FROM THE HISTORY OF PHYSICS PACS number: 01.65.+g On the history of the research into controlled thermonuclear fusion Contents 1. Shafranov V D "The initial period in the history of nuclear fusion research at the Kurchatov Institute" 835 2. Bondarenko B D "Role played by O A Lavrent'ev in the formulation of the problem and the initiation of research into controlled nuclear fusion in the USSR" 844 3. Goncharov G A "The 50th anniversary of the beginning of research in the USSR on the potential creation of a nuclear fusion reactor" 851 4. "From the Archive of the President, Russian Federation" 859 Physics - Uspekhi 44 (8) 835 - 865 (2001) ©2001 Uspekhi Fizicheskikh Nauk, Russian Academy of Sciences Abstract. The problem of controlled nuclear fusion (CNF) is a colossal scientific and technological challenge on a global scale; enormous teams of scientists in many countries are still trying to solve this problem. 50 years ago, on May 5, 1951, the USSR Council of Ministers Resolution enacted a governmental program, apparently the first in the world, "On conducting research and experimental work to clarify the feasibility of building a magnetic thermonuclear reactor". The three papers below briefly outline the history and sequence of events together with the evolution of ideas that led to the first governmental decisions to carry out the work that would clarify whether the creation of a controlled thermonuclear reactor was feasible, and also summarize the results of the first decades of research. DOI: 10.1070/PU2001v044n08ABEH001068 The initial period in the history of nuclear fusion research at the Kurchatov Institute V D Shafranov 1. Introduction The first period of research into controlled nuclear fusion (CNF) can be defined as the years 1951 - 1975. By the end of this period, tokamaks, i.e. machines with toroidal electric current and strong magnetic field, had become dominant among systems with magnetic plasma confinement. In view of the progress in thermonuclear weapons this research was absolutely top secret, which left a dramatic imprint on its history. We can mention, for example, that the adoption of the official programs on fusion research in the USA and the USSR was stimulated by the President of Argentina having announced an allegedly successful implementation of a controlled fusion reaction in his country. This article discusses the changes in the direction of fusion research at the Kurchatov Institute of Atomic Energy at the early stages, such as a switch from the already planned laboratory-scale toroidal models of a magnetic thermonuclear reactor (MTR) suggested by A D Sakharov and I E Tamm to linear pinches, and then the switch back from pinches to toroidal systems (tokamaks). The article mentions the first international contacts on fusion-related problems that took place prior to the 2nd Geneva Conference on Peaceful Uses of Atomic Energy (1958) where for the first time there were presentations of declassified research that was being conducted at the time in a number of countries. We mention a period of profound pessimism in the possibility of solving the CNF problems (roughly the second half of the 1950s). We discuss the first successes in plasma stabilization in open traps during the initial period of work, which instilled optimism into researchers. This success led to the concept of an 'average' magnetic potential well acting as a stabilizer of magnetohydrodynamic instabilities; the principle also worked in closed systems of plasma confinement (tokamaks and stellarators). The article ends with a brief outline of the strategic approach to participation in the international pilot project for a thermonuclear reactor based on the tokamak principle and certain innovative lines of research at the Russian Research Centre 'Kurchatov Institute'. 2. General remarks Nuclear fusion research began in the mid-20th century primarily in the countries occupied with building hydrogen bomb weaponry. The reason was simple: these were the countries with accumulated knowledge and experience of managing and implementing the high-cost projects required for this type of research. As for the unheard-of speed of organizing the research work, this is explained by the following two factors: (1) The original purpose of designing nuclear fusion reactors with deuterium plasma was primarily the generation of bomb-grade materials (charges) for thermonuclear weapons. Both in the USSR and the USA, the paramount stimulus for making a decision on launching the CNF project was a determination not to be left behind by the opposition in becoming armed with the most powerful weapons. (2) Successes in designing thermonuclear bombs led to confidence in a similarly fast solution to the problem of designing a nuclear fusion reactor. This hope proved too optimistic; however, a reorientation of the program to the