November 12, 2024
Soongsil University
Asia/Seoul timezone

Machine learning for nuclear masses in deformed relativistic Hartree-Bogoliubov theory in continuum

Nov 12, 2024, 2:20 PM
40m
Dasom Hall, Computer Science Institute Building, (Soongsil University)

Dasom Hall, Computer Science Institute Building,

Soongsil University

369, Sangdo-ro, Dongjak-gu, Seoul, Republic of Korea

Speaker

Dr Soonchul Choi (IBS)

Description

Most nuclei are deformed and deformations play an important role in various nuclear and astro-physical phenomena. Modern microscopic nuclear mass models have been/are being developed based on the covariant density functional theory to explore exotic nuclear properties. Among them we adopt the mass models based on the relativistic continuum Hartree-Bogoliubov theory (RCHB) with spherical symmetry and deformed relativistic Hartree-Bogoliubov theory in continuum (DRHBc) with axial symmetry to study possible effects of deformation on the rapid neutron-capture process
abundances. Since the DRHBc mass table is so far finished only for even-Z nuclei, we first study if a Deep Neutral Network (DNN) can be of use to complete the DRHBc mass table focusing on nuclear binding energies. To include information about odd-odd and odd-even isotopes to the DNN we also use the binding energies in AME2020 as a training set in addition to those of even-Z nuclei from the DRHBc mass table. After we obtain a reasonable mass table through a DNN study, we perform a sample sensitivity study of r-abundances to deformation or masses by using the RCHB⋆ and DRHBc⋆ mass tables. Here, ⋆ means the mass table is obtained by the DNN study. To see if such effects persist in various astrophysical sites of the r-process, we use the magnetohydrodynamic jets, collapsars and neutron star mergers. We find that the r-process abundances are highly sensitive to nuclear deformation, particularly in the mass region A= 80− 120.

Presentation materials

There are no materials yet.