Development of background simulation model for commercial radiation portal monitor using Monte Carlo code

At domestic ports and airports, radiation portal monitors (RPMs) have been installed and operated for detection of illegal radioactive materials in cargo containers. In RPM operation, however, false alarms occasionally sound due to naturally occurring radioactive materials. The alarm criterion is set based on the background count rate, which is a site-specific variable; therefore, optimization of the site-specific alarm criterion in consideration of site- and RPM-specific characteristics is essential for efficient operation of any RPM system. To this end, repetitive testing for various source conditions is required; such tests, however, are difficult to conduct, due to the 24/7 operation of airports and ports with their RPMs. Alternatively, Monte Carlo simulation could be the best option for RPM performance evaluation under various conditions for optimization of site-specific alarm criteria. In the present study, we employed iterative Monte Carlo simulations to precisely model a commercial RPM, based on which we were able to determine realistic gain and energy resolution parameters for a polyvinyl toluene (PVT)-based RPM. Background radiation contributing to RPM signals originated mainly from radioisotopes of the ²³⁸U and ²³²Th series as well as ⁴⁰K in the soil near where the RPM was installed; therefore, the corresponding nuclide activity ratios were determined by a high-purity germanium (HPGe) detector based on the measured energy spectrum, and were then applied to the background simulation model. The simulated background spectrum showed good agreement with the RPM-measured experimental data. Additionally, to improve calculation efficiency, two methods using a simple box enclosure and surface source were evaluated, and the results showed an approximately 370-times improvement in computation time. The developed realistic Monte Carlo-based background model will contribute to developing site-specific RPM alarm criteria.