Table of contents

Measurements of low power (≃ 1 mW) antenna loading are used to study the coupling of a compact loop antenna structure to plasmas in the divertor configuration in DIII-D heated by neutral beam injection (NBI) or electron cyclotron heating (ECH). When a transition to the H-mode regime occurs during NBI, the antenna loading resistance drops by approximately a factor of two. This coupling decrease is due to a steepening of the edge density profile near the separatrix, accompanied by a reduction in edge density in the scrape-off layer. During edge localized modes, the opposite effects occur, and the antenna coupling increases transiently. The loading measurements are compared with theoretical calculations which take into account the measured density profiles as well as the conducting side-walls of the recessed antenna housing. Absolute agreement between the theoretical and the experimental results is obtained, including the correct dependence on the density, antenna position, RF frequency and antenna geometry. The theoretical interpretation of the results is discussed, together with the technological implications for future high power experiments.

Auxiliary heating by neutral beam injection (NBI) or ion cyclotron resonance heating (ICRH) can broaden or steepen the plasma current density profile. A measure of changes to the current density profile J_ϕ is established by a combination of current moments which yields the profile broadening parameter ζ. JET experiments are demonstrated to increase ζ (broaden J_ϕ) when NBI is applied to discharges which exhibit either L-mode or H-mode confinement; pellet injection and ICRH or current decay are demonstrated to decrease ζ (peak J_ϕ). The time evolution of the parameter ζ is shown to correlate with the changes to the D_α emission seen in H-mode discharges with and without edge localized modes. The broadening of J_ϕ by NBI, especially for H-mode pulses, smarts in the edge region and it is identified with the non-inductive bootstrap current component as a consequence of steep density profiles. Associated with current profile broadening is an increase in the edge pressure and the total stored plasma energy W. An analysis of global confinement (variations of W with power P) is made by splitting W into three parts: the fast ion energy, W_f, the edge energy, W₁, which is empirically shown to be proportional to the current profile broadening parameter ζ, and the bulk plasma thermal energy, W_p. It is shown that the scaling for W_p in the L- and H-mode confinement regimes is similar.

In WT-3, lower hybrid waves are injected into a microwave discharge at the electron cyclotron resonance frequency to initiate the toroidal plasma current I_P and to sustain it by radiofrequency (RF) power alone without Ohmic heating power (this is called the 'RF tokamak'). First, fast electrons in the range of several kilo-electronvolts are generated by electron cyclotron and lower hybrid power which initiate a weak toroidal plasma current. Then, fast electrons above 100 keV are produced by the lower hybrid waves, and I_P increases rapidly (ΔI_p/Δt ∼ 780 kA·s⁻¹) to almost 13 kA. Subsequently the toroidal plasma current is ramped up slowly (ΔI_p/Δt ∼ 180 kA·s⁻¹) to > 25 kA for low densities (_e < 6 × 10¹⁸ m⁻³). The fraction ∈_R (≤8%) of injected lower hybrid power is converted into poloidal magnetic field energy. For high densities (_e = (6−12) × 10¹⁸ m⁻³), a quasi-steady-state discharge is obtained with I_P = 7−20 kA, and the figure of merit η_CD is about 0.05 × 10₁₉ A·m⁻²·W⁻¹. The experimental values of ∈_R and η_CD are explained by the direct loss of fast electrons.

The time evolution of neutron emission has been measured in sawtoothing discharges in the Frascati Tokamak (FT). The observed change in the neutron emission level during the sawtooth crash has been studied in terms of variations of the deuteron density and temperature. Changes of the peak values of these quantities as well as profile changes have been considered. It is shown that the occurrence of a net ion energy outflow through the mixing radius after a sawtooth collapse can explain the observed drop in the neutron emission. The model is also used to infer the ion thermal conductivity near the plasma centre which is found to be compatible with neoclassical predictions.

