P/2410 USA

Peaceful Uses of Fusion

By Edward Teller *

INTRODUCTORY REMARKS ticles involved in that explosion will necessarily move It is a great pleasure to discuss with you the topic with high velocity. of the day and I should like to begin by expressing The second and more important difference is the the same sentiments with which the previous speaker following: The fission chain reaction is regulated finished his paper. It is wonderful that over a large by assembling the correct amounts of active materials and important area of research we can now all talk and absorbers. If the correct amount is exceeded, and work together freely. I hope that this spirit the energy release rises in a rather slow manner due of cooperation will endure, that it will be generally to the action of delayed neutrons. In contrast, the exercised throughout the world in this field and that main regulating feature in a fusion reaction is the it will be extended also to other fields. temperature. When a certain ignition temperature has been exceeded, the reaction itself raises the It is remarkable how closely parallel the develop- temperature and this leads to an exceedingly rapid ments in the different countries are and this, of course, acceleration of further energy release. Slow energy is due to the fact that we all live in the same world release is possible only at low densities ; but this raises and obey the same laws of nature. I was particularly further problems. impressed by the wealth of detail given by the previous speaker. I should like to tell you a few generalities and, perhaps, a few little details which might help you HISTORY or might amuse you. Additional details can, of course, be found in other papers submitted to this The problem of controlled fusion is difficult but not conference. necessarily insoluble. Some of us discussed it in a rather detailed manner during the war in Los Alamos. FUSION AND FISSION This group included Enrico Fermi, John von Neumann, James Tuck and Luis Alvarez. The problem of how to release fusion energy in an explosion process was solved several years ago. Some elementary general facts were recognized at How to release this energy in a slow and controlled that time : That deuterium gas could react t above manner has proved to be a much harder question. an ignition point of approximately 35 kilovolts; that This situation is in sharp contrast with the history of deuterium-tritium mixtures could react î at a con- energy release from fission. One year of intensive siderably lower temperature of a few kilovolts; that the gas should be introduced at an exceedingly low work had sufficed to produce the first nuclear reactor 14 15 3 in the early winter of 1942, while several more years density of approximately 10 to 10 particles/cm were needed to perform the first successful nuclear in order to make the reaction rates sufficiently slow detonation. and in order to avoid excessively high pressures; that at these high temperatures the gas will be Among the reasons for this difference I should like completely ionized (such an ionized gas is called a to mention two which are simple and important. plasma) ; that with the presence of magnetic fields, the Fission energy is released by the mechanism of a ions in this plasma follow spiral paths and that by chain reaction. The chain-carrying neutrons can be appropriate arrangement of the magnetic fields the slowed easily. This facilitates a controlled reaction losses to the walls can be reduced; that the pressure not only because the time that the process takes of the plasma on the field leads to a thermal expansion becomes longer but also because slow neutrons are of the plasma which tends to stabilize the reaction; more easily controlled by specific absorbers and other that equilibrium with radiation is not established and means. On the other hand, release of fusion energy that the energy emitted as bremsstrahlung should be is accomplished by a heat explosion or, to use the treated as a loss ; that for this reason atoms other than technical term, a thermonuclear reaction. The par- t The reactions areD + D->T + H Mev and * University of California Radiation Laboratory, Livermore, D + D -> He3 + n + 3.25 Mev. California. t The reaction is D + T -* He3 + n + 18 Mev. 27 28 SESSION 4 P/2410 E. TELLER hydrogen isotopes must be eliminated as completely Laboratory, at the Los Alamos Scientific Laboratory as possible; and that even the equilibrium between and at the Naval Research Laboratory. electron and positive ion energies will not be complete at the highest temperatures. Very particularly, it was also noticed that in a simple closed field along a torus, A COMPARISON OF THE APPROACHES the particles will not continue to spiral indefinitely around the same magnetic line of force but that they Each of these proposals has its peculiar advan- will drift in a direction perpendicular to both the mag- tages and its peculiar difficulties. The pinch effect neticfield an d thefield gradient . This leads to smaller is based on an earlier idea of Willard Bennett \ who but nevertheless prohibitive wall losses. For the time derived the relations according to which a violent gas being, no experimental work was undertaken for the discharge