Table of contents

The current drive and heating properties of negative ion based NBI have been studied comprehensively in JT-60U. It has been confirmed from shine-through measurements of the injected beam (350 keV) that multistep ionization processes are essential in the ionization processes of high energy particles. The profile of the current density driven by a negative ion based NB (N-NB) has been determined experimentally. This is in good agreement with the theoretical prediction, and N-NB driven current reached 0.6 MA with E_B = 360 keV and P_INJ = 3.7 MW. The current drive efficiency η_CD is increased by increasing electron temperature and improved by increasing beam energy. The fast ions from N-NBs are well confined in the enhanced confinement core by the weak poloidal magnetic field of reversed shear plasmas. A clear H mode transition was obtained with N-NB dominant heating, where the net absorbed power required for an H mode transition seemed similar to the previous result obtained in JT-60U using a low energy beam (90 keV). With the strong electron heating by N-NBI (80% absorbed by electrons), an H factor ( = τ_E/τ^ITER-89PL_E) of 1.64 with T_e(0) = 1.4T_i(0) was obtained in the steady state ELMy phase.

In JT-60U lower hybrid current drive (LHCD) experiments, a reversed magnetic shear configuration that was accompanied by internal transport barriers was successfully maintained in a quasi-steady state almost entirely non-inductively by the LH driven current and the bootstrap current. The normalized β was kept near 1 and the neutron emission rate was nearly steady as well, indicating no accumulation of impurities in the plasma. Diagnostics showed that the profiles of the electron and ion temperatures, the electron density and the current were almost unchanged during the LHCD phase. The applicability of LHCD to H mode plasmas has also been investigated. It was found that the LH waves can be coupled to an H mode edge plasma with a plasma-wall distance of up to about 14 cm.

Current profile control experiments were carried out in TRIAM-1M by the combination of two lower hybrid waves (LHWs) at 8.2 and 2.45 GHz with different spectra of parallel refractive index N_||. In the combined lower hybrid current drive (LHCD) discharge, the current drive efficiency was improved by a factor of >1.5 compared with 8.2 GHz LHCD only and 2.45 GHz LHCD only. Measurements of the internal inductance and the radial profile of hard X ray emission indicated that the current profile can be controlled by changing the additional power of the 2.45 GHz LHW to the 8.2 GHz LHCD plasma. The improvement of the current drive efficiency and the modification of the current profile are considered to be attributable to the enhancement of the absorption of lower N_|| waves by higher N_|| waves in the two kinds of LHWs. Moreover, the controlled current profile can be successfully sustained for about two orders of magnitude longer than the current diffusion time.

In fast wave to ion Bernstein wave mode conversion experiments in DT supershot plasmas, localized efficient ion heating rather than electron heating was observed, which was due to Doppler broadened tritium cyclotron resonance overlap into the mode conversion region. The ion temperature heat pulse associated with RF power modulation in this regime could provide a diagnostic tool for measuring the local ion thermal conductivity in various confinement regimes. In direct launch ion Bernstein wave heating experiments, core power coupling was limited by the excitation of parasitic edge modes. However, a sheared poloidal flow was observed that is consistent in both magnitude and direction with theoretical models based on RF driven Reynolds stress. With the modest power coupled to the core (∼360 kW), the magnitude of the shear in the observed flow was estimated to be a factor of 3-4 too low to trigger transport barrier formation through localized shear suppression of turbulence.

The quasi-steady-state (t_H > 10 τ_Eoh) H mode with high plasma density (ELMy and ELM free) was routinely obtained by the injection of lower hybrid wave heating and lower hybrid current drive with a power threshold of 50 kW. The antenna spectrum was scanned over a wide range and τ_E was about 1.5-2.0 times that of the L mode scaling. The density increases by almost a factor of 3 during the H phase by gas puffing and the particle confinement time increases by more than this factor even with a line averaged density of 3 × 10¹³cm^-3, which is about 60% of the Greenwald density limit. A hollow T_e profile was achieved in the high density case. The experimental results reproducibly show a good agreement with the theoretical prediction for the LH off-axis power deposition profile. When a certain fraction of the plasma current is non-inductively sustained by the LH waves, a hollow current density profile is formed and the magnetic shear is reversed. This off-axis hollow profile and enhanced confinement improvement are attributed to a strong reduction of the electron thermal diffusivity in the reversed shear region.

