Search

The Korean Superconducting Tokamak Advanced Research has been focused on exploring the key physics and engineering issues for future fusion reactors by demonstrating the long pulse operation of high beta steady-state discharge. Advanced scenarios are being developed with the goal for steady-state operation, and significant progress has been made in high ℓ _i , hybrid and high beta scenarios with β _N of 3. In the new operation scenario called fast ion regulated enhanced (FIRE), fast ions play an essential role in confinement enhancement. GK simulations show a significant reduction of the thermal energy flux when the thermal ion fraction decreases and the main ion density gradient is reversed by the fast ions in FIRE mode. Optimization of 3D magnetic field techniques, including adaptive control and real-time machine learning control algorithm, enabled long-pulse operation and high-performance ELM-suppressed discharge. Symmetric multiple shattered pellet injections (SPIs) and real-time disruption event characterization and forecasting are being performed to mitigate and avoid the disruptions associated with high-performance, long-pulse ITER-like scenarios. Finally, the near-term research plan will be addressed with the actively cooled tungsten divertor, a major upgrade of the NBI and helicon current drive heating, and transition to a full metallic wall.

Intense bursts of suprathermal radiation, with spectral peaks at frequencies corresponding to the deuteron cyclotron frequency in the outer midplane edge region, are often detected from deuterium plasmas in the KSTAR tokamak that are heated by tangential neutral beam injection (NBI) of deuterons at 100 keV. Identifying the physical process by which this deuterium ion cyclotron emission (ICE) is generated, typically during the crash of edge localised modes, assists the understanding of collective energetic ion behaviour in tokamak plasmas. In the context of KSTAR deuterium plasmas, it is also important to distinguish deuterium ICE from the ICE at cyclotron harmonics of fusion-born protons examined by Chapman et al (2017 Nucl. Fusion57 124004; 2018 Nucl. Fusion58 096027). We use particle orbit studies in KSTAR-relevant magnetic field geometry, combined with a linear analytical treatment of the magnetoacoustic cyclotron instability (MCI), to identify the sub-population of freshly ionised NBI deuterons that is likely to excite deuterium ICE. These deuterons are then represented as an energetic minority, together with the majority thermal deuteron population and electrons, in first principles kinetic particle-in-cell (PIC) computational studies. By solving the Maxwell–Lorentz equations directly for hundreds of millions of interacting particles with resolved gyro-orbits, together with the self-consistent electric and magnetic fields, the PIC approach enables us to study the collective relaxation of the energetic deuterons through the linear phase and deep into the saturated regime. The Fourier transform of the excited fields displays strong spectral peaks at multiple successive deuteron cyclotron harmonics, mapping well to the observed KSTAR deuterium ICE spectra. This outcome, combined with the time-evolution of the energy densities of the different particle populations and electric and magnetic field components seen in the PIC computations, supports our identification of the driving sub-population of NBI deuterons, and the hypothesis that its relaxation through the MCI generates the observed deuterium ICE signal. We conclude that the physical origin of this signal in KSTAR is indeed distinct from that of KSTAR proton ICE, and is in the same category as the NBI-driven ICE seen notably in the TFTR tokamak and LHD heliotron–stellarator plasmas. ICE has been proposed as a potential passive diagnostic of energetic particle populations in ITER plasmas; this is assisted by clarifying and extending the physics basis of ICE in contemporary magnetically confined plasmas. [ABSTRACT FROM AUTHOR]

