Table of contents

A remote steering antenna was recently developed for electron cyclotron heating and current drive (ECH/ECCD) experiments on the TRIAM-1M tokamak. This is the first application of the remote steering antenna concept for ECH/ECCD experiments, which have conditions relevant to the International Thermonuclear Experimental Reactor (ITER). Fundamental ECH and ECCD experiments were conducted in the ITER frequency from the low field using this antenna system. In addition to the angles near 0°, the launcher was a symmetric direction antenna with an extended steering-angle capability of ±(8°–19°). The output beam from the antenna was a well-defined Gaussian with a proper steering angle. The Gaussian content and the steering-angle accuracy were 0.85 and −0.5°, respectively. The high power tests measured the antenna transmission efficiency at 0.90–0.94. The efficiencies obtained in the low and high power tests were consistent with the calculations using higher-order modes. In order to excite the pure O/X-modes in the oblique injection, two polarizers were used to control the elliptical polarization of the incident beam for the ECCD experiments. The fundamental O/X-mode ECH/ECCD was applied to lower hyrid current drive plasmas at the optimized incident polarization. In the X-mode experiment, at medium density (∼1 × 10¹⁹ m⁻³), clear differences in the plasma current and the hard x-ray intensity were observed between the co- and counter-steering injections due to the ECCD effect on the coupling of forward fast electrons.

Anomalous electron transport is determined by turbulence, which in turn is affected by magnetic shear. A novel application of electron cyclotron current drive (ECCD), aiming at localized shear modulation, has been applied on the TCV tokamak for experiments on shear-dependent electron transport. Pairs of EC beams, absorbed at the same radius, with one oriented for co- and the other for counter-injection, are modulated out of phase in order to force a local modulation of current–density at constant input power. Off-axis deposition (ρ_dep = 0.24) is performed to avoid the central region, where the low heat flux would make transport analysis difficult. In addition some sawteeth control is achieved in this way. A significant impact on local shear is achieved with I_ECCD ≈ 0.1I_OH, even when the modulation period is much shorter than the current diffusion time across the whole plasma radius. The main result is that although source (heat and particle) terms are constant, both electron density and temperature are modulated during alternated ECCD. Once equilibrium effects are taken into account for appropriate mapping of Thomson scattering measurements onto flux coordinates, modulation of T_e and electron pressure, peaked on-axis, is confirmed at all radii internal to EC deposition. The best confinement occurs for co-injection, in which case a local decrease (≈55%) in the magnetic shear causes a decrease in the electron thermal diffusivity of a similar amount (≈65%).

The electrostatic and density fluctuation are measured simultaneously with a heavy ion beam probe. The electrostatic fluctuation with the geodesic acoustic mode (GAM) frequency is observed in L-mode plasmas and not in H-mode plasmas. The poloidal and radial structure is consistent with the GAM. So the fluctuation is concluded to be the GAM.

The amplitude of the GAM changes in the radial direction; it is small near the separatrix, has a maximum at 3 cm inside the separatrix and decreases again to 5 cm inside the separatrix.

The GAM and the temporal behaviour of the ambient density fluctuation show a significant coherence, and the phase of modulation of the ambient density fluctuation tends to delay the potential oscillation of the GAM. It is clearly verified that the GAM affects ambient fluctuation and also the local particle transport through modulation of the amplitude of the ambient fluctuation.

Coupling to the electron Bernstein wave (EBW) via a phased array of waveguides is experimentally investigated in the Madison symmetric torus reversed field pinch (RFP). EBWs promise to provide localized heating and current drive in overdense plasmas such as those in the RFP, provided a technique can be developed to efficiently couple power from an externally launched electromagnetic wave to the electrostatic EBW. The choice of antenna structure and launched wave polarization are important factors in coupling waves with frequencies in the electron cyclotron range to EBW, especially on the RFP where the mode conversion to the EBW can take place in the near-field of the antenna. In this paper, a recently developed theory of coupling to EBW from a phased array of waveguides is tested against experiment. The theory predicts that coupling efficiency will vary with launch angle and that coupling depends sensitively upon the edge density profile. Amplitudes and phases of reflected power in each of the waveguides are measured experimentally and compared with predictions based upon density profiles measured by a Langmuir probe in the edge of the plasma. The parametric dependence of reflection has been studied for different polarizations, different launch angles, time varying density profiles and several frequencies of the launched wave. An asymmetry in reflection versus the launch angle predicted by theory was found experimentally for the X-mode, while the O-mode was symmetric. The dependence on density scale length predicted by theory was confirmed in the experiment. The phase of the reflected signal is shown to contain reflectometry-based information about the edge density profile. For appropriate phasing, measurements show that the total power reflection coefficient can be lower than 15%.

