Search

In JT-60U tokamak, low frequency modes (LFMs) have been reported recently when the m/n = 2/1 neoclassical tearing mode grows sufficiently. LFMs show the low mode frequencies (< 20 Hz) with the rotation direction to the counter-current direction (ctr-direction) toroidally even if the strong injection power to the co-current direction (co-direction) is applied by tangential neutral beam injections. As a candidate to drive the rotation to the ctr-direction, the neoclassical toroidal viscosity (NTV) torque is investigated in this study. Indeed, when a LFM is observed, the estimated "offset velocity" of the NTV torque is larger than the observed toroidal velocity around the q = 2 surface and results in the torque to the ctr-direction (ctr-torque). The torque balance between the NTV torque and the torque induced by the eddy current on the resistive wall is investigated. It is found that the rotation to the ctr-direction having the low mode frequency is obtained as the result of the ctr-torque to the offset velocity. Our investigation suggests the requirement to include the NTV torque in the modelling on the mode locking of neoclassical tearing modes in tokamaks.

Active control and stabilization of locked modes (LM) via rotating external magnetic perturbations are numerically investigated under a realistic low resistivity condition. To explore plasma responses to rotating and/or static external magnetic perturbations, we have developed a resistive magnetohydrodynamic code ‘AEOLUS-IT’. By using AEOLUS-IT, dependencies of mode behavior on frequency and polarity of the rotating magnetic perturbation are successfully clarified. Here, the rotational direction of the rotating magnetic perturbation to the equilibrium plasma rotation in the laboratory frame is referred to as ‘polarity’. The rotating magnetic perturbation acts on the background rotating plasma in the presence of a static field. Under such circumstances, there exist bifurcated states of the background rotating plasma, which should be taken into account when studying the dependence of the mode behavior on the rotating magnetic perturbation. It is found that there exist an optimum frequency and polarity of the rotating magnetic perturbation to control the LM, and that the LM is effectively stabilized by a co- polarity magnetic perturbation in comparison with a counter-polarity one.

Overview of plasma facing components (PFCs) in JT-60SA is introduced, including upper divertor, lower divertor, inboard first wall and outboard first wall. These PFCs are upgraded in 4 stages along plasma experiment of JT-60SA; In the Stage1, only partial inboard first wall and upper divertor are installed with graphite tiles and inertial cooling. In the Stage 2, remaining PFCs including outboard first wall and lower divertor are installed, and the inboard surface is fully covered. In the Stage 3, vertical targets of lower divertor are replaced by CFC monoblock targets with water cooling. In the Stage 4, whole carbon wall is replaced by tungsten wall, and remote handling for lower divertor is introduced. Three R&D on major issues concerning PFCs are presented. Mock-ups of CFC monoblock target were fabricated and it showed good heat removal of the expected maximum heat load of 15 MW/m2 without degradation of performance. Samples tungsten coating on CFC were fabricated and they showed no severe damage by surface temperature up to 2000 C. Laser welded samples showed good appearance and no defect was found by RT.

We report the experimental finding of n = 1 helical cores (HCs) accompanied by saturated m/n = 2/1 tearing modes (TMs) with low mode frequencies in JT-60U. The HCs accompanied by TMs were observed after an increase in the mode amplitude and a decrease in the mode frequency of m/n = 2/1 precursors with tearing parity. The decreased mode frequency is typically lower than 20 Hz. With various diagnostics, the coupling of n = 1 HCs and m/n = 2/1 TMs has been clearly observed. Because the coherent oscillations in the ion temperature are observed in both the core region and the edge region, the flux surfaces including the m/n = 2/1 magnetic island appear to have m = 1 helical deformation. It has also been suggested that the m/n = 2/1 TM and the HC rotate in the electron diamagnetic direction keeping fm/n=1/1(HC) = 2fm/n=2/1(TM) in several plasmas. Here, fm/n=1/1(HC) is the mode frequency of HCs and fm/n=2/1(TM) is the mode frequency of TMs. In addition, the core seems to be shifted to the high-field side when the O-points of the m/n = 2/1 magnetic island line up in the midplane, which is confirmed by reconstructions of magnetohydrodynamic equilibria with motional Stark effect measurement and the MEUDAS code. Our observation of m/n = 2/1 TMs having HCs contributes to the understanding of the excitation mechanism of HCs in tokamak plasmas.

