SNM Detection with a Large Water Cerenkov Detector

Year
2009
Author(s)
N. S. Bowden - Lawrence Livermore National Laboratory
A. Bernstein - Lawrence Livermore National Laboratory
R. Svoboda - Lawrence Livermore National Laboratory
S. Dazeleya - Lawrence Livermore National Laboratory
Abstract
Special Nuclear Material (SNM) can either spontaneously fission, or be induced to do so. Either case results in neutron emission. Since neutrons are highly penetrating and difficult to shield, they could, potentially, be detected escaping even a well shielded cargo container. Obviously, if the shielding is sophisticated, detecting it would require a highly efficient detector with close to 4p solid angle coverage. Water Cerenkov detectors may be a cost effective way to achieve that goal if it can be shown that the neutron capture signature is large enough and if sufficient background rejection can be employed as detectors get larger. In 2008 the LLNL Advanced Detector Group reported the successful detection of neutrons with a ¼ ton gadolinium doped water Cerenkov prototype. We have now built a 4 ton version. This detector is not only bigger, it was designed with photon detection efficiency in mind from the beginning. We are employing increased photocathode coverage and more reflective walls, coated with PTFE. The increased efficiency should allow better energy resolution. We expect that the better diffusive wall reflectivity will reduce the overall dependence of the detector response on particle direction, again producing a more consistent response. We also believe that as detectors get larger, both uncorrelated and correlated backgrounds due to gamma-rays and cosmic ray interactions near the detector will increase. To prove the effectiveness of the technology we must develop new ways to reject these backgrounds while maintaining our sensitivity to SNM neutrons. Better energy resolution will enable us to reject more of the low energy gamma-ray backgrounds on this basis. Overcoming cosmic ray induced neutrons is perhaps an even larger concern as detectors get larger. Our detector is designed so that we can test various segmentation schemes – effectively dividing the detector up into smaller ones. In this presentation, we will describe our detector in detail.