Table of contents

Radiation effects in components and materials will be one of the most serious technological issues in nuclear fusion systems realizing burning-plasmas. Especially, diagnostic components, which should play a crucial role in controlling plasmas and understanding the physics of burning-plasmas, will be exposed to high-flux neutrons and gamma rays. Dynamic radiation effects will affect the performance of components substantially from the beginning of exposure to radiation environments, and accumulated radiation effects will gradually degrade their functioning abilities in the course of their service. High-power-density fission reactors will be the only realistic tools to simulate the radiation environments expected to occur in burning-plasma fusion machines such as the International Thermonuclear Experimental Reactor (ITER), at present. Some key diagnostic components, namely magnetic coils, bolometers, and optical fibres, were irradiation-tested in a fission reactor, to evaluate their performances in heavy radiation environments. Results indicate that ITER-relevant radiation-resistant diagnostic components could be developed in time, although there are still some technological problems to be overcome.

The injection performance of the negative-ion based NBI (N-NBI) system for JT-60U has been improved by correcting beamlet deflection and improving spatial uniformity of negative ion production. Beamlet deflection at the peripheral region of the grid segment due to the distorted electric field at the bottom of the extractor has been observed. This was corrected by modifying the surface geometry at the extractor to form a flat electric field. Moreover, beamlet deflection due to beamlet–beamlet repulsion caused by space charge was also compensated for by extruding the edge of the bottom extractor. This resulted in a reduction of the heat loading on the NBI port limiter. As a result of the improvement above, continuous injection of a 2.6 MW H⁰ beam at 355 keV has been achieved for 10 s. Thus, long pulse injection up to the nominal pulse duration of JT-60U was demonstrated. This has opened up the prospect of long pulse operation of the negative-ion based NBI system for a steady-state tokamak reactor. So far, a maximum injection power of 5.8 MW at 400 keV, with a deuterium beam, and 6.2 MW at 381 keV, with a hydrogen beam, have been achieved in the JT-60U N-NBI. Uniformity of negative ion production was improved by tuning the filament emission current so as to direct more arc power into the region where less negative ion current was extracted.

In order to verify the accuracy of the tritium production rate (TPR), neutron irradiation experiments have been performed with a mockup relevant to the fusion DEMO blanket consisting of F82H blocks, Li₂TiO₃ blocks with a ⁶Li enrichment of 40% and 95%, and beryllium blocks. Sample pellets of Li₂TiO₃ were irradiated and the TPR was measured by a liquid scintillation counter. The TPR was also calculated using the Monte Carlo code MCNP-4B with the nuclear data library JENDL-3.2 and ENDF-B/VI. The results agreed with experimental values within the statistical error (10%) of the experiment. Accordingly, it was clarified that the TPR could be evaluated within 10% uncertainty by the calculation code and the nuclear data. In order to estimate the induced activity caused by sequential reactions in cooling water pipes in the DEMO blanket, neutron irradiation experiments have been performed using test specimens simulating the pipes. Sample metals of Fe, W, Ti, Pb, Cu, V and reduced activation ferritic steel F82H were irradiated as typical fusion materials. The effective cross-sections needed to calculate the formation of the radioactive nuclei (⁵⁶Co, ¹⁸⁴Re, ⁴⁸V, ²⁰⁶Bi, ⁶⁵Zn and ⁵¹Cr) due to sequential reactions were measured. From the experimental results, it was found that the effective cross-sections increased remarkably while coming closer to polyethylene board, which was a substitute for water. As a result of this present study, it has become clear that the sequential reaction rates are important factors in the accurate evaluation of induced activity in fusion reactor design.