will contract into a narrow filament under specific purpose of carrying the theoretical considera- the influence of its own magnetic field. The great tions into practice. virtue of this scheme lies in the fact that starting from a rather moderate plasma density one can push In 1952 and 1953, under the stimulus of a successful the system to very much higher densities and pressures man-made release of thermonuclear energy, such an and at the same time expose the containing system experimental approach was started. To a very great to much more moderate pressures. One price that extent this was due to the initiative and confidence one has to pay is that the high densities are obtained of the chairman of the Atomic Energy Commission, only for short periods of time. In other words, we Lewis L. Strauss. It was recognized that the prob- are dealing with a pulsed machine. In order to avoid lems of heating and containing the plasma would be instability and breakup of the pinch, the original extremely difficult. For this reason, several parallel simplicity of the concept had to be compromised by approaches were undertaken which, between them, introducing magnetic fields parallel to the direction amounted to an attack on plasma research on a broad of the plasma current. The detailed theory of the front. pinch contraction, as well as the theory of pinch stabilization, was worked out by M. Rosenbluth at Los Alamos. THE PRESENT APPROACHES Rather early in the game it was noticed that the pinch effect in deuterium gives rise to copious neutron Two of these approaches, the pinch and the stella- emission. The first public announcement of this fact rator, were based on the idea of containing the plasma was made by I. V. Kurchatov.2 Unfortunately, these in an annular region. The main difference between neutrons do not appear to be of thermonuclear origin. these two approaches is that in the pinch an axial Considerable progress has been made by S. A. Colgate current is established with magnetic lines of force in explaining the mechanism by which these neutrons encircling the current, while in the stellarator the roles are produced. A general difficulty in the pinch effect of current and magnetic lines are reversed with the seems to arise from the circumstance that during the principal currents established outside the reacting formation of the high-current filament most of the region. In the United States, the work on the pinch original energy output is lost to the walls. It seems effect was started by James Tuck in Los Alamos. to be of the greatest importance to understand and Later, similar work was undertaken in the University prevent these losses. of California Radiation Laboratory by W. R. Baker The Astron is based on the following observation: and S. A. Colgate and still later at other sites. This In contrast to the instability of the pinch current, it is approach has also been very strongly emphasized by expected that a current of relativistic electrons will be workers in the United Kingdom and in the USSR. stable. If this proves to be correct then one can con- Nicholas С Christofilos proposed an ingenious steady- fine a plasma in the field which encircles this rela- state arrangement, the Astron, which bears some tivistic current for a longer period than in an unstable similarity to the pinch effect. The stellarator was pinch. This arrangement is under development at proposed by Lyman Spitzer at the very beginning of Livermore. the thermonuclear program and work on it was The thorough theoretical studies of the stellarator, carried out in Princeton. carried out by L. Spitzer, M. Kruskal, E. Frieman Taking an entirely different approach, the advan- and others, led to the recognition that the simplest tages of a confinement geometry based on externally arrangements of magnetic lines are unstable in this generated, cylindrically symmetric magnetic fields case also. In order to prevent these instabilities, were recognized by Richard Post and Herbert York. additional magnetic fields are being introduced in a Starting with these concepts, Post evolved an approach direction perpendicular to the original fields. Thus which has come to be called the " Mirror Machine ", the present pinch and stellarator designs are no longer which uses various applications of the magnetic as sharply different as they used to be. However, the mirror principle in its operation. In the magnetic stellarator continues to be a device in which steady- mirror machine the ions spiral along magnetic lines state operation will be attempted and in which the between two regions in which the magnetic field is main magnetic fields will originate in a region outside more intense and where the ions are reflected. This the plasma. The problem of heating the plasma in kind of machine is now under study at the Radiation the stellarator has turned out to be a rather difficult Laboratory at Livermore, at the Oak Ridge National one. The first step in the heating process is by PEACEFUL USES OF FUSION 29 ohmic current, but soon the conductivity of the The homopolar, conceived by William Baker at plasma becomes high and, instead of uniform heating, Berkeley, is another machine based upon mirror con- runaway electrons are produced