The ideal and resistive MHD free-boundary stability of m = 0, 1, 2 and 3 modes is calculated for the circular force-free reversed field pinch (RFP). Functions μ(r)= μ₀·/B² specifying the equilibrium configurations are chosen to approximate the deviations of experimental RFPs from the fully relaxed state. Families of unstable modes are identified. All are resistive in the usual RFP parameter range, unless the distance to the conducting shell is very large. The usually stable m = 2 and 3 modes, as well as m = 0 and m = 1, are unstable near the plasma edge for practical shell separations, and these modes might promote greater edge relaxation than if plasma and shell were in contact. When a small central peaking of μ(r) is present, the unstable mode spectrum hardly varies as the shell is moved outward. It is suggested that extreme closeness of the shell to the plasma might not be required. Moderate plasma-shell separation might also eliminate sawtoothing at high reversals. Supporting evidence from OHTE and HBTX 1C experiments is cited.

Toroidal rotation velocity profiles during neutral beam injection on ASDEX have been investigated by means of charge exchange recombination spectroscopy. For both co- and counter-injection of neutral beams the ratio of the injected power and the line averaged electron density is found to be an essential experimental, parameter for the central rotation speed. With counter-injection, a continuous rise of rotation velocity and density is observed, indicating strong improvement of angular momentum confinement. This improvement is found to be strongly correlated with the peakedness of the density profile. Simple model calculations of momentum transport indicate that the increase of momentum confinement with density peaking can be partly attributed to the variation of torque deposition with the shape of the density profile.

The authors study the mechanisms by which light (oxygen, carbon) and heavy (metal atoms) impurities are released from the walls in the Uragan-3 torsatron, in which radiofrequency (RF) heating is the only way of heating (and producing) the plasma. Having investigated the relationship between the quantity of impurities in the peripheral plasma, as well as the mean density, the confining magnetic field and the power input, they show that metal impurities appear as a result of sputtering from metal surfaces. Carbon, too, like the metal atoms, enters the plasma as a result of physical sputtering of carbon bearing materials. However, oxygen production occurs in a different way — through chemical interaction between hydrogen and metal oxides on the surface. By comparing impurity concentrations in the edge plasma and the fluxes of charge-exchange atoms as functions of _e, power input and magnetic field, the authors show that these fluxes do not play a decisive role, either in surface sputtering of the antenna and the magnet casing or in the chemical interaction responsible for the release of oxygen from the surfaces. Sputtering from the surface of the magnet casing that faces the confinement volume is shown to be governed mainly by the high frequency current associated with RF potential oscillations of the peripheral plasma. Sputtering from the surface of the RF antenna is due to its interaction with the quasi-constant ion current, with an energy of ≥ 100 eV produced during RF heating. The formation of water molecules on the surface is due to interaction of surface oxides with hydrogen atoms formed during the dissociation of H₂ molecules.

A visible bremsstrahlung detector array system for Z_eff measurements and a charge exchange recombination spectroscopy system for fully ionized impurity profile measurements have been installed on JIPP T-IIU to study impurity transport in neutral beam injection (NBI) and ion cyclotron resonance frequency (ICRF) heated plasmas. More impurities are sputtered during ICRF heating than during NBI and/or Ohmic heating. The contribution of carbon to Z_eff is 80-90% for NBI heated plasmas and 60% for NBI + ICRF heated plasmas. With carbon coating of the vacuum vessel, the contribution of carbon to Z_eff becomes 80-90% also for NBI + ICRF heated plasmas. From the radial profile and the time evolution of fully ionized carbon after the ICRF pulse is turned on, a diffusion coefficient D_a of 1.0 m²·s⁻¹ and a convective velocity v_a(a) of 13 m·s⁻¹ for the carbon impurity at the plasma edge are obtained.

