Future National Research Policies Within the Industrialized Nations: Report of a Symposium (1992)

President, Science Council of Japan

Japan is now spending about 3 percent of its GNP on R&D; 80 percent of the money comes from the private sector and 20 percent from government. Of the money spent, 10 percent goes to the universities, and another 10 percent goes to government-established institutions.

These figures indicate, as it is sometimes said, that Japan is spending its money mostly on applied science, and is spending only a very small amount of money on basic science. However, the situation is now changing. One example of this change is that the big steel industries recently established a laboratory for applied biology—for biotechnology. I do not think that microbiology can make steel, but this is the situation, and it shows that when a big industry has the money, it may choose to directly invest the money in basic science. Not all of them do, but the Japanese attitude toward science is now changing.

We established the oldest university in Japan, the University of Tokyo, in 1878 in order to build a focal point for the introduction of Western civilization. When we opened our country in the middle of the last century, we were very much surprised. It was about a century after the Industrial Revolution, and we were surprised at the progress of science and technology in the Western world. Therefore, we needed some point for introducing this progress into our country, so Tokyo University was established.

For a long time Japan had only seven big national universities. But today, we have 460 institutions of higher education, a number that has increased by about 10 times since before the Second World War. This indicates obviously that, after the war, development of higher education in Japan progressed very rapidly, and this is one reason why Japan's economy has expanded so quickly.

The number is now leveling off because the size of the population has become somewhat stable. In the future, because of the decreasing birth rate, we will have some difficulty getting good students at the university level. Furthermore, because the number of public universities has now increased to 99 (whereas before the war, as I said before, we had only seven), the government budget has been diluted in order to allocate to each new university. Therefore, we have the feeling that the university situation in Japan is becoming very bad. We must admit to ourselves the need to establish some centers of excellence among our universities.

How we can cope with this situation? It is rather difficult in Japanese universities to get funds from the private sector, for most universities are government establishments. All professors are government employees, and they are not permitted to receive money from the private sector. They cannot use such money as their own salary, to support their family. Therefore, university professors are not very aggressive about getting funds from the private sector. However, in some particular places that we call centers of excellence—for example, at the University of Tokyo we recently established a research center for advanced science and technology—we can introduce funding from industries. These we call crown professorships. We can attract universities in order to do some particular research, and we can obtain and use private money.

The Ministry of Education has allocated funds to some particular places called institutes for common use. Common use means the university professors can use the facility. The KEK laboratory at Tsukuba is one example. Also, we are planning to build an astronomical observatory in Mauna Keya with a telescope 7.5 meters in diameter, called the Japanese National Large Telescope. It is not possible to distribute such a facility among colleges or universities, but by accumulating a large amount of funds and then setting up the facility in a particular place, we can establish a new center of excellence. Other ministries, such as agriculture, forest and fisheries, would like to establish other institutes as centers of excellence.

In the future, maybe up to the year 2010, what will be some of the main subjects for Japanese scientists? It is very hard to predict the progress of science. However, I think one important trend is research on the environment. I believe that global environmental issues will continue for a long time. Next year, we will have in Brazil a world congress on environment and development. Since the Persian Gulf war, we have been faced with widespread damage to the environment in the Middle East region. We need environmental research on these problems, and we also need to find a way to cope with global warming—a very big issue.

I have been appointed as director-general of a new organization, called RITE (Research Institute for Innovative Technology for the Earth). This institute was established last July; as yet we do not have any laboratories, but they will be located near the city of Kyoto. An example of an idea being explored at RITE is the development of a process that would remove carbon dioxide from exhaust gases and convert it into methanol by adding hydrogen.

We also have another program that involves introducing sunlight by means of optical fibers and using the photosynthesis of microorganisms that we think can absorb carbon dioxide. We have already discovered useful marine microorganisms that can absorb a large amount of carbon dioxide. Another research program will look for a new generation of refrigerant to substitute for CFCs.

If the environment is one direction for future research, energy is another. We are very much interested in the international thermonuclear experimental reactors (ITER), and we will collaborate with the main industrialized countries to build them. Another issue that we are very much concerned about is how to transfer our technology to the developing countries.

