PASSIVE NMIS MEASUREMENTS TO ESTIMATE THE SHAPE OF PLUTONIUM ASSEMBLIES

Year
1999
Author(s)
T.E. Valentine - Oak Ridge National Laboratory
J.A. Mullens - Oak Ridge National Laboratory
J.T. Mihalczo - Oak Ridge National Laboratory
R.B. Perez - Oak Ridge National Laboratory
J.K. Mattingly - Oak Ridge National Laboratory
J. A. March-Leuba - Oak Ridge National Laboratory
Abstract
A new technique to estimate the shape attribute of plutonium assemblies using the nuclear materials identification system (NMIS) is described. THe proposed methos posssesses a number of advantages. It is passive: no externam radiation source is required to estimate the shape of plutonium assemblies. Instead, in herent gamma and neutron emissions from spontaneous fission of 240Pu and subsequent induced fission of 239Pu are detected to estimate the shape attrtibute. The technique is also stationary: shape is estimated without scanning the assembly by moving the detectors relative to the assembly. The proposed methos measures third order correlations between triplets of gamma/neutron-sensitive detectors. The real coincidence of a pair of gammas is used as a \"trigger\" to approximatly identify the time of a spontaneous of induced fission event. The spatial location of this fission event is inferred from the real coincidence of a subsequent neutron with the initial pair of correlated gammas by using the neutron's time-of-flight(approximatly the delay between the gamma pair and the neutron) and the fission neutron spectra of 240Pu and 239Pu. The spatial distribution of fission sites and hence the approximate shape of the plutonium assembly is thereby inferred by measuring the distribution of a large number of these correlated triplets. Proof0of0principle measurements were performed using 252Cf as a surrogate for 240Pu to demonstrate that the technique is feasible. For the simple shapes approximated with 252Cf sources, the measurements showed that the proposed method is capable of correctly identifying the shape and accuratly estimating its size to within a few percent of actual.