Evaluation of the 235 U resonance parameters to fit the standard recommended values

A great deal of effort has been dedicated to the revision of the standard values in connection with the neutron interaction for some actinides. While standard data compilation are available for decades nuclear data evaluations included in existing nuclear data libraries (ENDF, JEFF, JENDL, etc.) do not follow the standard recommended values. Indeed, the majority of evaluations for major actinides do not conform to the standards whatsoever. In particular, for the n + 235U interaction the only value in agreement with the standard is the thermal fission cross section. A resonance re-evaluation of the n + 235U interaction has been performed to address the issues regarding standard values in the energy range from 10−5 eV to 2250 eV. Recently, 235U fission cross-section measurements have been performed at the CERN Neutron Time-of-Flight facility (TOF), known as n TOF, in the energy range from 0.7 eV to 10 keV. The data were normalized according to the recommended standard of the fission integral in the energy range 7.8 eV to 11 eV. As a result, the n TOF averaged fission cross sections above 100 eV are in good agreement with the standard recommended values. The n TOF data were included in the 235U resonance analysis that was performed with the code SAMMY. In addition to the average standard values related to the fission cross section, standard thermal values for fission, capture, and elastic cross sections were also included in the evaluation. This paper presents the procedure used for re-evaluating the 235U resonance parameters including the recommended standard values as well as new cross section measurements.


Introduction
Prior to describing the issues with regard to the standards for 235 U it is worthwhile to present a succinct description of the 235 U evaluation work carried out over the last decades.In the later 1980s and early 1990s a 235 U Reich-Moore resonance evaluation was performed from thermal to 2.25 keV [1] using the code SAMMY [2].It was the first attempt to use a more rigorous resonance formalism to address issues with interference effects in the fission channels.The evaluation represented a huge improvement compared to previous 235 U evaluations for which the Single-Level Breit-Wigner (SLBW) formalism was used together with background cross sections to make up for the SLBW deficiency to represent fissile isotope.However, very little integral benchmark testing were carried out to assess the evaluation effectiveness prior its inclusion in the evaluated files.The evaluation was adopted in evaluated nuclear data projects and then underwent series of benchmark testing.The testing included sensitivity analysis and cross section adjustments based on benchmark experiments.The results demonstrated that the evaluation performed poorly mainly a e-mail: luiz.leal@irsn.frdue to concerns with the capture cross-section in the energy region 22.6 eV to 454 eV [3] indicating a need for increasing the capture cross section.However no issue with the fission cross-section was found.A close inspection of the problem revealed that a low value of the average gamma-capture width was responsible for the very low capture cross section.It should be pointed out that no reliable capture cross section measurements existed at the time in the energy range above 100 eV.There existed capture cross-section data but these data were not systematically included in the evaluation due to issues such as normalization and background.Hence the 235 U evaluation was revised on the basis of integral results and sensitivity analysis.The revised 235 U evaluation was made available and included in the ENDF, JEFF and JENDL libraries [4].The JENDL project adopted the evaluation up to 500 eV and used an unresolved resonance representation above 500 eV to help improving the results of fast critical assembly benchmark (FCA) [5].The revised evaluation gave a high capture cross-section that did not support the FCA benchmark results.A similar scenario observed with the FCA benchmark results were also indicated with the ICSBEP HEU-MET-INTER-006 benchmark series (ZEUS benchmarks) [6].These divergences with integral benchmark calculations prompted the proposal for a subgroup of the Working Party on International Nuclear Data (WPEC) to investigate the 235 U capture problem [7].The results of the WPEC criticality calculations showed an overestimation of the 235 U capture cross-section of about 10%.The WPEC recommendation was that rather than re-evaluating the 235 U resonance parameters based solely on integral benchmark results new capture cross-section measurements should be made to confirm the findings of the WPEC subgroup.Hence, TOF capture cross-section measurements were planned and performed independently at the Rensselaer Polytechnic Institute (RPI) [8] and at the Los Alamos National Laboratory (LANL) [9].These new measurements were used together with the computer code SAMMY to update the 235 U resonance parameters in the energy range from thermal to 2.25 keV.The results demonstrated an improvement in benchmark results.However, despite all efforts to address the capture issue in the resonance region the problem with the standard cross section still remained.Recently measured fission cross-section data carried out at the n TOF facility provided strong support to the fission standard values.Indeed, the normalization of the n TOF fission data in the energy range 7.8 eV to 11.0 eV supported the standard value in this energy range but also reinforced the standard averaged fission cross-section values [10] in the resonance region.Hence, the resonance parameter evaluation was revised with the inclusion of the n TOF experimental fission data.

235 U resonance evaluation in the energy range 10 −5 eV to 2.25 keV
In this section a description of the 235 U evaluation using the computer code SAMMY is presented including resonance analysis and external level determination.The experimental data base used in the evaluation is described with emphasis on new measurements added to the present evaluation.

External energy levels determination
Previous evaluations of the 235 U resonance parameters made use of several external resonance energies, namely 14 bound levels and 14 levels above 2.25 keV, which proved to be unnecessary to represent the interference effects in the resonance range 10 −5 eV to 2.25 keV.Actually, it was apparent that the issue in connection with the fitting of the standard fission cross section was intimately related to the external energy levels.The long-range interference effects inherent in the R-matrix methodology precluded finding a good fitting of the experimental fission data.It also had an impact on the elastic scattering cross section.The present evaluation contains five bound energy levels and five energy levels above 2.25 keV.Included in the bound levels there is a negative energy close to zero with a very tiny neutron width which is responsible for the bending of the energy dependence of the η(E) at low energy [11].The 10 external energy levels are listed in Table 1 in which for each the resonance energy E r , gamma width γ , neutron width n , two fission widths f1 and f2 and the spin and parity J π are shown.Negative signs associated with the fission partial widths f1 and f2 reflect the positive-