generation of electric energy using the inexhaustible and ecologically clean source of fusion reactions became a driving stimulus for attacking the CNF problem the world over. The 25 years in the history of the nuclear fusion research at the Kurchatov Institute that we are to describe cover the years from 1951 to 1975. It should be emphasized that as early as 1955 H Bhabha who chaired the First International Conference on Peaceful Uses of Atomic Energy dared to predict at its opening that "a method of controlled release of the energy of nuclear fusion would be discovered in the next 20 years", i.e. by 1975. In a certain sense, this prediction proved correct. By this time, a plasma at a temperature of the order of 1 keV was created on the T-3 tokamak and the modified T-4 device (1968 - 1969). At the beginning of the 1970s, a decisive turn to tokamaks occurred in many laboratories working on the magnetic plasma confinement. At the Kurchatov Institute, the year 1975 ended with the launch of a sufficiently large, for the time, T-10 tokamak (still operating today). A plasma with an ion temperature of 7 to 8 keV was generated somewhat later (1978) on the PLT tokamak of the same generation (launched the same year in Princeton) and using the injection of a fast deuterium atomic beam. For us, this initial period is mostly connected with the name of L A Artsimovich, the first head of the state program in CNF research; he died in 1973. At its early stages, the thermonuclear research was strictly classified even after its orientation was switched from support of military programs to peaceful uses of atomic energy. Inside the guarded territory of the Laboratory of Measuring Instruments of the USSR Academy of Sciences (LIPAN was the code name of the future Kurchatov Institute), only a small group of senior researchers knew what was going on in the new building of the Bureau of Electric Instruments (BÉP) which was situated not far from the building of the Department of Electric Equipment (OÉA) where L A Artsimovich orchestrated the techniques of electromagnetic isotope separation for the manufacturing of fissile material for atomic bombs. Even the top secret reports used for some time such misleading terms as 'goo' (designating the plasma), 'altitude' (temperature), 'jet' (magnetic field), etc. As a result, the phrase 'high-temperature plasma in a magnetic field' was coded by the word combination 'high-altitude goo in a jet'. All participants of the project took pride in advancing to the magnificent goal: the generation of energy 'out of water' (the potential resources of energy inherent in deuterons in 1 litre of water is three hundred times greater than in 1 litre of petrol). We awaited with impatience any information on CNF work from other countries, first of all from Britain and the USA. Curiously enough, scientists in each of the first three countries starting their CNF research based on a closed toroidal system discovered their own approaches to magnetic plasma confinement. Experiments with toroidal gas discharge in the UK created a field of 'toroidal pinches with reversed toroidal magnetic field', abbreviated to RFP (reversed field pinches). Currently, large machines of this type exist, one in Padova, Italy, and another in Boston, USA. The proposal made by A D Sakharov and I E Tamm for a 'magnetic thermonuclear reactor' led to 'tokamak' systems, which grew to dominate the world program of CNF research. L Spitzer invented the closed system of magnetic confinement with nested magnetic plasma surfaces in which each magnetic line of force extends along the system (the topological torus) while rotating by a certain angle ('rotational transform') and covers the whole closed toroidal surface; this approach generated the fundamental research field of steady-state 'stellarators' or 'helical' systems of magnetic plasma confinement. These systems suffered from delays in their progress owing to their greater complexity and unsuccessful experiments at the initial stage in their history. At the present moment they have gained a 'second life' alongside the traditional approach; the largest contemporary representative nowadays is the largest helical system, the LHD (Large Helical Device), in Japan. 'Advanced helical systems' are being developed as well; a 'live' representative is the large stellarator WVII-X that is being constructed in Greifswald (Germany). The stellarator research was started in the Princeton Plasma Physics Laboratory; at present this laboratory is implementing a compact tokamak and an innovative 'quasisymmetric' stellarator NCSX with a self-generated 'bootstrap current' (the current connected with a specific drift trajectory on the torus) which helps to improve the plasma parameters. In addition to closed systems, scientists in the USA and the USSR arrived independently at the concept of open magnetic systems with magnetic mirrors (this is the American term) or magnetic plugs (the Russian term). At the present moment this field has mostly survived in the research towns of Tsukuba in Japan and of Novosibirsk in Russia. 