with velocities greatly tainment, but with an enhanced containment due to exceeding the Maxwellian values. From this point the large induced rotation of the plasma. Similar on, heating is accomplished by transferring energy work on rotating plasmas (Ixion) is being done by from magnetohydrodynamic motions to the particles Keith Boyer at Los Alamos. Generally speaking, in the plasma. This so-called " magnetic pumping " there are many designs involving a rotating plasma process bears a similarity to the acceleration of cosmic trapped in a mirror machine. This is accomplished rays as described by, Fermi. Unfortunately, in the by using crossed electric and magnetic fields to induce stellar at or, as in the pinch, a very considerable a rotation, with consequent enhancement of the mirror fraction of the energy is lost during the heating process. action. The heating derived from hydromagnetic shocks has There is one additional design which has received been investigated by H. Grad at New York University, serious consideration but on which little experi- M. Rosenbluth at Los Alamos and others. mental work has proceeded. The name of this design In the evolution of the mirror machine the basic is the picket fence or cusp geometry considered by simplicity of the original design has been maintained. J. Tuck and H. Grad. It consists of a series of mag- In addition to this advantage, the straight geometry netic field systems so arranged that the magnetic makes it possible to operate with a relatively small fields curve at each point away from the space which volume. An obvious disadvantage of the magnetic is enclosed and which contains the plasma. At the mirror machine is the loss of particles through the juncture of any two of these systems, the magnetic mirrors. In fact, all particles get lost whose motion fields have a cusp through which the plasma can includes a sufficiently small angle with the magnetic escape. The difficulties connected with the rate of lines. In the course of time, more and more particles this escape have been the reason for not pursuing this will be scattered into these orbits and eventually all particular design any further. It is by no means clear, of the plasma escapes. Detailed calculations have however, that such a design could not be successful. shown that escape through the mirror is a serious In later sessions a more detailed discussion will be consideration but that it does not rule out the mirror given of the several designs. Here I should like to machine as an effective thermonuclear energy genera- mention a few general difficulties and to discuss the tor. Furthermore, the end losses can in principle be way and extent to which these difficulties have been reduced by inducing a rapid rotation of the plasma. overcome. The open ends of the machine also lead to an advan- tage: a simple possibility of injection. If one can introduce a sufficient quantity of high-velocity, SINGLE PARTICLE CONFINEMENT ionized material through these ends, then, in contrast One may consider the plasma from the point of with the other machines, one can circumvent the view of the individual charged particles in a fixed field. problem of heating a plasma which is already present. The first and simplest question is whether or not a The open ends have another advantage: they allow field of magnetic lines can retain ions on their orbits one to study the particles which escape and, therefore, for a sufficiently long time. Since the ions slowly to determine what energy particles must be added to drift in a direction perpendicular to the magnetic lines the plasma. In injecting fast ion beams, difficulties of force and to the gradient of these magnetic lines, it have appeared due to space-charge limitation in the is not clear that an effective confinement is actually injection mechanism. possible. In a simple torus with a longitudinal magnetic field, the charged particles spiral around As has been pointed out by the representative of the closed magnetic lines of force. The inhomogeneity of USSR, we are particularly anxious to inject at high the field causes them to wander off steadily to the energy because then the leak is not so serious. A walls. The stellarator is designed to avoid the diffi- radically different system of injection is being tried at culties associated with the drift of particles out of a the Oak Ridge National Laboratory. This system con- simple torus. One such method, for instance, is to twist sists in bombarding the interior of the mirror machine the torus: with D2+ ions. Inside the mirror machine the mole- cular ions are disassociated in a carbon arc (but we could use a hydrogen arc) and the more strongly curved D+ ions are trapped. Considerable progress has been made in developing this technique and it may lead to the establishment of a steady-state plasma which is suitable for detailed experimental investiga- tion. In any case, a completely satisfactory density This was proposed as early as 1951. and temperature need not be obtained at once by the In the stellarator the lines get twisted, with the injection into the mirror machine because simply result that a line no longer closes but instead returns, strengthening the magnetic fields or moving the