The article reviews the design and performance of the in-vessel components of ITER as developed for the Engineering Design Activities (EDA) Final Design Report. The double walled vacuum vessel is the first confinement boundary and is designed to maintain its integrity under all normal and off-normal conditions, e.g. the most intense vertical displacement events (VDEs) and seismic events. The shielding blanket consists of modules connected to a toroidal backplate by flexible connectors which allow differential displacements due to temperature non-uniformities. Breeding blanket modules replace the shield modules for the Enhanced Performance Phase. The divertor concept is based on a cassette structure which is convenient for remote installation and removal. High heat flux (HHF) components are mechanically attached and can be removed and replaced in the hot cell. Operation of the divertor is based on achieving partially detached plasma conditions along and near the separatrix. Nominal heat loads of 5-10 MW/m² are expected on the target. These are accommodated by HHF technology developed during the EDA. Disruptions and VDEs can lead to melting of the first wall armour but no damage to the underlying structure. Stresses in the main structural components remain within allowable ranges for all postulated disruption and seismic events.

Features incorporated in the design of the International Thermonuclear Experimental Reactor (ITER) tokamak and its ancillary and plasma diagnostic systems that facilitate operation and control of ignited and/or high Q DT plasmas are presented. Control methods based upon straightforward extrapolation of techniques employed in the present generation of tokamaks are found to be adequate and effective for ITER plasma control with fusion powers of up to 1.5 GW and burn durations of ⩾ 1000 s. Examples of simulations of key plasma control functions, including plasma magnetic configuration control and fusion burn (power) control, are given. The prospects for the creation and control of steady state plasmas sustained by non-inductive current drive and bootstrap current are also discussed.

The ITER site has not yet been determined. Nevertheless, to develop a construction plan and a cost estimate, it is necessary to have a detailed layout of the buildings, structures and outdoor equipment integrated with the balance of plant service systems prototypical of large fusion power plants. These services include electrical power for magnet feeds and plasma heating systems, cryogenic and conventional cooling systems, compressed air, gas supplies, demineralized water, steam and drainage. Nuclear grade facilities are provided to handle tritium fuel and activated waste, as well as to prevent radiation exposure of workers and the public. To prevent interference between services of different types and for efficient arrangement of buildings, structures and equipment within the site area, a plan was developed which segregated different classes of services to four quadrants surrounding the tokamak building, placed at the approximate geographical centre of the site. The locations of the buildings on the generic site were selected to meet all design requirements at minimum total project cost. A similar approach was used to determine the locations of services above, at and below grade. The generic site plan can be adapted to the site selected for ITER without significant changes to the buildings or equipment. Some rearrangements may be required by site topography, resulting primarily in changes to the length of services that link the buildings and equipment.

Since the main features of the design of the neutral beam (NB) system for the International Thermonuclear Experimental Reactor (ITER) were first reported, integration with the tokamak and with the rest of the plant has been the main priority. Moreover, operational requirements and maintainability have been considered in the evolution of the design. Each of the three NB injectors is connected to the tokamak vacuum vessel with the NB duct on an equatorial port. The article describes the integration of the NB port/duct with the blanket, the vacuum vessel, the toroidal field and poloidal field coils, the cryostat and the bioshield. Two main design modifications are reported. The insulation of the source, originally done with compressed gas, is now achieved with vacuum to limit the power losses caused by the radiation induced conductivity. Large cylindrical insulators are still required but their inner diameter has been reduced from 2.7 to 1.8 m. The improvements on the compensation system needed to reduce the magnetic field in the NB volume are also described. Finally, the progress in R&D for the ITER NB system is reported, including an overview of the achievements in the critical areas of negative ion production at high current density (tests of a large size, low pressure, steady state caesiated ion source), acceleration up to 1 MV (tests of two alternative accelerator concepts) and neutralization (tests of an experimental plasma neutralizer to investigate it as an alternative to the gas target neutralizer).

A plasma neutralizer (PN) could be an attractive upgrade for the NBI system of ITER. A multicusp magnetic trap with microwave discharge plasma generation has been proposed as a scheme for a PN. The experimental investigation of the scheme is being undertaken in the Kurchatov Institute. The preliminary experiments were done at a small (70 L of plasma volume, 2.45 GHz, 1.5 kW) installation, PNX-1, slightly modified during the experiments to PNX-2. The achieved plasma density of 5 × 10¹¹cm^-3 exceeded the cut-off density by a factor of 7. The next step in this experimental activity is PNX-U (500 L, 7 GHz, 50 kW). Some results of the PNX-U experiments are presented.

