T. Lunt

Divertor power load feedback with nitrogen seeding in ASDEX Upgrade

Feedback control of the divertor power load by means of nitrogen seeding has been developed into a routine operational tool in the all-tungsten clad ASDEX Upgrade tokamak. For heating powers above about 12?MW, its use has become inevitable to protect the divertor tungsten coating under boronized conditions. The use of nitrogen seeding is accompanied by improved energy confinement due to higher core plasma temperatures, which more than compensates the negative effect of plasma dilution by nitrogen on the neutron rate. This paper describes the technical details of the feedback controller. A simple model for its underlying physics allows the prediction of its behaviour and the optimization of the feedback gain coefficients used. Storage and release of nitrogen in tungsten surfaces were found to have substantial impact on the behaviour of the seeded plasma, resulting in increased nitrogen consumption with unloaded walls and a latency of nitrogen release over several discharges after its injection. Nitrogen is released from tungsten plasma facing components with moderate surface temperature in a sputtering-like process; therefore no uncontrolled excursions of the nitrogen wall release are observed. Overall, very stable operation of the high-Z tokamak is possible with nitrogen seeding, where core radiative losses are avoided due to its low atomic charge Z and a high ELM frequency is maintained.

Disruption studies in ASDEX Upgrade in view of ITER

Experiments on ASDEX Upgrade and other tokamaks have shown that the magnitude of mechanical forces and thermal loads during disruptions can be significantly reduced by raising the plasma density with massive injection of noble gases. This method should be applicable to ITER too. Nevertheless, the suppression of the runaway electron (RE) avalanche requires a much larger (two order of magnitude) density rise. This paper reports on recent experiments aimed at increasing the plasma density towards the critical value, needed for the collisional suppression of REs. An effective electron density equal to 24% of the critical density has been reached after injection of 3.3?bar?l of neon. However, the resultant large plasma density is very poloidally and toroidally asymmetric; this implies that several valves distributed around the plasma periphery become necessary at this level of massive gas injection to ensure a homogeneous density distribution.

Comparison between stellarator and tokamak divertor transport

This paper compares the essential divertor transport features of the poloidal divertor, which is well developed for tokamaks, and the non-axisymmetric divertors currently investigated on helical devices. It aims at surveying the fundamental similarities and differences in divertor concept and geometry, and their consequences on the functioning of a divertor. In particular, the importance of various transport terms governing axisymmetric and helical scrape-off layers (SOLs) is examined, with special attention being paid to energy, momentum and impurity transport. Tokamak and stellarator SOLs are compared by identifying key geometric parameters through which the governing physics can be illustrated by simple models and estimates. More quantitative assessments rely nevertheless on the modelling using the EMC3-EIRENE code. Most of the theoretical results are discussed in conjunction with experimental observations.

Fast-ion redistribution and loss due to edge perturbations in the ASDEX Upgrade, DIII-D and KSTAR tokamaks

The impact of edge localized modes (ELMs) and externally applied resonant and non-resonant magnetic perturbations (MPs) on fast-ion confinement/transport have been investigated in the ASDEX Upgrade (AUG), DIII-D and KSTAR tokamaks. Two phases with respect to the ELM cycle can be clearly distinguished in ELM-induced fast-ion losses. Inter-ELM losses are characterized by a coherent modulation of the plasma density around the separatrix while intra-ELM losses appear as well-defined bursts. In high collisionality plasmas with mitigated ELMs, externally applied MPs have little effect on kinetic profiles, including fast-ions, while a strong impact on kinetic profiles is observed in low-collisionality, low q95 plasmas with resonant and non-resonant MPs. In low-collisionality H-mode plasmas, the large fast-ion filaments observed during ELMs are replaced by a loss of fast-ions with a broad-band frequency and an amplitude of up to an order of magnitude higher than the neutral beam injection prompt loss signal without MPs. A clear synergy in the overall fast-ion transport is observed between MPs and neoclassical tearing modes. Measured fast-ion losses are typically on banana orbits that explore the entire pedestal/scrape-off layer. The fast-ion response to externally applied MPs presented here may be of general interest for the community to better understand the MP field penetration and overall plasma response.

First EMC3-Eirene simulations of the impact of the edge magnetic perturbations at ASDEX Upgrade compared with the experiment

The EMC3-Eirene code package was applied for the first time to simulate the edge plasma in an ASDEX Upgrade discharge, in which the newly installed magnetic perturbation (MP) coils were used to mitigate edge-localized modes (ELMs). Two different points in time during this discharge were simulated, the ELM-mitigated phase after turning-on of the MP coils and, as a reference, the ELMy H-mode phase before. The results were compared with the measurements of various edge and divertor diagnostics. Assuming the main chamber profiles to be shifted by 15?mm with respect to their calibrated positions, an agreement within a factor of 2 was found between the main chamber profiles outside the separatrix and those at the outer divertor target. The most important result is the observation of several maxima and minima in the particle flux and in particular in the power deposition pattern of both the simulation and the experiment for the case with MPs, an effect also known as strike-point splitting.