mir- after a circuit, into the neighborhood of its starting rors closer together will lead to an adiabatic compres- point. If many circuits are executed, a single mag- sion of the plasma which will increase both the density netic line will have closely covered a toroidal surface. and the temperature. Now an ionized particle, in following this more com- 30 SESSION 4 P/2410 E. TELLER plicated line of force, will be subject to displacements in the electric current ; while if they were perpendicular due to the inhomogeneity of the magnetic field. One to the magnetic lines they would cause the plasma to can show, however, that these displacements will move, which is contrary to our assumption of a static average to zero in good approximation. situation. In the mirror machine, an ion spiraling around a Now the stability of this situation may be decided magnetic line of force performs a quasi-periodic by investigating infinitesimal virtual displacements. motion due to oscillations between reflecting points. In such a displacement, the material of the plasma The periodicity of the orbit is not strict because the may slide along magnetic lines of force but, in any guiding center drifts to neighboring lines of force, as motion transverse to the lines, the plasma and the has been mentioned above. In a cylindrically sym- lines must move together. Furthermore, any expan- metric case, it is evident that the drift perpendicular sion or compression of the plasma accompanying the to the line of force will map out a surface of cylindrical displacement must be considered as adiabatic. If symmetry. But it is not evident where the particles some of these displacements lead to a linear change will drift if there exists a non-symmetric perturbation. of the energy of the plasma then the system is not in The following consideration shows, however, that equilibrium. If all changes depend quadratically on despite such a perturbation the particles do return to the displacement and if all infinitesimal displacements their original line of force with a high degree of are connected with positive energy changes, then the precision. plasma is stable. If some displacements are con- During the periodic motion in a static magnetic nected with negative quadratic terms we expect an field the energy will be strictly conserved. Two unstable equilibrium. Many such instabilities have " adiabatic invariants " are also very accurately pre- been discussed, and the discussion has been carried to served, even though they vary to a small extent. a detailed calculation of the way the particle distribu- These are the magnetic moment of the spiraling par- tion, magnetic fields and currents change during the ticle in a direction along the magnetic line and the displacement. However, the hydromagnetic con- action jp y dq. Herein is the momentum of the ion sideration given above appears to be a simple and (or the electron) along the magnetic line of force general description which in many cases is satis- and dq is an element along the magnetic line on which factory. the guiding center of the spiraling particle happens The first models considered for the pinch machine to move. The essential character of the orbit of a and for the stellarator turned out to be unstable. particle is determined if its energy and magnetic Appropriate modifications of the field distribution moment are known. From this, the action \p\\dq have given rise to reasonable hopes for circumventing follows uniquely. If now the particle were to shift to various troubles. It appears, in a great measure, to be an arbitrary neighboring line of force, conserving its a question of whether these instabilities are of a hydro- energy and magnetic moment, the action integral would magnetic origin or are connected with plasma vibrations in general be different. However, the action integral excited by runaway processes. We are more doubtful must be conserved also. Therefore, a shift to an about the plasma vibrations, which may be a very real arbitrary neighboring line is not possible, but the and more lasting stumbling block. In this kind of guiding center of the particle must remain on a sheet motion, it turns out that the faster the electron the consisting of a single continuous set of magnetic lines longer is the wavelength of the plasma vibration that of force. In this way, the particle will shift around can be excited, and the longer the wavelength the more within the machine, remaining always on a surface on energy it can take up. To summarize the situation very which the action integral is conserved. When the briefly: it has been suspected and partly found that particle arrives back in the original neighborhood it these plasma vibrations have the disagreeable conse- will, as a general rule, actually embrace the same quence that the plasma can rapidly diffuse and there magnetic line from which it had started. While this is a leak in the magnetic confinement. These are some statement is not absolutely correct, it holds with a of the difficulties in the stellarator and pinch devices. very high precision for small Larmor radii. Un- Work is proceeding and there are methods, partly fortunately, the conservation of the adiabatic in- hoped for and partly realized, which