Table of contents

In the aperture limiter shadow of the T-10 tokamak, the temporal evolution of plasma density and electron temperature during single discharges is studied by using a Langmuir double probe. It is observed that the plasma of the scrape-off layer (SOL) passes through a stationary phase when the Ohmic heat input U_loop I_p also does, suggesting that the stationarity of the SOL is correlated with the stationarity of the discharge power balance. During this stationary phase the plasma density at the aperture limiter edge was found to be approximately proportional to the square of the line averaged core plasma density. – It is shown in the paper that this quadratic density scaling can be easily understood within the framework of a stationary particle balance model of the recycling process, i.e. the interaction of plasma and neutral particles of the working gas in the edge region. Quadratic density scaling is expected to hold generally if the plasma density is high enough to prevent the neutrals from reaching the centre. For high densities, on the other hand, there is a density limit defined by the requirement that the discharge should supply enough power to sustain recycling.

A series of measurements of the plasma boundary parameters have been made on JET using an array of Langmuir probes and heat flux probes. The profiles decay exponentially, with typical e-folding lengths of 30–60 mm for the density and 50–100 mm for the temperature. The cross-field diffusion coefficient estimated from these data is 0.5–1 m²·s⁻¹. Using the particle fluxes and energies, the calculated impurity production is generally consistent with that observed spectroscopically. The measured profiles have also been used to calculate the screening of impurities from the wall due to ionization in the boundary layer. It is found that metals from the wall are ionized in the boundary layer but are likely to be deposited on the limiter from where they will be re-sputtered into the plasma. The light impurities from the wall, such as carbon and oxygen, are not well screened. From the data, estimates of the electric field in the plasma have been made which indicate significant × drifts in the poloidal direction.

The paper deals with the formulation of microinstability-based thermal transport coefficients (χ_j) for the purpose of modelling anomalous energy confinement properties in tokamak plasmas. Attention is primarily focused on ohmically heated discharges and the associated anomalous electron thermal transport. An appropriate expression for χ_e is developed which is consistent with reasonable global constraints on the current and electron temperature profiles as well as with the key properties of the kinetic instabilities most likely to be present. Comparisons of confinement scaling trends predicted by this model with the empirical Ohmic database indicate quite favourable agreement. The subject of anomalous ion thermal transport and its implications for high density Ohmic discharges and for auxiliary heated plasmas is also treated.

In the JIPP T-IIU tokamak an experiment to demonstrate the feasibility of fast wave current drive using five loop antennas has been successfully carried out with a relatively high density plasma (ω_pe²/ω_ce²∼5). The RF frequency is 40 MHz and the toroidal field is 2 kG, which corresponds to ω = 13ω_cH. The experiment is conducted in the density range _e ∼ 2 × 10¹⁸ m⁻³ where only the fast wave can propagate, eliminating the possibility of slow wave current drive. This density is two orders of magnitude higher than the density limit predicted for slow wave current drive. The dependence of the drive efficiency on the relative phase difference Δφ is clearly observed with a maximum of about Δφ = π/4. The plasma current was limited by MHD instability which begins to occur around q_a = 10.

A Fokker–Planck formalism has been developed to interpret experimental data from high power electron cyclotron heating of tokamak plasmas in which the electron distribution function can be substantially distorted. The Fokker–Planck equation is solved using a 2-D code which incorporates electron trapping, a steady Ohmic electric field and a bounce averaged electron cyclotron heating term. The quasilinear RF diffusion coefficient is calculated using a single particle model which includes the mildly relativistic resonance condition and takes into account localized RF power injection and tokamak rotational transform effects. Comparisons are presented of the calculated and observed soft X-ray spectra from 60 GHz second harmonic electron cyclotron heating experiments on the CLEO tokamak. Good agreement between theory and experiment is found. In addition, predicted RF current drive efficiencies for the COMPASS tokamak are presented.