Experimental data base
The novel feature of the work described in this paper, in contrast to previous work on the 235 U resonance parameter evaluation, is the addition of new cross-section measurements done at LANL, RPI and at n TOF.Capture cross section measurements done at LANL and RPI were central to unveil issues with the capture cross section above 100 eV.The fission cross-section measurements carried out at n TOF supported the standard cross section values.Furthermore, the evaluation was done including high-resolution transmission, fission cross-section, and eta measurements that were accounted for in previous 235 U evaluations.The experimental data used in the resonance parameter evaluation are displayed in Table 2 in which TOF length is indicated by the letter L, thickness by n, and temperature by T. The Reich-Moore approach of the code SAMMY was used for fitting the data.A total of 3170 resonance levels were identified in the energy range 0-2250 eV to reproduce the experimental data.

235 U Resonance parameter evaluation
Before starting the fitting of the data shown on Table 2 a careful examination of the experimental conditions was done.Experimental resolution, normalization, background, multiple-scattering, data alignment, etc., were inspected to assure consistency with the data set.A sequential analysis of the data shown in Table 2 was carried out with the code SAMMY to achieve a reasonable fit of the data with an acceptable χ 2 .Not only the resonance parameters were let to vary but also normalization, resolution parameters, etc. were also searched.The two experimental η(E) values at the low energy were fitted with the SAMMY code.The η(E) shape  observed in the experimental is followed by the fitting of the data.The bending effect perceived is the result of the interference effects of the bound level placed at the energy −3.657 × 10 −5 eV with a very tiny neutron width.The fitting is displayed in Fig. 2.
The n TOF fission data were normalized based on the fission integral in the energy range 7.8 eV to 11 eV.A decision was made to not use the normalized data but rather to include the fission integral value in the fitting process.It was noted that, as a result of this procedure, the average values of the fission cross section related to the standard were straightforwardly fitted.The results of the fitting of the Harvey [13], Spencer [12], and the Paradela [17] are shown in Fig. 3.An examination of the transmission data of Harvey revealed an inconsistency with the remaining data set around the energy 4.25 eV.It was discovered that the issue was due to an impurity of 181 Ta present in the target sample.Although the data reduction was suitably done it appears that the 181 Ta impurity was not completely removed.
It is interesting to note that the fission integral value in the energy region 7.8 eV to 11.0 eV led to a normalization of the Weston [15,16] and Gwin [14] fission data of about 2%.The fitting of the Weston [15] and Paradela [17] is shown in Fig. 4 in the energy range 100 eV to 400 eV.The resolution of the n TOF data is excellent displaying the details of the Porter-Thomas like fluctuations not seeing in the Weston data in this energy range.
Two measured capture data at LANL and RPI were used in the resonance evaluation above 100 eV.The capture   data of Perez [21] were used below 200 eV.The three data sets are displayed in Fig. 5 together with the SAMMY fit in the energy range 100 eV to 200 eV.As can be seen the resolution of the RPI data is excellent for the use in the resonance analysis.The time-of-flight lengths for the three measurements are about the same with the main difference on the neutron burst width.In the energy range 100 eV to 200 eV the SAMMY fit was based on the Perez, Danon, and Jandel capture data.Above 200 eV the fitting relied mainly on the RPI data.

Thermal values, average values and the standard data
As previously indicated the main motivation for revising the 235 U resonance parameter evaluation is to address issues with the quoted standard values [10].The two major changes in the present evaluation relative to the thermal cross section standard (energy of 0.0253 eV) are the capture and scattering thermal cross section values.Previous 235 U resonance parameters evaluation replicates the thermal fission cross-section standard values.The revision of the external resonance values, bound energy levels and energies above 2250 eV, allowed a quick convergence to the standard thermal values and the average values for the fission cross section.The thermal values calculated with SAMMY and the new resonance parameters are shown in Table 3.It is also listed the fission integral in the energy range 7.8 eV to 11 eV.The results presented in Table 3 are in agreement with the standard values.The fission cross section data measurements performed at CERN, Paradela [17], were fundamental in reproducing the standard values.Furthermore, the standard average fission integral from 100 eV up to 2000 eV were also replicated with calculations using the revised resonance parameters.The average fission cross section values compared with the standard are shown in Table 4.

Conclusions
Reevaluation of the 235 U resonance parameter was carried out with the code SAMMY to address issues with thermal cross section standards and average fission cross-section values.A new fission cross section measurement done at CERN was the chiefly factor for obtaining a good agreement between calculated and experimental fission values.The new set of resonance parameters include less external energy levels and provide a better way of calculating the interference effects on the fission channels.Thermal result values are within the indicated standard uncertainties.The fission integral in the energy range 7.8 eV to 11 eV is well reproduced.The normalization of the n TOF fission data in the energy range 7.8 eV to 11.0 eV supported the standard values above 100 eV up to 2250 eV.The SAMMY fit of the data replicated the average values within the quoted uncertainties.The new set

Figure 1 .
Figure 1.External levels contribution on the scattering crosssection in the energy ranges 10 −5 eV to 2.25 keV.

Figure 2 .
Figure 2. Comparison of the experimental and calculated η(E) in the thermal energy range.

Figure 3 .
Figure 3.Comparison of the SAMMY fit of the experimental data.

Figure 4 .
Figure 4. SAMMY fitting the fission cross section in the 100 eV to 400 eV energy range.

Figure 5 .
Figure 5. SAMMY fitting of the fission cross section in the 100 eV to 200 eV energy range.

Table 1 .
Energy bound levels and energies above 2.25 keV.

Table 3 .
Standard values and resonance parameters results for 0.0253 eV.