A. Hakola

Development of laser-based techniques for in situ characterization of the first wall in ITER and future fusion devices

Analysis and understanding of wall erosion, material transport and fuel retention are among the most important tasks for ITER and future devices, since these questions determine largely the lifetime and availability of the fusion reactor. These data are also of extreme value to improve the understanding and validate the models of the in vessel build-up of the T inventory in ITER and future D–T devices. So far, research in these areas is largely supported by post-mortem analysis of wall tiles. However, access to samples will be very much restricted in the next-generation devices (such as ITER, JT-60SA, W7-X, etc) with actively cooled plasma-facing components (PFC) and increasing duty cycle.This has motivated the development of methods to measure the deposition of material and retention of plasma fuel on the walls of fusion devices in situ, without removal of PFC samples. For this purpose, laser-based methods are the most promising candidates. Their feasibility has been assessed in a cooperative undertaking in various European associations under EFDA coordination. Different laser techniques have been explored both under laboratory and tokamak conditions with the emphasis to develop a conceptual design for a laser-based wall diagnostic which is integrated into an ITER port plug, aiming to characterize in situ relevant parts of the inner wall, the upper region of the inner divertor, part of the dome and the upper X-point region.

First results and surface analysis strategy for plasma-facing components after JET operation with the ITER-like wall

During the carbon wall operations of JET since 2001, an extensive post-mortem analysis programme has been carried out under the JET Task Force Fusion Technology and a similar analysis programme is underway for the JET-ILW tiles removed during the 2012 shutdown. The first post-mortem results from the JET ITER-like wall tiles have shown that the overall amount of deposition on the divertor tiles and on remote divertor areas has been reduced by more than an order of magnitude with respect to JET-C. In addition, the obtained data indicate a possible interaction between Be and W such as the formation of mixed Be–W layers. This could be due to the surface roughness of the tiles, or could be caused by diffusion or even alloying. Ion-beam analyses and secondary ion mass spectrometry techniques give only elemental information, so other techniques such as x-ray diffraction, x-ray photoelectron spectroscopy, secondary electron microscopy/energy dispersive x-ray spectroscopy and nuclear microprobing are required. Since the nature of deposition and erosion has changed during the JET-ILW operations, a change in the post-mortem analysis programme is needed. For example, no cross-sectional samples from the sloping parts of tiles 4 and 6 are required. A strategy for post-mortem analyses of the marker-coated tiles will be presented in this paper.

Overview of JET post-mortem results following the 2007–9 operational period, and comparisons with previous campaigns

In 2010, all the plasma-facing components were removed from JET so that the carbon-based surfaces could be replaced with beryllium (Be) or tungsten as part of the ITER-like wall (ILW) project. This gives unprecedented opportunities for post-mortem analyses of these plasma-facing surfaces; this paper reviews the data obtained so far and relates the information to studies of tiles removed during previous JET shutdowns. The general pattern of erosion/deposition at the JET divertor has been maintained, with deposition of impurities in the scrape-off layer (SOL) at the inner divertor and preferential removal of carbon and transport into the corner. However, the remaining films in the SOL contain very high Be/C ratios at the surface. The first measurements of erosion using a tile profiler have been completed, with up to 200 microns erosion being recorded at points on the inner wall guard limiters.

The effect of non-axisymmetric wall geometry on 13C transport in ASDEX Upgrade

We present the first results of 3D simulations of global 13C transport in ASDEX Upgrade (AUG) indicating that the deposition profile of 13C exhibits toroidal asymmetry in the main chamber.In 2007, the migration of carbon in AUG was studied with a methane (13CH4) injection experiment (A. Hakola et al and the ASDEX Upgrade Team 2010 Plasma Phys. Control. Fusion 52 065006). The total amount of deposited 13C was estimated by assuming toroidally symmetric deposition. Remarkably, the total number of deposited atoms was observed to be less than 10% of the number of injected atoms.The experiment has been simulated with the 3D orbit-following Monte Carlo code ASCOT using both a realistic 3D wall geometry of AUG and a 3D magnetic field with toroidal ripple. The simulations indicate that the non-axisymmetric wall geometry causes notable toroidal asymmetry in the deposition profile in the outer (low-field side) midplane region which can provide a partial explanation for the missing carbon inferred from post-mortem analysis of 13C deposition.

