Table of contents

Based upon latest results from JET and other tokamaks and upon model projections, the operating conditions of a fusion reactor are predicted. The technical and scientific issues involved in continuous operation of such a reactor using noninductive drive are identified. Candidate techniques include injecting beams of high energy neutral particles and radio waves at various frequencies (such as fast, lower hybrid and electron cyclotron waves). The basis for a continuously operating reactor is not apparent and a convincing demonstration would require either a high current drive efficiency (above 10²⁰ A/m²W) at a density above approximately 10²⁰ m³; or ignition and adequate impurity control at a density of 5*10¹⁹ m^-3 with moderate current drive efficiency ( approximately 0.5*10²⁰ A/m²W); or high power operation in a regime with a dominant bootstrap current. Semi-continuous operation with inductive current drive offers the only viable alternative for long pulse reactor operation. This could use either forward current or alternating current operation, provided the central solenoid was sufficiently large. A tokamak reactor operating semi-continuously would be simpler in construction, use re-circulating power more efficiently and would likely be more reliable in operation. It is proposed that the Next Step tokamak be based on inductive semi-continuous operation.

Co-injection (D^o to D⁺) applied to TEXTOR leads to a hot ion mode regime with enhanced confinement. A synergistic increase of the beam effects is observed with the addition of ICRH at omega =2 omega _cD= omega _cH (H minority heating scenario) resulting, beside other reviewed effects, in a significant increase of the ion temperature and of the beam driven current (respectively larger than 30% and 50% for the addition of an RF power comparable to the NBI one). The large ion heating efficiency of ICRH also remains when ICRH is added to balanced injection and the hot ion mode regime remains up to the maximum achieved beta (=2/3 of the Troyson limit with more than 6 MW of auxiliary heating). ICRH also leads to the formation of more 'isotropic' tail. These results re interpreted with the help of Fokker-Planck code which computes the beam distribution function in the presence of RF and of TRANSP simulations. The amount of RF absorption by the H minority, by the ion beam and the bulk plasma is theoretically evaluated.

The latest developments of the JET ICRH and LHCD systems are reviewed. Feedback controls have alleviated the traditional difficulties with plasma-antenna coupling. Experimental results on high plasma performance, basic confinement aspects and issues relevant to the next generation of tokamaks are described. ICRH allows a narrow power deposition profile to be localised at will independently of the plasma density and size. The authors demonstrate that in large tokamaks, the constraint to use a minority species can be removed and heating a balanced D/T mixture at omega = omega _CD in ITER is proposed. LHCD allows the highest current efficiency to be obtained. The conditions for wave accessibility are discussed in light of the JET results. LHCD is well adapted to save volt seconds in the current rise phase. The observed synergistic LHCD/ICRH acceleration of fast electrons is a promising route towards higher current drive efficiency but raises unresolved physics aspects which are analysed. They finally consider the prospects of a steady state fusion reactor.

Current drive experiments have been performed on the DIII-D tokamak using fast waves, electron cyclotron waves, and neutral injection. Fast wave experiments were performed using a 4-strap antenna with 1 MW of power at 60 MHz. These experiments showed effective heating of electrons, with a global heating efficiency equivalent to that of neutral injection even when the single pass damping was calculated to be as small as 5%. The damping was probably due to the effect of multiple passes of the wave through the plasma. Fast wave current drive experiments were performed with a toroidally directional phasing of the antenna straps. Currents driven by fast wave current drive (FWCD) in the direction of the main plasma current of up to 100 kA were found, not including a calculated 40 kA of bootstrap current. Experiments with FWCD in the counter current direction showed little current drive. In both cases, changes in the sawtooth behavior and the internal inductance qualitatively support the measurement of FWCD. Experiments on electron cyclotron current drive have shown that 100 kA of current can be driven by 1 MW of power at 60 GHz. Calculation with a Fokker-Planck code show that electron cyclotron current drive (ECCD) can be well predicted when the effects of electron trapping and of the residual electric field are included.

Radio frequency methods of heating and noninductive current drive have become well established and are likely to be part of any next-step Tokamak programme. The author reviews the development of antennae for ion cyclotron heating and developments to enhance the effectiveness of fast wave current drive systems. The performance achieved by systems enables the provision of an ICRH system for next step devices within the existing technology limits. The main lower hybrid current drive systems are also reviewed. Present operating limits suggest that the design power density at the grill in large multijunction launchers must be somewhat reduced due to peaking of the electric field. The resulting launcher for a next step machine based on present technology is a large is a large and highly complex device. Development of proposals such as the rod array or the hyperguide, in parallel with necessary improvements in the current drive efficiency would make lower hybrid a more attractive method of noninductive current drive for the next step machines.

Two themes in current drive theory in tokamaks are reviewed, both relevant to the progression of tokamak experiments toward the reactor regime. The author reviews the understanding of the physics of tail electrons. These electrons are capable of carrying enormous RF-driven electric current, and, in the course of current-drive experiments worldwide not only has the current drive effect been demonstrated, but the underlying physical description of these tail electrons has been established. Anticipating the presence of the energetic alpha particles that result from D-T reactions in a reactor, the author examine's certain mechanisms through which these alpha particles can be used to facilitate current-drive.

With lower hybrid current drive (LHCD), very long pulses can be obtained in TORE SUPRA. Volt-second saving studies by LHW assisted current ramp-up, current drive and electron heating are discussed. A 12 second stationary discharge with improved energy confinement time (40%) by peaking the current profile is presented with a central electron temperature up to Te(o) approximately 10 keV. A 1 minute flat top time duration discharge has been achieved with a plasma current I_p=1 MA:80% of this current is driven by LHW. During this discharge the core density is fed continuously by a deuterium gas puff indicating that the inner first wall is still pumping all along the shot. Finally results concerning the interaction of LHW with high density plasmas (peak density approximately 10²⁰ m^-3) and 3 MeV fusion protons are given.