The carbon concentration in the central region of TFTR discharges with high power neutral beam heating has been measured by charge exchange recombination spectroscopy (CXRS) of the C⁵⁺ n = 3–4 transition in the extreme ultraviolet region. The carbon concentrations were deduced from absolute measurements of the line brightness using a calculation of the beam attenuation and the appropriate cascade corrected line excitation rates. As a result of the high ion temperatures (20–30 keV) in most of the discharges, the contribution of beam halo neutrals to the line brightness was significant and therefore had to be included in the modelling of the data. Carbon concentrations have been measured in discharges with plasma currents I_p in the range 1.0-1.6 MA and beam power in the range 2.6–30 MW, including a number of supershots. The results are in good agreement with carbon concentrations deduced from the visible bremsstrahlung Z_eff and with metallic impurity concentrations measured by X-ray pulse height analysis, demonstrating the reliability of the atomic rates used in the beam attenuation and line excitation calculations. Carbon is the dominant impurity species in these discharges; the oxygen concentration measured via CXRS in a high beam power case was 0.0006 of n_e, compared to 0.04 for carbon. Trends with plasma current and beam power in the carbon concentration and the inferred deuteron concentration are presented. The carbon concentration is independent of plasma current and decreases from 0.13 at 2.6 MW beam power to 0.04 at 30 MW, while the deuteron concentration increases from 0.25 to 0.75 over the same range of beam power. These changes are primarily the result of beam particle fuelling, as the carbon density did not vary significantly with beam power. The time evolutions of the carbon and deuteron concentrations during two high power beam pulses, one which exhibited a carbon bloom (a sudden influx of carbon due to local heating of the limiter) and one which did not, are compared. In both types of discharge, the carbon concentration decreases early in the beam pulse as a result of beam particle fuelling, and the carbon density rises slowly during the beam pulse until the start of the bloom. The electron density rise during the bloom is primarily due to the increase in the carbon density.

The hard X-ray emission from the TRIAM-1M tokamak plasma during steady state lower hybrid current drive with a discharge duration of a few minutes was measured with sodium iodide scintillation spectrometers. The radial profiles of the X-ray emission were also measured and indicate that, in the low density regime (n_e = (1−3)×10¹² cm⁻³), the current carrying high energy electrons are mainly in the inner region of the plasma column and their radial profile remains unchanged during current drive. On the other hand, high density discharges (n_e = (3−6) × 10¹² cm⁻³) are always accompanied by an abrupt drop of the plasma current, and the X-ray emission profile changes from peaked to broad. This change can be attributed to the conditions of wave accessibility. As the electron density increases, the accessibility of the plasma to lower hybrid waves with low values of the parallel wave number n₁ is significantly reduced and high energy electrons resonating with the waves are produced at the plasma periphery. Interaction of these electrons with the limiters causes an increase of the electron density in this region; waves with low n₁ then become completely excluded from the inner part of the plasma column. This interpretation is supported by measurements of the density profile and impurity radiation, and has been confirmed in an investigation of discharges with additional gas puffing.

Impurity concentration profiles have been determined for H-mode discharges in the DIII-D tokamak from measured n_e, T_e, Z_eff and radiated emissivity profiles. The central impurity levels in DIII-D high current H-modes, as modelled using this technique, remain below those seen in L-modes (fractional nickel concentrations ≤0.02%) throughout the neutral beam heating pulse. In contrast to some other experiments (ASDEX [1], JET [2], JFT-2M [3]), the H-mode does not terminate because of excessive radiation in DIII-D discharges heated with co-injected neutral beams. For increasing plasma current, the global impurity concentrations decrease and the profiles become more dominated by edge radiation. H-modes as obtained with electron cyclotron heating and co-injected neutral beams at similar heating powers also have low impurity levels, but the impurity distribution is significantly more hollow in the case of neutral beam heating.

Electrostatic and magnetic turbulence has been investigated in the edge region of the TJ-I tokamak using Langmuir and magnetic probes. The relative levels for density fluctuations are in the range ñ/n ≈ 0.4–0.8 and for magnetic fluctuations /B_t ≈ 5 × 10⁻⁵. Density and potential fluctuations do not scale in the same way as a function of plasma radius (i.e. ñ/n ≠ e/T_e). The particle confinement time inferred from electrostatic fluctuations is found to increase with increasing magnetic field and electron density. The global particle confinement times estimated from absolute H_α emission intensity measurements are in good agreement with the particle confinement times inferred from electrostatic fluctuations. The contribution of magnetic fluctuations to the edge particle transport is negligible.