Migration and deposition of 13C in the full-tungsten ASDEX Upgrade tokamak

The migration of carbon in low-density, low-confinement plasmas of ASDEX Upgrade was studied by injecting 13C into the main chamber of the torus at the end of the 2007 experimental campaign. A selection of standard tungsten-coated lower-divertor and main-chamber tiles as well as a complete set of lower-divertor tiles with an uncoated poloidal marker stripe were removed from one poloidal cross section and analysed using secondary ion mass spectrometry. The poloidal deposition profiles of 13C on both the tungsten-coated tiles and on the uncoated graphite areas of the marker tiles were measured and compared. For the W-coated lower-divertor tiles, 13C was deposited mainly on the high-field side tiles, while barely detectable amounts of 13C were observed on low-field side samples. In contrast, on the uncoated marker stripes the deposition was equally pronounced in the high-field and low-field side divertor. The marker-tile results are in agreement with those obtained from graphite tiles after the 2003 and 2005 13C experiments in ASDEX Upgrade. In the case of W-coated tiles, the 13C measurements were complemented by determining the total amount of deposited carbon (12C) on the tiles, which also shows strong deposition at the inner parts of the lower divertor. The estimated deposition of 13C on W at the divertor areas was less than 1.5% of the injected amount of 13C atoms. The 13C analyses of the main-chamber tiles and small silicon samples mounted in remote areas revealed significant deposition in the upper divertor, in upper parts of the heat shield, in the limiter region close to the injection valve, and below the roof baffle. Approximately 8% of the injected 13C is estimated to have accumulated in these regions. Possible reasons for the different deposition patterns on W and on graphite in different regions of the torus are discussed.

Global migration of impurities in tokamaks

The migration of impurities in tokamaks has been studied with the help of tracer-injection (C-13 and N-15) experiments in JET and ASDEX Upgrade since 2001. We have identified a common pattern for t ...

Outer divertor of ASDEX Upgrade in low-density L-mode discharges in forward and reversed magnetic field: II. Analysis of local impurity migration

Part I (Aho-Mantila L. et al 2012 Nucl. Fusion 52 103006) presented a detailed analysis of outer divertor plasma conditions in low-density L-mode discharges in ASDEX Upgrade. In this paper, we analyse the local migration of carbon that originates from 13CH4 injected into these plasmas from the vertical outer target. Notable changes are observed in the local carbon deposition patterns when reversing the magnetic field in the experiments. Kinetic impurity-following simulations are performed using the 3D ERO code package with 2D background plasma solutions calculated with the SOLPS5.0 code package. The modelling shows that the measured changes are due to the changes in plasma collisionality, dissociation and ionization rates, and E × B drift of the impurities. These conditions affect the direction and rate of impurity migration inside and out of the divertor, having wider consequences on the global migration of impurities in a divertor tokamak. It is further shown that the migration pathways are largely determined by carbon ions and, hence, relevant for impurities in general. Neutral carbon and hydrocarbons are deposited only in the near vicinity of the injection, where they affect the local re-deposition efficiency. In this limited region, a perturbation of the local plasma conditions by the methane puff appears likely, yielding a significant uncertainty for interpreting the deposition efficiencies. The local deposition is largely influenced by the magnetic presheath electric field, the structure of which is the main uncertainty in the SOLPS5.0-ERO simulations.

Deposition of 13C tracer in the JET MkII-HD divertor

Migration of 13C has been investigated at JET by injecting 13C-labelled methane at the outer divertor base at the end of the 2009 campaign. The 13C deposition profiles on carbon fibre composite divertor tiles were measured by secondary ion mass spectrometry and Rutherford backscattering techniques. 13C was mainly deposited near the puffing location on the outer divertor base tiles. High amounts of 13C were also found at the outer vertical target: at the bottom of the lower and at the top of the upper plates. Thirty-three percent of puffed 13CH4 was instantly pumped out by the divertor cryopump, which is close to the pump duct entrance. Global 13C transport in the torus was modelled by the EDGE2D/EIRENE and DIVIMP codes, and local 13C migration in the vicinity of the injection location by the ERO code. The DIVIMP and EDGE2D simulations show strong prompt deposition of 13C directly adjacent to the injection point as well as in the far scrape-off layer (SOL) along both the inner and outer divertor targets. In addition, the measured 13C deposition along the outer divertor wall tiles is qualitatively reproduced. However, EDGE2D and DIVIMP do not predict any deposition along the divertor surfaces facing the private plasma on the inner floor tile and inboard of the outer strike point on tile 5. The ERO calculations also indicate that most of the deposition occurs close to the injection location on the vertical face of the LBSRP tile and the horizontal part of tile 6.

Deuterium trapping and release in JET ITER-like wall divertor tiles

A selected set of samples from JET-ILW divertor tiles exposed in 2011-2012 has been analysed using thermal desorption spectrometry (TDS). The highest amount of deuterium was found on the regions wi ...

Development of laser-induced breakdown spectroscopy for analyzing deposited layers in ITER

Laser-induced breakdown spectroscopy (LIBS) experiments on different ITER-relevant beryllium–tungsten mixtures were performed to develop LIBS for tokamak applications. Some of the samples were doped or implanted with deuterium to simulate fuel retention. The results indicate that beryllium and tungsten have a number of distinguishable spectral lines for diagnostics purposes, but detection of deuterium is challenged by its low concentration and low fluences used. By studying the depth profiles of the coatings, the removal rates of the layers were observed to depend on the thickness, composition and deuterium content of the coating. Quantitative studies of the compositions were made by calibration-free LIBS, and the results agreed well with those given by ion-beam methods.