will be discussed variants is not valid if the field varies with a suffi- in other sessions. ciently high frequency. Under these conditions, of The general question of whether a moving, steady- course, the energy is not conserved either. state, hydromagnetic system is stable or not has been discussed in some interesting special cases but has not been treated along similarly simple lines as has the INSTABILITIES corresponding static problem. These dynamic effects Let us consider a time-independent distribution of probably play a very important role during the magnetic lines of force in space, and let us also consider constriction of the electric current in the pinch a static distribution of a plasma with vanishing process and during the heating of the plasma in the electrical resistivity. The magnetic lines need not be stellarator. At the same time, the runaway electrons curl-free because we may assume that there are mentioned above probably give rise to plasma vibra- electric currents in the plasma to account for the curl. tions, which also might arise by coupling with the No electric fields will be present : if they were parallel hydromagnetic motions. All these motions can gene- to the magnetic fields they would give rise to a change rate enough disorder to destroy conservation of energy PEACEFUL USES OF FUSION 31

and conservation of the adiabatic invariants for the carried out and the appearance and behavior of single particles which are supposed to be confined in neutrons have led to additional knowledge about the the system. As a result, both plasma and energy can state of the plasma. Thus, our knowledge of plasma be lost to the wall. physics is increasing considerably with each passing We see, therefore, that heating, plasma injection year. In this way, important features of plasma and plasma compression do represent sizeable addi- theory have been experimentally verified. The inter- tional problems. These problems are, however, action between magnetic pressure and hydrodynamic closely connected with the more general question of motions have been observed and transport phenomena confinement. If the latter were solved completely, in the plasma have been investigated. These are then we could take time with the other processes and greatly complicated by the effects of collective probably execute them sufficiently carefully so that motions but, when properly attacked, have yielded the plasma could be brought to the appropriate con- satisfactory verification of our expectations. ditions. The actual fact is, however, that the problem Hydromagnetic motions have been extensively of plasma confinement is only partially solved and, studied, most particularly in observing the formation therefore, this problem and that of the establish- of a pinch. In connection with the same experiment, ment of the right temperatures and densities have to the evolution of instabilities has also been investigated. be handled together. In the stellarator, the mirror machine and the pinch discharge, the importance of impurities has been demonstrated by studies of the influence that the wall EFFECTS OF IMPURITIES materials have on the observed phenomena. In the Another difficulty that is quite generally encountered case of the stellarator, a very ingenious device, the is connected with impurities. The presence of im- divertor, has been developed which harmlessly elimi- purities leads to increased bremsstrahlung radiation and nates that part of the plasma which otherwise would the consequent rapid loss of energy from the plasma. At touch the wall and would give rise to the emission temperatures low compared with the ignition tempera- of disturbing atomic species. The use of the divertor ture, these losses are often aggravated by radiation from has greatly improved the operation of the machine. incompletely ionized atoms. These partially ionized After proper precautions, temperatures of 100 elec- systems are usually not in S aha equilibrium, since of tron volts or more have been reached in the various the four possible processes, collisional ionization, machines. This should be compared with the few radiative electron capture, recombination by triple thousand electron volts needed for the effective collisions and ionization by photons, only the first two burning of a deuterium-tritium mixture. More than occur. This somewhat involved problem has been 100 electron volts have been reached in the stellarator thoroughly investigated by R. F. Post and has led to and still higher temperatures in the mirror machine. the recognition that heating must be carried out Proved temperatures in pinch devices are a little lower rapidly, particularly in the low temperature regions. but, on the other hand, greater densities have been If neutral atoms are knocked off the walls, these obtained. cause an energy loss by an additional mechanism: Perhaps the most important observable variable is They can neutralize fast plasma ions by charge the containment time. In the pinch effect, the time exchange and these fast neutral particles can cross the during which a high density prevails is only a few tens magnetic barrier and reach the wall. Here they can of microseconds. On the other hand, the operation knock out more