An analysis of Intense Pulsed Active Detection (IPAD) for the detection of special nuclear materials

Abstract

This paper analyzes the Intense Pulsed Active Detection (IPAD) approach for active detection of fissile material, highlights some of its merits, and compares it to the multiple-pulse-train LINAC approach. In both cases, bremsstrahlung photons are used to induce fissions. IPAD's single, 100-ns bremsstrahlung pulse offers the possibility to utilize prompt as well as delayed neutrons emitted during the fission process. A combination of the ITS and MCNPX codes is used to characterize the dose to human tissue and both the prompt and delayed neutron emissions from the induced fissions for incident bremsstrahlung spectra between 6 and 50 MeV. No effects from shielding or from the interaction of the bremsstrahlung with the environment (water, air, etc) have been included in these computations, but these effects are a topic of ongoing work. Curves are developed that give the number of useful prompt neutrons per unit dose and the total number of delayed neutrons per unit dose that are independent of both the total electron charge delivered to the bremsstrahlung converter and the source-to-object distance. Useful prompt neutrons are defined to be those with energies exceeding the highest energy photo-neutron expected from <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">207</sup> Pb for a given bremsstrahlung endpoint energy. The analysis shows that endpoint energies between 9 and 13 MeV are optimum for the production of useful prompt neutrons/dose. Because the prompt signal is very fast, the detection of just two useful prompt neutrons gives a false alarm rate from cosmic-ray neutrons of less than 1 in 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> . The analysis for delayed neutrons shows that above 14 MeV endpoint energies, both the neutron production and the dose increase as the endpoint energy squared. This suggests that, for a fixed body dose, no additional delayed neutrons are created above an endpoint energy of 14 MeV. A comparison of the multiple-pulse pulse-train LINAC approach and the single-pulse IPAD approach for detecting delayed neutrons is also presented. Because the 3-s detection time for the IPAD system is so much shorter than the 120-s detection time for the LINAC system, the IPAD system produces a signal that is much easier to separate from the cosmic-ray neutron background. This shorter time allows an IPAD system to detect fissile objects with 100% detection rate at a false alarm rate from cosmic-ray neutrons of less than 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-6</sup> compared with a 50% detection rate for a 10% false alarm rate for the LINAC approach. Alternatively, the same detection efficiency as the LINAC approach is obtained with the detector about 2.7 times farther away from the fissile object.