Diffusion bonding between tungsten base tiles and copper for fusion applications

Le résumé fourni par la source

In the R&D of plasma facing components fabrication, it is required to develop joining technologies between dissimilar materials resistant to strong thermal gradients. When, as in the divertor or in a few positions of the First Wall, the thermal loads exceed 1 MW/m2, the choice of the heat sink material is almost mandatory: copper alloys such as CuCrZr ensure the necessary high thermal conductivity and yield strength. The preferred plasma-facing material is currently Tungsten (W) (even in ITER, which in its previous design envisaged a Beryllium as plasma facing material for the First Wall, has been decided to replace with W to avoid the complications and additional costs related to the toxicity of Be). The so-called ITER-like monoblock design is the reference design for the divertor targets of many fusion devices, already in operation, such as WEST or KSTAR, or in the design phase, such as DTT and the future W JT60-SA divertor . The monoblock design has demonstrated great reliability even with cyclic thermal loads up to 25 MW/m2. However, the monoblock is not always a viable solution, due to complex geometries as the W7X divertor, or necessary, when the expected loads do not exceed 5 MW/m2 as for some ITER FW modules or for the Dome. In these cases, the plasma-facing surface in W can be obtained with a flat geometry tile joined on CuCrZr heat sink by a more or less thick layer of pure Cu. The work presented in this paper focuses on qualifying the Diffusion Bonding (DB) process to join W tiles to a Cu substrate. A number of samples were manufactured and the junction was verified by ultrasonic test non-destructive examination. Two samples were shear tested at room temperature up to the test system limit load without being detected by either visual or ultrasonic test (UT) inspection. Other samples were used to fabricate an actively cooled mock-up to be tested for thermal fatigue up to 5 MW/m2.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.