Ion cyclotron emission (ICE) is detected from all large toroidal magnetically confined fusion (MCF) plasmas. It is a form of spontaneous suprathermal radiation, whose spectral peak frequencies correspond to sequential cyclotron harmonics of energetic ion species, evaluated at the emission location. In ICE phenomenology, an important parameter is the value of the ratio of energetic ion velocity to the local Alfvén speed . Here we focus on ICE measurements from heliotron-stellarator hydrogen plasmas, heated by energetic proton neutral beam injection (NBI) in the large helical device, for which takes values both larger (super-Alfvénic) and smaller (sub-Alfvénic) than unity. The collective relaxation of the NBI proton population, together with the thermal plasma, is studied using a particle-in-cell (PIC) code. This evolves the Maxwell–Lorentz system of equations for hundreds of millions of kinetic gyro-orbit-resolved ions and fluid electrons, self-consistently with the electric and magnetic fields. For LHD-relevant parameter sets, the spatiotemporal Fourier transforms of the fields yield, in the nonlinear saturated regime, good computational proxies for the observed ICE spectra in both the super-Alfvénic and sub-Alfvénic regimes for NBI protons. At early times in the PIC treatment, the computed growth rates correspond to analytical linear growth rates of the magnetoacoustic cyclotron instability (MCI), which was previously identified to underlie ICE from tokamak plasmas. The spatially localised PIC treatment does not include toroidal magnetic field geometry, nor background gradients in plasma parameters. Its success in simulating ICE spectra from both tokamak and, here, heliotron-stellarator plasmas suggests that the plasma parameters and ion energetic distribution at the emission location largely determine the ICE phenomenology. This is important for the future exploitation of ICE as a diagnostic for energetic ion populations in MCF plasmas. The capability to span the super-Alfvénic and sub-Alfvénic energetic ion regimes is a generic challenge in interpreting MCF plasma physics, and it is encouraging that this first principles computational treatment of ICE has now achieved this. [ABSTRACT FROM AUTHOR]

The bifurcation of perpendicular mode velocity near the pedestal top () at the onset of RMP-driven edge-localized mode (ELM)-crash suppression has been directly measured by using an electron cyclotron emission imaging diagnostic on KSTAR. The ELM crashes are suppressed along with a sudden reduction of , which synchronizes with the transition into and out of the ELM-crash suppression. The change of flow appears mainly responsible for the change of , whose magnitude is the smallest near the normalized flux surface during the ELM-crash suppression. The plasma response to the RMP field was most enhanced in the vicinity of during the ELM-crash suppression and showed a hysteresis behavior with respect to resonant field strength. With these changes, a strong nonlinear coupling between turbulent eddies was observed in the ELM-crash suppression, which is an important feature of the ELM-crash suppression. [ABSTRACT FROM AUTHOR]

Fast sawtooth crash of m  =  1 mode (tens of µs) is a common MHD phenomenon in the core of tokamak plasma. In KSTAR plasmas with localized heating and current drive close to the sawtooth inversion radius (ρinv), very slow sawtooth crashes are observed with time scales of a few ms. Scan of heating deposition along with power modulation revealed two different mechanisms of the slow crash with the help of detailed 2D imaging diagnostics. The slow crash appears with slow heat transport of the remained m  =  1 mode after the partial crash of outer multiple m/n  =  1 modes, or with the growth of an island formed on ρinv when the heating deposition locates just inside ρinv or just outside ρinv. In the heating power modulation experiment, the slow crash with the island was sustained for the redistribution time scale of the driven current by a very short blip of heating. [ABSTRACT FROM AUTHOR]

The radio frequency detection system on the KSTAR tokamak has exceptionally high spectral and temporal resolution. This enables measurement of previously undetected fast plasma phenomena in the ion cyclotron range of frequencies. Here we report and analyse a novel spectrally structured ion cyclotron emission (ICE) feature in the range 500 MHz to 900 MHz, which exhibits chirping on sub-microsecond timescales. Its spectral peaks correspond to harmonics l of the proton cyclotron frequency fcp at the outer midplane edge, where l  =  20–36. This frequency range exceeds estimates of the local lower hybrid frequency fLH in the KSTAR deuterium plasma. The new feature is time-shifted with respect to a brighter lower-frequency chirping ICE feature in the range 200 MHz (8fcp) to 500 MHz (20fcp), which is probably driven (Chapman et al 2017 Nucl. Fusion57 124004) by 3 MeV fusion-born protons undergoing collective relaxation by the magnetoacoustic cyclotron instability (MCI). Here we show that the new, fainter, higher-frequency chirping ICE feature is driven by nonlinear wave coupling between different neighbouring spectral peaks in the lower-frequency ICE feature. This follows from bispectral analysis of the measured KSTAR fields, and of the field amplitudes output from particle-in-cell (PIC) simulations of the KSTAR edge plasma containing fusion-born protons. This reinforces the identification of the MCI as the plasma physics process underlying proton harmonic ICE from KSTAR, while providing a novel instance of nonlinear wave coupling on very fast timescales. [ABSTRACT FROM AUTHOR]