3. CNF research: global status before Geneva-58 Until the 2nd Geneva Conference on Peaceful Uses of Atomic Energy (1958), all CNF research was strictly classified. What follows now is a brief chronology of the relevant significant events in the United Kingdom, the USA and the USSR. Great Britain 1946. G P Thomson (see paper [1]) and M Blackman [2] patented a toroidal nuclear fusion reactor using deuterium. The stated power of the reactor P_DD = 9 MW. The initial plasma heating was produced by a 500-kA alternating current. 1949. First experiments with toroidal discharges (P Thonemann (see Ref. [1]); implementation of the pinch effect at a current of J = 27 kA by S W Cousins and A A Ware [3]). 1955. The idea of stabilizing the discharge with a magnetic field (R J Bickerton) (see paper [1]). April 1956. I V Kurchatov's lecture at Harwell. 1958. A sensation (British newspapers of 25 January; a publication in Nature accompanied with American papers [4]): plasma with an ion temperature T_i ≈ 300 eV (!) was obtained on the large toroidal machine ZETA (the radii of the plasma torus a_p = 0.5 m, R = 1.5 m). This announcement proved to be an error. However, ten years later at the Third IAEA Conference (Novosibirsk, 1968) an announcement was made of the discovery on this machine of a self-organizing quiescent mode with the generation of a magnetic flux in the plasma, which exceeded the initial magnetic flux within the conducting housing: the magnetic flux outside the plasma was negative. Hence the term 'reversed field pinch' (RFP). USA 1945 - 1946. Edward Teller's seminars on CNF. Negative results of experiments with beams (J Tuck, S Ulam) (see book [5]). March 1951. The announcement by Argentina's President Juan Peron of the successful demonstration of controlled thermonuclear reaction by R Richter led L Spitzer to the invention of the stellarator as a solenoid shaped into a 3D figure of 8. May 11, 1951. L Spitzer's proposals are discussed at the Atomic Energy Commission (AEC). July 7, 1951. A contract for a research project is signed with Princeton University (the Matterhorn project). All works on CNF (on pinches at Los Alamos, on the mirror trap at Livermore and others) are merged somewhat later into the single Sherwood project. USSR 1950. O A Lavrent'ev wrote letters to Moscow with, among other things, a description of the idea of realizing controlled fusion of deuterium nuclei in an electrostatic field (sent from Sakhalin Island at the end of July). 1950. A D Sakharov responds to O A Lavrent'ev's proposal (August 18) with a remark that a 'highly reflective grid is required', 'having a thin current-carrying segment', in order to reflect almost all incident nuclei back into the reactor. August - September 1950. The idea of creating high-temperature plasma directly in the magnetic field. A D Sakharov and I E Tamm work on the MTR theory. October - December 1950. Higher echelons of the administration get acquainted with the idea of magnetic plasma confinement. January - February 1951. A sequence of discussions and the preparation of the draft of the governmental resolution for the work on the magnetic thermonuclear reactor (MTR). April 5, 1951. Stalin approved USSR Council of Ministers Order for the creation of a laboratory-scale model of the MTR [see below Goncharov G A Phys. Usp. 44 851 (2001)]. Mid-April 1951. Discussions on the logistics and management of the MTR problem intensify in response to the information received from President Juan Peron's speech (March 25). May 5, 1951. Stalin approved USSR Council of Ministers Resolution on the organization of MTR research (see below the Section "From the Archive of the President, Russian Federation"). 1951 - 1955. Experimental and theoretical research into toroidal and linear discharges; short-lived innovations (such as RF plasma confinement, etc.). 1955. Pilot tokamak (still using a ceramic discharge chamber): torus with a magnetic field (Russ. abbr. TMP). 3.1. 