Table of contents

The objective of nuclear fusion research is to find out whether or not the nuclear energy of the light elements, especially that released when hydrogen isotopes react to form helium, can be put to practical use to provide a virtually limitless source of energy. This research has been conducted on an increasing scale since 1946 when a patent was taken out by Thomson and Blackmann (1946) to produce high temperatures and D-D thermonuclear fusion reactions in a toroidal electric discharge of about 500 kA. Nowadays the research is carried out on a scale of several hundred million dollars a year in each of Western Europe, the U.S.A, Japan and the Soviet Union.

We overview and compare the energy confinement behaviour of tokamaks and stellarators at high heating powers. Tokamaks show in all parameter ranges an anomalous enhancement of electron energy losses. At high heating powers a bifurcation is observed in the plasma energy vs. heating power diagram, with the low energy (L-regime) branch showing a clear deterioration of confinement time with power. Divergent results exist concerning the corresponding behavior of the high energy (H-regime) branch. Experiments with ECR-heating of net-current-free stellarator plasmas can be to a large extent explained by neoclassical electron heat conduction losses, including ripple effects. In these experiments anomalous transport enhancement has been identified only in the outer plasma layers and at low temperatures and densities, which in this form should not affect reactor aspects. At a given density, neoclassical electron heat conduction losses scale however in a similar manner with heating power as the L-regime losses observed in tokamaks. For a better discrimination - important due to the differing variation with size expected for the two types of losses - density scans over a range not accessible with present day ECRH-systems are needed. Corresponding experiments with NBI-heating on the high shear Heliotron-E device indeed gave evidence for a confinement anomaly increasing with heating power, although not well fitted by tokamak L-regime expressions. For low shear systems, conclusive evidence is expected from NBI-heating experiments on W VII-AS.

The properties of carbon, with respect to its ability to absorb and release hydrogen, are reviewed and applied to the interpretation of density behaviour in Ohmic-, NBI- and ICRF-heated tokamak discharges. Based on the experimental observations, improvement of H-mode parameters in JET due to reduced hydrogen re-emission from the walls is predicted with a numerical model. The particle removal rates required for Q = 1 in JET are estimated.

Operational limits on JET in terms of density, current_and piasma beta are described. The results for various heating schemes are displayed in the τ_Ei vs _i plane against the contours of constant thermonuclear gain Q. The confinement data is compared with several scaling laws and the eventual performance of JET predicted. Finally the likelihood of ignition in the enhanced version of NET is assessed.

New long-pulse ion sources have been employed to extend the neutral beam pulse on TFTR from 0.5 sec to 2.0 sec. This made it possible to study the long-term evolution of supershots at constant current and to perform experiments in which the plasma current was ramped up during the heating pulse. Experiments were conducted with co and counter injection as well as with nearly balanced injection of deuterium beams up to a total power of 20 MW. The best results, i.e., central ion temperatures T_io > 25 keV and n_eo τ_ET_io values of 3 × 10²⁰ keV sec m^-3, were obtained with nearly balanced injection. The central toroidal plasma rotation velocity scales in a linear-offset fashion with beam power and density. The scaling of the inferred global momentum confinement time with plasma parameters is inconsistent with the predictions of the neoclassical theory of gyroviscous damping. An interesting plasma regime with properties similar to the H-mode has been observed for limiter plasmas with edge q_a just above 3 and 2.5.

A brief review of results of stellarator plasma equilibrium studies based on 2D and 3D codes as well as on analytical approaches is given. The most essential differences between stellarators and tokamaks are described from the viewpoint of equilibrium. A physical interpretation of the non-monotonic behaviour of the rotational transform t at a rather high plasma pressure in ordinary stellarators with a planar axis is given and the possibilities of correcting the behaviour of t by a transversal quadrupole magnetic field are considered. The approach to 30 simulation of equilibria in stellarators based on the "natural" and "co-natural" coordinate system concept and realized with the POLAR 3D code is described.

Pulsed laser and particle beams offer prospects of generating intense Planck radiation in mm-size cavities of high-Z material. In order to achieve isotropy and high temperature, confinement of the radiation through multiple reemission by the wall of the cavity is necessary. This requires heating of the wall material to a certain depth, necessitating energetic pulses from the source. In experiments performed at the Max-Planck-Institut für Quantenoptik (MPQ) at Garching and at the Institute of Laser Engineering (ILE) at Osaka with laser light of 1.3, 0.53 and 0.44 μm wavelength, brightness temperatures up to 2.2 × 10⁶ K have been achieved in gold cavities with one- to two- fold reemission of the x-rays. A more favourable laser-plasma interaction and x-ray conversion efficiency may be obtained by shorter wavelength lasers.

