A hybrid rate theory model of radiation-induced growth

Differences in the diffusivity of radiation-produced defects along zirconium’s prismatic and basal planes result in radiation-induced growth (RIG) and creep. The difference in anisotropy of diffusion (DAD) model explains this behavior by assuming vacancies and self-interstitial atoms (SIAs) possess dissimilar ratios of in-basal-plane vs out-basal-plane diffusion coefficients. The self-interstitial cluster diffusion (SICD) model was proposed as an alternative to DAD, and relates growth to the formation of self-interstitial clusters which diffuse solely along the 〈 a 〉 direction. Here, we propose a hybrid model, in which the contribution of both effects are considered, which provides additional flexibility as compared to DAD or SICD taken individually. In this model, both anisotropic diffusing point defects and 1-dimensional (1-D) glissile SIA clusters are considered, which are removed in the internal and boundary sinks. Thirty-two DAD, SICD and hybrid models are fitted to a set of thirty-two RIG experiments. Good agreement between DAD and the experiments can be reached only by assuming unrealistically large SIA bias factors. In 9 of the 32 scenarios, good agreement between SICD and experiments could not be reached, indicating that the model lacks flexibility. Good agreement between the hybrid model and the experiments can be reached using realistic SIA bias factors; however, it requires assuming considerably lower vacancy bias factors than those found by ab initio calculations aimed at mono-vacancies, suggesting small vacancy clusters and loops play a more important role in RIG than previously thought.