In situ transmission electron microscopy studies of 30 keV H2+ and He+ dual-beam irradiated Pd: defect generation and microstructural evolution under varied temperatures and foil thicknesses

To fully understand the tritium aging behavior of palladium (Pd), an important material applied in hydrogen isotope storage and tritium engineering, an in situ irradiation of 30 keV H-2(+) and He+ dual beam was performed for the investigation of the microstructural evolution in Pd. The irradiation dose, temperature, and sample thickness obviously affected the morphology, distribution characteristics, density, and size of dislocation loops, bubbles, and bubble-loop complexes. Highly dense loops were aggregating at lowered temperatures, whereas the loop density dropped dramatically, producing bubble-loop complexes at elevated temperatures. Most of loops were 1/3<111> Frank loops with quasi-circular under a low dose at 573 K and then transformed to 1/2<110> perfect loops with wavy and polygonal shape at a high dose. The difference in growth behavior between faulted and perfect loops suggested that there was a critical transform loop size similar to 32 nm in diameter. The heterogeneity in size and shape of bubbles could be attributed to the interaction between bubbles and the periphery loop. Moreover, the hydrogen resulted in the formation of larger bubbles when compared with the single-beam helium irradiation. The unfaulting transition of loops led to bubble redistribution. The corresponding mechanisms of in situ irradiation experimental results were discussed. (C) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This study investigated the microstructural evolution in palladium (Pd) through in situ irradiation experiments, revealing significant effects of temperature, irradiation dose, and sample thickness on the morphology and density of dislocation loops, bubbles, and bubble-loop complexes. Hydrogen resulted in the formation of larger bubbles, while the unfaulting transition of loops led to bubble redistribution. The difference in growth behavior between faulted and perfect loops was discussed, suggesting a critical transform loop size for Pd.