Calibration and validation of a Monte Carlo model for PGNAA of chlorine insoil

A prompt gamma-ray neutron activation analysis (PGNAA) system was used to c alibrate and validate a Monte Carlo model as a proof of principle for the q uantification of chlorine in soil. First, the response of an n-type HPGe de tector to point sources of Co-60 and Eu-152 was determined experimentally a nd used to calibrate an MCNP4a model of the detector. The refined MCNP4a de tector model can predict the absolute peak detection efficiency within 12% in the energy range of 120-1400 keV. Second, a PGNAA system consisting of a light-water moderated Cf-252 (1.06 mu g) neutron source, and the shielded and collimated HPGe detector was used to collect prompt gamma-ray spectra f rom Savannah River Site (SRS) soil spiked with chlorine. The spectra were u sed to calculate the minimum detectable concentration (MDC) of chlorine and the prompt gamma-ray detection probability. Using the 252Cf based PGNAA sy stem, the MDC for Cl in the SRS soil is 4400 mu g/g for an 1800-second irra diation based on the analysis of the 6110 keV prompt gamma-ray. MCNP4a was used to predict the PGNAA detection probability, which was accomplished by modeling the neutron and gamma-ray transport components separately. In the energy range of 788 to 6110 keV, the MCNP4a predictions of the prompt gamma -ray detection probability were generally within 60% of the experimental va lue, thus validating the Monte Carlo model.