Experimental Prompt Fission Neutron Spectra Comparisons for the 235 , 238 U( n , f ) and 239 Pu( n , f ) Reactions

,


Introduction
Prompt Fission Neutron Spectra (PFNS) measurements have been a focus of recent reviews [1], and renewed appreciation for their importance in evaluations and applications [2]. The Chi-Nu project at the Los Alamos Neutron Science Center (LANSCE) is a joint LANL-LLNL effort to accurately measure the PFNS of the major actinides, for incident neutron energies from 1 to 20 MeV. Particular emphasis in Chi-Nu has been placed on understanding and reporting the experimental uncertainties and their correlations, to best inform subsequent nuclear data evaluations.
The Chi-Nu project started out with a program to measure only the PFNS for neutron-induced fission of 239 Pu and 235 U. Both of these measurements have been completed and published [3,4]; these data are available in EX-FOR [5]. These data have also been used in recent evaluations [6,7]. The important isotope 238 U was added to the project later, and these data are almost ready for publication. An approximately 20 mg sample of 242 Pu was available at LANSCE, and following the construction of a suitable ionization chamber, data on its spontaneous fission PFNS were taken; to date, these data have only been used to investigate correlations among the fission neutrons and gamma-rays. PFNS measurements of 240 Pu have also been performed, both in-beam and out-of-beam looking at the spontaneous fission neutrons. 252 Cf(s f ) neutrons were used to validate the MCNP model of the experiment, using the ENDF/B-VII [8] PFNS for 252 Cf, which is taken from * e-mail: devlin@lanl.gov the evaluation done by Mannhart [9]. As a result, using Chi-Nu 252 Cf data to extract the PFNS of 252 Cf(s f ) would be at least partly circular, and such an analysis has not been done despite the availability of 252 Cf(s f ) data from Chi-Nu. 252 Cf(s f ) data have been used to investigate correlations among neutrons and gammas emitted in fission. The current status of these Chi-Nu PFNS measurements is shown in Table 1, with references to relevant publications where possible. In addition to these measurements, the Commissariat à l'Énergie Atomique (CEA) have performed PFNS measurements at Chi-Nu with fission chambers and a different data acquisition and analysis method, on 238 U(n, f ) [10], 239 Pu(n, f ) [11] and 235 U(n, f ); these measurements are not discussed further here. A comparison of the Chi-Nu and CEA 239 Pu(n, f ) PFNS measurements is discussed in detail in Ref. [12].
Ratios of prompt fission neutron spectra with the same equipment and methods allow for comparisons among various isotopes, particularly if the same techniques are used in each measurement. Some systematic uncertainties can then cancel out in the ratio. The ENDF/B-V.II [8] and V.III.0 [13] release publications, for example, highlight only two such ratios for the ratio of 239 Pu to that of 235 U, those of Sugimoto [14] and Lestone [15]. With the numerous Chi-Nu measurements becoming available, and given the similarity of these measurements in experimental and analysis methods, comparisons of the PFNS ratios for various different isotopes could provide some insight into prompt neutron emission in fission.

PFNS Ratios
Ratios of the measured PFNS at Chi-Nu for 239 Pu(n, f ) to that of 235 U(n, f ) as a function of incident neutron energy are shown in Figs. 1 -4. All of these figures display the two Chi-Nu results, one from each array, separately, and include a comparison with ENDF/B-V.III.0 for an appropriate energy range, including the uncertainty band in ENDF as a shaded area. Fig. 1 covers the incident neutron energy range of 1-2 MeV, and includes comparisons with the prior data of M. Sugimoto, et al. [14] and J.P. Lestone and E.F. Shores [15]. Note that the Sugimoto data were taken at a mean energy of 0.55 MeV and transformed to 1.5 MeV for this type of comparison. Figure 2 shows 239 Pu(n, f )/ 235 U(n, f ) PFNS ratio for the incident neutron energy range of 6-7 MeV, where second-chance fission effects are expected in both isotopes. The ENDF/B-V.III.0 evaluation is largely consistent with these data, but at higher outgoing neutron energies the experimental ratio continues rising. This rise is not seen in the ENDF prediction, though it is more like the lower-energy ratio seen in Fig. 1.
At higher incident neutron energies, the Chi-Nu data continue this trend: Figs . Prior data of Sugimoto [14] and Lestone [15] are also shown, as is a comparison with ENDF/B-VIII.0 [13] with its uncertainty band (filled area). broadly agrees with the Chi-Nu data below 6 MeV in outgoing neutron energy. The uncertainties shown in these figures are calculated from the original data uncertainties, with the fully correlated systematic elements removed. This is a simplified treatment, as components to the systematic uncertainty are assumed to be either fully correlated or uncorrelated. A more complete uncertainty treatment would involve applying a full correlation matrix between all of the uncertainty components. This treatment has not been applied here, but is being pursued for future comparisons.

Mean Energy Comparisions
Another comparison among the PFNS results from various nuclei is to look at the average PFNS energy trend with increasing incident neutron energy. Figure 5 shows this comparison for the three isotopes ( 239 Pu, 235 U and 238 U) discussed previously. All three clearly show the expected  drop in mean energy at the onset of second-chance fission, ≈6 MeV. These trends are similar to those expected in evaluations; see Refs. [3,4] for more discussion in the 239 Pu and 235 U(n, f ) cases, respectively. At the onset of third-chance fission, ≈12 -14 MeV depending on the isotope, 235 U and 238 U mean energies display the drop expected in ENDF/B-V.III.0 [13], but 239 Pu mean energy does not. This apparent absence of much third-chance fission in 239 Pu has been discussed in Ref. [3]. The 238 U average energy data are still preliminary, particularly with respect to their over all normalization, and are plotted 0.2 MeV above their values to avoid overlap. The dip in the 238 U average energy at ≈4 MeV is neither expected nor explained, and could be a statistical fluctuation.

Summary
We have used the Chi-Nu measurements of prompt fission neutron energy spectra for neutron-induced fission of 239 Pu, 235 U and 238 U(n, f ) to investigate the ratios of   Figure 5. Mean PFNS energies as a function of incident neutron energies for 239 Pu, 235 U and 238 U(n, f ). The onset of second-fission are clearly seen in the all of these isotopes, but third-chance fission is clearly seen in the U isotopes and is only weakly seen in 239 Pu. Note that the 238 U data are preliminary, particularly with respect to their overall mean energy scale.
these results and to compare them to the expected values in the evaluated nuclear database ENDF/B-V.III.0. The respective mean PFNS energies as a function of incident neutron energies for these isotopes were also compared side-by-side, illustrating their similarity, until the lack of a strong third-chance fission mean energy drop in 239 Pu(n, f ) stands out in the comparison. Since the Chi-Nu project has produced similar PFNS data for spontaneous fission in some other isotopes, and is also measuring the prompt fission neutron spectra from 240 Pu(n, f ), further comparison should be available in the near future.