An experiment has been undertaken to see if chemical sputtering of carbon can be directly observed under tokamak operating conditions. The test material consists of a well instrumented graphite probe limiter which can be heated up to 900°C under steady state conditions and up to >1200° C with additional plasma heating during a discharge. The carbon released from the limiter is monitored spectroscopically by using a visible spectrometer and a CCD camera with interference filters. Measurements have been made from 20°C to 700°C in steady state conditions and up to 1200°C with plasma heating. Within experimental error, there is no change in the carbon production rate over the whole temperature range. The inferred chemical sputtering yields are low, in agreement with beam simulations that indicate reduced yields at high incident current densities, low energies and in the presence of metal surface contaminants.

A one-dimensional radial model for tandem mirror transport is developed for the study of the effect of temperature gradients. Induced azimuthal electric fields are neglected for simplicity. In the limit of large collisional energy exchange between electrons and ions – often appropriate to present-day tandem mirror experiments – the confinement time scaling law τ_⊥ ∼ (ΔΦ)^−3/2 is obtained for TMX-U.

Observations of impurity behaviour are presented from ICRF heating experiments at 180 MHz performed for a variety of conditions on the Alcator C tokamak, using graphite limiters and stainless steel antenna Faraday shields. Spectroscopic observations revealed significant increases in metal impurity concentrations during the RF pulse, with iron levels increasing by as much as a factor of twelve at the highest RF powers (about 350–400 kW). Analysis of the inferred iron source rates shows an approximately linear dependence on RF power up to 400 kW, with no clear dependence on resonance conditions or bulk plasma parameters. However, a sharp temperature increase in the limiter shadow region was observed during the ICRF pulse, which was well correlated with the iron influx rate. From this and other evidence it is concluded that physical sputtering of the Faraday shield due to an elevated sheath potential is the primary source of metal impurities during ICRF heating on Alcator C. The same process, occurring at the graphite limiter, is believed to be the dominant source of carbon and oxygen. The calculated sputtering yields obtained from an edge erosion code demonstrate the plausibility of this model.

The thermal stability of a collisional, radiating edge plasma in magnetic confinement experiments with respect to marfe-type perturbations is investigated analytically and numerically. The fluid description used includes a non-local electron heat flux model and non-equilibrium impurity radiation cooling. The dispersion relation of a homogeneous plasma is derived analytically, and stability criteria are given. Various experimental scenarios are simulated with a sophisticated numerical model including all the relevant dynamics along field lines and a rough approximation of cross-field transport. The formation and history of global 'marfes' and the onset of relaxation oscillations under appropriate conditions are demonstrated. The possible occurrence of more localized perturbations is discussed. The results compare well with those of experiments.

Experimental flux, energy and particle lifetimes are reported for a wide range of plasma conditions (T_i = 100–800 eV, r_s = 3.5–7 cm) in the slow risetime TRX-2 field reversed theta pinch. The lifetimes exhibit a scaling with the parameter r_s/ρ_i0^1/2 to an exponent of between 2.7 and 3.6. This favourable scaling, especially for the flux lifetime, implies an average plasma resistivity which scales as the drift parameter v_E/v_i to some power greater than unity. For present small devices, typical flux diffusion coefficients are about 4 m²·s⁻¹ and, based on the above scaling, would decrease rapidly in larger devices with smaller pressure gradient scalelengths. The above scaling is obtained only for well formed FRCs with low levels of formation turbulence, and it remains to be seen if this low turbulence level can be maintained in larger, less kinetic, FRCs.

The trapped electron drift mode has been studied analytically, including effects of toroidal couplings and finite beta, and the results are in good agreement with numerical solutions. The mode is found to be unstable for beta values corresponding to both the first and the second stability region for MHD ballooning modes. For beta values above the second stability limit the solutions are found to come close to that in the electrostatic limit while the mode becomes more localized to the outside of the torus.