The processes by which impurities are produced and enter the discharge are discussed. Emphasis is placed on sputtering at the limiter and an analytical global model is described which incorporates the self-stabilizing effects which control the edge temperature. Predictions of the scaling of edge temperature and of total radiated power are compared with experimental data from JET and other tokamaks operating with limiters. Under many conditions the scaling of the edge conditions and of the radiated power is accurately predicted. Impurity transport in the boundary and the question of how to control the boundary layer is then discussed. The example of the Impurity Control Limiter on DITE is described.

The results of the electron cyclotron heating (ECH) experiments in T-10 are presented. An 11-tube gyrotron set-up with a total power of 4 MW was used in the experiments. The set-up consists of two types of gyrotrons with different wavelengths. The dependence of the energy confinement on the RF power deposition profile was investigated. An electron temperature of reactor level was obtained.

Recent advances in Reversed Field Pinch (RFP) research are described. These include experiment and theory on a non-Spitzer loop voltage and non-collisional ion heating, both important for RFP confinement, and brief mention of results from transport theory and 3-D codes. Other experimental studies include the improvements in confinement obtained by reducing field errors and the equilibrium displacement, preliminary experiments on current drive, the successful operation of an RFP with a thin shell and studies of a non-circular pinch. Topics such as scaling, density control using pellet injection and impurity behaviour are briefly mentioned and the paper concludes with a discussion of how RFP research is likely to develop in the future, both on the existing devices and on new large machines being constructed.

In the present work the status of the open systen studies at the Novosibirsk Institute of Nuclear Physics (INP) is described.

For the heating of dense plasnas in open systens of the multi-mirror type the possibility is being studied to use high-current relativistic electron beans (REBS). At n_e 3 - 10¹⁵ cm^-3, j_b ~ 2.5 kA/cm², the experiments indicate that for fields up to B = 5 T the interaction efficiency sbstantially increases with increasing field strength. With the above aim in view, the work being performed at the Institute concerns the construction of high power generators of microsecond REBS. At the generator U-1 a beam with 130 kJ energy content was produced. The construction of the generator U-2 is finished. In this generator, a foilless diode with a band geometry beam has been employed. In a similar geometry, the beams from several generators can be injected sequentially into the same magnetic track.

The results, obtained on a gas-dynamical trap (GDT) with large mirror ratio (R > 80) for the case when the ion mean free path λ_ii satisfied the inequality: L > λ_ii(lnR/R) (L is the longitudinal size of the system), have shown that the plasma dynamics in the trap is consistent with gas-dynamical estimations. At the stage of GDT filling, the plasma in it proves to be stable. At a favourable curvature of the magnetic field lines in expanders the plasma turns out to be MHD stable with the plasma source switched off as well. In the latter case the plasma decay time is in reasonable agreement with the gas-dynanical estimation: τ ~ LR/v_Ti.

The project of an ambipolar trap (tanden mirror) with axisymmetrical magnetic field has been developed. For MHD stabilization on the outside of the end plugs, two semi cusps with hot ion plasma (T = 15 keV) are planned to be mounted. High mode number MHD modes are assumed to be stabilized by the finite Larmor radius (FLR) effect, while the mode m = 1 will be stabilized by a favourable curvature of the magnetic field. The basic parameters of the central mirror trap are: L = 14 m, B = 0.45 T, n_e = 10¹³ cm^-3. The end plugs have the following paraneters: L_p = 1.8 m, R = 4, B_max = 6 T, n_e = 3 × 10¹³ cm^-3.

The relations between plasma wall interaction and plasma core properties are studied by spectroscopic methods applied to the emission of neutral particles at the plasma boundary. The interdependence between the structure of the scrape-off layer (SOL), particle-fluxes and impurity release are analyzed. In particular, the importance of molecule formation for impurity release and particle recycling in a machine with an "all-carbon" surrounding is shown. These mechanisms can lead to a stationary discharge with a cold plasma mantle ("detached plasma"), the properties of which are discussed. Local heating inside the SOL with ICRH is evident and some consequences for gas release and impurity generation are drawn.

Achieving high energy gain in laser fusion will require the use of moderately-thin pellet shells. But thinner shells are expected to be more susceptible to break-up from the Rayleigh-Taylor instability. Calculations and experimental data thus far indicate that the growth rate of the Rayleigh-Taylor instability is reduced well below the classical value when the surface of the pellet shell is directly ablated by laser light. In addition, the "mix" of the cold DT shell with the hot ignitor may only slightly degrade the pellet yield. It may therefore be possible to successfully implode these moderately thin shells and produce gains of a few hundred, using a few-megajoule KrF laser driver.

