Delayed Hydride Cracking in Irradiated Zircaloy

Sammanfattning: Under some circumstances nuclear fuel cladding tubes made from zirconium based alloys may develop long axial cracks. The formation of these cracks is mainly thought to be connected with the oxiditian and hydriding of the cladding which takes plage after the the formation of a small primary defect. One mechanism proposed to be responsible for the propagation of the axial cracks is delayed hydride cracking, DHC. DHC is a process where hydrogen diffuses upwards the tensile stress gradient tbat exisur in front of an existing crack or flaw. This cancentrates the hydrogen solved in the matrix to the area ahead of the growing crack. When ihe solubility limit is passed in front of the crack, hydrides are precipitated. The hydrides are assumed to be brittle in their behaviour at temperatums up to 300°ree;C. When a certain critical size is passed, the hydrides or hydride-package fracture in a brittle matmer if the lotal stress intensity leve1 is above a threshold value. allowing the crack to grow the distance of the hydridel hydride-package. The process then repeats itself at the new location of the crack tip. The aim of the thesis was to determine if regular BWR Zircaloy-2 cladding was susceptible to crack growth due to DHC or a mechanism similar to DHC in its axial direction. To enable testing on actual spent fuel cladding, a test tcchnique was developed and applied both to unirradiated and irradiated material. The specimen is similar to a normal centre cracked tension, CCT-, specimen. The test program has included investigations on the crack growth rates at 200" and 300°ree;C, the threshold stress intensity level, KIH below which no crack growth occurs and the intubation period bcfore cracking starts. The experimental work has focused on hydrogen tontents above 5OOppmH.In the unirradiated case the maximum crack growth was found to be in the vicinity of 6.10-7 m/s, while the irradiated case demonstrated crack growth rates close to 10-6 rn/s. The threshold stress intensity leve1 was found to be strongly dependent on the yield strength of the material. such that higher yield strength resulted in lower Km. The intubation period was found to be fairly constant, regardless of the hydrogen tontent and yield strength but dependent on the temperature at which the specitic experiment was conducted.The obtained crack growth rates indicates that the growth of long axial cracks in nuclear fuel cladding can be described by a mechanism similar to delayed hydride cracking at hydragen levels above 500 ppm lotally. Whether the mechanism of crack growth is DHC as described in the case high strength Zr-2.5 wt% Nb and low hydrogen tontents or a process similar to that can however not be verified experimentally sinte most of the evidente is indirect. The temperature dependence is consistent with an activation energy for crack growth close the tbeoretically derived value of 69.5 Idlmole for crack growth controlled by hydrogen diffusion in a stress induced potential gradient. Thus the crack growth can be described by an Arrhenius relationship in the steady state region with reference to applied K, stage II.