The Monte-Carlo code ERO2.0 was used to simulate steady-state erosion and transport of beryllium (Be) in the ITER main chamber. Various plasma scenarios were tested, including a variation of the main species (hydrogen, deuterium, helium), plasma conditions (density, temperature, flow velocity) and magnetic configurations. The study provides valuable predictions for the Be transport to the divertor, where it is expected to be an important contributor to dust formation and fuel retention due to build-up of co-deposited layers. The Be gross and net erosion rates provided by this study can help identifying first wall regions with potentially critical armour lifetime.

A sequence of fuel recovery methods was tested in JET, equipped with the ITER-like beryllium main chamber wall and tungsten divertor, to reduce the plasma deuterium concentration to less than 1% in preparation for operation with tritium. This was also a key activity with regard to refining the clean-up strategy to be implemented at the end of the 2nd DT campaign in JET (DTE2) and to assess the tools that are envisaged to mitigate the tritium inventory build-up in ITER. The sequence began with 4 days of main chamber baking at 320 °C, followed by a further 4 days in which Ion Cyclotron Wall Conditioning (ICWC) and Glow Discharge Conditioning (GDC) were applied with hydrogen fuelling, still at 320 °C, followed by more ICWC while the vessel cooled gradually from 320 °C to 225 °C on the 4th day. While baking alone is very efficient at recovering fuel from the main chamber, the ICWC and GDC sessions at 320 °C still removed slightly higher amounts of fuel than found previously in isotopic changeover experiments at 200 °C in JET. Finally, GDC and ICWC are found to have similar removal efficiency per unit of discharge energy. The baking week with ICWC and GDC was followed by plasma discharges to remove deposited fuel from the divertor. Raising the inner divertor strike point up to the uppermost accessible point allowed local heating of the surfaces to at least 800 °C for the duration of this discharge configuration (typically 18 s), according to infra-red thermography measurements. In laboratory thermal desorption measurements, maintaining this temperature level for several minutes depletes thick co-deposit samples of fuel. The fuel removal by 14 diverted plasma discharges is analysed, of which 9, for 160 s in total, with raised inner strike point. The initial D content in these discharges started at the low value of 3%–5%, due to the preceding baking and conditioning sequence, and reduced further to 1%, depending on the applied configuration, thus meeting the experimental target.

A tomographic method for reconstructing the axially symmetric D _α emission profile in the ITER scrape-off layer and divertor from D _α intensity measurements in the fields of view (FoV) of Vis/IR TV and Divertor Impurity Monitor diagnostics is under development. The method takes into account the strong background signal due to stray light from divertor produced by reflections from the metal wall of the vacuum chamber. The method allows to filter out the reflections in D _α signals measured in the main chamber. The accuracy of this method is tested within the framework of a synthetic diagnostic, which uses the Raysect and Cherab numerical codes for ray-tracing simulations. The D _α emission profiles, simulated with the SOLPS and OEDGE codes, are used as synthetic experimental data. Uncertainty in light reflection properties of the first wall (FW) is the major factor affecting the accuracy of the D _α emission reconstruction. The dependence on the uncertainty in the FW surface roughness and on the ratio of diffusive to specular reflection of the error in recovering useful D _α signals (the direct light without reflections) in the FoV of the ITER main chamber H-alpha and Visible Spectroscopy Diagnostic is obtained.

A wide class of non-stationary superdiffusive transport on a uniform background with a power-law decay at large distances of the step-length probability distribution function (PDF) is shown to possess an approximate automodel solution. The solution for the Green's function is constructed using the scaling laws for the propagation front (relevant-to-superdiffusion average displacement) and asymptotic solutions far beyond and far in advance of the propagation front. These scaling laws are determined essentially by the long-free-path carriers (Lévy flights). The validity of the suggested automodel solution is proved by its comparison with numerical solutions in the one-dimensional (1D) case of the transport equation with a simple long-tailed PDF with various power-law exponents and in the 3D case of the Biberman–Holstein equation of the resonance radiation transfer for various (Doppler, Lorentz, Voigt and Holtsmark) spectral line shapes.