The integrated framework consisting of TOPICS, VMEC and FORTEC-3D, has been applied to provide an estimate of the neoclassical toroidal viscosity (NTV) produced by the applied magnetic perturbations (MPs). Six picture-frame error field correction coils in each row in JT-60SA realize the various patterns of the applied magnetic field spectrum. The NTV is investigated for various toroidal phase differences. As long as the same coil current is applied, the n = 3 MPs generate the same NTV, regardless of even or odd parity, or energizing the middle coils or not. The n = 1 MPs give rise to larger NTV than the n = 3 ones. The NTV invoked by the n = 1 resonant MPs is virtually equivalent to that by the n = 1 nonresonant MPs with five-fold coil current.

In order to account for the effect of field perturbations on the transport of fast ions in integrated codes used for the simulation of operational scenarios, it is crucial to develop computationally efficient reduced transport models. Such modeling efforts may greatly benefit from a simple method that determines the width of the island-like structures, which are produced by resonant perturbations in the phase space of the fast ion guiding centers and are known to play a key role for fast ion transport enhancement. In this paper, we present a method for estimating the widths of such ‘orbit islands’ for passing particles in the presence of static magnetic perturbations. The method consists of mapping the boundaries of magnetic islands from magnetic flux space (ψ) into the canonical angular momentum space (Pφ) of the fast ions. As a working example, we consider co-passing neutral beam (NB) ions subject to a resonant magnetic perturbation (RMP) in a KSTAR tokamak plasma. The estimated orbit island width deviates by less than 25% from the value obtained from Poincaré plots of the actual guiding center trajectories, even when the magnetic drifts are large (here, up to 50% of the minor radius). Our analysis also shows that most of the fast ion transport can be attributed to the effect of isolated islands, which means that stochastization of particle trajectories due to resonance overlaps does not play a major role in the case studied here. The island mapping method proposed here eliminates the need to compute and analyze Poincaré maps of particle trajectories, so that computation times can be reduced tremendously by several orders of magnitude. A further speed-up may be achieved by the development of a method for estimating the width of magnetic islands under realistic conditions.

The integrated simulation framework for toroidal momentum transport is developed, which self-consistently calculates the neoclassical toroidal viscosity (NTV), the radial electric field and the resultant toroidal rotation together with the scrape-off-layer (SOL) physics-based boundary model. The coupling of three codes, the 1.5D transport code TOPICS, the three-dimensional (3D) equilibrium code VMEC and the 3D drift-kinetic equation solver FORTEC-3D, makes it possible to calculate the NTV due to the non-axisymmetric perturbed magnetic field caused by toroidal field coils. Analyses reveal that the NTV significantly influences in JT-60U and holds the key to determine the NTV profile. The sensitivity of the profile to the boundary rotation necessitates a boundary condition modelling for toroidal momentum. Owing to the high-resolution measurement system in JT-60U, the gradient is found to be virtually zero at the separatrix regardless of toroidal rotation velocities. Focusing on , the boundary model of toroidal momentum is developed in conjunction with the SOL/divertor plasma code D5PM. This modelling realizes self-consistent predictive simulations for operation scenario development in ITER.