50 years ago May 2001 was thus the 50th anniversary of the official launching of CNF research in the USSR and the USA. The history of this research is both amusing and dramatic. In the USSR, this story began with a letter written by a Red Army sergeant Oleg Lavrent'ev, who was posted at Sakhalin Island, to the Central Committee of the Communist Party of the Soviet Union. The letter suggested creating the hydrogen bomb using the atomic bomb to prime it and even more interestingly, there was a suggestion for the electrostatic confinement of deuterium nuclei for industrial-scale generation of energy; two spherical grids placed under a negative and positive potentials were to be used for the purpose. The letter was sent for review to A D Sakharov who wrote that "the author formulates a very important and not necessarily hopeless problem". He mentioned a number of difficulties in realizing the electrostatic confinement and pointed out that the grid must have "wide meshes and a thin current-carrying part which will have to reflect almost all incident nuclei back into the reactor. In all likelihood, this requirement is incompatible with the mechanical strength of the device". However, the reviewer continued that "it cannot be excluded that certain changes in the project may correct this difficulty". At the end of his review, Sakharov emphasized that regardless of the results of further discussion "at this point, we must not overlook the creative initiative of the author". It is remarkable that the letter was sent from Sakhalin on July 29, 1950 and A D Sakharov's review was signed as soon as August 18, 1950. By that time O A Lavrent'ev, having passed examinations for the last three years of high school and demobilized from the army, had already enrolled in Moscow State University. Lavrent'ev's letter prompted A D Sakharov to the idea of the magnetic thermal insulation of high-temperature plasma. By October, A D Sakharov and I E Tamm completed the first evaluations of a magnetic thermonuclear reactor. In January 1951, the MTR project suggested by A D Sakharov and I E Tamm gained approval and O A Lavrent'ev, who stimulated A D Sakharov to the idea of 'magnetized' plasma, had obtained accelerated graduation from the university with the help of the authorities. Having graduated from Moscow State University, O A Lavrent'ev received an invitation to work, on L A Artsimovich's recommendation, at the Khar'kov Physico-Technical Institute, where he still works. The experiments he was running on electrostatic reflection of the electrons that escape from an axisymmetric 'anti-mirror' trap across an annular magnetic gap caused great interest at the Kurchatov Institute. A special trap which was toroidal (so that it could remove the loss of particles along the axis) got the acronym ATOLL [Anti-mirror TOroidal Lavrent'ev trap] and was built and operated in the 1981 - 1985 by M S Ioffe. 3.2 The idea of magnetic thermal insulation of the plasma As A D Sakharov saw it, the main new feature of O A Lavrentiev's idea was the low density of the confined particles. However, he was not happy with the long ranges of particles, which would inevitably lead to undesirable interactions of high-energy particles with the construction materials. Would it be possible to arrange for the trajectories of freely moving particles not to leave a prescribed volume? It would! A charged particle in a strong magnetic field follows a helix along a magnetic line of force. Therefore, a high-temperature plasma must be created in a toroidal solenoid. If the curvature of the solenoid is neglected, particles will impact the chamber walls only as a result of interparticle collisions — that is, as a result of diffusion across the magnetic field. However, the trajectory of a particle can shift after each collision only by a distance on the order of the Larmor radius (about 1 cm for deuterium ions and less then 1 mm for electrons at B = 50 kG and plasma temperature T_p ~ 50 keV). Therefore, the energy transfer to the construction elements of the reactor is greatly reduced. A D Sakharov discussed this problem with I E Tamm who had just returned from his holidays. Even though very busy at that moment with the work on the hydrogen bomb, they started to discuss the physical problems arising in the new field and to evaluate the parameters of the magnetic thermonuclear reactor (the name was suggested by I E Tamm); the curvature of the plasma torus was neglected. By the end of October 1950, the idea of the MTR was communicated to I V Kurchatov and his young deputy I N Golovin. On his short visits to LIPAN, A D Sakharov discussed a method of eliminating the vertical drift of charged particles in the toroidal magnetic field relative to the torus plane. First he suggested suspending a coil on the chamber axis, carrying a toroidal current whose magnetic field would convert magnetic lines of force into helical lines, thus creating a system of nested toroidal magnetic surfaces. Later he chose to generate such a current in the plasma itself by induction. To sustain the current-carrying plasma ring in equilibrium, he suggested a toroidal copper housing cut in two places: along the torus to allow the introduction of the toroidal magnetic field, and across the torus for the introduction of toroidal emf which would generate and sustain electric current in the plasma. In 1957, this system was given the name 'tokamak'. In January 1951, I V Kurchatov organized a workshop on the MTR with the leading designers of atomic weapons and, having gained their support, began preparations for a governmental resolution on launching the MTR work. In February 1951, the project was forwarded to L P Beriya. 