Search

To support the deployment of the DEMO wall protection strategy, the development of comprehensive analyses is essential to understand the implications of transient perturbations on plasma shape control and on vertical stability, some of the most critical aspects to be considered in elongated plasmas. The design activities of the DEMO limiter structures require a deep understanding of the effects induced by transient plasma perturbations coupled with one of the most severe load conditions to occur in tokamaks, the vertical displacement event (VDE). Since electromagnetic (EM) loads during VDE phases are among the DEMO limiter’s design drivers, this study focuses on predictive simulations of the final plasma position and of EM loads following a VDE. For this purpose, a multi-tokamak study, supported by the construction of an inter-machine database containing experimental transient plasma perturbations and VDEs from JET and ASDEX Upgrade (AUG), has been carried out. It aims to characterize some transient plasma perturbations that may lead to high control efforts by the vertical stability system in terms of variations of the plasma’s internal parameters and vertical displacements. Consequently, such experimental transient plasma perturbations have been properly scaled to DEMO reference geometries with different magnetic configurations, to be simulated in terms of plasma dynamical behaviour by means of MAXFEA code. Finally, initial predictive EM loads on DEMO limiter structures are discussed in the case of VDEs following plasma perturbations.

Organic NE213 liquid scintillator neutron detectors are commonly used at accelerator facilities for neutron beam diagnostics. In recent years, they have also been installed at nuclear fusion facilities to measure the neutron energy spectra from Deuterium (D) and Deuterium-Tritium plasmas, e.g., at the ASDEX Upgrade (Garching, Germany) and at the Joint European Torus (JET, Culham, UK). The NE213 equivalent scintillating material (now BC501A) of the detector is sensitive to neutron and gamma radiation, so pulse discrimination techniques are applied in order to obtain the measured neutron pulse height spectrum (PHS). If the neutron detector is stable, controlled, and well-characterized (i.e., its response function to incoming neutrons of known energy is determined with high accuracy), it can be used as a neutron spectrometer. The measured PHS can then be analyzed using standard techniques such as unfolding to determine the incoming neutron energy spectrum. This article focuses on the unfolding of neutron data measured in D plasma experiments at JET by a compact broadband neutron spectrometer using the maximum entropy unfolding code MAXED. A general method for optimizing MAXED input parameters is described and applied to the measured PHS to diagnose the effects of the auxiliary heating of JET D plasma on the neutron energy spectra.

Nuclear fusion is one of the best options to achieve a virtually limitless energy source in the future. However, sustaining burning plasma reactions is very challenging because disruptive events cause the loss of plasma confinement and damages to the tokamak machine. Thus, reproducing a nuclear fusion reaction on Earth represents an actual scientific and technical challenge. Next generation machines must be able to predict, mitigate or avoid the disruptions but the present understanding of the phenomenon has not yet gone so far as to provide a physical model describing their root cause. On the contrary, a large quantity of experimental disruption data exists. For these reason, in the last decades, machine learning techniques have been applied to design alarm systems both for mitigating or to actively avoid approaching disruptions. More recently, the disruption predictor concept has been evolving toward a more complex system, able to detect the proximity of the plasma state to the boundaries of the operational space free from disruptions in order to schedule avoidance strategies rather than mitigation actions. This system should also be able to determine which type of disruption is about to occur in order to efficiently take suitable disruption avoidance actions. This paper reports an overview of machine learning algorithms as tools for disruption prediction and classification at JET, the world's largest operational plasma physics experiment, located at Culham Centre for Fusion Energy in Oxfordshire, UK. Both traditional neural network models for disruption prediction and manifold learning approaches as a tool for plasma operational space mapping and visualization, disruption prediction and classification, are described.