This paper presents a review of experiments on the production and heating of currentless plasmas by electron cyclotron (EC) waves in the Heliotron-E, Wendelstein VII-A, and L-2 stellarators, in wide ranges of the magnetic field (1.0 - 2.5 T), frequency (28 - 70 GHz) and microwave power (100 - 500 kW). EC heating of low and high-density currentless plasmas as well as of ohmically heated (OH) plasmas is discussed. Nonlinear ion heating and plasma current generation were observed to accompany EC plasma heating. It was found that a high efficiency of microwave absorption, a high electron temperature (1 - 1.5 keV) and a considerable energy confinement, time are characteristic of EC plasma heating.

This paper reviews recent developments in the understanding of the scrape-off layer structure. After the discussion of the general properties of a scrape-off layer formed by an axisymmetric limiter the more common case of non-axisymmetry is analysed. Original results are presented concernng a three-dimensional model of the SOL and related experiments. In the high-density case a specific structdre develops, which may be described as a neutral cloud shielding the limiter. The role of a structured SOL for the accountability of energy and particle fluxes is shortly evaluated.

Advances of JT-60 experiments in current drive and high power heating are presented. Fully noninductive current drive of 2 MA is achieved by LHRF without OH assistance. The current drive efficiency η_CD, defined as n-bar_e(10¹⁹ m^-3)I_RF(MA)R_p(m)/P_LH(MW), is in the range 0.8 - 1.7 and is improved to 1.7 - 3.0 when NB heating is applied to the LHCD plasma. Confinement improvement is also obtained when LHCD and NB heating are combined. The MHD fluctuations in the central plasma region of such a discharge are suppressed by a factor of 10 compared to the case of NB heating only. In the electron heating mode of LHRF, T_e(0) of 6±1 keV is obtained with a high heating efficiency of 2- 3 × 10¹⁹ keV/MWm³, and L-mode-like behavior is seen in the energy confinement in such plasmas. High ion temperatures of T_i(O) = 11 ± 2 keV are attained in hydrogen plasmas at densities of n-bar_e = 2 - 3 × 10¹⁹ /m³ with NB heating. Substantial beam acceleration is observed when NB and ICRF heating are combined. Charge exchange neutral spectra show that high energy ion tail of up to at least 160 keV is generated compared to the NB acceleration voltage of 60 kV.

The paper describes the main results of currentless operation in stellarator/heliotron experiments. Various heating methods (ECRH, ICRH and NBI) with a power up to 4MW have been applied in Wendelstein VII-A, Heliotron E, L-2 and Uragan 3. Strong increase of impurity radiation terminates the discharge in NBI-heated plasmas. Toroidal currents, driven either by the plasma pressure or the heating mechanism, have been observed in several experiments. New experiments, ATF and Wendelstein VII-AS will start operation in 1987. Main aim of theoretical efforts is to increase β-limits of equilibrium and MHD-stability. This led to the Heliac concept and the concept of the Advanced Stellarator. Both in the field of torsatrons/heliotrons and advanced stellarators activity has started to plan the next generation of major experiments.

The paper reviews present understanding of H-mode physics by summarising relevant experimental observations and discussing possible interpretation. The most important features of the H-mode are a minimum threshold edge temperature (threshold input power) required to achieve the H-mode; the bifurcation nature of the H-transition with instantaneous changes at the plasma edge; and the formation of a transport barrier at the plasma edge leading to pedestals in the density and temperature profiles. Global energy confinement times, are typically 2× to 3× longer in H- than in L-mode plasmas, reaching, for example, almost 1s in 3MA JET X-point discharges. τ_E is found to increase linearly with plasma current. Results of the variation of τ_E with input power are somewhat contradictory: no power dependence is found in ASDEX and DIII-D, whereas a degradation with power is indicated in JET and in JFT-2M limiter H-modes. Correspondingly, predictions for full power (40MW) 6MA X-point discharges in JET range from 0.6s to >1s, depending upon which scaling is adopted. Two main theoretical models have been proposed to explain the H-mode with its heat barrier at the plasma edge. Such a barrier is predicted by the stability properties of ballooning modes close to a magnetic separatrix, corresponding to the transition to a second stable region above a certain threshold power. It could also arise from a critical temperature gradient model based on self-consistent stochasticity.