Using a high-speed and broadband radio frequency (RF) (0.1–1 GHz) spectrum analyzer developed on the KSTAR tokamak, it is found that several distinct stages of RF emission appear at the pedestal collapse in high confinement discharges. Comparison with 2D electron cyclotron emission (ECE) images has revealed that each stage is related to the instantaneous condition at the outboard mid-plane edge. First, high-harmonic ion cyclotron emissions (ICE) are intensified with the appearance of a non-modal filamentary perturbation in the edge within several tens of microseconds before the collapse. Then, the RF emission becomes broad toward high-frequency range (<500 MHz) at the burst onset of the non-modal filament. During the pedestal collapse initiated by the filament burst, rapid chirping (1–3 μs) appear with additional filament bursts. The strong correlation between the RF spectra and the perturbation structure provides important clues on the stability of edge-localized modes and on the ion dynamics in the plasma boundary. [ABSTRACT FROM AUTHOR]

The central safety factor (q0) during sawtooth oscillation has been measured with a great accuracy with the motional Stark effect (MSE) system on KSTAR and the measured value was However, this measurement alone cannot validate the disputed full and partial reconnection models definitively due to non-trivial off-set error (~0.05). Supplemental experiment of the excited m  =  2, m  =  3 modes that are extremely sensitive to the background q0 and core magnetic shear definitively validates the ‘full reconnection model’. The radial position of the excited modes right after the crash and time evolution into the 1/1 kink mode before the crash in a sawtoothing plasma suggests that in the MHD quiescent period after the crash and before the crash. Additional measurement of the long lived m  =  3, m  =  5 modes in a non-sawtoothing discharge (presumably ) further validates the ‘full reconnection model’. [ABSTRACT FROM AUTHOR]

Broadband density fluctuations with peak frequency ranging from 150 to 400 kHz were measured using a multichannel microwave imaging reflectometer in core region of the low-density L-mode plasmas heated by neutral beam injection on KSTAR. These fluctuations have been studied by comparing the dominant mode scales estimated from the measurement with those predicted from linear gyrokinetic simulation. The measured poloidal wavenumbers are qualitatively comparable to those of the ‘fastest growing modes’ from simulations, whereas they are larger than those of the ‘transport-dominant modes’ by about a factor of three. The agreement on wavenumbers between the measurement and linear simulation (for the fastest growing modes) is probably due to sufficiently weak flow shear compared to the maximum linear growth rate. Meanwhile, the transport-dominant modes seem to be related to the fluctuations in lower frequencies (∼80–150 kHz) observed in some of the measurement. [ABSTRACT FROM AUTHOR]