The Joint European Torus (JET) presently operates with the ITER-like wall (ILW) including tungsten divertor. Analyses of the soft X-ray radiation (SXR) data present an important tool for studies of transport of tungsten into JET plasmas. SXR spatial distribution can be measured by horizontal (H) and two vertical (V,T) pinhole cameras. Tomographic reconstruction of SXR emissivity from their data is challenging due to different spectral sensitivities of the cameras which can cause significant inconsistencies due to spectral properties of tungsten radiation. In this contribution an advanced correction based on evaluation of the sensitivity differences for individual lines of view of the SXR detectors is presented. The adjustments take into account atomic data of tungsten as well as the SXR intensity perturbation due to tungsten influx. First results are presented and discussed.

Hydrocarbon film deposition was studied with cavity samples in remote areas of the inner and outer JET divertor and below the divertor septum during the 1999–2001 and 2005–2009 campaigns. Thick hydrocarbon films were formed inside the cavities. These deposited hydrocarbon layers have high D/C ratios close to 1. The formation of these films is mainly due to sticking of hydrocarbon particles with high surface loss probabilities >0.6. The observed surface loss probabilities depend on the position in the divertor and vary during different campaigns. The particles responsible for hydrocarbon layer formation originate from the divertor strike points. Except for the septum cavity the deposition of beryllium was very low and showed a very different distribution from that of deuterium and carbon.

Effects of N-seeding in L-mode experiments in ASDEX Upgrade and JET are analysed numerically with the SOLPS5.0 code package. The modelling yields 3 qualitatively different radiative regimes with increasing N concentration, when initially attached outer divertor conditions are studied. The radiation pattern is observed to evolve asymmetrically, with radiation increasing first in the inner divertor, then in the outer divertor, and finally on closed field lines above the X-point. The properties of these radiative regimes are observed to be sensitive to cross-field drifts and they differ between the two devices. The modelled scaling of the divertor radiated power with the divertor neutral pressure is similar to an experimental scaling law for H-mode radiation. The same parametric dependencies are not observed in simulations without drifts.

Selected Ion Beam Analysis techniques applicable for detecting deuterium and heavier impurities have been used in the post-mortem analyses of tiles removed after the first JET ITER-Like Wall (JET-ILW) campaign. Over half of the retained fuel was measured in the divertor region. The highest figures for fuel retention were obtained from regions with the thickest deposited layers, i.e. in the inner divertor on top of tile 1 and on the High Field Gap Closure tile, which resides deep in the plasma scrape-off layer. Least retention was found in the main chamber high erosion regions, i.e. in the mid-plane of Inner Wall Guard Limiter. The fuel retention values found typically varied with deposition layer thicknesses. The reported retention values support the observed decrease in fuel retention obtained with gas balance experiments of JET-ILW.

Recent JET-ILW [1] , [2] experiments reiterated the importance of tuning the plasma fuelling in order to optimize ion cyclotron resonance frequency (ICRF) heating in high power H-mode discharges. By fuelling the plasma from gas injection modules (GIMs) located in the mid-plane and on the top of the machine instead of adopting the more standardly used divertor GIMs, a considerable increase of the ICRF antenna coupling resistances was achieved with moderate gas injection rates (