The use of radiofrequency (RF) fields in the megahertz range to produce the initial plasmas for nuclear fusion experiments can provide a broader operating regime for many stellarator devices and can result in a saving of volt-seconds in tokamaks. This process has been used for many years in various devices at the Institute of Physics and Technology at Khar'kov. The initial breakdown of neutral gas by non-resonant RF near-fields can be understood in terms of RF electric fields parallel to the static magnetic field and a ponderomotive trap with a long scale length parallel to the static magnetic field. If the parallel fields are not too strong for a given antenna geometry, the concept of a ponderomotive trap is valid and the trapped electrons can oscillate in the near-field region for many RF periods. Efficient breakdown can begin if the oscillation energy of these trapped electrons is sufficient to ionize the gas. Good agreement between the experimental observations and an analytic model for the early phase of the breakdown process is presented.

The authors apply a two-step tokamak regression procedure to L-mode confinement data. The resulting scaling expression for τ_E: is similar to other published scaling expressions. Using the random coeffcient model, tokamak-to-tokamak variations are included in the estimate of the uncertainty. The uncertainty factor in the predicted confinement times is calculated to be 19% for ITER and 35% for CIT. The confinement uncertainties are dominated by the propagation of uncertainties arising from the tokamak-to-tokamak variation in scalings with q and n to the second-step regression and, for CIT, by the lack of high magnetic field, moderate size devices in the L-mode database.

The paper presents a study of the effect of a perturbing magnetic field on a field whose magnetic surfaces are tori nested around a closed central line. This perturbation effect creates magnetic islands around surfaces with rational rotational transform. These islands are investigated analytically, which makes it possible to evaluate their size. The resulting turbulence of the medium can then be studied by calculating the interaction of two neighbouring islands.

The large multiplicity of scaling laws which almost fit the data is associated with the near degeneracy of the data covariance matrix. The traditional analysis assumes that only discharge to discharge variations are important. Using Swamy's random coefficient model, tokamak to tokamak variations are included in the statistical analysis.

The setup conditions for a very low q (surface safety factor less than 2) tokamak have been studied. For the startup of a very low q discharge, the rate of rise of the current should be higher than a limiting value; this value is a function of the plasma density and is almost independent of the strength of the toroidal magnetic field. Because of the relatively high ramping rate of the current, the distribution of the safety factor assumes a hollow profile during the startup phase.

It is shown that a plateau-like scaling gives a good overall fit to energy confinement data of Ohmic and L-regime discharges for devices ranging from Pulsator to JET. Systematic deviations become evident in density scans in single experiments which, at least for neutral injection heated discharges, might be explained by a compensation of confinement trends by penetration effects.

The mechanism that modifies the current profile during lower hybrid current drive has been investigated numerically for JT-60. A one-path absorption model of lower hybrid waves with a bi-Maxwellian electron distribution can simulate the experimentally observed current profile flattening. The analysis of the MHD activity of the simulated current profile is consistent with the measurements.

The first IAEA Technical Committee Meeting on Physics and Diagnostics of Fusion Produced Alpha Particles was hosted by the Kiev Institute for Nuclear Research (INR, Academy of Sciences of the Ukrainian SSR). The programme and organizing committee consisted of the following members: F. Engelmann (ITER, Garching), Ya.I. Kolesnichenko (Chairman, INR, Kiev), D.J. Sigmar (Massachusetts Institute of Technology, Cambridge), K. Tani (Japan Atomic Energy Research Institute) and H. Wilhelmsson (Chalmers University of Technology, Goteborg). The meeting was attended by more than 50 participants from nine countries; more than 40 papers were presented. The purpose of the meeting was to discuss alpha particle physics issues of magnetically confined fusion plasmas, to sum up the present theoretical understanding of alpha particle physics and the planned experiments with D-T plasma operation, and to point out the most important alpha particle problems which should be solved in view of the projection to fusion reactors.