at mutually satisfactory agreements. Under MITI contracting rules, for example, 50 percent of all royalties resulting from research belong to the Japanese government. Furthermore, all funded participants are required to share some background information with the other participants. AIST is currently reviewing its contracting procedures to find ways to encourage greater international participation in MITI programs. While MMC-funded work is considered precompetitive, foreground rights are 50 percent owned by NEDO. Other Japanese companies that pay taxes may also have access to this information. However, since they lack practical experience, the members do not believe that they can work effectively. All of the equipment purchased with MITI funds is owned by NEDO and "is returned to the government when the project is complete."

The project is designed so that the first five years will be used to investigate its research requirements. During this phase, changes to the project can be proposed to MITI. After the first five years, the project will be evaluated. MMC will use an external evaluation team that will be chaired by a representative from industry and will consist of representatives from universities, industry (producers and suppliers), and government. NEDO, AIST, and the Technology Committee of MMC will also review the project. AIST will not only evaluate the program's technical goals, but also the project's effectiveness in attaining national goals. MMC's schedule for implementation and evaluation of its program is shown in Figure 7.10.


Figure 7.10. MMC program schedule.

AIST evaluates projects on a two-tier basis: (1) management/supervisory -- whether there was an appropriate use of tax monies, a budget analysis, and so forth; and (2) technical evaluation -- whether goals were achieved. AIST is also considering another metric for this program -- methodology/metrology for evaluation. This would address the most fundamental question of what constitutes success on the micron level.

The national micromachine program is structured to develop the base technology for MEMS. It has a very wide definition of MEMS, with only 15 to 20 percent of the research in silicon IC technology. The program relies on the companies to determine any real applications for the technology and to develop commercial products for that market. The program ensures that there is a focus project, and that the project is broad enough to explore many avenues of the MEMS arena. MITI encourages the start of a program before the final goal is really well understood or in some cases well defined. MITI appears to rely on revisiting those goals during the first half of the program, and refining the objectives as more knowledge and information is obtained.

The communication infrastructure in Japan is excellent, both in its formal and informal aspects. It appears that the effort made by MITI to involve as large a number of institutions and professionals as possible in its programs has fostered a very effective informal network.

Japanese prefectures are also very active in forming and funding institutions (e.g., KAST) that support work in MEMS. These institutions actively allow companies to work there using equipment developed by other companies. An example is the microcar that was built at KAST on a Toshiba machine by a Nippondenso employee (Teshigahara, Hisanaga, and Hattori 1992).

Universities in Japan do not provide stipends to graduate students to the extent that U.S. universities do. Japanese universities encourage companies to send their engineers and scientists to the universities to do graduate work. In most cases, the scientists and engineers work on company-related issues. The United States has very limited situations where this is done.

The Japanese national university system is in the process of establishing Nagoya as a center for micromachining. This will cause the reorganization of departments, resulting in a strong focus on micromachining. Thus, there are at least two Japanese government agencies (MITI and the Ministry of Education) involved in funding MEMS research at a national level, in addition to efforts by local prefectures and private companies.

Similarly, several different U.S. government agencies have recognized the importance of MEMS and are involved in funding this field. The National Science Foundation (NSF) was instrumental in advancing MEMS by establishing an Industry/University Cooperative Research Center (I/UCRC) at the University of California/Berkeley. The mission of this center is to develop a science and engineering base for microsensors, microactuators, mechanical microstructures and microdynamic systems. NSF has also provided support for student attendance at conferences. The NSF's I/UCRC program may be funded by both government and industry in a membership format. In most cases, there are no long-term commitments by companies, and typically funding is reviewed on an annual basis. Other government agencies, including the Advance Research Projects Agency, National Institutes of Health, and Department of Defense, have supported MEMS work at universities to support their specific needs, and have also supported student attendance at conferences. Through these and other activities, an infrastructure has been established in the United States to exchange information, develop generic techniques for MEMS, and train students. Coordination of U.S. government efforts in MEMS comes about through informal networking of program managers at individual agencies.

Most of the effort in the United States has been directed toward IC silicon-based MEMS and some LIGA. There has been no effort in what the Japanese would call "milli" or conventional machining.

The professional engineering societies have been extremely supportive of MEMS work. They established the first workshop on solid state sensors and actuators, and held the first MEMS conference. Both of these conferences have become ongoing activities (involving students and researchers from the United States, Japan, and other countries) that allow for exchange of information among those active in the MEMS field. Many of those first students are now active professionals in the field.

Government agencies have been very active in promoting cooperative research between industry and universities. The I/UCRC program requires that the bulk of funding comes from outside of NSF. Conversion monies, those redirected from national defense, require that a commercial goal be established. MEMS in the United States has been very product oriented. Products have come from large companies and small companies, as well. The concept of a foundry system is receiving government support as a way to expedite the availability of facilities to more universities and companies. This should allow for faster development of concepts and devices.

The other area being supported by the U.S. government is the CAD area for MEMS. To be able to do MEMS modeling requires the ability to do both mechanical and electrical simulation. The present quality of MEMS modeling is one of the major hindrances, for all involved, to fast commercialization of MEMS devices.

REFERENCES

Higuchi, T., Y. Yamagata, and K. Kudoh. 1990. "Precise Positioning Mechanisms Utilizing Rapid Deformations of Piezoelectric Elements." In Proc. IEEE MEMS 1990 Workshop, pp. 222-226.

Morikawa, T., Y. Nonomura, K. Tsukuda, M. Takeuchi, A. Honsono, and M. Kawai. 1993. "3-Dimensional Piezoresistive FEM Analysis for a New Combustion Pressure Sensor." In Transducers '93, pp. 598-601.

Sampsell, J.B. 1993. "The Digital Micromirror Device and its Application to Projection Displays." In Transducers '93, pp. 24-27.

Teshigahara, A., M. Hisanaga, and T. Hattori. 1992. "Fabrication of a Shell Body Microcar." In Proc. Third International Symposium on Micro Machine and Human Science, pp. 137-141.


Published: September 1994; WTEC Hyper-Librarian