Optimized performance for neutron interrogation to detect SNM

Abstract

A program of simulations and validating experiments was utilized to evaluate a concept for neutron interrogation of commercial cargo containers that would reliably detect special nuclear material (SNM). The goals were to develop an interrogation system capable of detecting a 5 kg solid sphere of high-enriched uranium (HEU) even when deeply embedded in commercial cargo. Performance goals included a minimum detection probability, P{sub d} {ge} 95%, a maximum occurrence of false positive indications, P{sub fA} {le} 0.001, and maximum scan duration of t {le} 1 min. The conditions necessary to meet these goals were demonstrated in experimental measurements even when the SNM is deeply buried in any commercial cargo, and are projected to be met successfully in the most challenging cases of steel or hydrocarbons at areal density {rho}L {le} 150 g/cm{sup 2}. Optimal performance was obtained with a collimated ({Delta}{Theta} = {+-} 15{sup o}) neutron beam at energy E{sub n} = 7 MeV produced by the D(d,n) reaction with the deuteron energy E{sub d} = 4 MeV. Two fission product signatures are utilized to uniquely identify SNM, including delayed neutrons detected in a large array of polyethylene moderated 3He proportional counters and high energy {beta}-delayed fission product {gamma}-radiation detectedmore » in a large array of 61 x 61 x 25 cm{sup 3} plastic scintillators. The latter detectors are nearly blind to normal terrestrial background radiation by setting an energy threshold on the detection at E{sub min} {ge} 3 MeV. Detection goals were attained with a low beam current (I{sub d} = 15-65 {micro}A) source up to {rho}L = 75 g/cm{sup 2} utilizing long irradiations, T = 30 sec, and long counting times, t = 30-100 sec. Projecting to a higher beam current, I{sub d} {ge} 600 {micro}A and larger detector array the detection and false alarm goals would be attained even with intervening cargo overburden as large as {rho}L {le} 150 g/cm{sup 2}. The latter cargo thickness corresponds to 8 ft of hydrogenous or metallic cargo at the highest density allowed by the weight limit of the container. Simulations support the efficacy of this technique in the most challenging cases and experimental measurements are shown validating these predictions. Signal and background levels have been assessed and utilized to predict error rates due to false positive and false negative results. The laboratory system demonstrates the ability to detect HEU in amounts as small as m {ge} 250 g buried in the middle of a maximum density cargo and to do so with error rates that meet the goals given above. Higher beam current allows reliable SNM detection in shorter irradiation and/or counting times and with more challenging cargo threat scenarios.« less