Table of contents

The accretion theory of `spontaneous' toroidal rotation connects this phenomenon to the energy and particle transport properties of the plasma column and to the relevant collective modes. The consequent prediction that an inversion of the velocity direction in the transition from the H to the L regime should occur has been verified by the experiments. The theory is consistent also with the observation that the velocity is depressed when a peaked density profile appears as a result of a transport barrier. The fact that the rotation velocity increases with the total energy content is explained by the fact that the inflow of angular momentum, whose source is at the edge of the plasma column, results from the excitation of modes driven by the plasma pressure gradient. A quasi-linear derivation of the angular momentum transport produced by these modes, whose novel feature is the inflow, is given. A model of the relevant (transport) equation that can be easily solved is discussed. Fluctuations at the edge of the plasma column are considered responsible for the scattering out of confinement of particles that transfer to the surrounding material wall angular momentum in the same direction as that of the phase velocity of the prevalent modes. Thus the fact that the plasma rotates in the direction of the ion diamagnetic velocity in the H regime, when the prevalent modes are expected to have phase velocity in the direction of the electron diamagnetic velocity, can be explained. The rate of rotation decrease observed when the plasma current is increased is also justified.

Resistive instabilities and wall stabilization of ideal kink modes with low toroidal mode number n are investigated in JT-60U reversed shear discharges. Localized burst-like MHD activity with n⩽3 is observed in the negative shear region with large pressure gradient near the internal transport barrier; this activity is identified as the resistive interchange mode. The burst-like MHD activity is benign with respect to the internal transport barrier and no clear degradation of plasma stored energy by the burst-like MHD activity is observed. The resistive interchange mode, however, results in a major collapse through non-linear coupling with a tearing mode in the positive shear region. A resistive wall mode (RWM) followed by a major collapse is observed in wall stabilized high β discharges exceeding the ideal stability limit with the wall at infinity (β_N>β_N^no-wall). No clear continuous slowing down of the plasma toroidal rotation is observed before the growth of an RWM.

The physics of the ITER edge and divertor plasma is strongly coupled with the divertor and the fuel cycle design. Owing to the limited space available the design as well as the remote maintenance approach for the ITER divertor are highly optimized to allow maximum space for the divertor plasma. Several auxiliary systems (e.g., in-vessel viewing instruments and glow discharge electrodes) as well as a part of the pumping and fuelling system have to be integrated together with the divertor into the lower level of ITER. Two main options exist for the choice of the plasma facing material in the divertor, i.e. tungsten and CFC. On the basis of already existing R&D results it is likely that the material choice will be mainly determined by physics considerations and material issues (e.g., C-T co-deposition). The requirements for the ITER fuel cycle arise from plasma physics as well as from the envisaged operation scenarios. Owing to the complex dynamic relationship of the fuel cycle subsystems among themselves and with the plasma, codes are employed for their optimization. These interacting issues are elaborated on the latest design status discussed.

The resistive X point mode is shown to be the dominant mode in boundary plasmas in X point divertor geometry. The poloidal fluctuation phase velocity from the simulation results of the resistive X point turbulence shows a structure across the separatrix that is experimentally measured in many fusion devices. The fluctuation phase velocity is larger than the E × B velocity in both L and H mode phases. It is also demonstrated that there is a strong poloidal asymmetry of particle flux in the proximity of the separatrix. Turbulence suppression in the L-H transition results when sources of energy and particles drive sufficient gradients, as in the experiments.

