Issued b y S andia Na tional Laboratories, o perated for the Un ited States Department of En ergy b y Sandia Corporation.
NOTICE:This report was prepared as an account of work sponsored by an agency of the United States G overnment. Neither th e Un ited S tates Government, n or a ny a gency th ereof, n or a ny o f their employees, n or a ny o f th eir contractors, subcontractors, or th eir employees, make any warranty, e xpress o r i mplied, o r a ssume a ny l egal li ability o r responsibility for th e a ccuracy, completeness, or us efulness of a ny i nformation, a pparatus, pr oduct, or process di sclosed, o r represent that its use would not infringe privately o wned rights. Reference herein to any specific commercial p roduct, p rocess, o r se rvice b y trad e n ame, trad emark, m anufacturer, o r o therwise, does not necessarily c onstitute or im ply it s e ndorsement, re commendation, o r f avoring b y th e United States Government, any agency thereof, or any of their contractors or subcontractors. The views and o pinions expressed herein do not n ecessarily state or reflect those o f th e United States Government, any agency thereof, or any of their contractors.
AbstractThe heightened focus on nuclear safeguards and accountability has increased the need to develop and verify simulation tools for modeling these applications. The ability to accurately simulate safeguards techniques, such as neutron multiplicity counting, aids in the design and development of future systems. This work focuses on validating the ability of the Monte Carlo code MCNPX-PoliMi to reproduce measured neutron multiplicity results for a highly multiplicative sample. The benchmark experiment for this validation consists of a 4.5-kg sphere of plutonium metal that was moderated by various thicknesses of polyethylene. The detector system was the nPod, which contains a bank of 15 3He detectors. Simulations of the experiments were compared to the actual measurements and several sources of potential bias in the simulation were evaluated. The analysis included the effects of detector dead time, sourcedetector distance, density, and adjustments made to the value of ν-bar in the data libraries. Based on this analysis it was observed that a 1.14% decrease in the evaluated value of ν-bar for 239Pu in the ENDF-VII library substantially improved the accuracy of the simulation.
4