Équipe 107 : Physique des plasmas chauds; International audience; In JET-ILW dedicated melt exposures were performed using a sequence of 3MA/2.9T H-Mode JET pulses with an input power of P-IN = 23 MW, a stored energy of similar to 6 MJ and regular type I ELMs at Delta W-ELM = 0.3 MJ and f(ELM) similar to 30 Hz. In order to assess the risk of starting ITER operations with a full W divertor, one of the task was to measure the consequences of W transients melting due to ELMs. JET is the only tokamak able to produce transients/ ELMs large enough (>300 kJ per ELM) to facilitate melting of tungsten. Such ELMs are comparable to mitigated ELMs expected in ITER. By moving the outer strike point (OSP) onto a dedicated leading edge the base temperature was raised within similar to 1 s to allow transient ELM-driven melting during the subsequent 0.5 s. Almost 1 mm (similar to 6 mm(3)) of W was moved by similar to 150 ELMs within 5 subsequent discharges. Significant material losses in terms of ejections into the plasma were not observed. There is indirect evidence that some small droplets (similar to 80 mu m) were ejected. The impact on the main plasma parameters is minor and no disruptions occurred. The W-melt gradually moved along the lamella edge towards the high field side, driven by j x B forces. The evaporation rate determined is 100 times less than expected from steady state melting and thus only consistent with transient melting during individual ELMs. IR data, spectroscopy, as well as melt modeling point to transient melting. Although the type of damage studied in these JET experiments is unlikely to be experienced in ITER, the results do strongly support the design strategy to avoid exposed edges in the ITER divertor. The JET experiments required a surface at normal incidence and considerable pre-heating to produce tungsten melting. They provide unique experimental evidence for the absence of significant melt splashing at events resembling mitigated ELMs on ITER and establish a unique experimental benchmark for the simulations being used to study transient shallow melting on ITER W divertor PFUs

The impact of the divertor geometry on global plasma confinement in type I ELMy H-mode has been investigated in the JET tokamak equipped with ITER-Like Wall. Discharges have been performed in which the position of the strike-points was changed while keeping the bulk plasma equilibrium essentially unchanged. Large variations of the global plasma confinement have been observed, the H98 factor changing from typically 0.7 when the outer strike-point is on the vertical or horizontal targets to 0.9 when it is located in the pump duct entrance. Profiles are mainly impacted in the pedestal but core gradient lengths, especially for the density, are also modified. Although substantial differences are observed in the divertor conditions, none seem to correlate directly with the confinement. Modelling with the EDGE2D-EIRENE and SOLEDGE2D-EIRENE transport codes exhibits differences in the energy losses due to neutrals inside the separatrix, but orders of magnitude are too low to explain simply the impact on the confinement.

The erosion of Be and W marker layers was investigated using long-term samples during the first ITER-like wall discharge campaign 2011-2012. The markers were mounted in Be coated Inconel tiles between the inner wall guard limiters (IWGL). They were analyzed using Rutherford backscattering (RBS) before and after exposure. All samples showed strong erosion. The results were compared to the data for Be and W erosion rates for the 2005-2009 and the 2001-2004 campaigns, respectively, when JET was operated with a carbon wall. In 2005-2009 Be and C samples were used, and W samples were used in 2001-2004. The mean W erosion rates and the toroidal and poloidal distributions of the W erosion were the same for the 2001-2004 and the 2011-2012 campaigns. The mean erosion rate of Be during the 2011-2012 campaign was smaller by a factor of about two compared to the 2005-2009 campaign and showed a different poloidal distribution. The mean erosion rate of the inner JET ITER-like wall was about 4-5 times smaller than the mean erosion rate of the carbon wall.

High-density discharges on JET with ITER-like Wall (ILW) have been analysed with the aim of establishing a mechanism for the H-mode density limit (DL) and compared with experiments in the JET carbon material configuration. The density limit is up to 20% higher in the JET-ILW than in the JET-C machine. The observed H-mode density limit is found close to the Greenwald limit. It is sensitive to the main plasma shape and is almost independent of the heating power. It has been observed that the transition from H-mode to L-mode is not always an abrupt event but may exhibit a series of H–L–H transitions, the so-called “dithering H-mode”. It was observed that detachment, as well as the X-point MARFE itself, does not trigger the H–L transition and thus does not present a limit on the plasma density and that it is the plasma confinement which is ultimately responsible for the H-mode DL.

An original spot detection method for large surveys of videos in Joint European Torus (JET) is presented. The method can be used for the automatic identification of spots in JET infra-red (IR) videos and for the assessment of the long-term trends in their evolution. This method has been applied to the analysis of a large database of JET IR images collected during the last campaigns of operation with the International Thermonuclear Experimental Reactor-like wall. The evolution of the spots and their properties, such as size and distribution, can be correlated with macroscopic events, in particular series of intentional disruptions. On the other hand, care must be taken in the interpretation of the results because there is some evidence of toroidal asymmetries in the spot distribution.