3.3 President Peron on Ronald Richter's 'success' On March 25, 1951, Argentina's President Juan Peron made an announcement that experiments by a German physicist Ronald Richter succeeded in a "controlled release of atomic energy at a superhigh temperature of millions of degrees without using uranium fuel"; Richter was working in a specially created secret laboratory on the Huemul island in Argentina. Having learned from D V Efremov about President Peron's announcement, I V Kurchatov immediately communicated with L P Beriya who urgently convened a panel to discuss organizational aspects on the subject and a draft of a governmental resolution submitted earlier. It was suggested that the CNF research program be headed by L A Artsimovich (who, without resigning from his job of launching a plant for electromagnetic separation of isotopes, was to spend one third of his time on the new problem of controlled nuclear fusion). M A Leontovich was to head the theoretical effort (on I E Tamm's recommendation). A detailed Resolution from the USSR Council of Ministers, which ordered heads of certain industrial plants to meet the demands of the CNF research team, was already signed by I V Stalin on May 5, 1951. In October 1951, A D Sakharov and I E Tamm prepared projects for their research programs for CNF (published later in the collection of articles [7], pp. 3, 20, 31). The parameters of an 'optimal' MTR in Sakharov's calculations for a cylindrical model were: the major and minor radii of the plasma torus were R = 12 m and a_p = 2 m, B = 50 kG, n = 10^14 cm^-3, T = 100 keV, and P_DD = 880 000 kW. According to calculations, this plant could produce up to 100 g of tritium or 80 times that of 233U per day. Sakharov noted at this point that the energy-production value of the 233U which could be burned in a conventional reactor would greatly exceed the heat liberation in the nuclear fusion reactor itself. These remarks clarify that the decisive factor for enacting the decision on the CNF program at the time was the possibility of manufacturing charges for hydrogen or atomic bombs. It must again be emphasized that the curvature of the torus was not taken into account in the reactor calculations. In fact, the toroidal current suggested by Sakharov to compensate for the toroidal drift of charged particles leads to crucial changes in the physics of toroidal plasma confinement. One of them is that the theory of transport phenomena must especially take into account the drift trajectories of charged particles in the presence of the azimuthal component of the magnetic field (the future 'neoclassics' of A A Galeev and R Z Sagdeev!) — this was already mentioned in I E Tamm's paper. 3.4 The road from the idea to its implementation An analysis of closed toroidal systems highlighted the problem of toroidal drift of charged particles. We mentioned above that A D Sakharov suggested two methods to close the drift trajectories of charged particles inside the chamber: (1) To add a poloidal magnetic field created by an internal current ring suspended by cables or by a horizontal magnetic field; (2) To induce a high-frequency current in the plasma itself. The second of these techniques was more realistic and led to experiments with a single-pulse discharge sent from capacitor batteries. At the Kurchatov Institute, the need to inject toroidal current led to a proposal to forgo the toroidal magnetic field completely. The main effort was first concentrated on pinches in which, according to the Bennett relation J^2 = 4c^2NT, the plasma temperature must grow in proportion to the square of the current, T ~ J^2! It seemed that this approach promised a fast resolution of the problem. Only a small group headed by I N Golovin and N A Yavlinskii continued to do research along the ideas of Sakharov and Tamm. 3.5 Rapid discharges Experimenters in the meantime were 'storming' direct discharges but without visible success. It seemed that better vacuum conditions, a change in the scenario of preparing the discharge and so on should lead to success. Finally, on July 4, 1952, counters started to detect neutrons in N V Filippov's group: the deuterium plasma pinch generated neutrons! Hope appeared that modifications of the experimental program might gradually increase the plasma temperature. However, L A Artsimovich's demand that everything be checked extremely thoroughly killed the euphoria. This was soon followed by profound disappointment: the pinch instability did not allow the temperature to rise with increasing current. 