Quick Buzz Feed
Using a novel simulation model based on machine learning, an international research team at GSI/FAIR has succeeded in gaining a deeper understanding of element formation in stellar events such as neutron star mergers. For the first time, the scientists used deep learning with a neural network to model the energy release during r-process nucleosynthesis in hydrodynamic simulations. The results are published in the journal Physical Review D.
Many chemical elements originate from energetic stellar events like star explosions or neutron star mergers. These events generate immense energy, facilitating the rapid neutron-capture (r-process), where free neutrons are absorbed by existing nuclei, leading to the creation of heavier atomic nuclei.
Scientists worldwide aim to comprehend these intricate stellar reactions through theoretical simulations. However, modeling all necessary parameters demands vast computing power, often leading to simplified models. A new artificial intelligence-powered model, RHINE, offers an efficient alternative to overcome this computational barrier.
RHINE (r-process heating implementation in hydrodynamic simulations with neural networks) utilizes deep learning and neural networks to accurately describe the energy release from nuclear reactions within the r-process during hydrodynamic simulations. This 'heating' critically influences the dynamics and velocity of ejected material, impacting observable electromagnetic radiation such as kilonovae from neutron star mergers. The model was rigorously trained with extensive reference calculations and validated, demonstrating its ability to approximate heating rates with minimal effort and significant computing time savings, while emphasizing the importance of r-process heating.
The adoption of the RHINE model is expected to enable more detailed and precise simulations in the future. These advanced simulations could directly bridge the gap between experimental outcomes from the upcoming FAIR facility and actual astronomical observations of stellar explosions and neutron star mergers, providing a more comprehensive understanding of these cosmic phenomena.
The source code for the RHINE model has been made publicly accessible for wider use by the scientific community. The project itself received financial support from various entities, including co-funding from the European Research Council (ERC).