Table of contents

High-speed pellet fuelling experiments have been performed on the ISX-B device in a new regime characterized by large global density rise in both Ohmically and neutral-beam heated discharges. Hydrogen pellets of 1 mm in diameter were injected in the plasma midplane at velocities exceeding 1 km·s⁻¹. In low-temperature Ohmic discharges, pellets penetrate beyond the magnetic axis, and in such cases a sharp decrease in ablation is observed as the pellet passes the plasma centre. This behaviour can be accounted for by an ablation model that includes dynamic cooling of the target plasma while the ablation proceeds. Complete penetration can be prevented by operation in low-density regimes where runaway electrons are thought to be responsible for high ablation. A similar effect is observed with moderate to large amounts of neutral-beam injection. There is a strong enhancement of the ablation rate in the outer 10-cm plasma region even for short heating intervals, which can be explained by the presence of multi-kilo-electron volt ions in the discharge. Density increases of ∼300% have been observed without degrading plasma stability or confinement. Energy confinement time increases in agreement with the empirical scaling τ_E ∼ n_e and central ion temperature increases as a result of improved ion-electron coupling. Laser-Thomson scattering and radiometer measurements indicate that the pellet interaction with the plasma is adiabatic. The low level of power emission from the pellet-plasma interaction region is consistent with negligible charge-exchange losses; within the experimental accuracy, nearly all of the pellet mass can be accounted for in the initial plasma density rise. Penetration to r/a ∼ 0.15 is optimal, in which case large-amplitude sawtooth oscillations are observed and the density remains elevated. Gross plasma stability is dependent roughly on the amount of pellet penetration and can be correlated with the expected temporal evolution of the current density profile.

The problem of the helical equilibrium of a plasma column with a free boundary is considered. An expression for the maintaining field and the condition for helical equilibrium in the absence of additional helical windings is obtained. The method of moments is applied to a solution of the equilibrium equation for a plasma column with elliptical cross-section.

A simple model describing the scaling laws of tokamak fusion reactors is used to derive the dependence of capital cost on the basic parameters involved. The beta value required for each set of parameters is normalized to an equivalent beta in JET to provide a uniform measure of feasibility. The capital cost is related to the engineered volume per unit of power. The simplicity of the model and the uncertainty of the assumptions exclude a precise evaluation, but do allow an overview of the essential features arising in a reactor assessment.

The one-dimensional Vlasov distribution function in terms of the adiabatic invariants of the Vlasov equation for a single, non-linear, large-amplitude travelling wave excited by an external source is derived. The result takes into account the adiabatic turn-on of the pump and is valid to all orders in the electric field amplitude everywhere in phase-space, except for a narrow region near the separatrix. All electrons, including the trapped ones, are adiabatic and the processes are reversible. The non-linearity is found to lead to the generation of a steady-state current whose properties are examined. The theory is applied to study the steady-state currents that could be driven by the lower hybrid wave in present-day tokamak experiments. The trapped electrons play a very significant role in determining the nature of this current.

In the asymptotic treatment of tokamak MHD stability at small inverse aspect ratio , the special case of poloidal wave number m = 0 has been treated improperly in the literature for both axisymmetric and non-axisymmetric modes. In axisymmetric stability, a contribution to the perturbational vacuum field is either omitted or cancelled. In a variational stability analysis this field contribution provides δ²W with a correction term proportional to (ln )⁻¹, which may change the asymptotic range of stability and improve agreement with numerical finite-aspect-ratio results. In non-axisymmetric stability, for the perturbational vacuum field of the m = 0 modes, usually the wrong of two possible solutions is chosen. It is shown why in many cases this wrong choice has no consequences on the correctness of the stability results, and circumstances are pointed out under which consequences may arise.

The subject of this review is the field of intense pulsed ion beam generation and its potential application to fusion research. Considerable progress has been made in the past six years. Power levels of the order of 1000 MW/cm² have been obtained for pulse-lengths from 10⁻⁸ to 10⁻⁶ s. Light-ion beams with currents approaching 1 MA and energies exceeding 1 MeV have been produced. The first half of the paper treats the physics and technology of beam generation. Following a unified discussion of the theory of ion injectors, experimental work since 1973 is reviewed. Diagnostic techniques appropriate to high-flux ion beams are summarized. The general problem of high-current ion beam transport in vacuum and through plasmas is also considered. The second part of the paper is devoted to applications to fusion research. Applications in both magnetic confinement and inertial confinement fusion are considered. Intense ion beams have a number of features which make them ideally suited as inertial fusion drivers. Discussions are given of the physics of inertial fusion targets, options for ion beam production, experiments on light-ion beam focusing, and high-current multi-stage accelerators.

A heuristic model for classical ion confinement in a magnetic mirror is presented. The competing influences of electron drag and ion-ion scattering motivated us to develop this model. Its predictions are compared with computational results from a Fokker-Planck code. For a range of plasma parameters typical of tandem mirror reactor plugs, the model is accurate to within ± 10%.

As a result of well-organized co-operation in controlled nuclear fusion research, each day sees the rapid spread of knowledge and results gained in this field throughout the world. Thus, the knowledge is steadily accumulating. However, perhaps for psychological reasons, views on the status of the problem change rather abruptly. With the same corpus of knowledge, we are now beginning to recognize clearly the nature of a problem which was shrouded in mist only yesterday, and yet tomorrow we may be unable to believe that the mist ever existed. One eminent English writer, I believe, neatly expressed this thought in an aphorism to the effect that every truth enjoys only a brief moment of triumph between one period during which it is rejected as a paradox and another during which it is accepted as trivial. To experience these brief moments, we gather together at IAEA Conferences which afford us so much joy.

The 11 th Symposium on Fusion Technology, hosted by the Culham Laboratory of the UKAEA, was held at Oxford from 15–19 September 1980. The traditional aims of this biennial European Symposium have been to provide a forum for discussion of technical aspects of fusion research not often aired at more academic conferences, emphasizing the successes and shortcomings in the development of the technology required for the economic generation of power by controlled thermonuclear fusion. Recent successes generally stem from the designs of JET, TFTR and JT-60, from reactor studies, and from operational machines which now include (since the 10th Symposium which was held at Padua) ASDEX, PDX, Doublet III, ETA-BETA, TMX and also the superconducting T-7, which was unfortunately not represented at this 1980 Symposium. The most serious shortcomings were in remote maintenance and tritium systems, with only a few papers on these subjects. The dozen papers on materials technology, the most vital topic of all although a new topic for SOFT, only partially alleviated the impression gained by participants that materials capable of providing a reactor with an economic working life are not yet available.