First dust study in JET with the ITER-like wall: sampling, analysis and classification

Abstract

Results of the first dust survey in JET with the ITER-Like Wall (JET-ILW) are presented. The sampling was performed using adhesive stickers from the divertor tiles where the greatest material deposition was detected after the first JET-ILW campaign in 2011–2012. The emphasis was especially on sampling and analysis of metal particles (Be and W) with the aim to determine the composition, size, surface topography and internal dust structure using a large set of methods: high-resolution scanning and transmission electron microscopy, focused ion beam, electron diffraction and also wavelength and energy dispersive x-ray spectroscopy. The most important was the identification of beryllium dust both in the form of flakes and droplets with dimensions in the micrometer range. Tungsten, molybdenum, inconel constituents were identified along with many impurity species. The particles are categorised and the origin of the various constituents discussed.

Survival and in-vessel redistribution of beryllium droplets after ITER disruptions

“…Although numerical modelling of metallic melt layer dynamics under ITER-relevant conditions is available [8,[24][25][26], to our knowledge, no experimental validations currently exist regarding the size, speed and angle distributions of the Be droplets which may be ejected from the melt layers. Experimental observations of Be PFC exposure to VDEs in JET [27,28] suggest a droplet size of the order of a few μm, but the extrapolation of these results to ITER is uncertain due to the upscaling of plasma energy. More detailed experiments on W droplets in plasma guns report typical sizes of several tens of μm [29][30][31][32][33][34] and most probable ejection speeds of a few m s −1 , which are expected to increase by a factor 3 in the case of Be due to the main scaling with mass density [8,24,30,31,33,35].…”

Ion Microbeam Analyses of Dust Particles and Codeposits from JET with the ITER-Like Wall

“…All these contributions were taken together and are provided in Table 2. However, we also have to keep in mind that in the analysis we assumed that our dust particles are flat, thick and homogeneous, which is certainly only an approximation, as shown in the SEM analysis and in previous reports 6,10,11 ; however, this was justified with our Be and D peak shapes in the NRA simulations. In the cited reports, particles with rough irregular surfaces, fractures, voids and nanometer-sized W inclusions inside the material were observed.…”

“…Since K X-rays of O are overlapped with the L X-ray lines of Cr, characterization of O is not possible. The O in the samples could originate from plasma impurities and exposure to air during sampling and storage 10,58 . Our 3 He NRA measurements are also blind to O.…”

“…For that reason, comprehensive characterization of the wall components [7][8][9] and dust has been conducted in current fusion devices. A list of dust studies published until year 2011 mostly for carbon wall machines has been compiled by B. Braams 3 , whereas recent studies have involved JET-ILW 6,10,11 and ASDEX Upgrade 12 . The first dust study for JET-ILW was reported in 2015 by Baron-Wiechec et al 10 .…”

“…Sampling from the upper divertor region, which is the most critical area from a deposition point of view, showed the presence of two types of Be particles: flakes of co-deposits and small droplets. Both of these particle types are of considerable importance for ITER 10 . The composition and internal structure of the dust collected on carbon stickers from various regions of the divertor were studied after the second JET experimental campaign with the ILW (2013-14) and were reported by Fortuna-Zaleśna et al 6,11 .…”