J.N. Wagner, “The Development of a Method to Predict the Background Neutron Response in Nuclear Security Systems”, Ph.D. Dissertation, Nuclear Engineering, Texas A&M University, College Station, TX (2021).
Nuclear security operators frequently employ gamma spectroscopy and neutron counting techniques to locate radiological materials, particularly those that may be used for nuclear terrorism. A common employment method is the mobile radiological search system (MRSS). One problem with an MRSS is that it is difficult to determine when a neutron detector is observing background, but operational data suggested a relationship between the two detectors’ background responses. Specifically, the neutron count rate showed a linear correlation with the gamma spectrometer’s high energy (>3 MeV) count rate. Such a relationship would enable a method to estimate the neutron background in an MRSS and improve its ability to alert to non-background responses. Therefore, this work sought to characterize this relationship and determine its ability to predict the background neutron count rate in an MRSS for producing such alarms. Using locally available detectors, MRSS measurements of parking garages confirmed the presence of that linear correlation. To determine the correlation’s cause, this work performed spectral analysis of the high energy (4-85 MeV) responses from common scintillators, which found that they were the result of muons (negatively charged leptons with 200 times the mass of electrons). This result indicates that the correlation between the two detectors is a significant relationship: muons and neutrons are part of natural background and both are produced by cosmic ray interactions with the Earth’s atmosphere. Additional MRSS measurements sought to characterize this relationship by measuring locations throughout the Central United States, sampling variables that would affect the neutron and muon fluxes. The measurements’ results showed that the neutron count rate followed a power law function of the muon count rate, indicating the neutron count rate indeed could be predicted. This prediction method was then applied to MRSS measurements of a parking garage while 252Cf was present. These measurements found that the alarm method which came from the neutron prediction function performed properly when alerting for the presence of that radioactive material. These results show that MRSS operators can use similar systems to make such predictions, and the resulting alarm method should improve an MRSS’s sensitivity to man-made radiological materials.