A new high frequency pellet injector, part of the JET programme in support of ITER, has been installed on JET at the end of 2007. Its main objective is the mitigation of the Edge Localized Modes (ELMs), responsible for unacceptable thermal loads on the wall when their amplitude is too high. The injector was also required to have the capability to inject pellets for plasma fuelling. To reach this double goal, the injector has to be capable to produce and accelerate either small pellets to trigger ELMs (pace making), allowing to control their frequency and thus their amplitude, or large pellets to fuel the plasma. Operational since the beginning of the 2009 JET experimental campaign, the injector, based on the screw extruder technology, suffered from a general degradation of its performance linked to extrusion instability. After modifications of the nozzle assembly, re-commissioning on plasma has been undertaken during the first half of 2012 and successful pellet ELM pacing was achieved, rising the intrinsic ELM frequency up to 4.5 times. © 2013 EURATOM.

The current configuration of JET with ITER-like Wall (ILW) is the best available proxy for the ITER first wall. Beryllium redistribution in JET-ILW can be used for estimates of its migration in ITE ...

Tungsten erosion in the outer divertor of the JET ITER like wall was quantified by spectroscopy. Effective sputtering yields of typically 10−4 were measured in L-mode at ∼30 eV attached divertor conditions and beryllium was identified as the main cause of sputtering. The signature of prompt redeposition was observed in the analysis of WI 400.9 nm and WII 364 nm line ratios and indicative of >50% redeposition fractions. Inter- and intra-ELM sputtering were compared for an example of 10 Hz ELMs with 13 MW NBI heating, in which intra-ELM sputtering was found to dominate by a factor of 5. Nitrogen seeding initially increased the tungsten sputtering threefold due to higher extrinsic impurity levels and effectively reduced the tungsten sputtering when the divertor plasma temperature was decreased from the initial 25 eV down to 15 eV.

The Active Gas Handling System (AGHS) at JET is a unique facility enabling JET to perform reactor like, DT operations. As a future DT experimental campaign (DTE2) is scheduled for 2017 this paper provides a brief overview of the AGHS and a summary of ongoing work supporting the currently JET experimental campaign. In order to improve tritium accountancy a solid state based detector for tritium is being developed. Another important upgrade concerns tritium injection, 4 existing GIMs (Tritium Gas Introduction Module) will inject a mix of D and T rather than T{sub 2} in the divertor region rather than in the torus mid plane enabling a far better control and variability of the introduction of tritium into the plasma. An overview of the scale of DTE2 is included as well as an example of some of the upgrades currently being undertaken to fully exploit the learning opportunities for ITER and DEMO DTE2 provides. (authors)

The aim of the Exhaust Detritiation System (EDS) of the JET Active Gas Handling System (AGHS) is to convert all Q-based species (Q{sub 2}, Q-hydrocarbons) into Q{sub 2}O (Q being indifferently H, D or T) which is then trapped on molecular sieve beds (MSB). Regenerating the saturated MSBs leads to the production of tritiated water which is stored in Briggs drums. An alternative disposal solution to offsite shipping, is to process the tritiated water onsite via the implementation of a Water Detritiation System (WDS) based, in part, on the combination of an electrolyser and a cryo-distillation (CD) facility. The CD system will separate a Q{sub 2} mixture into a de-tritiated hydrogen stream for safe release and a tritiated stream for further processing on existing AGHS subsystems. A sensitivity study of the Souers' model using the simulation program ProSimPlus (edited by ProSim S.A.) has then been undertaken in order to perform an optimised dimensioning of the cryo-distillation system in terms of available cooling technologies, cost of investment, cost of operations, process performance and safety. (authors)

We study tungsten prompt re-deposition processes at the divertor plates via kinetic modelling of the JET SOL for different divertor plasma parameters. Our simulations demonstrate that the electric field and not the Lorentz Force is the major contributor to the prompt re-deposition process. The fraction of tungsten ions escaping from the divertor plasma is defined by the number of tungsten atoms ionized outside the magnetic sheath and does not exceed 3% of the ions sputtered from the divertor surface. We derived the corresponding fit function for estimation of the re-deposition coefficient.

