arXiv subjects

N. C. Shu

Publications and source records attributed to N. C. Shu.

6 recordsLinked to original sources

arXiv2026

Physics Informed Bayesian Machine Learning of Sparse and Imperfect Nuclear Data

The prevailing data-driven machine learning has been plagued by the absence of physics knowledge and the scarcity of data. We implement the physics-model informed prior into Bayesian machine learning to evaluate the energy dependence of independent fission product yields, which are crucial for advanced nuclear energy applications but only sparse and imperfect experimental data are available. The informative prior is the posterior after learning the generated data from fission models. Furthermore, cumulative fission yields are included as a physical constraint via a conversion matrix to provide augmented energy dependence. Our work demonstrated a truly Bayesian machine learning by incorporating comprehensive physics knowledges as a powerful tool to exploit the sparse but expensive nuclear data.

nucl-th↗

arXiv2021

Bayesian Data Fusion of Imperfect Fission Yields for Augmented Evaluations

We demonstrate that Bayesian machine learning can be used to treat the vast amount of experimental fission data which are noisy, incomplete, discrepant, and correlated. As an example, the two-dimensional cumulative fission yields (CFY) of neutron-induced fission of $^{238}$U are evaluated with energy dependencies and uncertainty qualifications. For independent fission yields (IFY) with very few experimental data, the heterogeneous data fusion of CFY and IFY is employed to interpolate the energy dependence. This work shows that Bayesian data fusion can facilitate the further utilization of imperfect raw nuclear data.

nucl-th↗

arXiv2021

Bayesian evaluation of charge yields of fission fragments of 239U

Recent experiments [Phys. Rev. Lett. 123, 092503(2019); Phys. Rev. Lett. 118, 222501 (2017)] have made remarkable progress in measurements of the isotopic fission-fragment yields of the compound nucleus $^{239}$U, which is of great interests for fast-neutron reactors and for benchmarks of fission models. We apply the Bayesian neural network (BNN) approach to learn existing evaluated charge yields and infer the incomplete charge yields of $^{239}$U. We found the two-layer BNN is improved compared to the single-layer BNN for the overall performance. Our results support the normal charge yields of $^{239}$U around Sn and Mo isotopes. The role of odd-even effects in charge yields has also been studied.

nucl-th↗

arXiv2012

Determination of astrophysical 12N(p,g)13O reaction rate from the 2H(12N, 13O)n reaction and its astrophysical implications

The evolution of massive stars with very low-metallicities depends critically on the amount of CNO nuclides which they produce. The $^{12}$N($p$,\,$γ$)$^{13}$O reaction is an important branching point in the rap-processes, which are believed to be alternative paths to the slow 3$α$ process for producing CNO seed nuclei and thus could change the fate of massive stars. In the present work, the angular distribution of the $^2$H($^{12}$N,\,$^{13}$O)$n$ proton transfer reaction at $E_{\mathrm{c.m.}}$ = 8.4 MeV has been measured for the first time. Based on the Johnson-Soper approach, the square of the asymptotic normalization coefficient (ANC) for the virtual decay of $^{13}$O$_\mathrm{g.s.}$ $\rightarrow$ $^{12}$N + $p$ was extracted to be 3.92 $\pm$ 1.47 fm$^{-1}$ from the measured angular distribution and utilized to compute the direct component in the $^{12}$N($p$,\,$γ$)$^{13}$O reaction. The direct astrophysical S-factor at zero energy was then found to be 0.39 $\pm$ 0.15 keV b. By considering the direct capture into the ground state of $^{13}$O, the resonant capture via the first excited state of $^{13}$O and their interference, we determined the total astrophysical S-factors and rates of the $^{12}$N($p$,\,$γ$)$^{13}$O reaction. The new rate is two orders of magnitude slower than that from the REACLIB compilation. Our reaction network calculations with the present rate imply that $^{12}$N($p,\,γ$)$^{13}$O will only compete successfully with the $β^+$ decay of $^{12}$N at higher ($\sim$two orders of magnitude) densities than initially predicted.

nucl-ex↗

arXiv2012

New determination of the 13C(a, n)16O reaction rate and its influence on the s-process nucleosynthesis in AGB stars

We present a new measurement of the $α$-spectroscopic factor ($S_α$) and the asymptotic normalization coefficient (ANC) for the 6.356 MeV 1/2$^+$ subthreshold state of $^{17}$O through the $^{13}$C($^{11}$B, $^{7}$Li)$^{17}$O transfer reaction and we determine the $α$-width of this state. This is believed to have a strong effect on the rate of the $^{13}$C($α$, $n$)$^{16}$O reaction, the main neutron source for {\it slow} neutron captures (the $s$-process) in asymptotic giant branch (AGB) stars. Based on the new width we derive the astrophysical S-factor and the stellar rate of the $^{13}$C($α$, $n$)$^{16}$O reaction. At a temperature of 100 MK our rate is roughly two times larger than that by \citet{cau88} and two times smaller than that recommended by the NACRE compilation. We use the new rate and different rates available in the literature as input in simulations of AGB stars to study their influence on the abundances of selected $s$-process elements and isotopic ratios. There are no changes in the final results using the different rates for the $^{13}$C($α$, $n$)$^{16}$O reaction when the $^{13}$C burns completely in radiative conditions. When the $^{13}$C burns in convective conditions, as in stars of initial mass lower than $\sim$2 $M_\sun$ and in post-AGB stars, some changes are to be expected, e.g., of up to 25% for Pb in our models. These variations will have to be carefully analyzed when more accurate stellar mixing models and more precise observational constraints are available.

astro-ph.SR↗

arXiv2009

Asymptotic normalization coefficient from the 12C(7Li,6He)13N reaction and the astrophysical 12C(p,g)13N reaction rate

Angular distribution of the 12C(7Li,6He)13N reaction at E(7Li) = 44.0 MeV was measured at the HI-13 tandem accelerator of Beijing, China. Asymptotic normalization coefficient (ANC) of 13N = 12C + p was derived to be 1.64 $\pm$ 0.11 fm$^{-1/2}$ through distorted wave Born approximation (DWBA) analysis. The ANC was then used to deduce the astrophysical $S(E)$ factors and reaction rates for direct capture in 12C(p,g)13N at energies of astrophysical relevance.

nucl-ex↗