A cooperation framework for analyses and predictions of the neoclassical toroidal viscosity (NTV) and the resultant toroidal flow is developed among the TOPICS, VMEC and FORTEC-3D codes. With the real geometry in JT-60U taken into account, it is found that the NTV is one of the cardinal torque sources especially in the edge region irrespective of the insertion of the ferritic steel tiles (FSTs) that reduce the toroidal field ripple amplitude and is essential to numerically reproduce the measured toroidal rotation profile in the edge. The up–down asymmetric component of the NTV is damped due to the FSTs and the NTV profile correlates with the profile of the radial electric field Er. Predictive simulations for JT-60SA H-mode scenarios are also performed to investigate the effects of the NTV on toroidal rotation. The NTV reversal is observed in the pedestal region where the steep pressure gradient is formed, due to the dependence of the NTV on Er.

The plasma current decay during the initial phase of the current quench in JT-60U disruption has been calculated by a disruption simulation code (DINA) using the measured electron temperature Te profile. In the case of fast plasma current decay, Te has a peaked profile immediately after a thermal quench and the Te profile does not change significantly during the initial phase of the current quench. On the other hand, in the case of a slow plasma current decay, the Te profile is broader immediately after the thermal quench, and the Te profile shrinks. The results of the DINA simulation show that the plasma internal inductance Li increases during the initial phase of the current quench, whereas the plasma external inductance Le does not change over time. If the plasma is represented by a simple electrical circuit, the time derivative of Li functions as a resistance on the plasma current decay. It was confirmed that the increase in Li is caused by current diffusion toward the core plasma due to the decrease in Te in the intermediate and edge regions.

This paper reports the highlights of the RFX-mod fusion science programme since the last 2010 IAEA Fusion Energy Conference. The RFX-mod fusion science programme focused on two main goals: exploring the fusion potential of the reversed field pinch (RFP) magnetic configuration and contributing to the solution of key science and technology problems in the roadmap to ITER. Active control of several plasma parameters has been a key tool in this endeavour. New upgrades on the system for active control of magnetohydrodynamic (MHD) stability are underway and will be presented in this paper. Unique among the existing fusion devices, RFX-mod has been operated both as an RFP and as a tokamak. The latter operation has allowed the exploration of edge safety factor q edge < 2 with active control of MHD stability and studies concerning basic energy and flow transport mechanisms. Strong interaction has continued with the stellarator community in particular on the physics of helical states and on three-dimensional codes. © 2013 IAEA, Vienna.

The first detailed measurements of ion-impurity dynamics for NBI-heated ELMy H-modes at the edge of the JT-60U tokamak are reported. We investigated the ability of external momentum/power input to modify and control the radial electric field, E r , and pedestal structures. The relationship between E r and pedestal structures of ion-impurity density, n i, and temperature, T i, during the ELMing H-mode phase for various momentum input directions (i.e. co-, balanced- and counter-NBI) and input powers from perpendicular NBI are compared with the ELM-free phase. The observed trend is that the edge E r -well width increases in the co-NBI discharge, while the E r value at the base of the E r -well becomes more negative in the counter-NBI discharge. The scale length for both n i and T i in the pedestal is ∼2 cm and values are ∼1 for both ELM-free and ELMing phases with different magnitudes of E r (and/or E r shear). Characteristics of the turbulent density fluctuation, in addition to a uniform toroidal MHD oscillation (i.e. n = 0), during both ELM-free and ELMing phases are also reported.

The JT-60SA device has been designed as a highly shaped large superconducting tokamak with a variety of plasma actuators (heating, current drive, momentum input, stability control coils, resonant magnetic perturbation coils, W-shaped divertor, fuelling, pumping, etc) in order to satisfy the central research needs for ITER and DEMO. In the ITER- and DEMO-relevant plasma parameter regimes and with DEMO-equivalent plasma shapes, JT-60SA quantifies the operation limits, plasma responses and operational margins in terms of MHD stability, plasma transport and confinement, high-energy particle behaviour, pedestal structures, scrape-off layer and divertor characteristics. By integrating advanced studies in these research fields, the project proceeds ‘simultaneous and steady-state sustainment of the key performances required for DEMO’ with integrated control scenario development applicable to the highly self-regulating burning high-β high bootstrap current fraction plasmas.