Externally launched ion Bernstein wave (IBW) experiments have demonstrated localized electron heating, sheared flows and transport barriers in several tokamaks. Experiments in the tokamak fusion test reactor (TFTR) showed that IBW waves launched from low-field side IBW antennas could drive a velocity shear layer in the central plasma, but the power coupled to the IBW was not sufficient to achieve a transport barrier. This experiment raised important questions concerning where the radio-frequency (rf) power went and whether the anomalous loss channels are more important in larger machines. Recently, it was proposed that the power loss was due to a coaxial electron plasma wave (EPW) mode excited in the low density plasma halo near the vessel wall (Myra et al 2000 Phys. Plasmas7 283). This mode could dissipate a significant power fraction by sheath and collisional mechanisms, fits more easily in larger machines like TFTR and has the phasing dependence observed in the experiments. Here we extend that work by demonstrating the existence and phasing dependence of the coaxial mode (CM) in a realistic rf coupling calculation. A three-dimensional finite-element electromagnetic code couples a detailed model of the antenna geometry with a plasma dielectric model that retains CM physics. Quantitative results show the dependence of the CM rf fields and power dissipation on the phasing of the multiple-strap array. Unlike conventional rf coupling codes, this paper enables the antenna limiters to be immersed in tenuous plasma, an important feature for correctly modelling parasitic coupling to the CM.

A new nonlinear destabilization process is found in the nonlinear phase of the double tearing mode (DTM) by using reduced MHD equations with helical symmetry. The nonlinear destabilization causes the abrupt growth of the DTM and subsequent collapse after long timescale evolution in the Rutherford-type regime. The nonlinear growth of the DTM is suddenly triggered, when the triangular deformation of magnetic islands with a sharp current point at the X-point around the outer rational surface exceeds a certain value. Decreasing the resistivity increases the sharpness of the triangularity and the spontaneous growth rate in the abrupt growth phase is almost independent of the resistivity. Current point formation is also confirmed in multi-helicity simulations, where the magnetic fields become stochastic between two rational surfaces.

Various types of MHD events including internal reconnection events are studied on the TST-2 spherical tokamak. In weak MHD events no positive current spike was observed, but in strong MHD events with positive current spikes, a rapid and significant impurity ion temperature increase was observed. The decrease in the poloidal magnetic energy is the most probable energy source for ion heating. The plasma current shows a stepwise change. The magnitude of this step correlates with the temperature increase and is found to be a good indicator of the strength of each event.

The resistive pressure gradient driven instability model is used to study the effect of plasma on a vacuum magnetic field island. Conditions under which the magnetic island is amplified or reduced have been identified. Starting with a set of reference parameters, we have varied these parameters by increasing either the density or the electron temperature. These scans lead to very different results. When β increases because of the increase in density, so does the island width. However, in cases in which β is increased by increasing the electron temperature, we observe a decrease in the island width. The main mechanism for island reduction seems to be the generation of strong sheared flow associated with the magnetic island. These results seem to reflect some observations in the Large Helical Device.

A high confinement plasma (including core plasma and edge plasma) produced by using lower hybrid current drive (LHCD) has been obtained on the HT-7 superconducting tokamak. An internal transport barrier in the core plasma was formed. The energy confinement time increases from 14.6 ms (Ohmic (OH) phase) to 24.5 ms (LHCD phase), which is close to the value calculated using the ITER93ELM free scaling law. The confinement factor H₈₉ increases from 0.78 (OH phase) to 1.42 (LHCD phase). The experimental results were in good agreement with the simulations calculated with a ray tracing code and a two-dimensional Fokker–Planck equation. The edge plasma characteristics around the last closed flux surface were investigated using Langmuir probes. Turbulence and transport of the edge plasma were suppressed greatly by the lower hybrid wave. Studies show that the enhanced confinement plasma may be ascribed to a shear flow resulting from the shear of the radial electric field.

In JET, advanced tokamak research mainly focuses on plasmas with internal transport barriers (ITBs) that are strongly influenced by the current density profile. A previously developed optimized shear regime with low magnetic shear in the plasma centre has been extended to deeply negative magnetic shear configurations. High fusion performance with wide ITBs has been obtained transiently with negative central magnetic shear configuration: H_IPB98(y,2) ∼ 1.9, β_N = 2.4 at I_p = 2.5 MA. At somewhat reduced performance, electron and ion ITBs have been sustained in full current drive operation with 1 MA of bootstrap current: H_IPB98(y,2) ∼ 1, β_N = 1.7 at I_p = 2.0 MA. The ITBs were maintained for up to 11 s for the latter case. This duration, much larger than the energy confinement time (37 times larger), is already approaching a current resistive time. New real-time measurements and feedback control algorithms have been developed and implemented in JET for successfully controlling the ITB dynamics and the current density profile in the highly non-inductive current regime.

