skip to main content


Title: Validation and uncertainty quantification of detector response functions for a 1"×2" NaI collimated detector intended for inverse radioisotope source mapping applications

Detector response functions (DRFs) are often used for inverse analysis. We compute the DRF of a sodium iodide (NaI) nuclear material holdup field detector using the code named g03 developed by the Center for Engineering Applications of Radioisotopes (CEAR) at NC State University. Three measurement campaigns were performed in order to validate the DRF’s constructed by g03: on-axis detection of calibration sources, off-axis measurements of a highly enriched uranium (HEU) disk, and on-axis measurements of the HEU disk with steel plates inserted between the source and the detector to provide attenuation. Furthermore, this work quantifies the uncertainty of the Monte Carlo simulations used in and with g03, as well as the uncertainties associated with each semi-empirical model employed in the full DRF rep-resentation. Overall, for the calibration source measurements, the response computed by the DRF for the prediction of the full-energy peak region of responses was good, i.e. within two standard deviations of the experimental response. In contrast, the DRF tended to overestimate the Compton continuum by about 45–65% due to inadequate tuning of the electron range multiplier fit variable that empirically represents physics associated with electron transport that is not modeled explicitly in g03. For the HEU disk mea-surements,more » computed DRF responses tended to significantly underestimate (more than 20%) the sec-ondary full-energy peaks (any peak of lower energy than the highest-energy peak computed) due to scattering in the detector collimator and aluminum can, which is not included in the g03 model. We ran a sufficiently large number of histories to ensure for all of the Monte Carlo simulations that the sta-tistical uncertainties were lower than their experimental counterpart’s Poisson uncertainties. The uncer-tainties associated with least-squares fits to the experimental data tended to have parameter relative standard deviations lower than the peak channel relative standard deviation in most cases and good reduced chi-square values. The highest sources of uncertainty were identified as the energy calibration polynomial factor (due to limited source availability and NaI resolution) and the Ba-133 peak fit (only a very weak source was available), which were 20% and 10%, respectively.« less
 [1] ;  [1] ;  [1] ;  [1] ;  [2] ;  [3] ;  [3]
  1. North Carolina State Univ., Raleigh, NC (United States). Dept. of Nuclear Engineering
  2. North Carolina State Univ., Raleigh, NC (United States). Dept. of Mathematics
  3. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Publication Date:
Grant/Contract Number:
AC05-00OR22725; 127981; NA0002576
Accepted Manuscript
Journal Name:
Nuclear Instruments and Methods in Physics Research. Section B, Beam Interactions with Materials and Atoms
Additional Journal Information:
Journal Volume: 410; Journal Issue: C; Journal ID: ISSN 0168-583X
Research Org:
North Carolina State Univ., Raleigh, NC (United States). Consortium for Nonproliferation Enabling Capabilities (CNEC); Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org:
USDOE National Nuclear Security Administration (NNSA), Office of Nonproliferation and Verification Research and Development (NA-22)
Country of Publication:
United States
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; NaI collimated detector; Response function; Validation; Uncertainty quantification; Holdup measurements
OSTI Identifier:
Alternate Identifier(s):
OSTI ID: 1438020