Table of contents

Three methods of density modulation have been used in FTU discharges (gas puffing, pellet injection and current rampup). The analysis of particle balance confirms that the neoclassical theory is not adequate and the experimental results obtained by the three methods are used to verify the validity of an empirical expression for the radial particle flux. Such an expression is relevant for the design of future experiments and constitutes a constraint for any satisfactory theory

The influence of radial and poloidal E × B drifts in the scrape-off layer (SOL) of a divertor tokamak is considered. For B in the `forward' direction (the ∇B ion drift towards a lower X point, single null divertor configuration), the poloidal drift tends to increase the plasma pressure and the recycling rates at the outer target, lowering the temperature there - as compared with the inner target. When the radial drift dominates, these asymmetries are reversed. The radial drift is estimated to be more important than the poloidal drift for high n<sub>e</sub> operation (the `high recycling' regime). The poloidal drift case is analysed by using a simple, isothermal model and boundary conditions at the targets which allow for the effect of drift. The radial drift case is modelled on the basis of a heuristic analysis of the viscous drag experienced by the return parallel flow along the SOL which is induced by the radial drift fluxes. Poloidal drifts can result in substantially reduced plasma contact at one target (`detachment'); simultaneous detachment at both targets due to poloidal drifts appears to be less likely. Radial drifts could result in detachment at one or both targets, but the type of detachment involved does not appear to be consistent with that experimentally observed. The experimental observations made in the Tokamak de Varennes, with divertor biasing arranged to increase/decrease the radial electric field in the SOL, relative to its natural level, are broadly consistent with the poloidal drift model

A simple analytical model is presented that describes impurity ion production and transport in the tokamak scrap-off layer (SOL). The equations of the model are solved analytically in the test particle approximation. The solution, as a function of different plasma parameters and target materials, is discussed in the case in which the background plasma is described by the simple SOL model and a comparison between the model and the numerical results of a 2-D multifluid code is presented

A self-consistent analysis of the impurity dynamics for bulk and scrape-off layer (SOL) plasmas in the Frascati Tokamak Upgrade (FTU) is presented. The model is based on a O-D description of the plasma transport in the tokamak centre and on a l-D model of the SOL plasma dynamics. The FTU configuration with a poloidal Inconel limiter is considered. Preliminary calculations for the configuration with the molybdenum and tungsten limiter plates are also presented. A comparison with the experimental data shows good agreement with the results of the model

Parasitic coupling to the ion Bernstein wave during fast wave heating is investigated using the WIGS computer code. The WIGS code models ion cyclotron range of frequencies (ICRF) antenna coupling to tokamak plasmas for phased coil antenna arrays in a recessed cavity. The antenna radiation resistance and the spectrum of the ICRF power coupled to the plasma are calculated. Parasitic coupling to the ion Bernstein wave is shown to be important for low parallel wavenumbers during minority ion heating with large gradients in the plasma density near the edge of the plasma. Results are presented for TFTR antenna and plasma parameters using experimental data

A vacuum magnetic field structure in the edge region of Wendelstein 7-X lending itself to the island divertor concept is studied for three rotational transform values, l=5/6, 5/5 and 5/4. An estimate of the power load on the divertor plates is given for these magnetic field configurations by using a simplified transport model in conjunction with complementary Monte Carlo methods allowing simulation of transport across field lines

Fourier analysis is proposed as a tool for the recognition of non-diffusive features of heat pulse propagation in fusion plasmas. Application to a simulated ballistic sawtooth heat pulse shows that Fourier analysis can identify the plasma region (often well beyond the mixing radius) where the heat pulse is diffusive. In this region the initial temperature perturbation, as recorded by a diagnostic with moderate (≃ 1 kHz) time resolution, is negligible. Fourier analysis of an example of JET sawtooth heat pulse diagnosed with the new ECE heterodyne diagnostic is consistent with the results from the simulated ballistic heat pulse. It cannot, however, be used as evidence of a ballistic effect owing to the limited diagnostic time resolution. When both the perturbed (χ_e^p) and the power balance (χe^PB) diffusivity can be determined reliably in recent JET sawtoothing plasmas, their ratio is χe^p/χe^PB ≃ 2-4 in agreement with previous results

The injection of impurities into the scrape-off layer (SOL) to stimulate radiative power exhaust from the mantle and SOL/divertor of the International Thermonuclear Experimental Reactor (ITER) is investigated. Intermediate Z impurities (e.g. argon to krypton) are able to radiate more than two thirds of the plasma heating power, without significant effect on the plasma power balance or the shear profile and without triggering an H to L mode transition

The loss region in the initial velocity space of the direct orbit loss ions near the plasma edge of tokamaks with the divertor configuration is studied analytically. The results of this analysis are compared with the numerical results of the loss region in the JET case obtained by Chankin and McCracken (1993). The results agree with each other semiquantitatively in several cases involving the presence of a radial electric field. A measure of the direct ion orbit loss Gamma is calculated from the given loss region in the initial velocity space for JET, JT-60U and ITER. When the initial position of an ion is located in the outside torus (r>R_x, where R_x is the radius at the null X point), the dependence of Gamma on the radial electric field shows the existence of a local maximum and a local minimum in the negative region of the radial electric field

Direct losses of charged particles are investigated for a torsatron with additional toroidal field coils. The effect of the appropriately unbalanced currents in the toroidal field coils on particle confinement in a torsatron is compared with the effect on it of an inward shift of the magnetic axis. For this purpose, the equations of motion of the guiding centre of the charged particles are integrated for different configurations of the URAGAN-2M torsatron and the Large Helical Device. The distribution functions of lost particles versus pitch values as well as fractions of losses are obtained