Ninth International Conference on Plasma Physics
Baltimore, 1-8 September 1982

Science Council of Japan
Tokyo, Japan

Distinguished Guests, Ladies and Gentlemen:

It is my greatest honour and pleasure to be given the opportunity of delivering the Artsimovich Memorial Lecture at this 9th International Conference on Plasma Physics and Controlled Nuclear Fusion Research, organized by the International Atomic Energy Agency in Baltimore.

First I would like to give some personal reminiscences about Academician Artsimovich, and this can be done best by introducing myself. I am at present serving the entire Japanese scientific community as president of the Science Council of Japan and, in particular, serving the Japanese fusion community as a member of the Fusion Council, under the control of the Japan Atomic Energy Commission. It is already nine years since I retired from the directorship of the Institute of Plasma Physics, Nagoya University. This Institute was established in 1961 and I was chosen as the first director. 1961 was the very year when the 1st International Conference on Plasma Physics and Controlled Nuclear Fusion Research was held by the IAEA in Salzburg, and, the late Academician Lev Andreevich Artsimovich, at the end of the Conference, delivered the famous review address in which he used the word "purgatory" to describe the not necessarily bright state of affairs in fusion research at that time. I was among the somewhat bewildered audience. Three years later I attended another conference, this time the 3rd United Nations Conference on the Peaceful Uses of Atomic Energy at Geneva. This was the occasion of my first personal contact with Academician Artsimovich, and I dared to ask him to allow me to visit the Kurchatov Institute in Moscow. He said he had no power to deal with such matters, but he was extremely kind and introduced me to Mr. Petroshants, Chairman of the State Committee for Atomic Energy. I succeeded in getting a letter of invitation directly from Petroshants. This letter had the magical power of obtaining a visa and an enjoyable tour in Moscow and Leningrad.

I met Academician Artsimovich for the second time in his office at the Kurchatov Institute, and he gave me a very extensive explanation of what was going on as well as a tour of the laboratories. I saw an array of different tokamak models, OGRA and a mirror with Ioffe' bars. The latter topic, the validity of the minimum-B principle, had been a favourite one at the Geneva Conference, and hence naturally became the main subject of our discussion. I learned that the idea of minimum-B had been formulated by Artsimovich many years ago and was contained in his textbook. However, to my great regret, we didn't enter into the details of tokamak performances. In those days the tokamak concept had not yet been fully appreciated. I felt greatly privileged to have personal contact with recognized genius. I found we had much in common in spite of our apparent differences: among other things, we were both deeply concerned with international co-operation and shared a firm intention to contribute to world peace by this means.

Now I must curtail my reminiscences about Academician Artsimovich, although our friendship continued. I want to make some criticisms of our past failures. I said I had failed to recognize the importance of the tokamak concept. I think I was not alone, and shared this opinion with many other plasma scientists. There may be a certain value in critically examining this failure. In the early 1 960s a great variety of magnetic confinement schemes were proposed, each inevitably in its infancy, but each striving eventually to grow into a fusion reactor. I, together with others, cancelled the tokamak entry from the list of proposals simply because its plasma current should sooner or later decay; a fusion reactor based on the tokamak principle would therefore operate intermittently and there was no hope of materials that could withstand the consequent thermal and mechanical fatigue. This disfavour proved to be unjustified in the early 1970s, when the tokamak became the only viable machine that could produce sufficiently hot dense fusion plasmas. This failure to recognize the value of the tokamak in its early phase teaches us that we ought to evaluate ideas in their proper historical perspective. It is worthy of note that the value of the tokamak was recognized as the result of a dramatic international collaboration. After the Novosibirsk Conference in 1 968 a British team with laser diagnosis technology, headed by Peacock, joined the Soviet group working on the T-3 machine, to demonstrate that T-3 actually produced plasmas of temperatures of the order of kilovolts. Now, however, when we can produce fusion plasmas and experiment with them, when we can deal with an almost ideal plasma state which can and must be interpreted in the light of theory, and when we are nearing the stage of actually designing fusion reactors, the story is quite different. We must correct the most obvious defect of the tokamak, the pulsed operation! Can we maintain or sustain the plasma current once induced, or, more directly, can we drive plasma current without using the induction mechanism? An obvious idea is to apply a running RF wave round the torus. A pioneering trial in this direction was made by Matsnura and others at IPP, Nagoya, many years ago using a toroidal machine, Synchromak. But their results did not achieve wide recognition. I believe this was the first experiment with the aim of maintaining toroidal current by RF. More recently, Fisch predicted the feasibility of RF-driven current theoretically, and Wong made preliminary experiments. In 1980 the group at JAERI were able to demonstrate currents of 0.1 kA with RF powers of = 1 kW in their tokamak JFT-2. This experiment was followed by JIPP-T II in Nagoya, by WT-II in Kyoto, by Versator-II at MIT, and the most conspicuous result was obtained in the PLT tokamak at Princeton, 200 kA by 100 kW of LHW. Their results also indicated a limitation; it does not work for denser plasma beyond 1013 cm-3. Another proposal by Okano and others, and independently by Bhadra and others, is to use ICRF in interaction with alpha-particles produced by fusion reactions. A more recent paper by Dawson and Kaw discussed the possibility of utilizing synchrotron radiation emitted by hot magnetized plasma itself. This paper particularly attracted me because it dealt with the positive role of radiation in hot plasma for the first time as far as I am aware. Until now, we have considered radiation only as a necessary evil and neglected its active side.