During edge localised mode (ELM) crashes in KSTAR deuterium plasmas, bursts of spectrally structured ion cyclotron emission (ICE) are detected. Usually the ICE spectrum chirps downwards during an ELM crash, on sub-microsecond timescales. For KSTAR ICE where the separation of spectral peak frequencies is close to the proton cyclotron frequency at the outer plasma edge, we show that the driving population of energetic ions is likely to be a subset of the 3 MeV fusion protons, born centrally on deeply passing orbits which drift from the core to the edge plasma. We report first principles modelling of this scenario using a particle-in-cell code, which evolves the full orbit dynamics of large numbers of energetic protons, thermal deuterons, and electrons self-consistently with the electric and magnetic fields. The Fourier transform of the excited fields in the nonlinear saturated regime of the simulations is the theoretical counterpart to the measured ICE spectra. Multiple simulation runs for different, adjacent, values of the plasma density under KSTAR edge conditions enable us to infer the theoretical dependence of ICE spectral structure on the local electron number density. By matching this density dependence to the observed time-dependence of chirping ICE spectra in KSTAR, we obtain sub-microsecond time resolution of the evolving local electron number density during the ELM crash. [ABSTRACT FROM AUTHOR]

Multiscale interaction between the magnetic island and turbulence has been demonstrated through simultaneous two-dimensional measurements of turbulence and temperature and flow profiles. The magnetic island and turbulence can mutually interact via coupling between the electron temperature (Te) gradient, the Te turbulence, and the poloidal flow. The Te gradient altered by the magnetic island steepens outside and flattens inside the island. The Te turbulence can appear in increased Te gradient regions. The combined effects of the Te gradient and the poloidal flow shear determines the two-dimensional distribution of the Te turbulence. When the poloidal vortex flow forms, it can maintain the steepest Te gradient and the magnetic island acts more like an electron heat transport barrier. Interestingly, when the Te gradient, the Te turbulence, and the vortex flow shear increase beyond critical levels, the magnetic island turns into a fast electron heat transport channel, which directly leads to the minor disruption. [ABSTRACT FROM AUTHOR]

This article assesses current understanding of hysteresis in transport relations, and its impact on the field. The rapid changes of fluxes compared to slow changes of plasma parameters are overviewed for both core and edge plasmas. The modulation ECH experiment is explained, in which the heating power cycles on-and-off periodically, revealing hysteresis and fast changes in the gradient–flux relation. The key finding is that hystereses were observed simultaneously in both the the gradient–flux and gradient–fluctuation relations. Hysteresis with rapid timescale exists in the channels of energy, electron and impurity densities, and plausibly in momentum. Advanced methods of data analysis are explained. Transport hysteresis can be studied by observing the higher harmonics of temperature perturbation in heating modulation experiments. The hysteresis introduces the term , which depends on the harmonic number m in an algebraic manner (not exponential decay). Next, the causes of hysteresis and its fast timescale are discussed. The nonlocal-in-space coupling works here, but does not suffice. One mechanism for ‘the heating heats turbulence’ is that the external source S in phase space for heating has its fluctuation in turbulent plasma. This coupling can induce the direct input of heating power into fluctuations. The height of the jump in transport hysteresis is smaller for heavier hydrogen isotopes, and could be one of the origins of isotope effects on confinement. Finally, the impacts of transport hysteresis on the control system are assessed. Control systems must be designed so as to protect the system from sudden plasma loss. [ABSTRACT FROM AUTHOR]

In this contribution we analyze modulation electron cyclotron resonance heating (MECH) experiment and discuss higher harmonic frequency dependence of transport coefficients. We use the bidirectional heat pulse propagation method, in which both inward propagating heat pulse and outward propagating heat pulse are analyzed at a radial range, in order to distinguish frequency dependence of transport coefficients due to hysteresis from that due to other reasons, such as radially dependent transport coefficients, a finite damping term, or boundary effects. The method is applied to MECH experiments performed in various helical and tokamak devices, i.e. Large Helical Device (LHD), TJ-II, Korea Superconducting Tokamak Advanced Research (KSTAR), and Doublet III-D (DIII-D) with different plasma conditions. The frequency dependence of transport coefficients are clearly observed, showing a possibility of existence of transport hysteresis in flux-gradient relation. [ABSTRACT FROM AUTHOR]