Edge profiles of electron temperature and density are measured in ASDEX Upgrade with a high spatial resolution of 2–3 mm with Thomson scattering. In the region of the edge transport barrier in ELMy H-mode, the gradient lengths of T_e and n_e are found closely coupled, with the temperature decay length two times shorter than the density decay length corresponding to η_e ≈ 2. The η_e constraint allows us to calculate the electron temperature and density profiles from the pressure profile if the density and temperature values are known at one spatial position. The edge density in the region of the barrier foot is closely coupled to the main chamber recycling, with no strong dependence on other parameters. In contrast, the density rise from the outer barrier foot to the pedestal exhibits pronounced dependence on plasma current and shaping, indicating quite different mechanisms determining the absolute density and its gradient.

A model for the ion-temperature-gradient-driven (ITG) instability is derived from the Braginskii magnetohydrodynamic equations of ions. The heat flow of electrons carried by the current density J_|| in collision-dominant plasmas is introduced into the model. It is found that, in contrast to the finite Larmor radius effect, the current density suppresses the ITG instabilities with small wavenumber k_⊥ as well as large wavenumber k_||. In addition, the effects of J_|| on the critical stability thresholds for the ITG mode are studied, and two critical values, η_ic1 ⩽ η_ic2, for the temperature gradient parameter η_i = d ln T_i/d ln n are identified. For typical tokamak parameters the stability condition η_i ⩽ η_ic1 may be realized when the parameter ε_n = −(d ln n/dr)⁻¹/R is large. However, the other stable regime η_i ⩾ η_ic2 can only be reached when the safety factor is large enough. Similarly, there are two critical values for the safety factor, q_c1 and q_c2. It is shown that, under specific conditions, q_c1 = q_c2 so that the ITG modes are stable for any q value.

In JFT-2M, ferritic steel plates (FPs) were installed on almost the whole inner surface of the vacuum vessel. This arrangement is called the ferritic inside wall (FIW), and is the third step of the advanced material tokamak experiment programme. The toroidal field (TF) ripple was reduced by optimizing the thickness of FPs but the total ripple structure has become more complex, with a non-periodic feature in the toroidal direction, because of the existence of ports and other components that limit the periodic installation of FPs. We investigated the effect of this complex ripple on the heat flux onto the first wall due to fast ion loss. The ripple trapped loss was reduced as a result of the reduced magnetic ripple of the FIW. Additional FPs were also installed outside the vacuum vessel to produce a localized larger ripple. A small ripple trapped loss was observed when the shallow ripple well exists in the poloidal cross section, and a large ripple trapped loss was observed when the ripple well extends deep into the plasma region. Experimental results were almost consistent with computation with a newly developed fully three-dimensional magnetic field orbit-following Monte-Carlo code including the three-dimensional complex structure of the TF ripple and the non-axisymmetric first wall geometry.

The stability of α-particle driven shear Alfvén eigenmodes (AE) for nominal burning plasma (BP) parameters in the proposed international tokamak experimental reactor (ITER), fusion ignition research experiment (FIRE) and IGNITOR tokamaks is studied. JET plasma, where fusion αs were generated in tritium experiments, is also studied to compare the numerical predictions with the existing experiments. An analytic assessment of toroidal AE (TAE) stability is first presented, where the α-particle β due to the fusion reaction rate and electron drag is simply and accurately estimated in plasmas with central temperature in the range of 7–20 keV. In this assessment the hot particle drive is balanced against ion-Landau damping of the background deuterons, and electron collision effects and stability boundaries are determined. Then two numerical studies of AE instability are presented. In one, the HIgh-n STability (HINST) code is used to predict the instabilities of low and moderately high frequency Alfvén modes. HINST computes the non-perturbative solutions of the AE including effects of ion finite Larmor radius, orbit width, trapped electrons etc. The stability calculations are repeated using the global code NOVAK. We show that for these plasmas the spectrum of the least stable AE modes is at medium-/high-n numbers. In HINST, TAEs are locally unstable due to the α pressure gradient in all the devices under consideration except IGNITOR. However, NOVAK calculations show that the global mode structure enhances the damping mechanisms and produces stability for the nominal FIRE proposal and near-marginal stability for the nominal ITER proposal. NBI ions produce a strong stabilizing effect for JET. However, in ITER, the beam energies needed to penetrate to the core must be high (∼1 MeV) so that a diamagnetic drift frequency comparable to that of α-particles is produced by the beam ions which induces a destabilizing effect. A serious question remains whether the perturbation theory used in NOVAK overestimates the stability predictions, so that it is premature to conclude that the nominal operation of all three BP proposals without neutral beam injection are stable (or marginally stable) to AEs.