Charge exchange recombination spectroscopy measurements with high spatial and temporal resolution have made the evaluation of the radial electric field (E r ) in the JT-60U tokamak peripheral region possible with a better signal-to-noise ratio. In a very low density (19 m−3) L-mode phase, the gradient in E r (or, equivalently, in poloidal rotation, V θ) was found to start increasing prior to an increase in the ion temperature (T i). As the E r values in the region r/a > 0.94 started to become negative, down to −40 kV m−1 (similar to V ϕ), the steepening of the T i profile in the same region accelerated. Even though the numerical value of V ϕ was considerably smaller than that of V θ, its contribution to shear could be significant: while for r/a ⩾ 0.90 V ϕ became increasingly negative; it simultaneously became increasingly positive for r/a < 0.90. Clean bifurcations in the edge E r values between weak negative and strong negative ones were observed in the later ELM-free H-mode phase, with essentially the same confinement properties corresponding to two very different E × B shearing rates. The ‘normal’ ELMy H-mode took place only after the bifurcations. The results indicate that energy confinement improvement may not necessarily be connected to E × B shear suppression of turbulence.

We have observed a fishbone-like mode in high-β plasmas above the ideal β limit without a conducting wall. The mode frequency chirps down, and its initial value is close to the precession frequency of trapped fast ions produced by perpendicularly injected neutral beams. This mode can often induce the resistive wall mode (RWM). The mode is observed mainly in the high-βN plasma where an ideal kink-ballooning mode (IKBM) and the RWM are marginally stable. Since this mode and the RWM were simultaneously observed, the mode is attributed to the interaction between the trapped fast ions and a marginally stable IKBM stabilized by a conducting wall. Actually, the MARG2D analysis shows that the IKBM is wall-stabilized and has a real frequency of the same order as the observed mode frequency. From these results, the observed mode was named ‘energetic particle driven wall mode (EWM)’. Moreover, the EWM can change the edge localized mode (ELM) behaviour. For example, it was observed that the ELM was synchronized with the EWM. An energy loss due to the EWM-triggered ELM is smaller, and its frequency becomes higher compared with the usual ELM.

AnL/Rmodel that predicts the current decay time from the circuit equation is essentially used for the design of the International Thermonuclear Experimental Reactor. In order to verify the validity of theL/Rmodel in the determination of current decay time during disruption, the plasma current decay time in the JT-60U tokamak is studied using experimental plasma resistance and inductance. The plasma resistance during the initial phase of current quench is estimated from the electron temperature profile measured using the electron cyclotron emission diagnostic system and by measuring the He I line emission intensity ratios and plasma inductance is estimated by the Cauchy-Condition surface method using magnetic sensor signals. Further, the radiation-induced disruptive plasma discharges with massive neon gas puffing are also analysed. The observed area-normalized current decay times have a weak dependence on the electron temperature, particularly in a small decay time region (5–10 ms m−2). The observed decay times are lesser by one order of magnitude than the decay times estimated by theL/Rmodel. However, a novel model for decay time prediction, which takes into account the time derivative of the plasma inductance, wields results that are extremely consistent with the experimental decay time.

Rseduction of heat loading appropriate for the plasma facing components such as the divertor is crucial for a fusion reactor. Power handling by large radiative power loss has been studied in long pulse ELMy H-mode discharges on JT-60U (τd = 30–35 s). Case 1 is argon (Ar) seeding into standard ELMy H-mode plasmas, where large radiation loss in the confined region of the main plasma caused a change in ELM characteristics from Type-I to Type-III. Case 2 is a combination of Ar and nitrogen (Ne) gas seeding into Type-I ELMy H-mode plasmas with an internal transport barrier (ITB). For case 1, large radiation loss both from the main plasma and from the divertor was produced, and operation of Type-III ELMs was preferable to a reduction in ELM energy loss fraction (W ELM/W dia) to 0.15%. Both transient and steady-state heat loadings were reduced. Relatively good energy confinement (H H98y2 = 0.87 − 0.75) with large f rad (P rad/P abs > 0.8) and divertor plasma detachment was sustained continuously for 13.5 s. For case 2, with reduced Ar seeding to the main plasma and increased divertor radiation with Ne seeding, the ELMy H-mode plasma with an ITB had better energy confinement (H H98y2 = 0.95 − 0.8), which was sustained continuously for 12 s. The radiated power was increased primarily in the divertor ( ), which was produced both by seeded Ne ions and by carbon influx due to transient (ELM) and steady-state heat loadings in the attached divertor. Reduction in the heat loading was not enough, thus enhancement of the radiated power in the divertor will be necessary for the formation of the divertor detachment.

