Table of contents

Triton production from one branch of the deuteron-deuteron fusion reaction is routinely measured at 6 ms time intervals in JET plasma discharges by recording the 2.5 MeV neutrons produced in the other branch using a set of calibrated fission chambers. The burnup of the tritons is measured by detecting the 14 MeV t-d neutrons with a 0.2 cm³ Si(Li) diode. The 2.5 MeV neutron flux can be used in a simple time dependent calculation based on classical slowing-down theory to predict the 14 MeV neutron flux. The measured flux and the triton slowing-down time are systematically lower than the values estimated from the key plasma parameters but the differences are within the experimental errors.

The process of catalytic burning in DD plasma induced by injection of T⁶Li pellets (amounting to 20% of the total density) in a magnetic confinement facility is studied theoretically. It is shown that the injection of a T⁶Li pellet may open up a simple way of igniting a self-sustained thermonuclear reaction in a pure DD plasma at an initial temperature of T⁰ ≤ 10 keV.

The paper presents a method for determining the potential of a reactor design to reach ignition conditions. The method makes use of a 'generalized' contour map with a three-dimensional representation. On this map, the height of (with P_ht the total heating power density and τ_E the energy confinement time) at the saddle point provides a criterion for reaching ignition. By a combination of this map with Goldston's L-mode scaling law, it is found that the operating point of the tokamak moves along a line of constant ; the value of for H-mode scaling can be increased with auxiliary heating power so that the saddle point can be reached. Confinement degradation effects due to alpha particle heating can be treated by replacing by (f_α is the fraction of the alpha particle heating P_α causing confinement degradation, η_α is the fraction of the alpha heating power trapped in the plasma). The saddle point becomes higher and shifts to a larger nτ_E regime by this alpha degradation effect.

An analytical investigation is made of the confinement and subsequent burnup of the fusion produced tritons in a deuterium tokamak plasma. Explicit approximations are obtained for the triton confinement factor, clearly displaying the scaling with physical parameters. The importance of pitch angle scattering losses during the triton slowing-down is also estimated. A comparison with experiments and numerical calculations on the FT tokamak shows good qualitative agreement.

A tokamak edge plasma significantly absorbs ICRF waves when a cutoff-resonance-cutoff is present in the evanescent region. Two ion component plasmas in a torus are considered although the plasmas are modelled by a slab. The absorption rate A_b is roughly given by M² for 1 ≫ M² > S² and S⁴ for 1 ≫ S² > M², where M = k_yℓ and S ≅ k_zℓ and ℓ is a scale-length of the order of the plasma minor radius, and k_y and k_z are the perpendicular and the parallel components of the wave vector. It is noted that M and S become of the order of unity. Since A_b is not very sensitive to the density ratio of the two ion species, a few per cent of impurities may cause significant absorption. As the mass ratio of the two ion species comes close to unity, the triplet forms readily. Therefore, a D-T plasma seems to suffer more easily from this kind of resonance absorption than a D-H plasma.

The evolution of a linear z-pinch in which the current rises at a constant rate is analysed in terms of a quasi-static pinch model. It is shown that there is a maximum current in hydrogen or deuterium equal to approximately 3^1/2 times the Pease-Braginskij current which cannot be exceeded because of radiative collapse. Higher currents may be obtained in D-T because heating by the alpha particles prevents collapse until most of the fusion fuel is consumed, at which point the pinch collapses like a star. The implications for the pinch as a fusion system are discussed and it is shown that, if the pinch remains stable while the current is rising, a dense z-pinch in D-T might form the basis of a small fusion reactor.

Local transport and ideal ballooning stability in the L-phase and ELM-free H-phase in ASDEX are analysed by computer modelling. It is found that the diffusivities χ_e and D at the edge are reduced by a factor of six a few milliseconds after the H-transition. Local transport in the inner plasma improves at an early stage by a typical factor of two. A change in the collisionality regime of electrons and ions does not take place. During the L-phase and the quiescent H-phase ideal ballooning modes are found to be stable. Computer experiments further show that a significant reduction in the particle flux at the separatrix takes place which is closely connected with the H-transition process. This explains the observed buildup of a density shoulder on a millisecond time-scale and the drop of the particle flow into the divertor. A strong decrease of the electron heat conduction flux at the separatrix is, however, ruled out in ELM-free periods. On the assumption of electrostatic turbulence induced transport, these results are consistent with measured density fluctuation levels near the separatrix.

In the thermal barrier tandem mirror GAMMA 10, the scaling law governing the enhancement of the ion confining potential, ϕ_c, resulting from thermal barrier formation, is obtained experimentally, and is consistently interpreted in terms of the weak and strong ECH theories set up by Cohen and co-workers. The scaling law on the axial particle confinement time, τ_p||, related to this ϕ_c formation, is also demonstrated in detail; it is in good agreement with the Pastukhov theory as modified by Cohen and co-workers. This scaling is verified at any radial position in the core plasma region and at any time through the various stages of a discharge; this indicates a scaling with drastic improvement of τ_p||, due to the potential formation in the tandem mirror plasma.