Fusion experiments with wall cladding of different chemical elements face the problem of material mixing by the plasma–wall interaction. Understanding of experiments that investigate this long term material re-distribution require an iterative approach that takes the local erosion, the global transport of material by the scrape-off layer (SOL) plasma and the local re-deposition into account. We present a model that combines parameterized DIVIMP calculations and a simplified sputtering model to simulate the time evolution of first wall material migration in a fusion experiment. The simulations are benchmarked by Be evaporation experiments performed previously at JET. The time evolution of the spectroscopically measured Be erosion fluxes for 800 s total plasma time at the main wall and inner divertor can be reproduced by the model using an extended calculation grid for DIVIMP and an increased cross-field diffusivity for impurity ions in the SOL. Discrepancies were found in the outer divertor region.

This paper presents an overview of the R&D activities for the ITER-like Wall (ILW) Project in JET which has been launched in 2005 and will be completed early 2011. A full replacement of the first wall materials in JET will be done to a ITER like wall composition with Be in the main chamber and W in the divertor as foreseen for the second activated phase in ITER. The project is directed to deliver answers to urgent questions on plasma surface interactions such as tritium retention and Be–W behaviour and to provide in general operational experience in steady state and transient conditions with ITER wall materials under relevant geometry and relevant plasma parameters.

An optical flow method is applied to the study of several fusion plasma relevant issues, including plasma wall interactions. A multi-resolution coarse-to-fine procedure is used in order to cope with large displacements of objects between consecutive frames, characteristic of plasma images captured by JET fast visible camera. Occlusion modeling is also implemented. The method is able to provide good results for JET fast visible camera images which can be affected by saturation, discontinuous movement, reshaping of image objects, low gray-level in-depth resolution. Significant experimental cases concerning pellet injection, plasma filaments and MARFEs are analysed. The method is able to provide the real velocity for objects moving close to structures.

Runaway electrons generated during ITER disruptions are of concern for the integrity of the plasma facing components. It is expected that a power of up to 8 GW is exposed to ITER PFCs. We present ill this article observations from JET and TEXTOR on the generation of runaways and the heat load deposition. Suppression techniques like massive gas injection and resonant magnetic perturbations are discussed. (C) 2009 Elsevier B.V. All rights reserved.

As a result of the R&D phase of the ITER-like Wall Project, combined magnetron sputtering and ion implantation (CMSII) technique was selected for 10 μm W coating of approx. 1000 CFC tiles for the new JET first wall. This technique involves simultaneous magnetron sputtering and high energy ion bombardment. A high voltage pulse discharge is superposed over the magnetron deposition and by this way positive ions are accelerated, bombarding initially the substrate and then the coating itself during its growth. Based on this method, industrial equipment with a deposition chamber of Φ800 mm × 750 mm and 24 magnetrons was designed, manufactured and commissioned. The coating productivity is about 1 m2/week. Tungsten coatings with a thickness of up to 17 μm and multilayer structures Mo/W/Mo/W with a thickness of ∼25 μm were produced and successfully tested at 100 pulses of 16.5 MW/m2 for 1.5 s.

