Search

Recent experimental studies of H-mode plasmas in JET have revealed an n = 0 m = 1 magnetic oscillation starting immediately at the L to H transition (called M-mode for brevity). While the oscillation is present a weak ELM-less H-mode regime is obtained (medium confinement), with a clear increase of pedestal density (ne,ped) and a weak temperature pedestal (Te,ped). In ICRH heated plasmas or low power, low density NBI plasmas the mode and the pedestal pressure can remain steady for the duration of the heating phase, of order 10s or more. The observed axisymmetric MHD oscillation has period ~1­2ms, and odd parity across the mid-plane: m = 1 with sin(q) dependency. Electron Cyclotron Emission, interferometry, reflectometry, SXR and fast Li beam measurements locate the mode in the pedestal region. Da, fast infrared camera and Langmuir probe measurements show that the mode modulates heat and particle flux to the target. Similar phenomena are described elsewhere, lately described as an "I-phase". From a physics point of view, our fundamental observation is that the M-mode is magnetic, not electrostatic, and its frequency scales with pedestal poloidal Alfvén velocity.

For the design and the operation of large fusion devices a detailed understanding of the power exhaust processes is necessary. This paper will give an overview of the current research on the divertor power load mechanisms. The results shown are obtained in JET with the ITER like wall and ASDEX Upgrade with tungsten coated plasma facing components. The challenges of infrared thermography on an ITER like bulk tungsten divertor are presented. For the steady state heat load, the power fall-off length q in JET-ILW is compared to an empirical scaling found in JET and ASDEX Upgrade with carbon plasma facing components. A first attempt to scale the divertor broadening S in ASDEX Upgrade with tungsten plasma facing components is shown. The ELM duration tELM in JET-C and JET-ILW is compared. For similar pedestal conditions (Te,ped and ne,ped), similar ELM durations are found in JET-C and JET-ILW. For higher ne,ped at the same pedestal pressure pe,ped, longer ELM durations are found in JET-ILW. The pedestal pressure pe,ped is found to be a good qualifier for the ELM energy fluency in both JET-C and JET-ILW. Improved diagnostic capabilities reveal ELM substructures on the divertor target occurring a few milliseconds before the ELM crash.

The ITER-Like Wall recently installed in JET comprises solid beryllium limiters and a combination of bulk tungsten and tungsten-coated carbon fibre composite divertor tiles without active cooling. During a beryllium power handling qualification experiment performed in limiter configuration with 5MW NBI input power, accidental beryllium melt events, melt layer motion and splashing were observed locally on few beryllium limiters at the plasma contact areas. The Lorentz force is responsible for the observed melt layer movement. To move liquid beryllium against the gravity force, the current flowing from the plasma perpendicularly to the limiter surface must be higher than 6kA/m2. The thermo-emission current at the melting point of beryllium is much lower. The upward motion of the liquid beryllium against gravity can be due to a combination of the Lorentz force from the secondary electron emission and plasma pressure force.

The JET Scrape-Off Layer (SOL) has been characterized with a reciprocating probe in inner wall, IW, and outer wall, OW, limited plasmas. Experiments revealed that SOL profiles are substantially broader (by a factor of ~5 ­ 7.5 in the power e-folding length) for IW limited than in OW limited plasmas. Results are consistent with the larger radial turbulent transport found for IW limited plasmas. Major differences are observed between IW and OW limited plasmas on the density and electron temperature e-folding lengths, parallel flow, radial turbulent transport as well as on the temporal and spatial characteristics of the fluctuations. Experimental findings on JET suggest that the differences in the SOL characteristics for both configurations are due to a combination of a poloidal asymmetry in radial transport with a reduced cross-field transport across the last closed flux surface associated with the confinement improvement observed for OW limited plasmas. The dependence of the SOL power e-folding length on the main plasma parameters was also investigated for IW limited plasmas and a modest negative dependence on both the plasma current and the line-averaged density found. Finally, it is shown that the SOL radial transport and the amplitude of the fluctuations increase with plasma current and decrease with line-averaged density for IW limited plasmas.

Installation of the JET ITER-Like Wall (ILW)has brought on the need of a reliable protection system as too high a temperature of the plasma facing components (PFCs) can result in beryllium (Be) melting or thermal fatigue of tungsten (W). A proper temperature measurement relies on the knowledge of three different parameters of equal importance: the sensitivity of the camera sensor, the transmission factor of the optics used and the emissivity of the monitored PFCs. In this paper we first present the calibration process of a wide angle infrared (IR) camera, which has involved the capabilities of the remote handling for introducing a calibrated hot source into the JET vacuum vessel. Then, we describe how the PFCs equipped with embedded thermocouples have been used to assess the emissivity of both Be and W.