A system composed of a palladium diffuser and an electrolytic reactor has been proposed and then developed for the fuel cleanup system of ITER. The performance of the system has been studied in detail in a stand-alone test. A fuel simulation loop of ITER was constructed by connecting the fuel cleanup and hydrogen isotope separation systems developed; and the function of each system in the loop has been demonstrated. For tritium recovery from the exhaust gas during helium glow discharge cleaning of the vacuum chamber of ITER, a cryogenic molecular sieve bed system has been proposed and demonstrated.

In order to improve the safe handling and control of tritium for the ITER fuel cycle, effective in situ tritium accounting methods have been developed at the Tritium Process Laboratory in the Japan Atomic Energy Research Institute under one of the ITER-EDA R&D tasks. The remote and multilocation analysis of process gases by an application of laser Raman spectroscopy developed and tested could provide a measurement of hydrogen isotope gases with a detection limit of 0.3 kPa analytical periods of 120 s. An in situ tritium inventory measurement by application of a `self-assaying' storage bed with 25 g tritium capacity could provide a measurement with the required detection limit of less than 1% and a design proof of a bed with 100 g tritium capacity.

The thermal hydraulic characteristics in the vacuum vessel (VV) of a fusion reactor under an ingress of coolant event (ICE) and a loss of vacuum event (LOVA) were investigated quantitatively using preliminary experimental apparatuses. In the ICE experiments, pressure rise characteristics in the VV were clarified for experimental parameters of the wall temperature and water temperature and for cases with and without a blowdown tank. In addition, the functional performance of a blowdown tank with and without a water cooling system was examined and it was confirmed that the blowdown tank with a water cooling system is effective for suppressing the pressure rise during the ICE. In the LOVA experiments, the saturation time in the VV from vacuum to atmosphere was investigated for various breach sizes and it was found that the saturation time is in inverse proportion to the breach size. In addition, the characteristics of exchange flow through breaches were clarified for the different breach positions on the VV. It was proven from the experimental results that the exchange flow became a counter-current flow when the breach was positioned on the top of the VV and a stratified flow when it was formed on the side wall of the VV, and that the exchange flow under the stratified flow condition was smoother than that of counter-current flow. On the basis of these results, the severest breach condition in ITER was changed from the top-break case to the side-break case. To predict with high accuracy the thermal hydraulic characteristics during ICEs and LOVAs under ITER conditions, a large scale test facility will be necessary. The current conceptual design of the combined ICE-LOVA test facility with a scaling factor of 1/1000 in comparison with the ITER volume is presented.

Laser imprinting has been studied and, in particular, saturation of areal density perturbations induced by near single mode laser imprinting was observed. Several issues important for the foam buffered direct drive scheme have been investigated. These studies included measurements of the absolute levels of stimulated Brillouin and Raman scattering observed from laser irradiated low density foam targets, either bare or overcoated with a thin layer of gold. A novel scheme is proposed to increase the pressure in indirectly driven targets. By heating a foam supersonically that is attached to a solid target the pressure generated is not only the ablation pressure but also the combined pressure due to ablation at the foam-foil interface and the heated foam material. Planar brominated plastic foil targets overcoated with a low density foam were irradiated by a soft X ray pulse. The pressure was obtained by comparing the rear side trajectory of the driven target observed by soft X ray radiography with one dimensional radiation hydrodynamic simulations. Observations were also carried out of the transition from supersonic to subsonic propagation of an ionization front in low density chlorinated foam targets irradiated by an intense soft X ray pulse. The diagnostic for these measurements was K shell point projection absorption spectroscopy. In the fast ignitor area the channelling and guiding of picosecond laser pulses through underdense plasmas, preformed density channels and microtubes were investigated. It was observed that a large fraction of the incident laser energy can be propagated. Megagauss magnetic fields were measured, with a polarimetric technique, during and after propagation of intense picosecond pulses in preionized plasmas. Two types of toroidal fields, of opposite orientation, were detected. In addition, the production and propagation of an electron beam through solid glass targets irradiated at intensities above 10¹⁹W/cm² were observed using optical probing techniques.

