Table of contents

The longest tokamak discharge, with a duration of 11 406 s (3 h 10 min), has been achieved. The global particle balance has been investigated. In the longest discharge, the global balance between the particle absorption and release of the wall was achieved at t ∼ 30 min. After that, the plasma density was maintained by the recycling flux alone until the end of the discharge. The maximum wall inventory is about 3.6 × 10²⁰ H at t ∼ 30 min, but it is finally released from the wall at the end of the discharge. The hydrogen release seems to be caused by the temperature increase in the whole toroidal area of the main chamber. Moreover, it has been observed that there is a large difference between the properties of wall recycling in the continuous gas feed case (i.e. static condition) and in the additional gas puff case (i.e. dynamic condition). In the static condition, the effective particle confinement time increases to ∼10 s during the 1 min discharge and it increases to ∼100 s before the global balance in the longest discharge. In the dynamic condition, the decay time of the electron density just after the gas puff, i.e. the effective particle confinement time, is constant at 0.2–0.3 s during the discharge. The large difference in the effective particle confinement time between the static and dynamic conditions seems to be caused by the reduction in the recycling coefficient due to the enhanced wall pumping resulting from the additional gas puff.

The results of studies on heating and current drive by the electron cyclotron (EC) waves in JT-60U are presented. An electron temperature of up to 26 keV, as measured by ECE diagnostics, was achieved by injecting EC waves into the centre of a reversed shear plasma produced by the lower hybrid (LH) waves. The electron temperature T_e exceeds 24 keV in the region ρ < 0.3, where ρ is the normalized minor radius. The EC driven current profile was measured at high T_e up to 21 keV without using LH waves, and was found to be spatially localized. The ECCD (current drive) efficiency η_CD increased with T_e and reached 0.42 × 10¹⁹ A W⁻¹ m⁻² at T_e = 21 keV. The dependence of normalized CD efficiency on deposition location was also studied to optimize the CD efficiency, since the trapped particle effect, which depends strongly on deposition location, is expected to reduce ζ. The effect was detected from a significant decrease in ζ in the lower magnetic field deposition, which is consistent with a linearized Fokker–Planck calculation. The measured CD efficiency ζ increased with electron density n_e for the same T_e, which showed a stronger dependence on n_e compared with the linear calculation. Further off-axis ECCD profile at about half of the minor radius was measured, showing fairly good agreement with linear calculation.

The late-time (t ≥ 80 ns) behaviour of hohlraums designed for the National Ignition Facility (NIF) is simulated using the multiphysics radiation hydrodynamics codes LASNEX and HYDRA. The spatial distribution of x-radiation outside the hohlraum is shown as a function of time. The energy spectrum of the x-ray emission is presented for various hohlraum viewing angles. We have made refinements to the grid motion algorithms in the arbitrary Lagrangian Eulerian (ALE) hydrodynamics code HYDRA to obtain the first late-time simulations of a hohlraum that can be used to give the spatial distribution of the vaporized hohlraum wall. The importance of late-time simulations in determining the lifetimes of debris shields on NIF is discussed.

The heavy ion beam (HIB) indirectly driven compression of a small amount of DT fuel, and its fast ignition by a beam of laser accelerated protons (PB) is considered (Roth M. et al 2001 Phys. Rev. Lett.86 436). First, a representative working point (0.8 mg of DT at 400 g cm⁻³) has been derived analytically from the energy balance of an hypothetical power plant. Second, through self-consistent two-dimensional radiation hydrodynamic simulations with the MULTI code (Ramis R. et al 1988 Comput. Phys. Commun.49 475) we define and optimize a HIB-driven hohlraum target to produce the above conditions; also, with ignition simulations, we determine the thermonuclear yield (≃83 MJ) and the minimum PB energy (≃33 kJ) needed to ignite the asymmetrically compressed fuel. The target is found to be a factor 5 less sensitive to implosion asymmetries than a conventional central spot one. This allows us to reduce the hohlraum size (≃6 mm), improve the driver coupling, reduce the compression energy to the 0.8–1.2 MJ range, and have enough clearance for the installation of the PB converter foil at distances (3–4 mm) appropriate for an efficient ignition.

