Benchmarking of tungsten steel joints for the plasma facing wall of DEMO Project ID: AWP17-ENR-FZJ-01 Provider: Euratom Period: Jan 01, 2017 - Dec 31, 2018 Recipient -- coordinator: Forschungszentrum Jülich GmbH -- Daniel Dorow-Gerspach Co-recepient(s) / PI(s): Institute of Plasma Physics, Czech Academy of Sciences, Prague -- Ing. Jiří MatÄ›jĂ­ÄŤek, Ph.D. Politecnico Torino -- Monica Ferraris Co-investigators from IPP M. VilĂ©mová, M. Janata, Z. KutĂ­lek, J. Veverka Aims of the project: Tungsten represents a viable solution for the manufacture of the plasma facing first wall components in future fusion devices (such as DEMO). Furthermore, reduced activation ferritic martensitic (RAFM) steels are candidate materials for supporting structures and breeding blankets. For the successful outcome of the joining between tungsten and steel, an appropriate interlayer must be inserted in order to limit the thermal mismatch between both materials. In several studies tungsten steel joints have been produced either by an interlayer or by functionally graded material (FGM). The used production methods were vacuum plasma spraying (VPS), atmospheric plasma spraying (APS), physical vapour deposition (PVD), spark plasma sintering (SPS), pulse plasma sintering (PPS), hot pressing and diffusion bonding. These preliminary experiments have shown the feasibility; however, their performance under thermal cycling still needs to be investigated more systematically. In the frame of this Enabling Research Program selected joining techniques shall be benchmarked in high heat flux tests at 2-3 MW/m². Therefore the same geometry and thermal conditions shall be used for all samples to ensure the best comparability of the joints. In particular, two innovative approaches - functionally graded interlayers between modified surfaces (e.g. using laser sculpturing or magnetron sputtering) and interlayers comprising expansion gaps - shall show their potential in comparison to existing techniques. Based on the past experience, the production of the joints will be optimized.