Table of contents

The plasma energy flux is frequently equated to the kinetic energy flux. However, the periodic exchange between kinetic and potential energy in the gyratory motion produces a spurious contribution, which is balanced by a potential energy flux. The total energy flux is more meaningful. This is equivalent to the guiding-centre energy flux. The convective part of this is (3/2) Gamma T, not (5/2) Gamma T as for the kinetic energy flux.

Equilibria of RF Tokamak plasmas during lower hybrid (LH) wave-driven current start-up experiments in the WT-2 and WT-3 Tokamaks are studied by comparing spatial profiles of tail and bulk electrons with simple model equilibria for the electron tail. When the plasma current I_p is smaller than the Alfven critical current I_A, the confinement areas of the tail and bulk electrons are not coincident. The cross-section of the outermost flux surface is always found to be larger than the tail cross-section. Particularly in the case of the small Tokamak WT-2 (a_L=9 cm) is the outermost flux surface coincident with the limiter circle and much larger than the tail cross-section, indicating that good confinement of bulk electrons which carry LH waves and provide electrons to the tail is crucial for RF current start-up.

For comparison with experiment, the expected production of 14-MeV neutrons from the burnup of tritons produced in the d(d,t)p reaction must be computed. An effort was undertaken to compare in detail the computer codes used for this purpose at TFTR and JET. The calculation of the confined fraction of tritons by the different codes agrees to within a few per cent. The high electron temperature in the experiments has raised the critical energy (where ion drag equals electron drag) of the tritons that are slowing down to near or above the peak of the D-T reactivity, making the ion drag terms more important. When the different codes use the same slowing-down formulas, the calculated burnup was within 6% for a case where orbit effects are expected to be small. Then results from codes with and without the effects of finite radial orbit excursions were compared for two test cases. For medium to high current discharges the finite radius effects are only of the order of 10%. A new version of the TFTR burnup code using an implicit Fokker-Planck solution was written to include the effects of energy diffusion and charge exchange.

The linearized MHD perturbation theory of a magnetically confined plasma is reconsidered, with special emphasis on systems where there is a spatially inhomogeneous magnetic field part generated by constant currents in fixed external conductors. Electromagnetic induction then leads to induced surface current effects at the plasma boundary, due to the inhomogeneity of the externally imposed magnetic field. The corresponding surface currents are physically different from those which arise from the pressure balance condition at the plasma boundary. The apparent contradiction between the electromagnetically induced surface current effect and the pressure balance condition at the boundary is resolved by a transfer of the force due to the surface current into a volume force acting on the entire plasma body. The integrated effect of the surface current is thus included in the form of an additional volume contribution to the change in potential energy, whereas all other parts of conventional theory and its boundary conditions are preserved in the present approach.

The effects of the dielectric properties of a relativistic magnetized plasma on the scattering of electromagnetic radiation by fluctuations in electron density are investigated. The origin of the density fluctuations is not considered. Expressions for the scattering cross-section and the scattered power accepted by the receiving antenna are derived for a plasma with spatial dispersion. The resulting expressions allow thermal motion to be included in the description of the plasma and remain valid for frequencies of the probing radiation in the region of omega _p and omega _ce, provided the absorption is small. Symmetry between variables relating to incident and scattered fields is demonstrated and shown to be in agreement with the reciprocity relation. Earlier results are confirmed in the cold plasma limit. Significant relativistic effects of practical importance to the scattering of millimeter waves in large Tokamaks, are predicted.

Thermal alpha particles are observed in JET during helium discharges using spectral emission in He II (n=4 to 3) near 4685 AA following charge transfer reactions along the path of the neutral deuterium heating beams. New and reappraised He²²/H charge transfer cross-sections are presented. The effects of cross-section energy dependence on temperatures, velocities and absolute densities deduced from thermal alpha particle charge exchange spectra are evaluated. The possibility of detecting fusion alpha particles produced at 3.5 MeV and slowing down by collisions with plasma electrons and ions using visible charge exchange spectroscopy is addressed. The spectral signature of slowing-down fusion alpha particles expected during the deuterium-tritium phase of JET is modelled and its identification against thermal alpha particle and background radiation is investigated.

Previous theoretical work on fast Alfven eigenmodes suggests that it is not possible for any single mode to satisfy boundary conditions in a rectangular cross-section torus, even though such modes are observed experimentally. The authors show that it is possible for fast wave modes to satisfy the boundary conditions, provided that the bulk plasma is separated from the vessel walls by a low-density edge plasma or vacuum region. The boundary conditions can then be satisfied by the presence of poloidally-evanescent wave fields. Toroidal geometry introduces a strong asymmetry in the wavefields between the high- and low-field sides of the torus, but the eigenfrequencies are close to those predicted using a simple cylindrical model.

Measurements are presented of the edge wave magnetic fields generated by fast wave antennas in the TORTUS Tokamak. All measurements were made in a deuterium plasma and at frequencies in the range 2< omega / omega _ci<4, where omega _ci is the ion cyclotron frequency. High Q cavity modes were observed only for the m=+1 mode, although many other cavity modes were expected on the basis of cold plasma calculations. High m modes were also observed, not as cavity modes, but in the form of an evanescent beam guided along steady magnetic field lines in the plasma edge. The three-dimensional structure of the beam was determined using toroidal and poloidal arrays of magnetic probes. The existence of the beam is attributed to Alfven resonance conditions at the plasma edge.

It is of great interest to determine the deuteron temperature or density in Tokamak fusion plasmas from total neutron rate measurements. This can be done easily during Ohmic heating or H⁰ injection, where the velocity distribution function of the plasma deuterons is Maxwellian. However, during D⁰ injection the distribution function is highly non-Maxwellian and until now deuteron temperatures or densities could only be evaluated with the aid of complex codes. The authors present a fast interpretation code using a Fokker-Planck formalism and identify the parameter ranges where deuteron temperatures or densities can be determined from neutron rate measurements during neutral-beam heating. As an example of its application, ion temperatures and densities are evaluated for JET discharges.

Previous calculations and computations of resistive magnetohydrodynamic equilibria have revealed helically-deformed steady states with flow, just above critical current thresholds. The poloidal flow pattern consists of paired macroscopic vortices, which effectively transport fluid elements from the center of the plasma column to the perimeter and back. The appearance of the helically-deformed states with vortices can be suppressed by a rigid poloidal rotation, entirely within the magnetohydrodynamic framework. Velocity shear is not required.

A simple analytical mode for the implosion of very thin spherical shell targets filled with fuel gas is developed. The shock trajectory in the fuel is described consistently with the shell acceleration, and two dimensionless parameters which govern the complete dynamics are found. The model applies to recent experiments focused on high neutron yield and provides a simple description of the main physical phenomena, which is in agreement with simulation and experiments.