Dynamics of ion internal transport barrier (ITB) formation and impurity transport both in the Large Helical Device (LHD) heliotron and in the JT-60U tokamak are described. Significant differences between heliotron and tokamak plasmas are observed. The location of the ITB moves outwards during the ITB formation regardless of the sign of magnetic shear in JT-60U, and the ITB becomes more localized in plasmas with negative magnetic shear. In LHD, a low T e/T i ratio (

H-modes operated at higher l i with the current ramp down have shown higher energy confinement with higher density in JT-60U. The H-factor evaluated for the core plasma depends strongly on l i with the relation of for the case without sawtooth activities. Centre peaked profiles of electron density and electron temperature are obtained in high l i H-modes. While the peripheral current density profiles are largely modified by the current ramp, the pedestal pressure is not significantly changed. The enhanced energy confinement in high l i H-modes is attributed to the core improvement with the peaked profiles of electron density and electron temperature while no explicit difference in pedestal profile is observed. The electron heat diffusivity is reduced at the plasma core in the high l i case, resulting in the centre peaked T e profile while the T i profiles are approximately unchanged.

The performance and sustained duration of long-pulse discharges for the ‘ITER hybrid scenario’ have been improved in JT-60U. The modification of power supply systems for three perpendicular neutral beam (NB) injections provides a long period of central NB heating up to 30 s, which is important for keeping the internal transport barrier (ITB). The peaked density profile in the core plasma can be maintained even when the density at the pedestal increased in the latter phase of the discharge due to the increase in the divertor recycling. Then, the peaked pressure profile attributed to the ITB can be kept constant through the discharge with the peaked power deposition profile. In these long-pulse discharges, MHD activity with toroidal mode number n = 1 is observed even when neoclassical tearing modes (NTMs) are suppressed. When the amplitude of the mode in the peripheral region becomes large, the pedestal pressure is degraded. The mode amplitude is sensitive to the toroidal magnetic field (or edge safety factor) and heating power. After the adjustment of the toroidal magnetic field so as to reduce the mode amplitude, a high normalized beta (βN) of 2.6 and a high thermal confinement enhancement factor (H H98(y,2) > 1) are sustained for 25 s (∼14τR, where τR is the current diffusion time) under the ITER relevant small toroidal rotation condition. The peaked pressure profile in low safety factor plasma (safety factor at 95% flux surface q 95 ∼ 3.2) is stable against NTMs up to βN ∼ 3. A high βN H H98(y,2) of 2.6 gives a high G-factor ( ) of 0.25 and a peaked pressure profile gives a large bootstrap current fraction (f BS > 0.43).