A new fusion capsule drive scheme was investigated. The capsule is illuminated by a low intensity thermal X ray pulse prior to the main drive pulse. This leads to a noticeable suppression of initial imprinting by the drive beam because of thermal smoothing in the preformed plasma. Of the several types of indirect-direct hybrid target, the authors investigated the hybrid effect for two types. One is a foam hybrid, in which the fuel capsule has a low density foam layer attached directly on its surface and where pulsed radiation generated from a thin high Z layer on the foam propagates through the foam, creating a preformed plasma. The other is an external hybrid, in which the capsule is illuminated by external X ray radiation generated using different beams from the capsule drive beams. The hybrid effect was demonstrated for both types by imposing an initial imprint on a planar target with an intensity modulated beam, and subsequent non-uniformity growth due to Rayleigh-Taylor instability was observed by means of face-on backlighting. The observed suppression due to the presence of the preformed plasma is interpreted by a cloudy day model for both hybrids. Capsule implosion experiments have also been started. The overall implosion dynamics observed is replicated by 1-D hydrocode simulations. Preliminary results from the implosion experiments are presented.

The National Spherical Torus Experiment (NSTX) is being built at Princeton Plasma Physics Laboratory to test the fusion physics principles for the spherical torus concept at the MA level. The NSTX nominal plasma parameters are R₀ = 85 cm, a = 67 cm, R/a ⩾ 1.26, B_t = 3 kG, I_p = 1 MA, q₉₅ = 14, elongation κ ⩽ 2.2, triangularity δ ⩽ 0.5 and a plasma pulse length of up to 5 s. The plasma heating/current drive tools are high harmonic fast wave (6 MW, 5 s), neutral beam injection (5 MW, 80 keV, 5 s) and coaxial helicity injection. Theoretical calculations predict that NSTX should provide exciting possibilities for exploring a number of important new physics regimes, including very high plasma β, naturally high plasma elongation, high bootstrap current fraction, absolute magnetic well and high pressure driven sheared flow. In addition, the NSTX programme plans to explore fully non-inductive plasma startup as well as a dispersive scrape-off layer for heat and particle flux handling.

Significant progress has been made in the development of new, compact stellarator devices that could test optimization principles for the design of a more attractive reactor. These are 3 and 4 field period low aspect ratio quasi-omnigenous stellarators based on an optimization method that aims at improved confinement, stability, ease of coil design, low aspect ratio and low bootstrap current. With respect to the analysis of these devices, the authors have developed improved measures of neoclassical transport, collisionless loss analyses of ICRF tail and α population confinement and methods for the rapid assessment of ballooning stability boundaries.

Optimization studies have been done for the helical axis heliotron configuration. One purpose is to find a configuration suitable for experimental studies of the basic properties of a helical axis heliotron. In the present study, the role of the bumpy field component (toroidal mirror ratio) in MHD stability and neoclassical confinement for this type of configuration is examined. The physical mechanism of the improvement of the neoclassical transport through control of the bumpy field component is clarified. The physics design and current status of the new helical axis heliotron device, Heliotron J, are also described.

The Korea Superconducting Tokamak Advanced Research (KSTAR) project is the major effort of the national fusion programme of the Republic of Korea. Its aim is to develop a steady state capable advanced superconducting tokamak to establish a scientific and technological basis for an attractive fusion reactor. The major parameters of the tokamak are: major radius 1.8 m, minor radius 0.5 m, toroidal field 3.5 T and plasma current 2 MA, with a strongly shaped plasma cross-section and double null divertor. The initial pulse length provided by the poloidal magnet system is 20 s, but the pulse length can be increased to 300 s through non-inductive current drive. The plasma heating and current drive system consists of neutral beams, ion cyclotron waves, lower hybrid waves and electron cyclotron waves for flexible profile control in advanced tokamak operating modes. A comprehensive set of diagnostics is planned for plasma control, performance evaluation and physics understanding. The project has completed its conceptual design and moved to the engineering design and construction phase. The target date for the first plasma is 2002.