Minor and major disruptions by explosive MHD instabilities were observed with the novel quasi 3D electron cyclotron emission imaging (ECEI) system in the KSTAR plasma. The fine electron temperature (Te) fluctuation images revealed two types of minor disruptions: a small minor disruption is a localized fast transport event due to a single m/n  =  2/1 magnetic island growth, while a large minor disruption is partial collapse of the region with two successive fast heat transport events by the correlated m/n  =  2/1 and m/n  =  1/1 instabilities. The m/n  =  2/1 magnetic island growth during the minor disruption is normally limited below the saturation width. However, as the additional interchange-like perturbation grows near the inner separatrix of the 2/1 island, the 2/1 island can expand beyond the limit through coupling with the cold bubble formed by the interchange-like perturbation. [ABSTRACT FROM AUTHOR]

The structure of the toroidal rotation profile with mixed heating by neutral beam injection (NBI) and electron cyclotron resonance heating (ECH) has been investigated in KSTAR L-mode plasmas. ECH with varying resonance layer positions was used for heating a mix control. The experimental results show that ECH causes a counter-current rotation increment both for off-axis and on-axis ECH heating. For L-mode plasmas, off-axis ECH produces larger counter-current rotation than on-axis ECH. Analysis of ion heat and momentum transport for the ECH L-mode plasmas shows that the electron temperature gradient is the main reason for the degradation of ion heat confinement and also the main driving force for the non-diffusive momentum flux. As a possible mechanism for the counter-current intrinsic torque with ECH, the transition of the turbulence mode from ion temperature gradient (ITG) to the trapped electron mode (TEM) with the resulting sign change of turbulence driven residual stress is suggested. A linear gyro-kinetic analysis shows the ITG  →  TEM transition occurs in a localized region during ECH injection, and the trend of TEM excitation is consistent with the observed macroscopic trend of the toroidal rotation. [ABSTRACT FROM AUTHOR]

Temporal and spatial modulation of the edge localized mode (ELM) structure has been observed during the inter-ELM-crash period by toroidally-separated two electron cyclotron emission imaging systems. The observed modulation is interpreted as a beat wave of two modes with adjacent toroidal mode number. An additional assumption is that each mode has to have a different poloidal rotation speed. In nonlinear simulation, the low-n mode can be driven by locking between the dominant modes. The modulation is reconstructed using beat waves not the locking of modes. [ABSTRACT FROM AUTHOR]

Multiple (two or more) flux tubes are commonly observed inside and/or near the q = 1 flux surface in KSTAR tokamak plasmas with localized electron cyclotron resonance heating and current drive (ECH/CD). Detailed 2D and quasi-3D images of the flux tubes obtained by an advanced imaging diagnostic system showed that the flux tubes are m/n = 1/1 field-aligned structures co-rotating around the magnetic axis. The flux tubes typically merge together and become like the internal kink mode of the usual sawtooth, which then collapses like a usual sawtooth crash. A systematic scan of ECH/CD beam position showed a strong correlation with the number of flux tubes. In the presence of multiple flux tubes close to the q = 1 surface, the radially outward heat transport was enhanced, which explains naturally temporal changes of electron temperature. We emphasize that the multiple flux tubes are a universal feature distinct from the internal kink instability and play a critical role in the control of sawteeth using ECH/CD. [ABSTRACT FROM AUTHOR]

A detailed study of edge-localized mode (ELM) dynamics in the KSTAR tokamak is performed using a two-dimensional (2D) electron cyclotron emission imaging (ECEI) diagnostic system. Highly coherent mode structures rotating in the poloidal view plane are routinely observed in the inter-ELM pedestal region where the optical thickness for ECE rapidly changes and the interpretation of emission intensity is complicated. To have confidence on the measurements, the observed images are compared with synthetic images of the ELM structure deduced from three-field BOUT++ simulations. The synthetic process considers instrumental effects of the ECEI diagnostic, intrinsic broadening of the ECE and background noise. The synthetic 2D images highly resemble the observed structure, providing confidence that the ELM dynamics can be visualized by ECEI. [ABSTRACT FROM AUTHOR]