G.P. Maddison, C. Giroud, B. Alper, G. Arnoux, I. Balboa, M.N.A. Beurskens, A. Boboc, S. Brezinsek, M. Brix, M. Clever, R. Coelho, J.W. Coenen, I. Coffey, P.C. da Silva Aresta Belo, S. Devaux, P. Devynck, T. Eich , R.C. Felton, J. Flanagan, L. Frassinetti, L. Garzotti, M. Groth, S. Jachmich, A. Järvinen, E. Joffrin, M.A.H. Kempenaars, U. Kruezi , K.D. Lawson, M. Lehnen, M.J. Leyland, Y. Liu, P.J. Lomas, C.G. Lowry, S. Marsen, G.F. Matthews, G.K. McCormick, A.G. Meigs, A.W. Morris, R. Neu, I.M. Nunes, M. Oberkofler, F.G. Rimini, S. Saarelma, B. Sieglin, A.C.C. Sips, A. Sirinelli , M.F. Stamp , G.J. van Rooij, D.J. Ward , M. Wischmeier and JET EFDA contributors The former all-carbon wall on JET has been replaced with beryllium in the main torus and tungsten in the divertor to mimic the surface materials envisaged for ITER. Comparisons are presented between Type I H-mode characteristics in each design by examining respective scans over deuterium fuelling and impurity seeding, required to ameliorate exhaust loads both in JET at full capability and in ITER. Attention is focused upon a common high-triangularity, single-null divertor configuration at 2.5MA, q95 3.5 yielding the most robust all-C performance. Contrasting results between the alternative linings are found firstly in unseeded plasmas, for which purity is improved and intrinsic radiation reduced in the ITER-like wall (ILW) but normalised energy confinement is 30% lower than in all-C counterparts, owing to a commensurately lower (electron) pedestal temperature. Divertor recycling is also radically altered, with slower, inboard-outboard asymmetric transients at ELMs and spontaneous oscillations in between them. Secondly, nitrogen seeding elicits opposite responses in the ILW to all-C experience, tending to raise plasma density, reduce ELM frequency, and above all to recover (electron) pedestal pressure, hence global confinement, almost back to previous levels. A hitherto unrecognised role of light impurities in pedestal stability and dynamics is consequently suggested. Thirdly, while heat loads on the divertor outboard target between ELMs are successfully reduced in proportion to the radiative cooling and ELM recurrence effects of N in both wall environments, more surprisingly, average power ejected by ELMs and estimated scrape-off-layer pressure ratio during the approach to outboard detachment also both decline in the same proportion for the ILW. Finally, inter-ELM W sources in the ILW divertor tend to fall with N input, although core accumulation possibly due to net ELM erosion still leads to significantly less steady conditions than in all-C plasmas. This limitation of ILW H-modes so far will be readdressed in future campaigns to continue progress towards a fully-integrated scenario suitable for D-T experiments on JET and for "baseline" operation on ITER. The diverse changes in behaviour between all-C and ILW contexts demonstrate essentially the strong impact which boundary conditions and intrinsic impurities can have on tokamak-plasma states.

JET is equipped with a first wall material combination comparable to the ITER selection acomprising beryllium (Be) in the main chamber and tungsten (W) in the divertor and some recessed wall areas. Be is selected owing to its low atomic number, its low tritium retention and excellent getter properties, but material erosion limits the lifetime of plasma-facing components PFCs made of Be. Initial ERO modelling of the shaped Be first wall modules close to the ITER separatrix predict high erosion rates and limited armour lifetime. However, uncertainties in the modelling, in particular in the atomic data and sputtering yields, still exist and further bench- mark under ITER-comparable tokamak conditions is required. The ITER-Like Wall in JET (JET-ILW) demonstrated successful plasma operation, strong reduction of the C content (x20), and high plasma purity (Zeff ~ 1.2). Equipped with its bulk Be limiters, the JET-ILW allows the study of Be erosion by optical emission spectroscopy and observation of various transitions of BeI (e.g. 457nm), BeII (e.g. 527nm) and the BeD A-X band under different plasma conditions and surface temperatures. The total Be sputtering consists of the bare physical sputtering and the chemical assisted physical sputtering ­ sometimes referred as swift chemical sputtering. However, the composition of the total sputtering, its dependence on the impact energy and temperature, the strength of the chemical assisted physical sputtering are not known for a high temperature plasma edge conditions as present in the JET-ILW or in future ITER. Here, deuterium plasmas in limiter configuration have been used to vary the local electron temperature (Te) in the scrape-off layer (SOL), or better, scanning the impact energy of the impinging deuterons (Ei), as well as, to vary the PFC surface temperature (Tsurf) by plasma impact as they are only inertially cooled. The increase of Tsurf is expected to inhibit the sputtering channel via BeD which thermally decomposes at about 540K according to studies in PISCES.

The ITER-like wall (ILW) at JET is a unique opportunity to study the combination of material (beryllium and tungsten) that will be used for the plasma facing components (PFC) in ITER. Both the limiters (Be) and divertor (CFC W coated and bulk W) have been designed to maximise their power handling capability. During the last experimental campaign (October 2010 to July 2011) this capability has been assessed and even challenged in the case of the beryllium wall. The Be limiters are composed of highly shaped, castellated tiles. They are optimised for a range of scrape-off layer (SOL) power decay length 5 < lq,design < 20mm. Its power handling capability (19MW/m2s1/2), predicted with a simple model, has been proven to be robust by the experiments despite unexpected power loads pattern. However, this capability has been pushed to its limit leading to Be melt events, which revealed that the power load is toroidally asymmetric. The toroidal asymmetry could be explained by a misalignment of the limiters by 1mm. The protection system of the ILW has now been partially commissioned and the main chamber protection was operational. It did not prevent melt events mainly because the protection. The bulk W divertor target performed as predicted. Operations were constrained by: 1) an energy load limit of 60MJ per stack ( 60MJ/m2), and 2) the limited number of cycles of the surface temperature above 1200oC in order to prevent thermal fatigue. This latter limit has been exceeded about 300 times and no signs of damage or thermal fatigue have been observed by the photogrammetric survey.