This paper describes the activities of installation and commissioning of the magnetic diagnostic enhancement for JET. The new system includes four sub-sets of probes located in the vessel in two opposite octants, for a total of 48 magnetic pick-up coils. The main goal of this enhancement is to improve the accuracy of the reconstruction of the plasma equilibrium and the performance of the real time feedback control of the plasma shape. After a brief description of the system, the paper focuses on the installation activities, accomplished during the 2007 shut-down using the JET remote handling system. The paper then concentrates on the analysis and interpretation of the data collected during the functional commissioning of the new system, carried out during the restart phase of the machine and the first part of the 2008 experimental campaigns, which proved the accuracy of the new signals and the achievement of the project goals.

The main objectives of the JET ITER-like Wall Project are to provide a beryllium main wall and tungsten divertor with at least a 4 year lifetime to allow full evaluation of the materials and related plasma scenarios for ITER. Tungsten coatings will be used over most of the divertor area and this paper describes the latest developments in the coating technology and an analysis of the implications for the coating lifetime and machine operation. Both steady state and transient heat loads are assessed. (C) 2009 Published by Elsevier B.V.

Over the last 7–8 years, two quantitative analyzing methods—accelerator mass spectrometry (AMS) and full combustion (FC) followed by scintillation detection have been applied for determining the tritium activity concentrations in JET divertor tiles. These methods have two main differences – the range of detection and the spatial resolution – and are thus complementary. However, these differences can also complicate the comparison of the two techniques for typical JET divertor samples. Therefore a cross comparison exercise for tritium measurements was performed between the two methods using specially produced identical standard samples. The cross comparison measurements were performed firstly on the identical standard samples and were extended then to selected samples from the JET divertor tile (laser treated and non-treated samples from tile 14ING3B exposed in 2001–2004). The results obtained by AMS and FC agreed within an error limit of 10% for the standard samples and of 18% for the JET divertor samples. The results of this comparison study provided also useful information concerning the efficiency of tritium removal from the plasma facing surface of the divertor tiles by means of laser ablation.

In the last years, several diagnostic systems have been installed on Joint European Torus (JET) providing new information that may be potentially useful for disruption prediction. The fast visible camera can deliver information about the occurrence of multifaceted asymmetric radiation from the edge (MARFE) instabilities that precede disruptions in density limit discharges. Two image processing methods-the sparse image representation using overcomplete dictionaries and the Histogram of oriented gradients (HOGs)-have been used for developing MARFE classifiers with supervised learning. The methods have been tested with JET experimental data and a good prediction rate has been obtained. The HOG method is able to provide predictions useful for online disruption prediction.

In this paper numerical simulations with COREDIV code of JET discharges with ITER-like wall are presented. We concentrate on the JET H-mode and hybrid scenarios with nitrogen seeding and all simulations have been performed with the same transport model and only the discharge input parameters like auxiliary heating Paux, line average plasma density nline, confinement factor H98, nitrogen input flux ΓpuffN were changed in the calcu lations. The separatrix density nsepe is an input parameter in our model and has been kept equal to 0.4÷0.5nlinein the simulations, with the recycling coefficient adjusted accordingly. It has been shown that COREDIV is able to reproduce basic parameters of nitrogen seeded discharges for both H-mode and hybrid scenarios. We have achieved reasonable agreement with global plasma parameters like radiations levels, Zeff and tungsten concentrations and the plasma profiles, including density, temperature and radiation are in very good agreement with experimental data. The agreement of the code results with the experimental data might be even better, if the simulations are further tuned taking into account uncertainties to the sputtering model, separatrix density or SOL transport. Simulations show that the observed Zeff level is defined mostly by the low Z impurity content, Be and N2 in the considered shots. It has been found that the tungsten radiation plays always very important role and can not be mitigated even by strong influx of nitrogen. (© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

