Progress report on research project [Parameters for calculation of nuclear reactions of relevance to non-energy nuclear applications (Reference Input Parameter Library: Phase III)]

Full text: RIPL-2 database was extensively used in the evaluations being prepared for the ENDF/B-VII library. As many as 69 totally new evaluations were prepared using EMPIRE-II system with RIPL-2 library using as default for all model parameters. Taking into account that about 25 nuclei are being involved in a single evaluations, and that there is a natural overlap of nuclides when an isotope chain is evaluated the total number of nuclides tested in these calculations was about 1000. It should be stressed that evaluation procedure implies more thorough level of testing than blind calculations across periodic table. In general, modification of the default parameters was necessary in order to reproduce experimental data but the initial values were quite good starting points and they were kept for the cases where no experimental cross sections were available. No critical errors (causing fatal error in the EMPIRE code) were encountered. The only annoying feature was discrete levels with undefined energy (+X) that caused problems in the ENDF-6 format and were distorting calculations. This problem has already been solved in the RIPL-3 database. A new evaluation of neutron resonances and associated quantities has been completed by Said Mughabghab. This analysis is a continuation of the well known BNL-325 compilation, of which the last edition was published in 1984, and takes into account all experimental data on neutron resonances published until 2005. Of interest to RIPL-3 will be average quantities such as neutron spacings for s- and p-waves and gamma widths. The initial comparison of neutron spacings shows that in a number of cases there are significant differences between RIPL-2 database and the new evaluation. The agreement is much better for the gamma widths. The new Atlas of neutron resonances has been submitted for publication and is expected to be available by March/April 2006. Once the data are public RIPL-3 should review discrepant cases and eventually update the current resonance segment. The FORTRAN module used to calculated EMPIRE-specific level densities has been extracted from the EMPIRE system and made into a stand-alone code. The EMPIRE-specific approach offers a number of advantages compared to the classic Gilbert-Cameron or Back-Shifted Fermi Gas models still preserving computational convenience of the closed-form approximation. Work is in progress on adding minimization routine in order to re-parameterize EMPIRE-specific level densities using RIPL- 3 recommended data including neutron resonance spacing, shell corrections and RIPL-3 methods. In particularly, a method to estimate the vibrational enhancement factor selected by the RIPL-3 participants along with table/systematics of vibrational states will be used in fitting. In addition, deviations from the rigid-body moments of inertia might be considered. The resulting level density parameters and associated systematics will be determined and provided along with the EMPIRE-specific level density FORTRAN module. (author)