PARALLEL PLATE COLLIMATOR OPTIMIZATION USING MCNP AUTOMATION

Year
2012
Author(s)
Graham V. Walford - The University of Tennessee
Sean J. Branney - Savannah River National Laboratory
Raymond Maynard - Savannah River National Laboratory
Jason Wilson - Savannah River Nuclear Solutions
Abstract
Parallel plate collimators can provide highly accurate source direction information, which may have application in certain safeguards applications such as hold up measurements in enrichment or reprocessing plants, or even gamma ray profiling of packages. The performance of such collimators is highly dependent on parameters such as plate material, depth, thickness, and spacing, source-detector distance and direction, as well as by characteristics of the radiation itself. All these aspects may be modeled using MCNP, but internally this program does not lend itself to automation. In order to find optimal collimator design specifications, an external program was used to automate the creation of over 200 MCNP design variations, which were then run separately. The resulting datasets were consequently collected and analyzed using Mathematica 8 to produce a range of values which allowed for precise characterization of collimator performance as a function of plate thickness, spacing, and source off-axis distance. In this paper, the general technique will be described, followed by the results for a NaI Detector + Collimator system of interest. This technique may be applied in general to other types of collimator designs as well. This work was supported by the Next Generation Safeguards Initiative (NGSI), Office of Nonproliferation and International Security (NIS), National Nuclear Security Administration (NNSA).