Executive Technical Advisor, Sony Corporation; Professor of Electronics, Tokai University

The other speakers at this symposium have given you a bird's-eye view of science and technology policy. I, however, because of my background, cannot give you a bird's-eye view; I can give you only a worm's-eye view that is based on my understanding through my own experience in 40 years of working on semiconductor devices and electronics.

I think it is very important to identify or recognize the historical position of electronic technology right now. Modern electronic technology started from the invention of the transistor in 1947 and in only 20 years created a revolution. The revolution changed almost everything because it involved a new concept, quite different from the concept of vacuum tubes. In recent years, however, the change has slowed down, and we are coming into a pattern of steady evolution.

Therefore, we now have two different tasks. The first is how to contribute to this steady evolution. The second is how to prepare for the next possible breakthrough. Over the next 10 to 15 years, I think three areas will provide important foundation stones for new electronic technologies. These are further progress in very large scale integration (VLSI), optoelectronics (the use of lasers in electronic circuits), and software.

If we break down just the first one, the progress of VLSI, we can identify three different fields. The first is higher capacity: how we can put more and more transistors or elements on one chip. This will be influenced by improved silicon technology and computer-assisted design (CAD) and simulation. A second way to improve VLSI is by developing higher-speed devices, perhaps through new types of semiconductor compounds, such as gallium arsenide. The third avenue for improvement is higher reliability with lower cost; this is related to the further progress in process technologies.

To prepare for future breakthroughs, we must meet some challenges. I believe two are vitally important. The first is to come up with brand new ideas for devices—new concepts. The second is to develop some new or unknown material. For example, when I graduated from university in 1948, semiconductors were an unknown kind of material; now they are really

well known. If we want to talk about future breakthroughs, we must look for unknown materials and try to understand them.

In Japan the next generation of basic technology is being addressed by some programs sponsored by the Ministry for International Trade and Industry (MITI). Topics include superconductive materials and devices, which started in 1988, photoreactive materials, and so on. I would like to mention something here about the mechanism of setting up projects related to MITI, because I feel there is often some overreaction or some misunderstanding among my American and European friends about how these things go.

As an example, I will talk about some of the interesting tasks now being tackled by Japanese scientists and engineers. A symposium on future electronic devices was sponsored by an association that is not purely government, but is, instead, between government and private industry. This symposium was sponsored as well by the Japan Industry Technological Association, another half-government association, and by the Agency of Industrial Science and Technology, again belonging to MITI. That was the ninth symposium held in 1990, for which I gave the keynote address. That organization also sponsored a workshop on future electronic devices in 1989, covering advanced crystal growth, characterization of microstructure, and other fine process technology.

Now I would like to mention my own views. I think the present is a very interesting time in the history of science and technology because feedback from technology to science is taking place. We have such experience— feedback from technology to science—in the history of science. For example, the technology of the telescope opened a new field in astronomy; likewise, the development of the microscope opened a new field in biology. That was feedback from technology to science. In the field of electronics, process technology is now giving tremendous feedback to the physics of new material and structures. One new technology being used in experiments performed in many universities and laboratories is the making of crystals in atomic dimensions by stacking atoms. As you remember, in 1955 Bell Laboratories developed solid-state diffusion, for the first time in history producing control on the order of one micron. But now, with the new technology, you can control on the order of angstroms—you can really control the stacking of atoms.

This makes it possible for us to produce superstructures or superlattices. In this way we are creating semiconductor lasers, high-speed transistors, and MQW devices, and we are trying many other things as well. Moreover, we are now facing an interesting step in the physics of integration. More and more transistors are now put on the same chip, so you can understand that the distance between electrodes is becoming more and more reduced. Indeed, the distance between electrodes has become submicron in these days.

Submicron distances have affected the fundamental current equation for semiconductors. This equation will not hold anymore, for it is based on many collisions over greater distances. This was formerly the ordinary condition, but now electrons can jump directly into the different electrodes, which are closely spaced.

The fundamental mechanism is changing, and the wave nature of electrons is now coming into practice. An example is the scanning tunneling microscope, which lets us see each atom. So, in this way, we are now doing research on so-called nano-devices or mesoscopic devices. It is a fundamental idea, and that is why we are trying to work with it: to take a first half-step into the new physics.