A Clemmov–Mullaly–Allis (CMA) type diagram, the ECW–CMA diagram, for representing the operational possibilities of electron cyclotron heating and current drive (ECRH/ECCD) systems for fusion plasmas is presented. In this diagram, with normalized density and normalized magnetic field coordinates, the parameter range in which it is possible to achieve a given task (e.g. O-mode current drive for stabilizing a neoclassical tearing mode) appears as a region. With also the Greenwald density limit shown, this diagram condenses the information on operational possibilities, facilitating the overview required at the design phase. At the operations phase it may also prove useful in setting up experimental scenarios by showing operational possibilities, avoiding the need for survey type ray-tracing at the initial planning stages. The diagram may also serve the purpose of communicating operational possibilities to non-experts. JET and ITER like plasmas are used, but the method is generic.

Studies of reactor designs for inertial fusion energy (IFE) have indicated that some of the material requirements can be alleviated if the target chamber is filled with a low-pressure inert gas (<1 Torr). The experiments described in this paper examine the effects of such an ambient gas on the laser-target interactions relevant to laser-driven IFE. These experiments used the Nike KrF laser facility at the Naval Research Laboratory to irradiate uncoated 78 µm thick polystyrene (CH) foil targets at focal intensities around 10¹⁴ W cm⁻². They compared the laser–target interactions in different inert gases (xenon, argon or krypton) at several ambient pressures up to 1 Torr with that of the vacuum (reference) case. The diagnostics included streaked rear-surface emission at wavelengths 460–540 nm and side-on shadowgraphy at 263 nm. The results of this study were encouraging. In spite of initial concerns about resonantly enhanced nonlinear optical effects in the Xe gas, we found no evidence of laser beam break-up, spreading, or attenuation. The shock breakout profiles (from emitted light) and the front and rear surface plasma profiles (from shadowgraphy) remained smooth and symmetric in all cases. The central shock breakout times, and thus the central laser irradiances, were about the same for the target in an inert gas as in vacuum. Our analysis indicates that the large nonlinear optical processes in atomic Xe are self-limiting because the converging beam ionizes the gas, thereby removing the atoms from the most intense part.

Extensive studies of the H-mode density limit (DL) in JET gas-puffed discharges have been performed in the past four years targeting at an improved database for extrapolation to ITER. This paper reviews the arguments for the particular DL definition (pedestal density at the H–L boundary), the logic underlying the choice of parameters under focus (toroidal field, major radius, triangularity, safety factor) as well as some improvement in the interpretation of typical JET density ramp-up signatures that led to a critical review of the existing data. An empirical scaling is derived and compared with existing empirical and model based scalings. ASDEX Upgrade data are included in this analysis to provide information on the size dependence. The main results are: earlier findings on the B_t, R and q₉₅-dependences are confirmed. The triangularity dependence, if any, is weak. The SOL-based BLS (Borrass, Lingertat, Schneider) scaling and the empirical scaling are virtually indistinguishable. The Greenwald scaling provides values in the right absolute range, but the overall fit is comparatively poor. The proposed scaling predicts ITER critical densities considerably below the reference value. Fuelling methods other than gas-puffing are outside the scope of this paper, but pellet fuelling, envisaged for ITER as an option to alleviate the situation, is discussed.

Small-scale quasi-coherent oscillations of the x-ray emissivity and poloidal magnetic field perturbations are identified in high density plasmas in the T-10 tokamak with the use of a toroidal view x-ray imaging system, in-vessel hard x-ray detector and fast magnetic probes. Small-scale x-ray oscillations are associated with forward bremsstrahlung emission of the suprathermal electrons (E_γ ∼ 25–100 keV) from the plasma region where the m = 2, n = 1 magnetohydrodynamic (MHD) mode is localized. The oscillations are typically represented by a solitary high frequency (15–85 kHz) harmonic in the spectrum of perturbations accompanying MHD modes during density limit disruptions. During a disruption energy quench, the small-scale oscillations are often transformed to repetitive bursts of the nonthermal x-ray radiation localized in the outer part of the plasma. Analysis indicates a possible connection of the oscillations with nonthermal electron beams induced during magnetic reconnection around X-points of the m = 2, n = 1 magnetic islands. It is argued that the small-scale oscillations can be generated due to the modulation of the electron beams moving through an equilibrium magnetic field with ripples during rotation of the m = 2, n = 1 mode.