The results of stabilizing neoclassical tearing modes (NTMs) with electron cyclotron current drive (ECCD) in JT-60U are described with emphasis on the effectiveness of the stabilization. The range of the minimum EC wave power needed for complete stabilization of an m/n = 2/1 NTM was experimentally identified for two regimes using unmodulated ECCD to clarify the NTM behaviours with different plasma parameters: 0.2 < j EC/j BS < 0.4 for W sat/d EC ∼ 3 and W sat/W marg ∼ 2, and 0.35 < j EC/j BS < 0.46 for W sat/d EC ∼ 1.5 and W sat/W marg ∼ 2. Here, m and n are the poloidal and toroidal mode numbers; j EC and j BS the EC-driven current density and bootstrap current density at the mode rational surface; W sat, W marg and d EC the full island width at saturation, marginal island width and full-width at half maximum of the ECCD deposition profile, respectively. Stabilization of a 2/1 NTM using modulated ECCD synchronized with a mode rotation of about 5 kHz was performed, in which it was found that the stabilization effect degrades when the phase of the modulation deviates from that of the ECCD at the island O-point. The decay time of the magnetic perturbation amplitude due to the ECCD increases by 50% with a phase shift of ±50° from the O-point ECCD, thus revealing the importance of the phasing of modulated ECCD. For near X-point ECCD, the NTM amplitude increases, revealing a destabilization effect. It was also found that modulated ECCD at the island O-point has a stronger stabilization effect than unmodulated ECCD by a factor of more than 2.

Transport analysis during the transient phase of heating (a dynamic transport study) applied to the plasma with internal transport barriers (ITBs) in the Large Helical Device (LHD) heliotron and the JT-60U tokamak is described. In the dynamic transport study the time of transition from the L-mode plasma to the ITB plasma is clearly determined by the onset of flattening of the temperature profile in the core region and a spontaneous phase transition from a zero curvature ITB (hyperbolic tangent shaped ITB) or a positive curvature ITB (concaved shaped ITB) to a negative curvature ITB (convex shaped ITB) and its back-transition are observed. The flattening of the core region of the ITB transition and the back-transition between a zero curvature ITB and a convex ITB suggest the strong interaction of turbulent transport in space.

In JT-60U, the filament structure of Edge Localized Modes (ELMs) is measured at three poloidal locations, and their temporal evolution is investigated at the inner (high magnetic-field-side, HFS) SOL as well as outer (low magnetic-field-side, LFS) SOL. At the inner SOL, filament structure with 7-8 temporal multi-peaks in ion-saturation probe signals and other diagnostics is, for the first time, measured close to the sepatratrix. The delay of the first peak after the start of MHD activity is faster than the characteristic time of the parallel convection from the outer midplane, and the Mach numbers of the plasma parallel flow reach ion sonic levels. These results show that ELM filaments extend from the outer to the inner plasma edge, and that a part of the filaments are ejected into the inner SOL. The toroidal mode numbers (n) and poloidal width of the filaments (δz) are evaluated from the interval of the multi-peaks and duration of each peak, yielding n ~ 18-44 and δz ~ 2-6 cm on the HFS SOL, while n and δz are comparable on the LFS SOL. After the multi-peaks appear, flow reversal of the SOL plasma toward is observed over a wide region of the inner SOL.

Aiming at optimization of current profile in high-β plasmas for higher confinement and stability, a real-time control system of the minimum of the safety factor (q min) using the off-axis current drive has been developed. The off-axis current drive can raise the safety factor in the centre and help to avoid instability that limits the performance of the plasma. The system controls the injection power of lower-hybrid waves, and hence its off-axis driven current in order to control q min. The real-time control of q min is demonstrated in a high-β plasma, where q min follows the temporally changing reference q min,ref from 1.3 to 1.7. Applying the control to another high-β discharge (βN = 1.7, βp = 1.5) with m/n = 2/1 neo-classical tearing mode (NTM), q min was raised above 2 and the NTM was suppressed. The stored energy increased by 16% with the NTM suppressed, since the resonant rational surface was eliminated. For the future use for current profile control, current density profile for off-axis neutral beam current drive (NBCD) is for the first time measured, using the motional Stark effect diagnostic. Spatially localized NBCD profile was clearly observed at the normalized minor radius ρ of about 0.6–0.8. The location was also confirmed by multi-chordal neutron emission profile measurement. The total amount of the measured beam driven current was consistent with the theoretical calculation using the ACCOME code. The CD location in the calculation was inward shifted than the measurement.