Impurity injection with noble gases has been used in DIII-D to increase radiation in the mantle region, with confinement enhancements above the ITERL-89P L mode scaling relation in both diverted and limited discharges. For discharges with an L mode edge, impurity injection produces a prompt increase in confinement and a more gradual increase in density. These changes occur at densities and radiated power fractions significantly lower than those observed in the TEXTOR tokamak device. ELMing H mode discharges with active pumping and high deuterium gas feed (puff and pump) exhibit an increase in density with no degradation in energy confinement after impurity injection, increasing to nearly the Greenwald density limit following a spontaneous transition several hundred milliseconds after impurity injection. The highest density phase of both L mode and ELMing H mode radiating mantle discharges is usually terminated after the onset of n = 2 MHD activity, identified as an m/n = 3/2 neoclassical tearing mode. A reduction in density fluctuations after impurity injection in the mantle region has been measured using beam emission spectroscopy in L mode discharges and is coincident with reductions in thermal diffusivity and increases in core toroidal rotation. The similarities and differences between these types of impurity seeded discharge will be presented.

An experimental study of the extraordinary mode (X mode) absorption at the third electron cyclotron harmonic frequency has been performed on the TCV tokamak in plasmas preheated by X mode at the second harmonic. Full single pass absorption of injected X3 power was measured with X2 preheating in co-current drive (CO-ECCD). The measured absorption exceeds that predicted by the linear ray tracing code TORAY-GA by more than a factor of 2 for the CO-ECCD case. Experimental evidence indicates that a large fraction of the X3 power is absorbed by electrons in an energetic tail created by the X2 ECCD preheating.

The Electric Tokamak, a low field ITER sized device with R = 5 m, has been operating with well equilibrated clean plasmas since January 2000. Short, 0.9 s, discharges with a central energy confinement time τ_E(0) = 150 ms are now routinely obtained at a toroidal field B = 0.1 T with kT_e, kT_i≃120 eV. The discharges are feedback controlled in up/down position and in plasma current. Biased electrode driven H modes have been obtained that compare well with the results obtained on CCT by R.J. Taylor and align with the `neoclassical bifurcation' theory of K.C. Shaing. Very successful second harmonic ion heating has been demonstrated with the ICRF antenna outside the vacuum system and with 50% single pass absorption. ICRF heated discharges indicate that poloidal rotation sufficient for edge bifurcation (H mode) may soon be achieved by ICRF induced fast ion losses. The threshold electrode biasing current required for bifurcated poloidal rotation has so far been reduced by 70% owing to ICRH driven ion orbit loss. The remaining critical ICRF item needed for the exploration of high beta plasma equilibria is the demonstration of the required current profile shaping. It is expected that mode conversion in the ion-ion hybrid regime, with high field side launch, will allow the current drive required to approach and exceed the Troyon beta limit. In 1-D full wave calculations, high harmonic current drive appears most promising at higher beta. Achieving the goal of plasma equilibration near unity beta will require 10 s long discharges (at kT = 3 keV, n_e = 3×10¹⁹ m^-3, B = 0.25 T) because of current profile shaping requirements.

Densities of up to 40% above the Greenwald limit are reproducibly achieved in high confinement (H_ITER89P = 2) ELMing H mode discharges. Simultaneous gas fuelling and divertor pumping were used to obtain these results. Confinement of these discharges, similar to moderate density H mode, is characterized by a stiff temperature profile, and is therefore sensitive to the density profile. A particle transport model is presented that explains the roles of divertor pumping and geometry for access to high densities. The energy loss per ELM at high density is a factor of five lower than the predictions of an earlier scaling, based on data from lower density discharges.

The effect of ELMs on the vertical position observer in TCV has been investigated and a new, optimized, observer which minimizes the influence of ELMs has been constructed. The new observer has been tested in TCV, under closed loop conditions, and it is shown to be considerably less sensitive to ELMs than the classical observer. In order to determine whether this method could also be used in JET, ELMy H mode discharges and vertical displacement events in JET are analysed and a modified vertical position observer is constructed, optimized for JET conditions. It is shown that, if this new observer had been used in JET, the perturbations caused by ELMs would have been greatly reduced.

