Desarrollo de aleaciones de wolframio autopasivantes para su aplicación en futuros reactores de fusión nuclear

Resumen

Tungsten is presently the main candidate material for the blanket first wall of DEMO. However, the use of W as first wall material implies an important safety concern in case of a loss-of-coolant accident with simultaneous air ingress into the reactor vessel. In such a situation, the high temperatures achieved in the in-vessel components within 10 to 30 days due to the decay heat would lead to fast W oxidation with the release of large amounts of volatile radioactive tungsten oxides. Besides, W oxidation is also a concern for the inner part of the thimble in the He-cooled modular divertor proposed by the EU, which can fail due to the formation of a thick oxide layer in a relatively short time due to the residual oxygen in the He coolant loop. A possible way for improving the oxidation resistance of W would be the addition of alloying elements forming stable oxides, so that at high temperature in the presence of oxygen a self-passivating layer is formed protecting the material from further oxidation. This idea was first proposed by the Max-Planck-Institute for Plasma Physics (IPP) in Garching, Germany, within the EU project ExtreMat (2004-2010). The results of this work demonstrated that thin film alloys of the system W-Cr-Si exhibit a three order of magnitude lower oxidation rate that pure W when exposed to air at temperatures up to 1000¢XC due to the formation of a stable Cr2O3 protective scale. However, the PVD route followed by IPP is not industrially relevant because for the first wall of DEMO coatings or tiles with a thickness of several mm are required, and such thicknesses are not feasible by this route. Thus, the powder metallurgical (PM) route was proposed, as alternative for the production of self-passivating W alloys. In this doctoral thesis the viability of the manufacturing of two different selfpassivating W alloy systems, W-Cr-Si and W-Cr-Ti, with ¿î¿n100% relative density, extremely fine microstructure and significantly reduced oxidation rate compared to pure W is shown. The oxidation mechanisms of the bulk alloys are slightly different than for thin films. In WCr10Si10 alloy, lower parabolic oxidation rates than thin films with similar composition are obtained, and the self-passivating layer consists of Cr2WO6 instead of Cr2O3, as observed in thin layers. On the other hand, in WCr12Ti2.5 alloy the parabolic oxidation rates are higher than for thin films, although the oxide layer formed at the surface has the same composition, Cr2O3. Furthermore, three bending point tests (performed at the Karlsruhe Institute of Technology KIT), revealed that both alloys exhibit brittle behavior up to high temperatures of the order of 900-1000 ¢XC.