Direct measurement of ${}^{59}\mathrm{Ni}(n,p){}^{59}\mathrm{Co}$ and ${}^{59}\mathrm{Ni}(n,\alpha ){}^{56}\mathrm{Fe}$ at fast-neutron energies from 500 keV to 10 MeV

Nuclear reaction data for neutron induced reactions on unstable nuclei are critical for a wide range of applications spanning studies of nuclear astrophysics, nuclear reactor designs, and radiochemistry diagnostics. However, nuclear data evaluations of the reaction cross sections are largely based on calculations due to the difficulty in performing this class of measurements and the resulting lack of experimental data. For neutron induced charged particle reactions at fast neutron energies, at the MeV scale, these cross section predictions are predominately driven by statistical Hauser-Feshbach calculations. In this work, we present partial and total ${}^{59}\mathrm{Ni}(n,p)$ and ${}^{59}\mathrm{Ni}(n,\alpha )$ cross sections, measured directly with a radioactive ${}^{59}\mathrm{Ni}$ target, and compare the results to the present nuclear data evaluations. In addition, the results from this work are compared to a recent study of the ${}^{59}\mathrm{Ni}(n,xp)$ reaction cross section that was performed via an indirect surrogate ratio method. The expected energy trend of the cross section, based on the current work, is inconsistent with that of the surrogate work. This calls into question the reliability of that application of the surrogate ratio method and highlights the need for direct measurements on unstable nuclei, when feasible.

Figure 1

The current status of ${}^{59}\mathrm{Ni}(n,p)$ (bottom panel) and ${}^{59}\mathrm{Ni}(n,\alpha )$ (top panel) cross sections are shown in comparison to evaluated data libraries [7, 8, 9] and a limited selection of experimental data [12, 18]. At fast-neutron energies, for which there are no experimental data available from studying the reactions directly, the evaluations are very discrepant.

Figure 2

The correlation between energy and time of flight is used to identify particular reaction channels for the (top panels) ${}^6\mathrm{LiF}+{}^{59}\mathrm{Ni}$ target, (middle panels) the ${}^{59}\mathrm{Ni}$ target, and the (bottom panels) Pt background target. The contributions due to frame-overlap, owing to the $1.8\mu \mathrm{s}$ LANSCE proton beam pulse-spacing, is most evident in the top panels due to ${}^6\mathrm{Li}(n,t)\alpha$ reactions occurring at low neutron energies.

Figure 3

The correlation between energy and time of flight for the LANSCE proton beam pulse-spacing set to $3.6\mu \mathrm{s}$ for the (top panels) ${}^6\mathrm{LiF}+{}^{59}\mathrm{Ni}$ target and the (bottom panels) ${}^{59}\mathrm{Ni}$ target.

Figure 4

Projections of the reconstructed reaction $Q$ value for the ${}^6\mathrm{Li}$ target, assuming the detection of a triton, at different neutron energy bins (left-right panels) and two different angles (top-bottom panels). The background contributions due to frame overlap are highlighted in the top-left panel.

Figure 5

Projections of the reconstructed reaction $Q$ value for the ${}^{59}\mathrm{Ni}$ target, assuming an outgoing proton and ${}^{59}\mathrm{Co}$ residual nucleus, at different neutron energy bins (top-bottom) and angle ranges (left-right). Rise-time pulse shape discrimination is used to discriminate between the $(n,p)$ and $(n,\alpha )$ reaction channels as illustrated by the “proton-gated” spectrum. Here the dominant peak at 1.8 MeV is from the ${}^{59}\mathrm{Ni}(n,p_0){}^{59}\mathrm{Co}_{gs}$ reaction channel. The vertical dashed lines are shown to illustrate the number of excited levels of ${}^{59}\mathrm{Co}$ that are populated up to a particular excitation cutoff, as indicated by the adjacent label.

Figure 6

MCNP simulation of the expected charged particle spectra measured in the upstream-most silicon detector. The simulation has been modified to use an input data library based on the work of Ref. [33], built on ENDF/B-VIII.0 [7], with improved outgoing charged particle spectra that includes the discrete population of the $(n,p)$ and $(n,\alpha )$ reactions leading to the ground state and excited states of ${}^{59}\mathrm{Co}$ and ${}^{56}\mathrm{Fe}$.

Figure 7

Top panel: Partial cross section for the ${}^{59}\mathrm{Ni}(n,p_0){}^{59}\mathrm{Co}_{gs}$ and the summed partial cross section for the excited states, up to an excitation energy of 2.35 MeV. Based on the RIPL database that is used in the calculations, this includes up to the 12th excited state of ${}^{59}\mathrm{Co}$. Bottom panel: The summed partial cross section of the ground state and first three excited states of ${}^{56}\mathrm{Fe}$ after being populated by the ${}^{59}\mathrm{Ni}(n,\alpha )$ reaction. The cross sections are compared to calculations using ${\mathrm{CoH}}_3$, with modified proton optical model and ${}^{59}\mathrm{Ni}$ level density parameters.

Figure 8

Integrated $(n,p)$ and $(n,\alpha )$ reaction cross sections are shown in the top and bottom panels, respectively. The results are shown in comparison to various evaluations [7, 8, 9].

Figure 9

Left: Past experimental data on the ${}^{59}\mathrm{Co}(p,n){}^{59}\mathrm{Ni}$ reaction, which is the inverse of the ${}^{59}\mathrm{Ni}(n,p){}^{59}\mathrm{Co}$ reaction, compared to ${\mathrm{CoH}}_3$ calculations with modified optical model parameters (see text). The dash-dotted green line shows the calculation (talys) using the optical model adjustment suggested by Pandey et al. [18] which performs worse in predicting the $(p,n)$ cross section below 8 MeV than the default parameters. By contrast, adjusting the same parameter in the opposite direction and by a smaller magnitude brings the calculation into better agreement with the $(p,n)$ data. The calculation that best reproduces the magnitude of the $(p,n)$ data also better reproduces the magnitude of the ${}^{59}\mathrm{Ni}(n,p_0)$ reaction measured in this work (right panel). At incident neutron energies around 2 MeV, the adjustment suggested by Pandey et al. leads to calculations that overpredict the measured cross section by a factor of 4–5.

Figure 10

Comparison of the integrated $(n,p)$ cross section (left), from this work, with the proton production cross section of Pandey et al. [18] that was determined using a surrogate ratio method. The expected energy trend of the calculations, after adjustment to match the magnitude of our experimental data, appears to be significantly inconsistent with the $(n,xp)$ data. In addition, our results are consistent within the $1\sigma$ bounds (orange band) of Refs. [9, 17], whereas the surrogate work is predominately outside these bounds. The curve labeled “Calculation Pandey2019” is adopted from the curve labeled “TALYS1.8 (enriched)” from Fig. 14 of Ref. [18]. For the $(n,\alpha )$ comparison (right), the same level density adjustment is shown based on the $(n,p)$ analysis. Here, no adjustment was made to the default alpha optical model parameters, whereas the proton optical model parameters were adjusted to reproduce the low energy $(n,p)$ cross sections.