After the code parameters have been fixed by the numerical modeling of a well diagnosed JET pulse, the electron density and the input power have been changed, resulting in 4 density scans (〈ne〉 in the range 3.8 – 8.2 x 1019m-3) at Pin = 17, 22, 27, 32 MW. At any given power level, W flux decreases with increasing 〈ne〉 as a consequence of the decrease in Te at the target plates. Also the W concentration in the core (cW) decreases, but this not necessarily leads to reduced core radiation. Indeed, while at high Pin the core radiation decreases with density, at low Pin it increases. At high 〈ne〉 the increase in the input power leads to enhanced PradPrad, leaving, however, nearly unchanged the power radiated fraction frad Indeed, the increase in frad withPin is observed only at low 〈ne〉, up to a level of about frad = 0.4. These numerical results, linked to the non-linear self-consistent physics of W production and transport, suggest the best conditions are achieved when the level of the electron density is adapted to the level of the available Pin. (© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

The evolution in the past years of Machine learning techniques, as well as the technological evolution of computer architectures and operating systems, are enabling new approaches for complex problems in different areas of industry and research, where a classical approach is nonviable due to lack of knowledge of the problem's nature. A typical example of this situation is the prediction of plasma disruptions in Tokamak devices. This paper shows the implementation of a real time disruption predictor. The predictor is based on a support vector machine (SVM). The implementation was done under the MARTe framework on a six core x86 architecture. The system is connected in JET's Real time Data Network (RTDN). Online results show a high degree of successful predictions and a low rate of false alarms thus, confirming its usefulness in a disruption mitigation scheme. The implementation shows a low computational load, which in an immediate future will be exploited to increase the prediction's temporal resolution.

A new family of techniques, called conformal predictors, have very recently been developed to hedge the estimates of machine learning methods, by providing two parameters, credibility and confidence, which can assess the level of trust that can be attributed to their outputs. In this paper, the main steps required to extend this approach to fuzzy logic classifiers are reported. The more delicate aspect is the definition of an appropriate nonconformity score, which has to be based on the fuzzy membership function to preserve the specificities of Fuzzy Logic. Various examples of increasing complexity are introduced, to describe the main properties of fuzzy logic based conformal predictors and to compare their performance with alternative approaches. The obtained results are quite promising, since conformal predictors based on fuzzy classifiers outperform solutions based on the nearest neighbour in terms of ambiguity, robustness and interpretability.

International audience; The in vessel tritium inventory control is one of the most ITER challenging issues which has to be performed to fulfil safety requirements. This is due mainly to the presence of Carbon as a constituent of Plasma Facing Components (PFCs) which leads to a high fuel permanent retention. For several years now, physics studies and technological developments have been undertaken worldwide in order to develop reliable techniques which could be used in ITER severe environment (Magnetic field, vacuum, high temperature) for in situ tritium recovery. The scope of this presentation is to review the present status of these achievements and define the remaining work to be done in order to propose a dedicated work program.Different treatment techniques (chemical treatments, photonic cleaning) will be reviewed. In the frame of ITER, they will be compared in term of fuel removal efficiency as well surface accessibility, type of production (gas or particulates), ability to clean mixed material. Ad last, consequences of bulk trapping observed in tokamak on the techniques currently under development will be addressed.

Chemical erosion of carbon can be described as function of the incident ion flux, the ion energy and the surface temperature. Spectroscopy on the hydrocarbon break-up products CH and C 2 is applied to determine in-situ hydrocarbon fluxes and quantify the chemical erosion by means of hydrocarbon injection. Present-day knowledge on critical issues concerning the erosion yield and its determination is presented: Effective inverse photon efficiencies are measured in TEXTOR for different hydrocarbon species and compared with calculations from HYDKIN. The underlying database for the methane break-up, used in different erosion/deposition models, has been put to test. The chemical erosion yield related to higher hydrocarbons is determined to be lower in the JET outer divertor than measured in previous experiments. In ASDEX Upgrade and DIII-D first in-situ calibrations of hydrocarbon fluxes in the detached outer divertor are performed.