So much for current-sustainment in tokamaks. I should like now to discuss the low-beta character of tokamaks and similar devices. Artsimovich remarked on several occasions that the low-b approach has an advantage over the high-beta approach in that the plasma structure and behaviour can be observed and understood theoretically in more detail than in the case of dense plasma. Has this ideal been reached? My opinion is that it has not. There are many theories to explain this-and-that particular phenomenon of plasma, but I think we still lack a comprehensive theory that can deal self-consistently with the many different aspects of plasma behavior in terms of a single model. Is a complete revision of our theoretical framework required? I don't know. Probably I am demanding too much. The difficulty might be inherent in the nature of plasma, where many effects of comparable magnitude are competing and where the usual way of understanding the overall behavior in terms of local parameters, such as thermal conductivity and diffusion coefficient, may be invalid because of strong correlations between spatially separated points. It is an urgent task for theoreticians to give solid theoretical foundation to the various semi-empirical scaling laws on the basis of which many future reactors are being designed.

It was a very nice idea to name our conference series by two key terms: plasma physics and nuclear fusion. Actually, almost all fusion research has so far been limited to pure plasma physics, and there has been scarcely any work on nuclear reactions. This was a historical necessity. But since we are now in a mature state with respect to the plasma, it is quite in order to start enquiring about the nuclear reactions, probably in close connection with the plasma aspects. Since fusion reactions bring energetic production of fusion energy. However, we must be extremely careful to avoid contaminating our environment by radioactivity. It was this prudence that led the Japanese fusion community to choose a pure hydrogen machine for JT-60, while other major tokamak projects dared to conduct D-T burning experiments. Now we are entering the phase where radioactivity must be dealt with, and with this in mind the people at IPP, Nagoya, are planning to construct a modest-sized tokamak in which D-T reactions can take place. The gravest problem we are facing in designing fusion reactors is, of course, that of materials exposed to neutron flux. In this kind of problem we are often tempted to learn from the experience obtained during the development of fission technology. I would like to stress, however, that fusion presents far more serious problems than fission because we have to deal with the fact that 80% of the energy released is carried by neutrons, while in fission only a small percentage of energy is carried by neutrons, the greater part being confined in the narrow space of the fuel rods. Materials scientists working on the design of fusion reactors speak about 100 d.p.a., which means that every atom of the material will be knocked out of its position a hundred times during the lifetime of the reactor. The role of nuclear phenomena is far more important in fusion than in fission, and we must study radiation effects in fusion reactors on their own, not in analogy with fission reactors.

Why are we at such a primitive stage of study of the nuclear reaction aspects? I think this is due to our state of mind at the infant stage of fusion research. During that phase we had great freedom of choice among various nuclear reactions between light nuclei, as well as among various confinement schemes. If someone raised an objection against using tritium because of its limited availability, then a fusion scientist would often reply by adopting another reaction, for example D+D. When one asks about the radioactivity nuisance produced by the fusion reactions, the answer may be the reaction p+ 11B, where only non-radioactive nuclei appear on the stage. We had too many options in those rosy, dreaming days, when we saw only the good points of various schemes. As we approach the realization stage, we are forced to choose only one option and thus give up the advantages of options left unused. We are still often tempted to make other choices than the D+T tokamak, but we shall probably continue to adhere to it because it is imperative for us to have as soon as possible a machine that can produce fusion plasma. There is always the reservation that this first fusion reactor may not be the final one.

I have touched upon various errors we have committed during the last quarter century. What was the greatest error? I think this was the ease with which we started. Everyone thought fusion would be realized within a decade or so. Artsimovich introduced the word "purgatory" in his speech in 1961 in order to warn fellow scientists of the difficulty of achieving the final goal of fusion research. I should like to repeat his warning in a more straight forward way. Fusion technology is an extremely difficult art! Again I would like to compare fusion with fission. In contrast, I dare say, fission is an extremely easy technology. Fission of heavy nuclei is liable to occur; it requires no input energy; fission is accompanied by emission of neutrons so that a chain reaction of fission is possible; chain reaction can proceed, increasing or decreasing exponentially according to the design or disposition of materials, and it can be regulated by managing neutron-absorbing rods (control rods); control is facilitated by the presence of delayed neutrons; fission reactors are inherently safe because of the Doppler effect; the impeding component 238U turned out to be useful if transmuted to plutonium; and so on. A rare combination of lucky circumstances! A proof of the ease of operating a fission reactor is provided by the discovery of a natural reactor in the Oklo mine in Africa. Nature was cunning enough to make a fission reactor long before Fermi invented it. My advice is: don't rely on the experience obtained in fission research. Fusion is a different technology.

In the introductory chapter of a recent book on fusion, E. Teller cited Niels Bohr's definition of an expert as a man who, through bitter experience, has learned what should not be done. With this definition Teller announces that we are nearing the stage of expertise. I agree with him as far as plasma physics is concerned, but hesitate to agree in other areas. At any rate, he found another way of expressing the extreme difficulty of fusion technology. In stressing this difficulty, I must be careful not to go too far. Difficulty does not cause despair; difficulty invites courage, particularly when the goal is splendid.

I have the privilege of being born in the same year as Artsimovich, in 1909. According to a newspaper report, he said before an audience of academicians that fusion would not be realized while he was alive. He died soon after this remark, and thus his prophecy came true. But I am still surviving and firmly believe that fusion will be realized, or at least the more important obstacles to fusion will be overcome, mainly through intensive international co-operation, well before I close my eyes.

Thank you for your attention.

Note: Someone remarked that the current induced by ion cyclotron waves was observed and analyzed theoretically very early by Yoshikawa and Yamato on the classical C-Stellarator (Phys. Fluids 9 ( 1966) 1814). I pretend in no way to be exhaustive in literature.

Back to List of Topics