High heat flux testing of actively cooled graphite- and tungsten-armoured JT-60SA flat tile divertor mock-ups

• Actively cooled JT-60SA flat-tile mock-ups with tungsten and graphite armour tiles diffusion bonded to a TZM heat sink are presented. • The mock-up performance is assessed in the high heat flux test facility GLADIS, with special focus on qualifying the diffusion bonding technology. • Diffusion bonding works well to join tungsten and graphite tiles to TZM. However, a potentially decreasing heat removal capability of the TZM heat sink during cyclic high heat flux loading and detrimental deep-cracking of the tungsten tiles are uncovered. During the process of developing actively cooled divertor plasma-facing components for JT-60SA, a fusion experiment in Japan built within the framework of the “Broader Approach Agreement” between the European Union and Japan, three small-scale divertor mock-ups have been manufactured. The mock-ups follow the flat-tile design, i.e., have plasma-facing armour tiles joined to an actively cooled heat sink. One mock-up has tungsten (W) armour tiles, and two mock-ups have carbon (C) armour tiles, with the heat sink material being the molybdenum alloy TZM. The joining was realized via diffusion bonding with a titanium interlayer. In this work, the thermo-mechanical behaviour of the mock-ups is assessed with the high heat flux test facility GLADIS in order to qualify the joining technology. This includes screening tests up to a heat flux of 15 MW/m 2 and cyclic loading with a heat flux of 10 MW/m 2 for 10 s. During high heat flux testing, pyrometer and thermocouple temperature measurements, digital camera images, and thermographic imaging were used to monitor the mock-ups. In addition, comparative infrared thermography tests and visual characterizations before and after high heat flux testing were performed, including the preparation of cross-sections for scanning electron microscopy. Debonding of the armour tiles did not occur during high heat flux testing, indicating that the diffusion bonding process is suitable. However, this work outlines challenges like a potentially decreasing heat removal capability of the TZM heat sink during cyclic loading or the occurrence of detrimental deep cracking in the case of W armour tiles.