First results of gyrokinetic analysis of JET ELMy H mode plasmas are presented. ELMy H mode plasmas form the basis of conservative performance predictions for tokamak reactors of the size of ITER. Relatively high performance for long durations has been achieved and the scaling appears to be favourable. It will be necessary to sustain low Z_eff and high density for high fusion yield. The article studies the degradation in confinement and the increase in the anomalous heat transport observed in two JET plasmas: one with an intense gas puff and the other with a spontaneous transition from type I to III ELMs at the heating power threshold. Linear gyrokinetic analysis gives the growth rate γ_lin of the fastest growing modes. The flow shearing rate γ_E×B and γ_lin are large near the top of the pedestal. Their ratio decreases approximately at the time when the confinement degrades and the transport increases. This suggests that tokamak reactors may require intense toroidal or poloidal torque input to maintain sufficiently high γ_E×B/γ_lin near the top of the pedestal for high confinement.

The dominant causes of the reduction in the enhancement factor of thermal energy confinement with increasing plasma density in ELMy H mode plasmas in JT-60U were analysed by separating the thermal stored energy into core and pedestal components. The enhancement factor of the core confinement evaluated by the offset non-linear scaling which was developed in the ITER H mode confinement database decreased significantly with increasing density, while the pedestal component remained almost constant over a wide range of density. The influence of the edge pedestal structure imposed by the destabilization of type I ELMs on thermal energy confinement of the core plasma was also quantitatively clarified. As the density increased, the saturation of the pressure at the pedestal shoulder, p_ped, caused by the type I ELM activities forced a reduction in the pedestal temperature. The core temperature for each species, in turn, decreased only by an approximately constant factor with a reduction in the pedestal temperature, resulting in the saturation of the thermal stored energy in the high density regime.

A detailed comparison of core and edge confinement with different heating methods - NBI and ICRH - has been carried out in the ELMy H mode regime in JET with the Gas Box divertor. Transport in the core and characteristics of the edge pedestal have been assessed in discharges at 2.0 MA/2.6-2.7 T at a total input power level of 11-12 MW. The thermal core confinement has been found to be higher by about 10% in ICRH dominated discharges. Although this difference is well within the uncertainties of the thermal confinement estimation, it has been consistently found in similar experiments in the past and it may be related to the more peaked power deposition provided by ICRH. Local transport analysis carried out with the TRANSP code indicates that, independently of the NBI versus ICRH mix, ion conduction losses are the dominant energy loss channel. In contrast to previous experiments, the gas flow and density of NBI and ICRH discharges have been closely matched. In these conditions it has been found that both types of heating yield similar values of edge density and temperature and produce similar ELMs. The fact that the edge fast ion concentration can be varied from roughly 0.4% up to 4% without producing significant changes in the edge pedestal parameters is an indication that fast ions do not always play a dominant role in the edge stabilization between ELMs, especially in discharges with strong gas fuelling.

Work on the acceleration of a compact toroid plasma configuration between coaxial electrodes is reported. In the experiment the maximum poloidal field component and the full width at half maximum of the poloidal field are shown to increase and decrease with the accelerator voltage, respectively. The velocity of the compact toroid is shown to increase with accelerator voltage and then saturate as the accelerator voltage is increased above approximately 11 kV. The saturation in the velocity and field components of the compact toroid is due to crowbarring of the accelerator insulator. The crowbarring of the insulator is consistent with the onset of the `blowby' effect, which is the most likely triggering source.

An on-line predictor of the time to disruption has been installed on the ASDEX Upgrade tokamak. It is suitable either for avoidance of disruptions or for mitigation of those that are unavoidable. The prediction uses a neural network trained on eight plasma parameters and their time derivatives extracted from 99 disruptive discharges. The network was tested off-line over 500 discharges and was found to work reliably and to be able to predict the majority of the disruptions. The trained network was installed on-line, tested over 128 discharges and used to inject killer pellets to mitigate the disruption loads.

In Huguet M 2001 Nuclear Fusion41 1503, on page 1509, Fig. 7. Replace with the figure shown here.

Figure 7. CS conductor jacket options: Incoloy 908 (left) with reinforcing strip, and double jacket (Ti-SS) (right).