The scientific parameters and the technology issues for a modest size spherical torus (ST) at 10 MA plasma current are discussed. This class of devices includes a DT-capable ST experiment (DTST, R₀ = 1.2 m) for extended plasma performance tests for limited pulse lengths and neutron fluences, and a volume neutron source (VNS, R₀ = 1.1 m) for steady state energy technology testing to high neutron fluences. The scientific issues of interest for DTST include non-inductive ramp-up of plasma current on a limited timescale (∼30 s), the confinement needed for high Q burn, the behaviour of energetic particles, the physics and techniques to handle intense plasma exhaust, and the possibility of high performance plasma regimes free of disruptions or large disruption impact. Of further interest for the VNS would be steady state operation using large external current drive, possibly at a modest Q (∼1-2), achieving significant neutron wall loading (∼1 MW/m²) and a configuration relatively amenable to remote maintenance. A much longer timescale would be permitted in a VNS for non-inductive current ramp-up. The centre leg of the toroidal field coils, possibly multiturn for DTST and necessarily single turn for a VNS without significant nuclear shielding, presents technical and material issues of unique importance to the ST. Positive ion neutral beam injection and high harmonic fast wave (∼80 MHz) heating and current drive systems already available are likely to be adequate for DTST following pulse length extension to ∼50 s. Given an adequate physics database, the remaining enabling technologies needed for the VNS appear largely similar in nature to those of the ITER EDA design.

The article reports on the progress made by the ITER European Home Team in strong interaction with the ITER Joint Central Team and the Japan Atomic Energy Research Institute regarding several key aspects of the beam source for the ITER injectors: (1) Integration of the SINGAP accelerator into the ITER injector design. This is a substantially simpler concept than the multiaperture, multigap (MAMuG) accelerator of the ITER NBI reference design that has the potential for significant cost savings and that avoids some of the weaknesses of the reference design such as the need for intermediate high voltage potentials from the high voltage power supply and pressurized gas insulation. (2) High energy negative ion acceleration using a SINGAP accelerator. (3) Long pulse (i.e. >1000 s) negative ion source operation in deuterium. (4) RF source development, which could reduce the scheduled maintenance of the ITER injectors (as it uses no filaments), and simplify the transmission line and the auxiliary power supplies for the ion source.

In March 1998, the LHD project finally completed its eight year construction schedule. LHD is a superconducting (SC) heliotron type device with R = 3.9 m, a_p = 0.6 m and B = 3 T, which has simple and continuous large helical coils. The major mission of LHD is to demonstrate the high potential of currentless helical-toroidal plasmas, which are free from current disruption and have an intrinsic potential for steady state operation. After intensive physics design studies in the 1980s, the necessary programmes of SC engineering R&D was carried out, and as a result, LHD fabrication technologies were successfully developed. In this process, a significant database on fusion engineering has been established. Achievements have been made in various areas, such as the technologies of SC conductor development, SC coil fabrication, liquid He and supercritical He cryogenics, development of low temperature structural materials and welding, operation and control, and power supply systems and related SC coil protection schemes. They are integrated, and nowadays comprise a major part of the LHD relevant fusion technology area. These issues correspond to the technological database necessary for the next step of future reactor designs. In addition, this database could be increased with successful commissioning tests just after the completion of the LHD machine assembly phase, which consisted of a vacuum leak test, an LHe cooldown test and a coil current excitation test. These LHD relevant engineering developments are recapitulated and highlighted. To summarize the construction of LHD as an SC device, the critical design with NbTi SC material has been successfully accomplished by these R&D activities, which enable a new regime of fusion experiments to be entered.

The recent operational phase in JET in which DT fuel was used (DTE1) resulted in record breaking fusion performance. In addition to important contributions in plasma physics, the JET Team also made major advances in demonstrating the viability of some of the key technologies required for the realization of future fusion power. Two of the most important technological areas which have been successfully demonstrated in JET are the ITER scale tritium processing plant and the exchange of the divertor and maintenance of the interior of JET by totally remote means. The experiment also provided the first data on tritium retention and co-deposition in a diverted tokamak. Of the 35 g of tritium injected into the JET torus, about 6 g remained in the tokamak, residing mainly on cool surfaces at the inboard divertor side. The precise, safe and timely execution of the remote handling shutdown proved that the design, function, performance and operational methodology of the remote handling equipment prepared over the years at JET are appropriate for the successful and rapid replacement of components in an activated tokamak environment.