3.6 Toroidal discharge stabilized by a strong magnetic field The theory of pinch stabilization by a longitudinal magnetic field again reoriented the studies towards A D Sakharov's suggestion: to use both the toroidal magnetic field and the toroidal current. In 1955, the first tokamak-like machine was built — the TMP. It still had a ceramic discharge chamber with a helical metallic insert. 3.7 Period of pessimism: until the beginning of the 1970s A temperature not exceeding 30 eV was typical for a long time. There was no progress either in pinches or in toroidal systems. Searching for other scenarios of plasma confinement, Gersh I Budker came up with the idea of a direct axisymmetric magnetic system with enhanced magnetic fields at its ends (the 'probkotron'). Some time later I N Golovin decided to focus on this relatively simple option and design OGRA, a large trap with magnetic plugs (mirrors). 3.8 All-Union Conference of 1955 This was the situation in which I V Kurchatov decided that the CNF program needed declassifying. He organized the All-Union conference in 1955 to discuss CNF work. Presentations were made by L A Artsimovich and M A Leontovich. This conference also inspired B B Kadomtsev to join the group. 3.9 First steps to international cooperation on CNF In April 1956, I V Kurchatov, as a member of a Soviet governmental delegation headed by N S Khrushchev, arrived in Great Britain and gave a lecture "On the feasibility of the thermonuclear reaction in gas discharge" at the atomic research center at Harwell. This was the first real step towards international cooperation in CNF. In April 1956, the first delegation from abroad visited the Kurchatov Institute (members of the Swedish Academy of Sciences). As a response gesture, H Alfvén invited L A Artsimovich and I N Golovin to Stockholm in autumn 1956. In June 1957, a large number of papers related to CNF were presented to the conference on "Ionization phenomena in gases" in Venice. 3.10 The ZETA facility and the boom it caused Unexpectedly, British papers announced in January 1958 the sensational news that a temperature of 300 eV had been achieved at Harwell on the ZETA facility. S I Braginskii and V D Shafranov analyzed the reports and concluded it was a tokamak-like system, but the January issue of Nature showed it was a pinch with trapped magnetic field (later understood as reversed field pinch). The announced 300 eV result proved erroneous, but it spurred international momentum. 4. CNF before and after the Geneva Conference of 1958 The Kurchatov Institute's papers on CNF were declassified before the start of the 2nd Geneva Conference (1958) and published in the 4 green volumes 'Plasma Physics and Problems of Controlled Thermonuclear Reactions'. At the conference, L Spitzer's stellarator was the 'star' of the program. However, comparison by Shafranov and Braginskii showed advantages of tokamaks over stellarators at the time due to larger minor radius and higher ohmic heating current. 5. Pessimism defeated 1961. Experiments by M S Ioffe on plasma stabilization in open traps were reported at the 1st IAEA Conference in Salzburg. Kadomtsev explained this stabilization through the concept of a magnetic well, demonstrating that Bohm diffusion was not universal. 1962. Success of tokamaks with vertical magnetic field correction for plasma positioning. 1965. 2nd IAEA Conference in Culham: Artsimovich reported confinement times exceeding the Bohm limit by nearly tenfold. 1968. 3rd IAEA Conference in Novosibirsk: Mean electron temperature <T_e> on T-3 approached 1 keV without Bohm diffusion. Artsimovich invited Culham physicists with laser diagnostics (Thomson scattering) to verify the results. 1969. 2nd workshop on toroidal systems in Dubna: D Robinson confirmed T-3 local electron temperature measurements. Tokamaks achieved international recognition. 1970s. Global redirection of fusion programs toward tokamaks. T-10 in Moscow (1975) and PLT in Princeton achieved multi-keV temperatures. Main results of the first period in CNF history in the Kurchatov Institute: (1) The possibility of confining a high-temperature plasma has been proved. (2) The tokamak as a sufficiently simple system with ohmic heating became a leader that opened the way to CNF in other types of system. Current status of CNF research in the Institute of Nuclear Fusion at RRC 'Kurchatov Institute': (1) Operation of T-10 tokamak (transport barriers). (2) Design of T-15M tokamak. (3) Pinch program revitalization. (4) Investigation of D-3He fuels. (5) Broad theoretical plasma research. 6. Conclusions The first 20 years of CNF were marked by uncertainty, which was dispelled by the tokamak. Controlled thermonuclear reaction feasibility in magnetic confinement is now proven, shifting focus to technological, engineering, and materials challenges (including ITER-FEAT). References include seminal papers by Carruthers, Blackman, Cousins & Ware, Bishop, Gamow, Sakharov & Tamm, Golovin, Bennett, Mukhovatov, and Razumova.