Table of contents

Density, temperature and plasma potential are measured simultaneously at the same spatial point in the SOL plasma of FTU at a rate faster than the typical turbulence frequency. This is done through a harmonic analysis of the current drawn on a single Langmuir probe swept in voltage at high rate (≈ 0.9 MHz). The current signal is detected with a down frequency conversion by means of a homodyne technique. The fluctuation levels are around 30% for all three quantities, but the autocorrelation times of temperature and potential are shorter than that of density: <3 μs against ≈ 13 μs. The magnitude and mutual coherence of the fluctuations are affected by the connection length L_c of the flux tube connected to the probe, offering a possible explanation of the variation of both the particle and the energy transport coefficient with L_c found experimentally.

Empirical scaling laws (Goldston, Kaye-Goldston, ITER89P and ITER93H) for the confinement time τ_E in tokamaks depend on dimensionless global plasma parameters in a way which is difficult to associate with some general plasma physics model. It is shown that such dependences can arise from artefacts in the database used to derive the scaling expression. The artefacts involve inner relationships, for example, collinearities between variables. For the ITER L mode and H mode databases, it is shown that subsets of the scaling expressions are approximately constant. An alternative description of L mode, ELM-free, small ELM and giant ELM H mode data is given. The alternative description does not include the dependences upon plasma β, elongation and safety factor which are present in the above scaling laws; extrapolations to ITER parameters by this alternative data description are virtually the same as those by ITER89P and ITER93H. The alternative description is based on a simple electrostatic physics model (plateau scaling) in which confinement is degraded by non-collisional processes. It can describe global confinement data for all four confinement regimes; the common scaling features a multiplier C which is assigned a different value for each regime. The complex physics associated with MHD instabilities (ELMs, sawteeth, outer modes) is hidden in the multiplier C, and attempts are made to uncover such physics. In these attempts global confinement is made to scale with physics models rather than physics variables.

The stability of the vertical position control system in TCV is analysed using two different plasma models, a deformable plasma model (DPM) and a rigid plasma model (RPM). The open loop growth rates and closed loop stability are computed, as functions of various plasma parameters, for both models. It is found that the difference between the open loop growth rates predicted by the two models depends significantly on plasma elongation and triangularity. Growth rates are also computed using the NOVA-W code, and good agreement is found with the DPM results. The stability of the closed loop system is analysed as a function of the feedback gain settings. The size of the stable domain in gain space is evaluated for the DPM and RPM models as a function of plasma elongation and triangularity, and the effect of varying the vertical position observer is shown. The stable domains, as predicted by the two models, are consistent with the experimental observations obtained from a highly elongated, D shaped plasma in TCV (κ = 2.55).

Scaling of the ignition energy threshold E_ig with the implosion velocity v_im and isentrope parameter α of imploding spherical DT shells is investigated by performing one dimensional (1-D) hydrodynamic simulations of the implosion and hot spot formation dynamics. It is found that the a and b exponents in the power law approximation E_ig ∝ α^av_im^-b depend crucially on the subset of initial configurations chosen to establish the scaling law. When the initial states are generated in the same way as in the Livermore study (W.K. Levedahl, J.D. Lindl, Nucl. Fusion 37 (1997) 165), the same scaling, E_ig ∝ α^1.7 v_im^-5.5, is recovered. If, however, the initial states are generated by rescaling the parent configuration according to the hydrodynamic similarity laws, a different scaling is obtained, E_ig ∝ α^3.0 v_im^-9.1, which is very close to the α³v_im^-10 dependence predicted by the simple isobaric model for assembled fuel states. The latter is more favourable than the Livermore scaling when rescaling the fusion capsules to higher implosion velocities, but requires the peak drive pressure to be increased as P ∝ v_im⁵.

The fluxes and energy distributions of the charge exchange (CX) neutrals as measured at specific locations at the tokamak ASDEX Upgrade (AUG) and the stellarator W7-AS at Garching are discussed as a function of the discharge conditions. For the evaluation of the plasma-wall interaction, knowledge of the energetic neutral fluxes and their energy and angular distributions at all poloidal and toroidal locations are required. For AUG these are obtained from B2-EIRENE computer simulations, taking the experimental results into account. The CX fluxes and the shapes of the spectra vary greatly around a poloidal cross-section. This has a strong effect on the wall erosion by sputtering and hydrogen isotope implantation into the vessel walls. The sputtering of the actual carbon wall, and the effect on possible wall materials, such as tungsten, beryllium, TiC and SiC, are discussed. The depth distribution of the CX neutrals implanted into the walls is calculated for two cases using the TRIMSP code. Saturation, release by glow discharges and permanent retention of the implanted hydrogen isotopes are discussed. This is important for the expected tritium content of the vessel walls of a future D-T fusion device.