neutral particles and a loss mechanism of this machine is in least need of protracted confine- can be built up. ment times. In the stellarator, confinement times of a millisecond or more have been established. In the mirror machine, confinement times of several milli- EXPERIMENTAL DEVELOPMENT seconds are not unusual. This is particularly pleasing because of the small size (a few cubic centimeters) The above enumeration of difficulties and problems and high electron temperature (more than 10 kev) is by no means complete. Neither does this theoretical of the plasma. Furthermore, in this machine no discussion do justice to the experimental problems of difficulties have arisen in connection with instabilities machine design and to the diagnostic work which has or disordered motion. I would therefore term this been carried out. The development of extremely quite a hopeful approach. These times have to be large, pulsed capacitor banks for creating the large compared with a period of approximately one second volume, high magnetic fields required, and the opera- in which a successful thermonuclear reaction would tion of large " baked out " vacuum systems for high double the original input energy. plasma purity at pressures as low as 10~~10 mm Hg are major achievements of the experimentalists. The observation of gross current and electric field behavior NEUTRONS and the determination of magnetic field and plasma pressure by means of probes introduced into a system Throughout the development, it has been believed have given valuable data. Plasma densities have that production of true thermonuclear neutrons would been measured by experiments involving absorption be a milestone in the development of a controlled and transmission of electromagnetic waves in the thermonuclear reactor. The observation of such radar range. Spectroscopic observations have been neutrons would not be merely encouraging but would, 32 SESSION 4 P/2410 E. TELLER at the same time, give us a sensitive method of measur- and, when once brought under control, are not likely ing temperature distributions within the machine: to be subjected to dangerous excursions. Therefore, the slightest rise in temperature would give an enorm- they can be operated more safely than fissionreactors . ous rise of neutron production. In " thermonuclear Finally, the interaction of a hot plasma with machines " neutrons have been repeatedly observed. magnetic fields opens up the way to the direct pro- Unfortunately, these neutrons may be due to some duction of electrical energy. This may be of great form of organized motion rather than to the random practical advantage since high-temperature heat ex- thermal agitation. This shows in a most direct changers and many moving parts could be eliminated. manner that we are still very far from a successful Now I have a question: Can all this be done? thermonuclear machine. Furthermore, even when I think we are at a stage similar to the stage at which true thermonuclear neutrons are obtained, their effects flying was about one hundred years ago. There are could at first be obscured by the presence of the some wise people now, as there were at that time, who neutrons due to collective motions. have proved that it cannot be done. I should like to say that those people were perhaps better off because at least they saw the birds. All we can see are the sun THE FUTURE OF THERMONUCLEAR POWER and the stars. The sun produces thermonuclear I should now like to ask: Where are we going? energy by brute force or, what is worse, by sheer I believe that thermonuclear energy generation is inertia. Other people will say that the sun does it possible. Whether it will be in precisely seventeen with the help of infinite patience. I do not think years, as our Chairman predicted three years ago, or any physicist wants to go along either of those direc- at some other time is a matter that I think he will not tions. argue with me and I shall not argue with him. How- ever, I will say this: The problem is not quite easy. I will also say that on the path there may be some PLOWSHARE little flowers to be picked. Plasma physics has This survey should not be concluded without men- importance in the cosmic arrangement of things, as tioning another way in which thermonuclear energy we heard in Professor Alfvén's paper during this can be made to serve peaceful and constructive session. It may have important technical applica- purposes. Our recent investigations have led to the tions other than energy production. I wish I could increasing expectation that thermonuclear explosions know and tell you more about this, and I hope can be brought sufficiently under control to be of help suggestions will come forward; because this needs in earth-moving, in mining and also, conceivably, imagination and I am lacking in imagination. If we even in the production of energy. We can also under- want to shoot for the jackpot, for energy production, take projects which I should like to designate as geo- I think that it can be done, but do not believe that in graphical engineering. We can dig canals, we can this century it will be a thing of practical importance. make harbors, we can open up water reservoirs for This view may be much too