Triangle Nuclear Theory Colloquium - Bridging Nuclei and Stars: Constraining Nuclear Mass Models with the r-Process Observables

Presenter: Dr Menke Li - University of Notre Dame

Understanding the formation of heavy elements relies not only on extreme astrophysical environments but also on nuclear properties that are often unknown far from stability. Predicting these unmeasured properties with machine learning (ML) is a promising new approach. While ML models can reproduce known data well, they struggle in neutron-rich regions where experimental input is scarce. Yet, these regions are accessed in nature through the rapid neutron capture process (r-process)-the mechanism responsible for producing roughly half of the universe's heavy elements. In this talk, I'll show how r-process abundance patterns-observed in the Solar System and metal-poor stars-can be used to constrain ML-based nuclear mass models. By combining nuclear data with astrophysical observables through a multi-objective optimization framework, we refine mass predictions and deepen our understanding of the nuclei that shape the r-process.

Contact

Steffen Bass
bass@duke.edu