The MHD current driven instabilities of Reversed Field Pinch (RFP) configurations are analysed for a plasma in contact with a perfectly conducting wall. The equilibrium distributions are obtained on the basis of a model which allows a parametric variation of the value of the safety factor on the axis, q(0), and of the current density distribution. – The RFP is found to be stable in a wide range of parameters. However, when the toroidal field reversal becomes too deep, unstable modes, resonant outside the reversal surface, are found in analogy to the stability limit at θ = 1.56 for the Taylor's theoretical, fully relaxed states. Nevertheless, it is shown that these modes are not very significant in that they arise in a region of the parameters space which is only rarely approached in experiments. On the other hand, for peaked current density distributions, new unstable tearing and kink modes are found when q(0) drops below the limit q(0) ≈ 2a/(3R). These modes are resonant inside the reversal surface and may include the mode resonant on the axis. – The results of the MHD stability for the internal modes and in particular the limit on the value of q on axis that has been found are discussed in connection with experimental observations on mean field profiles and related oscillations.

Using LASNEX, the authors investigate the effect of initial fuel aspect ratio on the hydrodynamic efficiency and implosion stability of 1 mg ICF capsules directly driven with 0.27 μm light. It is concluded that these capsules will be either hydrodynamically inefficient, with less than 8% of the incident laser energy converted into fuel kinetic energy, or hydrodynamically unstable, with an in-flight aspect ratio greater than 80, unless the instability growth rate can be reduced to a small fraction of its classical value.

This letter reports on two experiments undertaken to evaluate the retention of gaseous and target produced impurities in the ASDEX divertor. The retention for gaseous impurities was determined by puffing Ar into the main chamber and simulating the time behaviour of the Ar XVI line intensity with a time dependent impurity transport code including a simple divertor model. During Ohmic heating a factor of 3 and 4.5 enhancement of impurity retention if found relative to the vacuum time constant (90 ms) of the divertor chamber, for n_e = 2 × 10¹³ cm⁻³ and n_e = 3.5 × 10¹³ cm⁻³, respectively, while a drastic breakdown of the retention occurs during high power NI heating. – To deduce the retention of impurities generated at the divertor plates, a segment (3.5%) of the plates was covered with copper, a metal previously not used in ASDEX. By measuring the Cu influx at the target plates and the line intensity of the Cu XX line (11.38 Å) in the core plasma and by using the transport code, it is found that during NI heating (n_e ≤ 2 × 10¹³ cm⁻³) Cu atoms originating from the target plates have a ≤ 3.5 times higher probability to penetrate into the core plasma than if they had when originating from the main chamber walls.

It is argued that transport models with sufficiently small central (relative to edge) thermal conductivity, so as to allow sawteeth at the highest q, can easily explain 'profile consistency' in tokamaks. This includes the observed increase in temperature profile peakedness T(0)/⟨T⟩ with increasing q and the relatively small variation of the peakedness with variation in heating profile. In fact, it is proven that if q(0) = 1 and the Spitzer relation between current and temperature profiles holds, then the peakedness is bounded by q^2/3 ≲ T(0)/⟨T⟩ ≲ q. A distinction is made between the central sawtooth mechanism which 'tops' the profiles and possible q dependent edge mechanisms which 'shrink' profiles. The latter are not precluded from contributing to profile consistency and may be required to understand confinement time scaling.

It is shown that the injection of suitable amount of electron cyclotron power can stabilize the Parail–Pogutse instability driven by lower hybrid waves in the current drive regime. The computation is explicitly performed for the ordinary mode in perpendicular propagation.

The International Symposium on Heavy Ion Fusion held 27–29 May 1986 in Washington, D.C., was the third in a series of symposia previously held in Darmstadt (1982) and Tokyo (1984). These symposia continued the strong tradition of international co-operation fostered at a preceding series of workshops on heavy ion fusion held in the USA (Berkeley, 1976 and 1979; Brookhaven National Laboratory, 1977; and Argonne National Laboratory, 1978).