A fully superconducting tokamak named JT-60SC is designed for the modification programme of JT-60 to enhance economical and environmental attractiveness in tokamak fusion reactors. JT-60SC aims at realizing high-β steady-state operation in the use of low radio-activation ferritic steel in a low ν* and ρ* regime relevant to the reactor plasmas. Objectives, research issues, plasma control schemes and a conceptual design for JT-60SC are presented.

A method of reduction of the two-dimensional equation for the potential to a one-dimensional ordinary equation is suggested and implemented in the B2SOLPS5.0 transport code to simulate a divertor tokamak. The one-dimensional version, which gives similar results as the full two-dimensional version, provides better understanding of the role of various mechanisms of perpendicular conductivity and reduces computational time. The scheme of poloidal (parallel) current calculation is presented and compared to the two-dimensional code results both for the core region and for the scrape-off layer.

Density, temperature and potential fluctuations have been measured in the edge region and the scrape off layer (SOL) of the SINP tokamak. The triple probe technique has been used to obtain the instantaneous value of these quantities simultaneously and virtually at the same spatial position. The fluctuation levels thus found are significant (_e/T_e≈20–35% and ñ/n≈35–55%) and the spectra are broadband. The effect of temperature fluctuation on the turbulent transport has also been investigated and found to be considerable. The measured diffusion coefficient comes out to be about twice the Bohm diffusion coefficient. Radial profiles show sharp density and temperature gradients and there exists a self-generated E_r×B velocity shear layer near the limiter. The results have been compared with the existing theories of edge turbulence in tokamaks and the rippling mode driven by the resistivity gradient has been identified as the most probable mechanism.

The interaction of short and intense laser pulses with plasmas is a very efficient source of relativistic electrons with tunable properties. In low-density plasmas, we observed bunches of electrons up to 200 MeV, accelerated in the wakefield of the laser pulse. Less energetic electrons (tens of megaelectronvolt) have been obtained, albeit with a higher efficiency, during the interaction with a pre-exploded foil or a solid target. When these relativistic electrons slow down in a thick tungsten target, they emit very energetic Bremsstrahlung photons which have been diagnosed directly with photoconductors, and indirectly through photonuclear activation measurements. Dose, photoactivation, and photofission measurements are reported. These results are in reasonable agreement, over three orders of magnitude, with a model built on laser–plasma interaction and electron transport numerical simulations.

Recent experiments on DIII-D have demonstrated the ability to sustain plasma conditions that integrate and sustain the key ingredients of advanced tokamak (AT) operation: high β with 1.5<q _min <2.5, good energy confinement, and high current drive efficiency. Utilizing off-axis (ρ = 0.4) electron cyclotron current drive (ECCD) to modify the current density profile in a plasma operating near the no-wall ideal stability limit with q _min >2.0, plasmas with β≈2.9% and 90% of the plasma current driven non-inductively have been sustained for nearly 2 s (limited only by the duration of the ECCD pulse). Negative central magnetic shear is produced by the ECCD, leading to the formation of a weak internal transport barrier even in the presence of Type I ELMs. Separate experiments have demonstrated the ability to sustain a steady current density profile using ECCD for periods as long as 1 s with β = 3.3% and >90% of the current driven non-inductively. In addition, stable operation well above the ideal no-wall β limit has been sustained for several energy confinement times with the duration only limited by resistive relaxation of the current profile to an unstable state. Stability analysis indicates that the experimental β limit depends on the degree to which the no-wall limit can be exceeded and weakly on the actual no-wall limit. Achieving the necessary density levels required for adequate ECCD efficiency requires active divertor exhaust and reducing the wall inventory buildup prior to the high performance phase. Simulation studies indicate that the successful integration of high β operation with current profile control consistent with these experimental results should result in high β, fully non-inductive plasma operation.