There is a global consensus among materials scientists and engineers that the qualification of materials in an appropriate test environment is indispensable for the design, construction and safe operation of demonstration fusion reactors as well as for the calibration of data generated from fission reactor and accelerator irradiations. In an evaluation process based on a series of technical workshops it was concluded that an accelerator driven D-Li stripping source would be the best choice to fulfil the requirements within a realistic time frame. In response to this need, an international design team with members from the European Union, Japan, the United States of America and the Russian Federation has developed, under the auspices of the International Energy Agency during a conceptual design activity phase (1994-1996), a suitable and feasible concept for an accelerator driven D-Li stripping source. This reference design for the International Fusion Materials Irradiation Facility (IFMIF) is based on conservative linac technology and two parallel operating 125 mA, 40 MeV deuteron beams that are focused onto a common liquid Li target with a beam footprint of 50 mm by 200 mm. The materials testing volume behind the Li target is subdivided into different flux regions: the high flux test region (0.5 L, 20-55 dpa/full power year), the medium flux test region (6 L, 1-20 dpa/full power year), and the low flux test region (>100 L, <1 dpa/full power year). The design developed was the basis for the conceptual design evaluation phase (1997-1998) and for subsequent engineering oriented activities. On the basis of comprehensive neutron transport calculations, an evaluation of the irradiation parameters and the available testing volumes has shown that the users' requirements can be fulfilled. Major engineering efforts have been undertaken to establish an IFMIF design that is based on available and already proven technologies. The design developed is based on extensive reliability, availability, maintainability and safety studies and is conceived for long term operation with a total annual facility availability of at least 70%.

A low aspect ratio tokamak transmutation system is proposed as an alternative application of fusion energy on the basis of a review of previous studies. This system includes: (1) a low aspect ratio tokamak as fusion neutron driver, (2) a radioactivity-clean nuclear power system as blanket, and (3) a novel concept of liquid metal centre conductor post as part of the toroidal field coils. In the conceptual design, a driver of 100 MW fusion power under 1 MW/m² neutron wall loading can transmute the amount of high level waste (including minor actinides and fission products) produced by ten standard pressurized water reactors of 1 GW electrical power output. Meanwhile, the system can produce tritium on a self-sustaining basis and an output of about 2 GW of electrical energy. After 30 years of operation, the biological hazard potential level of the whole system will decrease by two orders of magnitude.

The compact reversed shear tokamak CREST is a cost competitive reactor concept based on a reversed shear high β plasma and water cooled ferritic steel components. The moderate aspect ratio A = 3.4 and the elongation κ = 2.0 of CREST are very similar to the case of the ITER advanced mode plasma. Presentation of such a concept based on the ITER project should be worth while for formulating a fusion development strategy. The achievement of a competitive cost of electricity (COE) is the first priority for electric power industries. High β and high thermal efficiency are the most effective parameters for achieving a competitive COE. In order to achieve a high efficiency power plant, a superheated steam cycle has been adopted which permits a high thermal efficiency (η = 41%). Current profile control and high speed plasma rotation by neutral beam current drive stabilize the ideal MHD activity up to the Troyon coefficient β_N = 5.5. A cost assessment has shown that CREST could generate about 1.16 GW(e) electric power at a competitive cost.

A `pebble drop' divertor concept is proposed for future commercial fusion reactors. The marked feature of this system is the use of multilayer pebbles that consist of a central kernel and some coating layers. By using multilayer pebbles the pebble drop divertor has advantages such as steady state wall pumping with low bulk tritium retention. The performance of the whole divertor system depends on the characteristics of the multilayer pebble. In particular, the maximum heat load of the system is determined by the dimensions, the layer structure and the material of the kernel. The kernel also has an important role in determining surface temperature, which affects the wall pumping efficiency. Numerical results for the maximum allowable heat load and the surface temperature of the divertor pebble are presented. From the numerical estimation of thermal stress and surface temperature, it is found that the radius of a divertor pebble with a ceramic kernel should be 0.5-1 mm.