Self-sustained detachment has been obtained in the Large Helical Device (LHD). Strong hydrogen gas puffing of ∼200 Pa m³ s⁻¹ after a density feedback phase detaches the plasma from the divertor plates with high reproducibility. High electron density of over 1 × 10²⁰ m⁻³ is sustained without gas puffing until the heating beam stops and a high-density flat top for 2 s has been demonstrated. Throughout the self-sustained detachment phase, the minor radius of the hot plasma column shrinks to ∼90% of the last-closed-flux-surface, which corresponds to the rational surface. This new state has been named the 'Serpens mode', for self-regulated plasma edge 'neath the last-closed-flux-surface. Global energy confinement of the Serpens mode is compared with the international stellarator scaling 1995 (ISS95) and the recently established scaling for high-density LHD plasmas (HD scaling), where shrinking confinement volume and shallow penetration of the heating beams are taken into account. Although the energy confinement of the Serpens mode seems deteriorated compared with ISS95, as in the case of high-density attached plasmas, it is consistent with the HD scaling. This suggests that the energy confinement properties of detached plasmas in LHD are similar to those in high-density attached plasmas.

The injection of a high pressure jet of noble gas is one possible approach to mitigating potential disruption damage to tokamaks. The efficacy of this approach given only modest penetration of the gas-jet into the plasma is of particular concern in a high absolute pressure device, like Alcator C-Mod (Hutchinson I.H. et al 1994 Phys. Plasmas1 1511) or ITER. To investigate the effectiveness of gas-jet penetration to various depths, a simple radiation model based on a corona equilibrium assumption is employed in resistive magnetohydrodynamic (MHD) simulations of an Alcator C-Mod plasma. Destabilization of an m = 2/n = 1 mode is key in each case to the onset of stochasticity producing rapid parallel thermal transport to the impurity cooled region just inside the equilibrium boundary (separatrix). Results indicate that a thermal quench is triggered even at very shallow penetration, but penetration of the impurities to the q = 2 surface affords a significant advantage for hastening the quench onset.

For the first time, the predictive capabilities of the mixed Bohm/GyroBohm, Weiland and 'retuned' GLF23 transport models are investigated with ITB discharges from the ITPA ITB database with fully predictive, time-dependent transport simulations. A range of plasma conditions is examined for JET, JT-60U and DIII-D discharges with internal transport barriers (ITBs). The simulations show that the Bohm/GyroBohm model is able to follow the time evolution of the discharge from the preheating phase without an ITB through the ITB onset phase until the high performance phase with fair accuracy in most cases in JET and JT-60U. This indicates the importance of the interplay between the magnetic shear and ω_E×B flow shear in ITB formation since these are the mechanisms that govern the ITB physics in the model. In order to achieve good agreement in DIII-D, the α-stabilization had to be included in the model, emphasizing the role played by the α-stabilization in the physics of the ITBs. The Weiland and GLF23 transport models show limited agreement between the model predictions and experimental time evolution of the ITB and kinetic plasma profiles. The Weiland model does not form a clear ITB in any of the three tokamaks despite varying plasma profiles, such as the q-profile. On the other hand, the average temperatures and density are often in fair agreement with experimental values. The GLF23 model often predicts an ITB, but its radial location is often too far inside the plasma, or shrinks as the simulations proceed in time. Consequently, the central temperatures at the end of the simulations during the high performance phase are usually underestimated. It is worth noting that GLF23 features in general better predictions of the T_e and T_i profiles outside the ITB than the other models studied. Achieving the quantitative capability to predict the multi-channel ITB dynamics with fully predictive, time-dependent transport simulations has turned out to be extremely challenging.