Power Distribution in the Snowflake Divertor in TCV

TCV experiments demonstrate the basic power exhaust properties of the snowflake (SF) plus and SF minus divertor configurations by measuring the heat fluxes at each of their four divertor legs. The measurements indicate an enhanced transport into the private flux region and a reduction of peak heat fluxes compared to a similar single null configuration. There are indications that this enhanced transport cannot be explained by the modified field line geometry alone and likely requires an additional or enhanced cross-field transport channel. The measurements, however, do not show a broadening of the scrape-off layer (SOL) and, hence, no increased cross-field transport in the common flux region. The observations are consistent with the spatial limitation of several characteristic SF properties, such as a low poloidal magnetic field in the divertor region and a long connection length to the inner part of the SOL closest to the separatrix. Although this limitation is typical in a medium sized tokamak like TCV, it does not apply to significantly larger devices where the SF properties are enhanced across the entire expected extent of the SOL.

Dust investigations at ASDEX Upgrade

Investigations on dust in the full tungsten divertor tokamak, ASDEX Upgrade, are summarized. Newly designed collectors are used to distinguish plasma-produced dust from other kinds of debris. Whereas submicron particles are often tungsten spheres, larger ones are agglomerates from tungsten in a boron/carbon matrix. Extrapolation of the measurement yields 80 mg of carbon and 1200 mg of tungsten in the dust, including 510 mg of tungsten spheres. In the case of its mobilization, the amount of tungsten observed may constitute a significant impurity source. Droplets produced by arcs seem to be the production mechanism for tungsten spheres. The strong light emission of dust particles penetrating the scrape off layer (SOL) is rarely observed by a fast camera during normal discharges, but is very frequently observed after disruptions. As the particles follow straight lines, we infer only a weak interaction with the SOL plasma. This is also confirmed by the observation of fragile agglomerates.

Blob properties in L- and H-mode from gas-puff imaging in ASDEX Upgrade

Blob properties are studied in the scrape-off layer of the tokamak ASDEX Upgrade with a fast camera. The gas-puff imaging technique is used to investigate the detection rate as well as the blob size and velocity scaling. The experiments were performed in L- and H-mode phases of the same discharges to study the change in blob properties after the L-H transition. In both regimes the detection rate is of the order of a few thousand blobs per second, which is compatible with the picture of blob generation by edge micro instabilities. The blob size increases in H-mode, while the radial velocity decreases slightly. The changes are, however, not indicating a drastic change in the blob dynamics in both phases. The experimentally found blob properties were compared to predictions from a novel blob model including effects due to a finite ion temperature, which should be more appropriate for the conditions in the SOL of fusion plasmas.

Ion energies and currents of Type I and mitigated ELMs in the ASDEX Upgrade far scrape-off layer

New measurements of ion energies and currents in type I and mitigated ELMs have been carried out in the ASDEX Upgrade far scrape-off layer using a retarding field analyser (RFA). The ion temperature averaged over an ELM, Ti?ELM measured 35?60?mm outside the separatrix (i.e. 15?25?mm in front of the outboard limiter) is in the range 20?200?eV, which is 5?50% of the ion temperature at the pedestal top. Ti?ELM decreases with the separatrix distance with the e-folding length of ~10?mm measured in the far SOL for a particular set of conditions, and increases with the ELM energy WELM. Lowest Ti?ELM is measured during mitigated type I ELMs. Likewise, the ELM-averaged ion current e-folding length increases with WELM, similar to the e-folding length of the heat flux density at the RFA probe head during an ELM, monitored by a fast IR camera. The most plausible explanation of observed trends is that on average the filaments of larger ELMs travel faster radially and have less time to dilute by parallel losses along field lines before reaching the far SOL. These observations provide further evidence that the fraction of the ELM energy deposited on the main chamber plasma-facing components increases with WELM.

Three-dimensional modeling of plasma edge transport and divertor fluxes during application of resonant magnetic perturbations on ITER

Results from three-dimensional modeling of plasma edge transport and plasma–wall interactions during application of resonant magnetic perturbation (RMP) fields for control of edge-localized modes in the ITER standard 15 MA Q = 10 H-mode are presented. The full 3D plasma fluid and kinetic neutral transport code EMC3-EIRENE is used for the modeling. Four characteristic perturbed magnetic topologies are considered and discussed with reference to the axisymmetric case without RMP fields. Two perturbation field amplitudes at full and half of the ITER ELM control coil current capability using the vacuum approximation are compared to a case including a strongly screening plasma response. In addition, a vacuum field case at high q 95 = 4.2 featuring increased magnetic shear has been modeled. Formation of a three-dimensional plasma boundary is seen for all four perturbed magnetic topologies. The resonant field amplitudes and the effective radial magnetic field at the separatrix define the shape and extension of the 3D plasma boundary. Opening of the magnetic field lines from inside the separatrix establishes scrape-off layer-like channels of direct parallel particle and heat flux towards the divertor yielding a reduction of the main plasma thermal and particle confinement. This impact on confinement is most accentuated at full RMP current and is strongly reduced when screened RMP fields are considered, as well as for the reduced coil current cases. The divertor fluxes are redirected into a three-dimensional pattern of helical magnetic footprints on the divertor target tiles. At maximum perturbation strength, these fingers stretch out as far as 60 cm across the divertor targets, yielding heat flux spreading and the reduction of peak heat fluxes by 30%. However, at the same time substantial and highly localized heat fluxes reach divertor areas well outside of the axisymmetric heat flux decay profile. Reduced RMP amplitudes due to screening or reduced RMP coil current yield a reduction of the width of the divertor flux spreading to about 20–25 cm and cause increased peak heat fluxes back to values similar to those in the axisymmetric case. The dependencies of these features on the divertor recycling regime and the perpendicular transport assumptions, as well as toroidal averaged effects mimicking rotation of the RMP field, are discussed in the paper.