Detection of hazardous materials in vehicles using neutron interrogation techniques
Abstract
The development of a neutron-based interrogation system is presented that non-intrusively, non-destructively detects threats and contraband within vehicles. It has been applied to the detection of vehicle borne improvised explosive devices. It utilizes two D-T neutron generators coupled with uranium reflectors and eight high-purity germanium (HPGe) gamma-ray detectors. The neutron source and gamma-ray detectors are housed in moving components that sweep the entire vehicle and allow for a single location re-scan. It was found that the coupling of the D-T neutron generator with a uranium reflector/amplifier produces a switchable source of 14 MeV neutrons and 1 MeV fission neutrons that optimizes the detection of gamma-ray signals produced in inelastic scattering and thermal capture reactions on nuclei in explosives. For every 14 MeV neutron entering the reflector/amplifier, more than one fission-like neutron is emitted in the direction of the vehicle. Experiments have shown that the reflector/amplifier enhances thermal signals in a sample by up to a factor of 15 in contrast to no reflector. A threat algorithm is presented that automatically detects the presence and location of explosives in a vehicle. The algorithm images each trigger element detected along the length of the vehicle and then between the detector towers. It clears vehicles when the elements occur in distinct parts of the vehicle, reducing the statistical false-positive rate to improve overall system performance. A 13 vehicle data set was collected which quantified the false-positive rate reduction to be 6×10-6, while keeping the explosives detection probability at greater than 99%.
Degree
Ph.D.
Advisors
Koltick, Purdue University.
Subject Area
Nuclear physics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.