While the results of early work on the density limit in tokamaks from the ORMAK and DITE groups have been useful over the years, results from recent experiments and the requirements for extrapolation to future experiments have prompted a new look at this subject. There are many physical processes which limit the attainable densities in tokamak plasmas. These processes include: (1) radiation from low Z impurities, convection, charge exchange and other losses at the plasma edge; (2) radiation from low or high Z impurities in the plasma core; (3) deterioration of particle confinement in the plasma core; and (4) inadequate fuelling, often exacerbated by strong pumping by walls, limiters or divertors. Depending upon the circumstances, any of these processes may dominate and determine a density limit. In general, these mechanisms do not show the same dependence on plasma parameters. The multiplicity of processes leading to density limits with a variety of scaling has led to some confusion when comparing density limits for different machines. The authors attempt to sort out the various limits and to extend the scaling law for one of them to include the important effects of plasma shaping, i.e. ;_e = k, where n_e is the line average electron density (10²⁰ m⁻³), κ is the plasma elongation and (MA·m⁻²) is the average plasma current density, defined as the total current divided by the plasma cross-sectional area. In a sense, this is the most important density limit since, together with the q-limit, it yields the maximum operating density for a tokamak plasma. It is shown that this limit may be caused by a dramatic deterioration in core particle confinement occurring as the density limit boundary is approached. This mechanism can help explain the disruptions and Marfes that are associated with the density limit.

The paper presents results from the impurity control limiter (ICL) experiment, which was installed in the DITE tokamak. The ICL was designed so that the majority of ions leaving the plasma strike the undersides of its graphite tiles, which are angled towards the wall. These experiments confirm that the impurities sputtered from the ICL are ionized predominantly in the scrape-off layer and that there is significant deposition of neutral carbon on the walls. Both the spatial distribution of ionization and the toroidal distribution of carbon on the walls are in good agreement with code predictions. In helium discharges, the insertion of the ICL resulted in a 50% reduction of the total radiation and a 37–66% reduction in the impurity contribution to Z_eff compared with the values obtained with rail limiters of conventional geometry. In deuterium discharges, the reduction in the total radiation was only 20%; this is attributed to the greater importance of the wall sources of carbon and oxygen in deuterium plasma relative to those in helium.

Characteristics of the internal mode frequency associated with the plasma rotation have been investigated for unidirectional neutral beam injection experiments into the JT-60 divertor plasma. The phase velocity inferred from the mode frequency as observed on the soft X-ray signals reached 1.2 × 10⁵ m·⁻¹ for counter-injection with 9 MW. The dependence of the velocity on the electron density and beam power is studied. Simple modelling calculations based on the incremental phase velocity show that the obtained toroidal momentum confinement time of ∼ 100 ms is of the order of the energy confinement time.

Alpha particle parameters in sub-ignited D-T tokamaks such as TFTR can be optimized in a high temperature 'alpha storage regime' in which the alpha particle thermalization time τ_α is long (≈1.0 s) and in which the alpha particle source rate S_α is enhanced because of beam-target and beam-beam reactions (by a factor of ≈2-4). Near reactor-level alpha instability parameters, β_α(O)≈ n_α(0)/n_e(0) ≈ 1%, are predicted by simulation codes when Q ≳ 0.5; present TFTR 'supershots' already have β_α(O) ≈ n_α(0)/n_e(0)≈ 0.1-0.2%. This alpha storage regime can be used for the first time to test theories of collective alpha instabilities and can also be used to provide a strong (but transient) alpha heating pulse. An experimental scenario to exploit this regime is described.

Absorption of ICRF waves by alpha particles is investigated. Higher harmonic cyclotron damping of the waves is examined not only for the slowing-down distribution of alpha particles but also for the distribution given by the Fokker-Planck calculation. The right and left hand circularly polarized amplitudes of the RF electric field are taken into account in the RF diffusion coefficient. The linear absorption rate for the alpha slowing-down distribution increases with decreasing ratio of electron plasma to electron cyclotron frequency and increasing cyclotron harmonic number. Unless this absorption rate is nearly unity, the bulk formation of alpha particles results in a further increase of the rate of absorption by alphas. Fusion reactivity enhancement due to the ICRF induced ion tail and fusion energy multiplication in the presence of alpha particles are also discussed.

Experiments were conducted in the ELMO Bumpy Torus (EBT) from 1973 until 1984. A number of papers have been published concerning various aspects of experiments during the final two years of operation. The paper summarizes the final experimental conclusions and discusses those issues which remain unresolved. The main conclusions are as follows: (1) measurements of the width of the hot electron rings showed them to be wider than previously suspected, diluting their diamagnetism and negating their ability to locally reverse the magnetic field gradient; (2) two-dimensional plots of the plasma potential revealed open potential contours in the C-mode and closed potential contours near the plasma centre in the T-mode; (3) an in–out asymmetry in the plasma potential was always observed, giving rise to electric fields which drove convective plasma loss.

The 7th International Conference on High Power Particle Beams – "BEAMS '88" – took place in Karlsruhe from 4 to 8 July 1988. It was organized by the Kernforschungszentrum Karlsruhe GmbH. About 270 participants convened in the recently constructed Congress Centre.