The radial electric field E_r(x, t), and particularly its gradient, has been invoked by various theories and empirical models as a crucial parameter 'per se' for determining the transition to high confinement regimes, such as the onset of an internal transport barrier (ITB) in the plasma core and of the H-mode pedestal at the plasma edge. This idea, however, does not consider the basic fact that in most experiments the transition to a steady-state higher confinement regimes is produced by applying sufficient additional heating onto a given target density and current profile. In order to test this ansatz on a more routine basis, we have developed here an analytical approximation to the neoclassical calculation of the radial electric field, adapted for the 2D toroidal geometry of JET to describe all collisionality regimes (banana, banana-plateau, Pfirsch–Schlüter) and to include averaging over the potato orbits. An analytic calculation of the error bars on E_r(x, t) has also been developed, which has allowed us to compare and successfully benchmark our calculations with the results of neoclassical codes such as JETTO and NCLASS. We are then able to demonstrate a striking similarity in the shape of E_r(x, t) in steady-state L-mode, H-mode and ITB plasmas when normalizing E_r(x, t) with respect to the total heating power flux. This clearly indicates that, experimentally, there is no direct causality relation between changes in E_r(x, t) and steady-state improved confinement, as these are brought about together by changes in the power deposition profile. Only two cases do not satisfy this general rule. First, localized and rapid transients (i.e. occurring on time scales much shorter than the momentum and energy confinement time) could be linked to non-neoclassical changes in E_r(x, t), possibly due to turbulence suppression mechanisms. Second, when comparing H-mode plasmas with forward and reversed ion ∇B-drift direction, we demonstrate the role of prompt fast ion losses in affecting E_r(x, t), most likely due to the different edge flows and their cascading effect onto the core plasma.

Theory-based turbulent transport models have been proposed and used in transport simulations to predict the performance of the International Thermonuclear Experimental Reactor (ITER). We have selected the current diffusive ballooning mode (CDBM), GLF23 and Weiland models as typical transport models. Taking experimental data from the ITPA profile database, simulations have been carried out to compare the results with each other using our one-dimensional diffusive transport code. It is found that each model has unique dependence on devices and operation modes, and most reasonable results among the models can be obtained by the CDBM model for L-mode discharges and by the GLF23 model for H-modes. From an analysis of T_e/T_i dependence, predictions of the CDBM and Weiland models could be better in the hot electron regime. The effect of the elliptic cross section of the plasma has been included in the CDBM model, and it is confirmed that an accuracy of the prediction is improved for H-mode discharges. Finally, heat transport simulations for a single particle species have indicated that the CDBM model reproduces T_i profiles more accurately than T_e profiles.

Calculations of the particle flux caused by neutrals originating from the edge are performed for TCV and JET using the one-dimensional kinetic transport code Kn1D. The analysis, confirmed by experimental evidence, shows that for TCV and JET edge fuelling as well as neutral beam fuelling cannot be responsible for the density gradient in the plasma bulk, confirming the presence of inward particle convection. The results corroborate the expectation of a peaked density profile in ITER, despite the lack of bulk particle fuelling.

A divertor magnetic configuration is an alternative method of obtaining interchange mode stability in the central cell of the GAMMA10 tandem mirror (Cho et al 2005 Phys. Rev. Lett.94 085002). Various divertor configurations are investigated in the central cell of GAMMA10 to determine the best configuration. The equilibrium in the divertor configuration is calculated using the generalized Grad–Shafranov equation with anisotropic plasma pressure to estimate how high a plasma pressure is sustained by the magnetic divertor.

Vanadium alloys, candidate fusion reactor blanket materials, consist of intrinsic low activation elements for 14 MeV fusion neutrons (e.g. vanadium, chromium and titanium). From 500 to 1000 K, the tensile strength of vanadium alloys is independent of temperature, enabling the fusion blanket to be operated at least 100 K higher than a blanket made from ferritic steel. However, most of the data have come from laboratory-scale fabricated vanadium alloy, so a feasibility study on large scale production of vanadium alloy products is required. In the present study, high-purity vanadium alloy products (e.g. plates, wires and tubes) were fabricated from reference high-purity V-4Cr-4Ti ingots designated as NIFS–HEAT, with the use of technologies applicable to industrial scale fabrication. A critical issue for NIFS–HEAT large-scale melting was to reduce the levels of interstitial impurities (e.g. C, N and O) that are known to deteriorate mechanical properties before and after neutron irradiation. Impurity behaviour during working and annealing processes and its effect on mechanical properties were investigated. Mechanical properties of the products were significantly improved by the control of Ti-C, N and O precipitation induced during the processes. Good mechanical properties comparable to laboratory-scale alloys were obtained at above 96% in cold working degree.