Global warming is one of the most serious problems which human beings are currently facing. Carbon dioxide (CO₂) from power plants is considered one of the major causes of global warming. In the present study, CO₂ emissions from tokamak fusion power plants are compared with those from present power generating technologies. Plasma parameters are calculated by a systems code that couples the ITER physics, toroidal field coil shape and cost calculation. CO₂ emissions from construction and operation are evaluated by multiplying component volume by the CO₂ emission intensities of the component materials. The reactor building, balance of plant, etc., are scaled from the ITER reference power reactor (`ITER-like') by use of the Generomak model. The most important finding is that CO₂ emissions from fusion reactors are less than those from photovoltaic systems and less than double those from fission reactors. The other findings are that: (i) Most CO₂ emissions from fusion reactors are from materials. (ii) CO₂ emissions from reactor construction account for almost 60-70% of the total, with the rest coming from reactor operation. (iii) The reversed shear reactor can reduce CO₂ emissions by half compared with the ITER-like reactor. It is concluded that tokamak fusion reactors are excellent for their low CO₂ emission intensity, and that they can be one of the effective energy supply technologies to solve global warming.

Results are presented from non-linear gyrokinetic simulations of toroidal ion temperature gradient turbulence and transport. The ion thermal fluxes are found to have an offset linear dependence on the temperature gradient and are significantly lower than gyrofluid or IFS-PPPL model predictions. A new phenomenon of non-linear effective critical gradients larger than the linear instability threshold gradients is observed and is associated with undamped flux surface averaged shear flows. The non-linear gyrokinetic codes have passed extensive tests, including comparison against independent linear calculations, a series of non-linear convergence tests and a comparison between two independent non-linear gyrokinetic codes. The most realistic simulations to date used actual reconstructed equilibria from experiments and a model for dilution by impurity and beam ions. These simulations highlight the importance of both self-generated and external E × B flow shear as well as the need for still more physics to be included.

It is becoming clear that tokamak anomalous transport is dominated by radially extended non-local modes which originate from strong toroidal coupling of rational surfaces in non-uniform plasmas. To aid in understanding the internal transport barrier (ITB) formed in reversed magnetic shear experiments, in addition to the well known shear flow effect, the article points out an important non-local effect and/or finite size effect which comes from the complex behaviour of the mode over a finite radial region around the minimum q (safety factor) surface. The non-local mode, which is characterized by its radial extent and the degree of tilting in the poloidal direction (Δr, θ₀), changes its structure depending on the sign of the magnetic shear, and as a result such modes are weakly excited across the q_min surface. This leads to a discontinuity or gap which disconnects the phase relation in the global wave structure across the q_min surface. Once such a discontinuity (or gap) is formed, transport suppression occurs and therefore a transport barrier can be expected near the q_min surface. The existence of this discontinuity is confirmed through use of a toroidal particle simulation. It is also shown that whether such a discontinuity is efficiently established depends on the presence of the radial electric field and the related plasma shear flow.

The requirements to stabilize microinstabilities with velocity shear in tokamaks are examined for both aspect ratio A = 3 and A = 1.4. A comprehensive linear gyrokinetic code is used to compute growth rates in realistic numerical equilibria. Growth rates for A = 3 and A = 1.4 are generally similar for electron drift modes and ion temperature gradient modes. Velocity shear is stronger at low aspect ratio; however, low A profiles have a stronger microtearing mode which is more difficult to shear stabilize. Nonetheless, low aspect ratio devices are predicted to have good confinement and may ignite at very small size. Profiles are presented which may allow high β, MHD stable operation at A = 1.4 with high confinement. The possibility of using a controlled application of non-turbulent loss processes to control profiles to prevent turbulent transport and to maximize β is suggested.

A theoretical model of an internal transport barrier (ITB) is developed. The transport model based on the self-sustained turbulence theory of the current diffusive ballooning mode is extended to include the effects of E × B rotation shear. Delayed formation of the ITB is observed in transport simulations. The influence of finite gyroradius is also discussed. Simulation of the current ramp-up experiment successfully described the radial profile of density, temperature and safety factor. A model of ITB collapse due to magnetic braiding is proposed. Sudden enhancement of transport triggered by overlapping of magnetic islands terminates the ITB. The possibility of destabilizing global low n modes is also discussed.

