Faire avancer la sûreté nucléaire

La Recherchev2


A new model on radial hydride precipitation in Zircaloy-4 claddings under decaying stress and temperature transient



Email :

Mot de passe :

International Conference on Management of Spent Fuel from Nuclear Power Reactors - An Integrated Approach to the Back-End of the Fuel Cycle / Vienne (Autriche), juin 2015

Type de document > *Congrès/colloque

Mots clés >

Unité de recherche > IRSN/PSN-RES/SEREX/LE2M


Date de publication > 15/06/2015


​In the context of spent fuel transportation and storage studies, the radial hydride precipitation is considered as a important issue for SRA Zircaloy-4 claddings. The fuel residual thermal power induces a temperature increase within the shipping cask cavity. As a consequence, the zirconium-hydrides contained within irradiated fuel claddings dissolve as solid-solution hydrogen. After transportation and during unloading in the spent fuel storage pool, the fuel rods are cooled down under internal gaseous pressure inducing mainly hoop stress within the cladding. At low hoop stress level, the precipitated hydrides are platelets oriented in the hoop and axial direction. When the hoop stress exceeds a threshold value, the precipitated hydrides rather develop along the radial-axial direction. These so-called radial hydrides strongly embrittle the cladding.

Laboratory experimental studies were performed to characterize the degree of radial hydride precipitation at room temperature under constant applied stress after a slow cool-down. The model is successfully compared to reference experimental data.

However, considering fuel assemblies transportation, the internal pressurization induced cladding stress decreases during the fuel cool down. There is a need for differential models providing step by step prediction of the radial hydride precipitation during the fuel cool down. A new modeling, describing the influence of an applied stress along with a temperature transient, has been consequently developped and compared to experimental data.