Analysis of the type-l ELM power load asymmetries using infra-red thermography and target current measurements is performed ASDEX Upgrade Upper Single Null and JET DOC-L type-I ELMy H-Mode discharges with 'normal' and,reversed' field direction, i.e. with the ion B x del B drift direction pointing towards the active X-point and the ion B x del B drift direction pointing away from the active X-point, respectively. The ELM power load towards the inner target plate is found to be larger as towards the outer target with 'normal' field direction and vice versa with 'reversed' field. Current measurements are performed in ASDEX Upgrade providing information that a net negative charge flows into the outer target and a net positive charge into the inner target during the ELM in 'normal' field and vice versa for discharges with 'reversed' field. The difference between the ELM energy load on the inner and outer target, E-outer - E-inner, is well correlated with the measured charge flowing through the targets due to the ELM. A comparison to JET data shows that in both devices the maximum asymmetry in energy load corresponds to values Of E-inner/E-outer approximate to 2. (c) 2007 Elsevier B.V. All rights reserved.

Équipe 107 : Physique des plasmas chauds; International audience; In the recent melt experiments with the JET tungsten divertor, we observe that the heat flux impacting on a leading edge is 3-10 times lower than a geometrical projection would predict. The surface temperature, tungsten vaporisation rate and melt motion measured during these experiments is consistent with the simulations using the MEMOS code, only if one applies the heat flux reduction. This unexpected observation is the result of our efforts to demonstrate that the tungsten lamella was melted by ELM induced transient heat loads only. This paper describes in details the measurements and data analysis method that led us to this strong conclusion. The reason for the reduced heat flux are yet to be clearly established and we provide some ideas to explore. Explaining the physics of this heat flux reduction would allow to understand whether it can be extrapolated to ITER.

The hydrogenic fuel retention in ITER will be dominated by co-deposition. Therefore predictions of retention in ITER require a global model of migration of low-Z (Be, C) impurities and their co-deposition with hydrogenic species. To track the global material erosion/deposition balance and the resulting formation of mixed material layers the WallDYN code has been developed. This paper describes the application of WallDYN to the interpretation of results on deposition and co-deposition in JET experiments both in the ITER like wall (ILW) and full carbon configuration. The calculations show qualitative agreement with the Be deposition patterns determined from post-campaign wall tile analysis. The calculated retention results for C and Be first wall configurations even show quantitative agreement with experimental gas balance measurements when long term outgassing is taken into account. Applying WallDYN to ITER for different first wall and plasma configurations shows that for the current (Be + W only) material choice, retention will not limit ITER operation whereas C would increase retention by factors 10–100, leading to unacceptably high fuel retention by co-deposition.

This article presents the development of a region of high electron density, being one order of magnitude larger than the separatrix density, in the inner divertor of ASDEX Upgrade (AUG) and JET. The high-field side high density (HFSHD) occurs about 10 × λ q away from the separatrix and about 20 cm above the X-point. To trigger the HFSHD, the inner divertor has to be detached while the outer one remains attached and the heating power has to be sufficiently high. The HFSHD, determined independently from spectroscopic and interferometric measurements, is clearly correlated with neutral fluxes in the far scrape-off layer; both increase with heating power. Injection of N2 into the divertor private flux region suppresses the HFSHD in AUG and JET. The density distribution in the divertor with and without N2 seeding is consistent with the total radiation and the D γ distribution.

With increasing size of a tokamak device and associated fusion power gain an increasing power flux density towards the divertor needs to be handled. A solution for handling this power flux is crucial for a safe and economic operation. Using purely geometric arguments in an ITER-like divertor this power flux can be reduced by approximately a factor 100. Based on a conservative extrapolation of current technology for an integrated engineering approach to remove power deposited on plasma facing components a further reduction of the power flux density via volumetric processes in the plasma by up to a factor of 50 is required. Our current ability to interpret existing power exhaust scenarios using numerical transport codes is analyzed and an operational scenario as a potential solution for ITER like divertors under high density and highly radiating reactor-relevant conditions is presented. Alternative concepts for risk mitigation as well as strategies for moving forward are outlined.