ECRH driven transport of suprathermal electrons is studied in non-axisymmetric plasmas using a new Monte Carlo simulation technique in 5-D phase space. Two different phases of the ECRH driven transport of suprathermal electrons can be seen. The first is a rapid convective phase due to the direct radial motion of trapped electrons and the second is a slower phase due to the collisional transport. The important role of the radial transport of suprathermal electrons in the broadening of the ECRH deposition profile in W7-AS is clarified. The ECRH driven flux is also evaluated and considered in relation to the `electron root' feature recently observed in W7-AS. It is found that, at low collisionalities, the ECRH driven flux due to the suprathermal electrons can play a dominant role in the condition of ambipolarity, and thus the observed electron root feature in W7-AS is thought to be driven by the radial (convective) flux of ECRH generated suprathermal electrons. A possible scenario for this type of electron root is considered for the LHD plasma.

The article presents an analysis of the typical features of shear Alfvén waves in tokamak plasmas in a frequency domain ranging from the `high' frequencies (ω ≅ v<sub>A</sub>/2qR<sub>0</sub>, where v<sub>A</sub> is the Alfvén speed and qR<sub>0</sub> is the tokamak connection length) of the toroidal gap to the `low' frequencies, comparable with the thermal ion diamagnetic frequency ω_*pi and/or the thermal ion transit frequency ω_ti = v_ti/qR₀ (where v_ti is the ion thermal speed).

Some self-consistent effects pertaining to feedback control of neoclassical tearing modes in high temperature large tokamaks are investigated. For the ECRH scheme of local electron heating, it is shown that the self-consistent bootstrap currents created by the driven pressure gradients within the island are comparable to those due to the usually considered resistivity change mechanism. Similar self-consistent currents can also arise from pressure gradients created by density and energy deposition from neutral beams, thereby offering a new possibility for neoclassical mode control. The stabilizing current in such an application of neutral beams is estimated. It is further shown that such a feedback scheme can be made even more effective through appropriate modulation of the beam source to match the phase variation arising from the island rotation.

A number of edge plasma physics phenomena are considered to determine tokamak performance: transport barrier, edge MHD instabilities and plasma flow. These phenomena are thought to be causally related: a spontaneous increase in the plasma flow (actually, its radial variation) suppresses heat and particle fluxes at the plasma edge to form a transport barrier; the edge pressure gradient steepens until limited by MHD instabilities, resulting in a temperature pedestal at the top of the steep gradient region; a number of core transport models predict enhanced confinement for higher values of the temperature pedestal. The article examines these phenomena and their interaction.

Three dimensional MHD simulations are executed in a full toroidal geometry to clarify the physical mechanisms of the internal reconnection event (IRE) which is observed in spherical tokamak experiments. The simulation results reproduce several main properties of IREs. A notable property in the linear growth of modes is the simultaneous excitation of multiple low n modes, which grow together with similar growth rates. A spontaneous phase alignment mechanism among these excited modes is found in the non-linear stage of the development, which causes a bulge-like deformation of the torus and subsequent expulsion of heat energy to the external region through the occurrence of external reconnection.

A situation in tokamaks is analysed in which the development of strong plasma current filamentation leads to non-linear magnetic island formation near resonant magnetic surfaces. It is shown that, in low magnetic shear configurations or in the case of strong current perturbations, positive islands can form in local regions with positive current modulation together with the usual negative islands, in which the modulation of the perturbation currents is negative. The positive magnetic islands can be a reason for the hot spot in the plasma centre during a sawtooth crash. The positive islands probably play an important role in m = 2 perturbation development during a major disruption and perhaps in locked mode dynamics.

Results are presented from the 3-D non-local electromagnetic turbulence code BOUT and the linearized shooting code BAL for studies of turbulence in tokamak boundary plasmas and its relationship to the L-H transition, in a realistic divertor plasma geometry. The key results include: (1) the identification of the dominant resistive X point mode in divertor geometry; (2) turbulence suppression in the L-H transition by shear in the E × B drift velocity, ion diamagnetism and finite polarization; (3) the fact that calculated values of the thermal diffusivities such as χ_i are comparable to those inferred from the experimentally measured plasma density and temperature profiles. On the basis of simulation results, a parameterization of the transport is given that includes the dependence on the relevant physical parameters.

A report is given of: (a) a substantial transport reduction by the turbulence driven E × B flows observed in three dimensional non-linear gyrokinetic simulations of microturbulence in magnetically confined toroidal plasmas; (b) an analytical derivation of the effective shearing rate for the time dependent E × B flow; (c) an interpretation of experimental data using linear gyrokinetic microinstability rotation models of E × B shear; (d) other developments in gyrokinetic theory and simulation.