The kinetic theory of radiofrequency current drive in tokamak plasmas is investigated. The problem of the impact of anomalous transport on the driven current profile and efficiency is considered in detail. Among the possible candidates for explaining anomalous transport in tokamaks, magnetic turbulence is known to have a strong influence on the dynamics of superthermal electrons and is assumed to be the basic mechanism responsible for radial diffusion of the RF-driven current. The 3-dimensional kinetic equation in the presence of RF heating and magnetic turbulence is studied. Its properties are first investigated by a nonlocal response function technique. Both the adjoint formalism and the Langevin equations method are extended to the case of radially diffusing electrons. The full kinetic equation is numerically solved by means of a 3-D Fokker-Planck code. Applications to lower-hybrid current drive are presented and several kinetic effects are discussed.

Numerical simulation of heating and current drive in the ion cyclotron frequency range has progressed to the point that both the propagation and absorption of IC waves on the one hand, and the evolution of ion and electron distribution functions due to interaction with the HF fields on the other hand, can be described in realistic geometry and to some extent consistently with each other. The author reviews work in this field, with particular emphasis on the derivation and numerical solution of adequate equations to describe wave propagation in hot plasmas in toroidal geometry. Codes solving the quasilinear kinetic equations and their interface to wave propagation codes are also briefly mentioned. The complete problem is so complex, however, that in most cases simplified models must be used for extensive numerical studies of IC heating and current drive scenarios. Plane stratified models are appropriate to investigate antenna loading and global power balance; ray tracing gives excellent results in large hot plasmas and are easily interfaced to quasilinear kinetic codes. Full-wave codes in toroidal geometry are often simplified by neglecting either finite Larmor radius effects or the poloidal component of the static magnetic field, or both.

The author discusses the results obtained in the field of transport properties of current driven plasmas. The power balance is analysed and the results on power absorption discussed. The energy content in the suprathermal tail that carries the plasma current can be accounted for only by collisional slowing down and therefore direct power losses from tail electrons are negligible. The confinement time of the thermal bulk of the plasma degrades with the input power but displays a positive scaling with plasma density. Control of the current profile and of the MHD activity can considerably improve thermal diffusivity especially in the plasma centre. The threshold for the onset of the H-mode can be lowered in the presence of the fast electron tail and the frequency of the ELMs can be influenced by current profile control.

The author reviews the progress in the application of RF to heating and current drive for nuclear fusion made by the Japanese RF community.

The author presents examples of high-power RF applications covering a frequency range of a few MHz up to microwave frequencies of several GHz. All examples refer to the RF generation of nonfusion plasmas meanwhile for many important technical processes. Some specific items are: inductively driven RF discharge for ion beam extraction (for space propulsion, ion surface treatment and ion implantation), RF-plate reactor (for etching and thin-film deposition in microelectronics), and newly developed microwave plasma sources (ECH discharges for large-area coatings of up to 200 m²/h, or for internal coating of pipes.

The enhancement of reactivity that can be gained from using auxiliary heating methods which transfer the power to fuel ions is investigated by using a transport code (JETTO) equipped with an RF package that computes the ion tails created by minority fundamental ion-cyclotron heating. The feasibility of ion-cyclotron heating scenarios is briefly examined from the point of view of absorption efficiency, power distribution among species and parasitic alpha -particle absorption. The main body of the work consists of the analysis of transport simulations of auxiliary heated plasmas. Methods resting on electron and ion heating are compared under three different confinement assumptions. In each case, the importance of RF-induced tail effects is also investigated. The differences between the electron and ion heating schemes are explained on the basis of a simple O-D power balance which summarizes the essential features of the JETTO simulations. It is concluded that significant benefit can be obtained from using ion heating methods in the case of reactors operating at moderate fusion Q's (10-40) and even more if the Q is lower. The relative inefficiency of electron heating results from the competition between electron transport losses and equipartition in the electron power channel.

The functions of external power injection on a Next Step tokamak and the ways to cover these functions are discussed. It is concluded that a combination of methods will be needed and that, before the selection of the systems to be used on the Next Step device, in addition to the ongoing research on tokamaks a consistent effort on system design and development must be made.

In recent years, impressive progress has been made both on the physics understanding and on the power handling of the main RF systems: Ion Cyclotron Resonance Frequency (ICRF), Lower Hybrid Waves (LH), Electron Cyclotron Resonance Frequency (ECRF). In many large experiments, RF techniques are now being considered as a tool to improve plasma performances and to understand the physics of tokamaks or stellarators. For instance, H-modes have been achieved with all methods, with low levels of impurities, ICRF has allowed to heat and to produce plasmas at very high densities with high n tau T, as well as plasmas where most of the current is bootstrap driven and off-axis ICRF and ECRH have been used to understand energy radial transport. Large plasma currents have been driven by LH waves, up to 2 MA in conjunction with ICRF or Neutral beams. Fast Wave Ion current drive has been shown to have large efffects on MHD activity. Use of synergistic effects between LH and ICRF waves appear to be a possible route to increase current drive efficiency.

More than 200 MJoules of ICRF and 100 MJoules of LH power have been launched recently on one minute long plasma discharges with plasma current up to 2 MA. The development of powerful ECRH sources at high frequency is steadily progressing. Steady state ignited plasmas require current drive efficiencies which are still exceeding present achievements, but a few approaches to the problem have been discussed. Heating to ignition with ICRF is very promising both on the physics and on the technical side since present power and energy handling capability are comparable to those needed for a reactor. In addition to reaching ignition in fusion reactors, other roles for RF waves, such as burn control and profile control, have been extensively discussed.