Probing the limits of statistical neutron capture for the ,[object Object], process: Experimental constraints on ,[object Object],Cs nuclear level densities
B. Greaves (Department of Physics, University of Guelph, 50 Stone Rd. E, Guelph, N1G 2W1, Ontario, Canada)
; D. Mücher (Department of Physics, University of Guelph, 50 Stone Rd. E, Guelph, N1G 2W1, Ontario, Canada, Institut für Kernphysik, Universität zu Köln, Zülpicher Str. 77, Köln, 50937, Germany); A. Spyrou (Department of Physics and Astronomy, Michigan State University, East Lansing, 48824, MI, USA, Facility for Rare Isotope Beams/National Superconducting Laboratory, 640 S Shaw Ln., East Lansing, 48824, MI, USA); S. Goriely (Institut d’Astronomie et d’Astrophysique, CP-226, Universite Libre de Bruxelles, Brussels, 1050, Belgium); D. Rochman (Reactor Physics and Systems Behavior Laboratory, Paul Scherrer Institute, Forschungsstrasse 111, Villigen, 5232, Switzerland); et al - Show all 27 authors
The r-process abundance peaks, particularly near mass number A ∼ 130, reflect underlying nuclear structure effects such as closed neutron shells, yet modeling the nucleosynthesis in this region remains hindered by uncertain neutron-capture rates. These rates are especially sensitive to nuclear level densities (NLDs) and γ-ray strength functions of neutron-rich nuclei, where experimental data are scarce. We present the first experimental constraint on the NLD of 141Cs using the β-Oslo method, extending sensitivity to the neutron-rich regime near the closed shell. Our data allow for critical calibration of microscopic NLD models and reveal that 141Cs lies near the limit of statistical model applicability. Using this experimental input, we evaluate radiative neutron-capture rates across neighboring isotones using both Hauser–Feshbach (HF) and High Fidelity Resonance (HFR) models. Our results show order-of-magnitude rate increases for nuclei along the line, signaling a transition to resonance-dominated capture in this region. These findings underscore the importance of constraining NLDs to improve r-process reaction network predictions, particularly in environments where the validity of statistical models breaks down.