In another area altogether, many of my friends in Japan hope to learn from biosystems. There are many different kinds of expressions, and there are many different kinds of challenges. But the fundamental point of view is that there is a great deal to learn from biosystems. They depend essentially on the nature of biology, but we can transfer fundamental mechanisms to electronic systems. This is the basis for biochips and neurodevices.

These projects are supported fully or partly by the government through three different channels. In many cases MITI does not directly give financial support to someone or to some company. There are classifications of national support, but I am concerned with big-scale R&D-type projects. In these cases, as I mentioned before, we have an intermediate step called an association—between MITI and the working research group. The head of the association comes from MITI—which can be a problem, but if the association can match the technological challenge, it can do some good. It depends on the particular case.

The second source or means of support is the Ministry of Education. In this case the ministry supports a project by directly supporting an individual professor at the university. The third and final method of support is the Agency of Science and Technology, which is more biased toward so-called fundamental research and gives more freedom to the group leader. In most cases, the group leaders are from the university.

The amount of support also depends on the particular situation. In some cases the project is support 100 percent by the government. Such was the case for Next Generation Industrial Technology, begun by MITI in 1981. Another example is the International Human Frontier project, which is very fundamental work actually done in the RIKEN Institute of Chemical Science. The project invites very prominent foreign scientists to head the effort. Other projects, however, are supported at 70 percent or lower, depending on the circumstances. You should also understand that this support by the government is not a large portion of the budget. Here is an example: at Sony's research center, we are talking about optical memory related to next generation basic technology. We have accepted 15 million yen per year for five years. Another area is superstructure devices; for 10 years we have about 15 million yen.

There is a lot of discussion that we Japanese are very strong in applied research but very poor in fundamental research. When my friends discuss this, I raise an objection to the idea. I do not believe that fundamental research and applied research are two distinct concepts. I see them instead as opposite ends of a continuous spectrum. That is my principle; even in presiding at my research center at Sony for 15 years, I treated everything on that spectrum.

An example of this can be seen in the case of the CCD that is used in the video cameras now being sold. CCD projects started in Sony in 1970. The first product came out in 1980, so it took 10 years to complete. In the early days of the CCD projects, we suffered from defects in the final picture, which resulted from crystal defects formed during the oxidation process. Because of this, I pulled the project away from the applied end of the spectrum to the fundamental research end. I let the physicists work on that end for 2 1/2 years to check the fundamental nature of this defect. After that, we could get good results, so I put the project back to the applied research side and returned it to the semiconductor group in Sony.

Therefore, I think we should treat Japan's situation in terms of that continuous spectrum. Right after the war, Japan had an imbalance toward the applied research side of the spectrum because we were very poor. But as our income grew, we began to extend the tail of the applied research toward the fundamental side of the spectrum. It goes slowly, and takes a long time, but I believe it is happening.

Question: It turns out that 80 percent of all R&D in Japan is privately financed. I wonder whether you could tell us something about what it is that Japan is doing to provide private industry with the incentive to support such a large fraction of all R&D spending.

Dr. Kondo: MITI is in charge of the development of Japanese industry. Let me give one example: chips for electronic devices. I think now most people feel that large memory chips are manufactured in Japan. But in the beginning the risk was very high, and, therefore, private industries could not afford to proceed by themselves. MITI organized a kind of association, and in this association the risk was taken by the government. When such an association is successful at good quality production, then MITI pulls back and lets them go.

I can give another, but opposite, example: production of large aircraft in Japan. We have manufactured aircraft successfully, but because the risk is so great, private companies are very much afraid to step in. The government is also very much afraid of the risk. So when it came to building the

next generation of passenger jets, we decided that it might be safer to collaborate with Boeing. This is another example. One is more successful, the other one not successful.

Question: Japanese firms show great interest in supporting foreign research and apparently only a modest interest in supporting research in the academic institutions at home. Does this indicate some continuing lack of confidence in Japanese research capabilities?

Dr. Kondo: I think that sometimes Japan is criticized for "buying brains" outside of our country. However, we do this because Japanese universities are not so cooperative with industry. As I said before, faculty have not been permitted to accept outside funds. But the situation is now changing. I believe that, in the future, Japanese industry will distribute funds not only outside our country but also inside our country.