Plasma impurity content depends on the impurity sources, fuelling efficiency, and confinement. In JET, carbon is the primary impurity, and its fuelling efficiency has been studied using methane gas injection and modelled with the scrape-off-layer (SOL) codes: DIVIMP and EDGE2D. In this paper, EDGE2D modelling of similar ASDEX Upgrade experiments and projections to ITER are described. The parameters that govern the size scaling of carbon screening have been identified.

Size scaling is complex. For carbon injected from the main chamber, the important factors include: the SOL temperature, the magnitude of the thermal force at the divertor entrance, and the parallel distance to the divertor. For carbon injected at the strike points, the intersection of the carbon ionization region with the region of strong thermal force determines the carbon fuelling efficiency.

ITER projects to have much better carbon screening than JET. The ITER SOL is hotter so that main chamber carbon is ionized further from the separatrix making the calculated carbon fuelling efficiency lower. Also, the carbon originating near the strike point has less chance of escaping the divertor since the ITER divertor is larger. The carbon sputtering is projected to be larger, making the ITER core contamination difficult to estimate. A general result is that the core contamination at fixed total sputtering rate and core impurity confinement increases when the fraction of carbon ionized in the main chamber SOL increases, and decreases for larger machine size and higher density operation.

Joint experiments have been conducted on JET and DIII-D to study m = 2, n = 1 neo-classical tearing modes (NTMs). Very similar instability behaviour is observed on both machines and the scaling of the mode island width with β and the observation of n = 1 islands at q = 2, strongly points to these modes being driven by neo-classical effects. Identity experiments using closely matched non-dimensional parameters (plasma shape, aspect ratio, q, ρ^* and collisionality) on JET and DIII-D show similar global β-limits for the NTM. Further scalings, which combine broader data sets from JET and DIII-D, show an approximately linear scaling, or slightly stronger, with ion gyro-radius and a relatively weak dependence on collisionality.

A radiation-preheated direct-drive target is examined where the fuel isentrope is varied by changing the high-Z dopant concentration in the ablator. As usual, the growth of the Rayleigh–Taylor instability as determined by a one-dimensional stability analysis can be decreased when the isentrope increases. The target performance is found to be quite similar to previously designed targets where the ablator and fuel isentropes were increased via an initial shock. However, the gain decreases slightly more because some of the laser input energy is carried away as radiation and does not contribute to the kinetic motion of the target on its way inward. This design adds to the options available for high-gain targets that are stable enough against hydrodynamic instability.

Edge polarization experiments were carried out on ISTTOK using an emissive electrode. The dependence of the bias current on the radial electric field, measured by a radial array of Langmuir probes, is in qualitative agreement with the expression derived by Stringer for the transitional region between plateau and collisional transport regimes (Stringer T.E. 1993 Nucl. Fusion33 1249). Improved particle confinement is clearly observed for negative bias associated with a large radial electric field. In some cases, a double-peaked profile of the radial electric field is observed just after biasing, evolving afterwards to a single-peaked profile as confinement improves. These types of profiles may be related to the multiple solutions of the non-linear equation for the radial electric field, recently put forward by Kasuya et al (2003 Nucl. Fusion43 244). Above a certain threshold of the bias current stronger improved confinement events are observed, during short periods, for both bias polarities.

On lines 3 and 5 in the right-hand column of page 433, the following corrections should be made:

(1 - β_pR²0) should be (1 - β _p)R²₀

β_p (=0.60) and R₀ (=3.32 m) should be β_p (=0.70) and R₀ (=3.50 m)