Absorption of ion Bernstein (IB) waves by electrons is investigated. These waves are excited by linear mode conversion in tokamak plasmas during fast wave (FW) heating and current drive experiments in the ion cyclotron range of frequencies. Near mode conversion, electromagnetic corrections to the local dispersion relation largely suppress electron Landau damping of these waves, which becomes important again, however, when their wavelength is comparable to the ion Larmor radius or shorter. The small Larmor radius wave equations solved by most numerical codes do not correctly describe the onset of electron Landau damping at very short wavelengths, and these codes, therefore, predict very little damping of IB waves, in contrast to what one would expect from the local dispersion relation. We present a heuristic, but quantitatively accurate, model which allows account to be taken of electron Landau damping of IB waves in such codes, without affecting the damping of the compressional wave or the efficiency of mode conversion. The possibilities and limitations of this approach are discussed on the basis of a few examples, obtained by implementing this model in the toroidal axisymmetric full wave code TORIC.

Intense bursts of millimetre wave emission are emitted from plasmas in TFTR when the graphite limiter is well conditioned with elemental lithium through the injection of lithium pellets. The bursts have been observed in ohmic and neutral beam heated plasmas fuelled with deuterium only, a mixture of deuterium and tritium or tritium only. The bursting is characterized by 5-10 μs duration emission spikes of randomly varying amplitude. Spectral analysis of the millimetre wave emission reveals two components; broadband emission, which appears at frequencies below the second harmonic electron cyclotron frequency, and narrowband emission (Δf < 3 GHz), with an equivalent radiation temperature of over 1 MeV and a frequency corresponding to the upper hybrid frequency near the low field plasma edge. The equivalent radiation temperature of the bursts increases with decreasing edge density, decreasing limiter recycling and increasing plasma stored energy.

An accurate and simple technique for plasma boundary determination in a tokamak from external magnetic measurements is described. The method uses the filamentary current model of the plasma current profile, optimized to comply with the requirement of 1 cm accuracy in the determination of the plasma boundary in ITER. An error analysis is performed by numerical simulation of the ITER plasma configuration. The conclusion is that the method can provide this accuracy provided that the level of relative measurement errors can be kept below 1%.

A series of controlled experiments has been carried out in DIII-D to induce a bulk ion flow in the SOL and evaluate its effect on the localization of impurities in the divertor. This induced SOL flow was created by simultaneous deuterium puffing and divertor exhaust using a divertor cryopump, and the impurity enrichment was measured directly. The experiments were designed to compare enrichment in discharges with and without induced flow having otherwise similiar divertor parameters. Significant increases in impurity compression and enrichment are observed when flow is induced, and the degree of impurity enrichment in the divertor is found to be dependent on the impurity of interest. Detailed particle measurements made possible by the direct measurement of impurity densities in several reservoirs indicate reasonable particle balance for helium throughout the duration of the discharge. Conversely, while the total input of neon is balanced by the total exhaust by the end of a discharge, particle balance is not observed during the course of the discharge. A significant wall inventory with a short release time (∼10 ms) is surmised.

The physics of sawtooth oscillations in plasmas produced by the Alcator C-Mod machine is discussed by identifying the regimes that characterize these oscillations, which occur in the central value of the electron temperature, as measured by X ray and/or ECE diagnostics. Sawteeth are generally present in ohmic as well as in RF heated plasmas. It is shown by numerical analysis that these plasmas are stable to ideal internal, pressure driven modes. Hence, the modes responsible for the crash phase of the sawteeth have to be resistive in nature. This is confirmed by numerical computations using a resistive code. The role of energetic ions, produced by ion cyclotron heating, is also discussed.

An equation is derived for fast magnetoacoustic eigenmodes (FMEs) in a tokamak fusion plasma with an elongated cross-section. It is shown that edge localized fast magnetoacoustic waves (FMWs) with both positive and negative poloidal wavenumbers are possible, provided that the product a² exceeds a certain critical magnitude N_cr, where is the plasma density at a point near the plasma edge and a is the plasma radius. Fulfilment of this condition in various devices is analysed, which is important in explaining the fine structure of the spectrum of suprathermal ICE from tokamak plasmas. It is found that a small population of high energy ions with a non-equilibrium distribution function can significantly affect the FMEs.

Recent experiments in the RTP tokamak have shown that the electron temperature, T_e, profile in EC heated L mode discharges can have only a small number of distinct profile shapes, depending on the location of the additional heating. The sharp transitions between the profile shapes are associated with the loss or gain of a low rational q surface. These observations suggest that the electron thermal diffusivity χ_e is a direct function of the safety factor q, with alternating layers of low and high χ_e. It is shown that such a model not only excellently describes the distinct profile shapes but also predicts other, observed but so far unexplained, phenomena, such as sharp off-axis maxima of T_e observed with off-axis heating at specific radii.