conservative and may be better water supply of arid or semi-arid regions. In proved wrong within the coming decade, but I shall this field, people from the USSR have made very early tell you why I believe what I believe. statements which I believe raise very early hope, It is likely that we shall be dealing with an intricate hope which I believe, in principle, to be correct. machine which is inaccessible to human hands because Recently, an underground detonation of a fission of radiation and on which all control and maintenance device in Nevada has demonstrated the feasibility of must proceed by remote control. The irradiation of the confinement of the energy of the device. Further- materials by neutrons and gamma rays will cause the more, this experiment has shown that the bulk of the properties of these materials to change. Surfaces radioactive fission products were entrapped in the bombarded by the bremsstrahlung radiation will get glass formed by the molten rock. The energy in this heated more fiercelytha n is the case in any portion of case was primarily dissipated at a low temperature in our present nuclear reactors. You can operate the vaporizing the relatively large water content of the machine to the extent that this one surface can be rock but, in addition, a large quantity of rock was cooled, the rest of the machine being at a relatively crushed. With the equivalent of one ton of high low temperature. These and other difficulties are explosive we can pulverize more than one hundred likely to make the released energy so costly that an tons of rock. This could be useful in mining and for economic exploitation of controlled thermonuclear the enhancement of the oil flow in large low-specific- reactions may not turn out to be possible before the yield formations. Therefore, as has been the case so end of the 20th century. often in history, a method for destruction has turned Nevertheless, the ultimate goals toward which we in a direction where it can be made useful for the most are working are apt to be highly rewarding. When important and, in the long run, cooperative effort economic thermonuclear energy production becomes between the nations. feasible we shall reap a number of important benefits. In the peaceful use of nuclear explosives, the fusion The fuel of the thermonuclear reactor is cheap and reaction is likely to play a prominent role for two practically inexhaustible so that, if I may put it this reasons. One is that the fusion fuel is exceedingly way, we have deuterium to burn. Thermonuclear cheap. The other reason is that fusion reactions give reactors produce Jess dangerous radioactive materials rise to less radioactive contaminations than do PEACEFUL USES OF FUSION 33 fission reactions. In fact, a careful planning of the the long run, the inherent advantages of controlled fusion explosion can reduce the activity to a point fusion in a steady-state system will probably dominate where its effects are quite small compared with the our planning for the far future. At any rate, in the effects from an equivalent fission reaction. This is direction of controlled explosions, as well as in the particularly important in earth-moving applications great field of plasma physics, I see a very fruitful designed to build harbors, canals, change the distribu- field of research for physicists, for groups of physicists, tion of water or lay bare mineral deposits. for scientists in a whole country and for scientists in There is no difficulty, in principle, of releasing large the whole world. I also believe that one thing which amounts of fusion energy underground, in the con- science and technology can do for us is to draw us all version of this energy into high pressure steam and in closer together into mutual help and into better under- the use of this steam for the production of electricity. standing if we are guided by a search for the things It is, however, important to find a cheap way of con- that can be accomplished by active work, which we fining the explosion in order that the economical communicate to one another and in which we cooperate. exploitation of explosive fusion energy for the pro- In any case, I am convinced that the extraordinary duction of electricity may become possible. It is not and abundant power of fusion energy can be used in unreasonable to expect that explosive fusion energy peaceful pursuits for the benefit of mankind. When can be harnessed for the production of electricity this has been accomplished it may well turn out before the same feat can be accomplished in the con- to be one of the most important advances of our age. trolled fusion process. It will also be noticed that both the cheapness of the fuel and the reduction of radioactive contamination are the same factors which REFERENCES also play an important role in our future expectations concerning controlled fusion reactions. 1. W. H. Bennett, Magnetically Self-Focusing Streams, Phys. Rev., 45, 890-7 (1934). Even though the first application of thermonuclear 2. I. V. Kurchatov, On the Possibility of Producing Thermo- power may be by using controlled explosions—and nuclear Reactions in a Gas Discharge, Harwell Conference this would be a very great achievement indeed—in (1956).