Loading...
Loading...
Contributors:
Dataset URI:
10.6084/m9.figshare.32648526.v1this is the dataset of the paper:Tensile degradation and defect evolution in helium-bubble-containing palladium: high-throughput molecular dynamics and physics-informed machine learning. ABSTRACT:Helium bubbles are a major microstructural cause of mechanical degradation in metallic materials exposed to fusion and tritium storage environments. In palla dium, a representative tritium storage material, helium produced by tritium decay can aggregate into nanoscale bubbles, inducing local stress concentrations, dislo cation activity, and plastic instability. Here, high-throughput molecular dynamics simulations and physics-informed machine learning are combined to investigate the tensile response and defect evolution of helium-bubble-containing palladium under the coupled effects of temperature, strain rate, bubble radius, and helium to-vacancy ratio (He/V). Increasing temperature markedly reduces the peak stress, whereas larger helium bubbles and higher helium-to-vacancy ratios intensify local structural instability and promote defect propagation. Higher strain rates pro duce pronounced strain-rate strengthening. Atomistic analyses further show that helium bubbles promote local dislocation nucleation, vacancy aggregation, and stacking fault tetrahedron formation, thereby driving the transition from elastic load bearing to plastic damage. A physics-informed support vector regression model, incorporating Johnson–Cook-type constitutive trends, is then developed to predict peak stress across complex conditions. Compared with conventional support vector regression, which shows poor extrapolation performance (R2 = −2.200, root-mean-square error = 2.425 GPa), the physics-informed support vector regres sion model achieves substantially improved extrapolation stability and physical consistency (R2 = 0.910, root-mean-square error = 0.406 GPa). These findings clarify the coupled mechanisms of helium-bubble-induced degradation in palla dium and provide a physically constrained framework for limited-data property prediction in materials for nuclear environments.
Keyword(s):
helium bubbles; machine learning; Metals and alloy materials; molecular dynamics; Palladium.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)