Modification for long pulse operation up to 65 s with a neutral beam heating power of 12 MW (30 s) has provided an opportunity for studies on plasma-wall interaction on a time scale of tens of seconds. During a long pulse, ELMy H-mode discharge, the net wall-pumping rate, evaluated by a particle balance analysis, gradually decreases, and subsequently, becomes zero in ∼20 s after the neutral beam injection starts. This situation is interpreted as 'wall saturation'. Wall saturation is observed in repetitive long pulse discharges. Although the outgas due to the increase in the divertor plate temperature and increase in the static retention with pulses also contribute to the particle balance, the dynamic retention process dominates the particle balance and determines the wall saturation. Under the condition of wall saturation, the plasma density continuously increases due to insufficient divertor-pumping rate. This density rise results in detachment of the outer divertor plasma, followed by an X-point MARFE. Results from the calculation of a 2-dimensional plasma transport code indicate that higher divertor-pumping efficiency by a factor of 1.5 is required to avoid the detachment. Note that no negative effects on impurity behaviour such as carbon bloom, or significant dilution of the core plasma has been observed even in the case of the total input energy reaching 350 MJ.

The National Spherical Torus Experiment (NSTX) has demonstrated the advantages of low aspect ratio geometry in accessing high toroidal and normalized plasma beta, and β_N ≡ 10⁸⟨β_t⟩ aB₀/I_p. Experiments have reached β_t = 39% and β_N = 7.2 through boundary and profile optimization. High β_N plasmas can exceed the ideal no-wall stability limit, β_Nno−wall, for periods much greater than the wall eddy current decay time. Resistive wall mode (RWM) physics is studied to understand mode stabilization in these plasmas. The toroidal mode spectrum of unstable RWMs has been measured with mode number n up to 3. The critical rotation frequency of Bondeson–Chu, Ω_crit = ω_A/(4q²), describes well the RWM stability of NSTX plasmas when applied over the entire rotation profile and in conjunction with the ideal stability criterion. Rotation damping and global rotation collapse observed in plasmas exceeding β_Nno−wall differs from the damping observed during tearing mode activity and can be described qualitatively by drag due to neoclassical toroidal viscosity in the helically perturbed field of an ideal displacement. Resonant field amplification of an applied n = 1 field perturbation has been measured and increases with increasing β_N. Equilibria are reconstructed including measured ion and electron pressure, toroidal rotation and flux isotherm constraint in plasmas with core rotation ω_ϕ/ω_A up to 0.48. Peak pressure shifts of 18% of the minor radius from the magnetic axis have been reconstructed.

The problem of aliasing in relation to the use of Wigner distributions to process discrete time (DT) non-stationary signals from fusion plasma diagnostics is comprehensively addressed. Three DT Wigner distributions, two of which are alias free (AF), are thoroughly studied and compared, their properties being provided as well as, making use of so-called pseudo-Wigner (p-Wigner) distributions, formulae for their efficient computation. Of the three, one is singled out as that AF form which simultaneously obeys an arguably more complete set of desirable properties for time–frequency distributions and which has a lower computational complexity. Such an AF DT Wigner distribution has been known in quantum mechanics as the rotational Wigner function, where it has been introduced to treat rotation angles and their canonically conjugate angular momenta, whose spectra have the same discrete nature as DT signals. As an illustration of the advantages and shortcomings of each of these three distributions, they are applied to the time–frequency analysis of MHD activity data. Thus, the three DT Wigner distributions, more precisely their p-Wigner counterparts, together with the spectrogram for the sake of comparison, are used to obtain representations in the time–frequency plane of non-stationary events from neoclassical tearing mode (NTM) and toroidal Alfvén eigenmode (TAE) experiments in the JET tokamak. Besides aliasing, the appearance in the time–frequency plane of spurious images due to cross-term interference between different signal components is also discussed. If such undesirable artefacts are generated by negative frequencies interfering with positive frequencies, it is shown that the analytic signal can be of great help in reducing cross-term effects.