Characteristics and Performance of an Integrated Portable High Efficiency Neutron Multiplicity Counter for Detection of Illicit Neutron Sources

In the interdiction of illicitly trafficked Special Nuclear Material (SNM) the neutron signal is a valuable adjunct to the gamma-ray signature. Neutrons are not attenuated by heavy metal shielding which might be used to shield the gamma rays emitted by the source. However, “gross-counting” neutron detectors are unable to distinguish neutrons in the environmental background from those emitted by a neutron source. The most challenging interdiction situation for neutron detection is the case where the neutron count rate is up to ten times the average background. Count rates higher than this are readily detectable with gross counting instruments. Below this level however, gross counters have major problems due to innocent alarms. A portable instrument has been developed which incorporates 30 moderated He tubes, multiplicity counting electronics and software capable of discriminating against noncorrelated neutron events, thereby greatly increasing the sensitivity of detection of SNM. The instrument is packaged in a way that is easily transported in a motor vehicle and to be rapidly deployed as needed. The instrument is described and performance data are presented. Introduction Neutron Multiplicity counting is a technique familiar to those in the business of nuclear materials assay, but less so to those engaged in homeland security applications. The fact that SNM emits multiple (coincident) neutrons is a big factor in its utility as reactor fuel or a nuclear weapon. In nuclear materials assay, the fact that SNM emits multiple neutrons simultaneously can be used to derive equations which allow for the determination of sample mass without requiring an absolute efficiency calibration for the neutron coincidence counter. Less well known is that the time-coincident or correlated nature of neutrons emitted from SNM can be used as a way of discriminating from non-correlated sources or sources correlated differently, thereby helping greatly in the identification of threat neutron sources which might be encountered by security organizations. The most challenging interdiction situation is the case where the neutron count rate is between average background and about ten times average background, although the higher neutron fluxes are readily detectable with a portable instrument such as the ORTEC Detective-EX. This “slightly-elevated” count-rate regime is challenging because legitimate, non-fissile cargo can cause increases in background. This change is due to the interaction of cosmic rays with nearby metal such as iron and can increase the background by a factor of 10, making the detection of fission sources for example on board a ship extremely difficult. Because of these variations, attempts to detect the presence of a man-made neutron source based on an increase in count rate above background are not very effective, unless the source is so strong that the count rate increases to many times background. It is clearly a challenge to discriminate man-made neutron sources from background when the overall count rate is less than ten counts per second. Ten counts per second may represent an increase over the typical background of about three times, but is still difficult to discriminate with the typical handheld or backpack search instrument, because of the lower efficiency of these instruments. A transportable instrument is reported here which, through the use of multiplicity counting, is able to separate cosmic from non-cosmic neutron sources. It can provide supplemental data to expert teams who can use more sophisticated analysis techniques to fully characterize suspicious neutron sources encountered at a border crossings or searches. Neutron Multiplicity Distributions There are several explanations of neutron multiplicity in the literature (see e.g. reference (1)); a very short review is given here. The key neutron signature for SNM results from the fact that the spontaneous fission process emits multiple neutrons in closely spaced temporal groupings. The number of neutrons emitted in spontaneous fission can vary from zero to six or more. The process is random, or statistical, in nature, and the probability distribution of the number neutrons is referred to as the “neutron multiplicity distribution”. Total neutron counting counts all the emitted neutrons without further analysis. Neutron coincidence counting looks for pairs of neutrons within a small time window. Multiplicity counting counts separately the number neutrons detected within a time gate (e.g., none, 1, 2, 3, 4, 5, 6, 7...). The average spontaneous fission multiplicity associated with Pu is reported to be about 2.16 and is characteristic of the material. In an actual Pu source, there is also the possibility of induced fission which changes the observed multiplicity. The observed multiplicity depends on the efficiency of detection. Figure 1 Multiplicity distribution of spontaneous fission neutrons in Pu Figure 2 (from ref 1) shows the count distribution of an AmLi random neutron source. The AmLi source emits uncorrelated neutrons because they are produced by (α,n) reactions where the α-particles emitted by the Am source expel neutrons from the Li target. The count distribution for uncorrelated neutrons follows Poisson statistics. The observed distribution is assumed to be the sum of the individual distributions contributing to the neutron flux. A simple way to characterize a count distribution is the variance-tomean ratio, R, given by where is the first moment and is the second moment of the count distribution. For a random (Poisson) distribution the ratio is unity. If correlation is present, the ratio, R, is not 1. (The Feynman Variance (Y2F), is defined as (R-1)/2.) To determine the presence of an SNM source therefore it is required that we determine if R can be distinguished from random (Poisson) background or from correlated, but not SNM, sources such as are produced in cosmic ray showers. If the multiplicity distribution can be unfolded from the other interferences in the distribution, then the identity of the fission source can be determined or at least an “educated guess” can be made. Description of the Instrument The purpose of the instrument is to identify when a small increase above background in the neutron count rate of the detector due to man made sources by recording the single and multiple neutron counts. The design objective was to be sensitive enough to see the correlation in natural background in a reasonable count time. The system, “Fission MeterTM”, is designed to achieve high neutron detection efficiency in an instrument which is reasonably portable. Time-correlation is used as a way of distinguishing the different types of neutrons present in the flux on the detector. The instrument had to be large enough to provide reasonable efficiency for detection of multiple time-correlated, but not spatially correlated, neutrons. The instrument is arranged in a folding format so that it can be “wrapped around” a suspect package for optimum counting geometry. Each panel has a “thin side” and a “thick side” HDPE moderator, giving optimal detection of fast or partially thermalized neutrons. The He tubes are 1 inch diameter with 19 inches active length and 7.5 ATM gas pressure. Each panel contains 15 He tubes specially selected for low microphonics. Figure 2 Count Distribution of an AmLi random neutron source The detector sub-system includes the HV supplies for the He tubes and the preamplifier/discriminator units required to record the neutron events. Located at the top of one of the two detector panels, the integrated electronics subsystem is powered by readily available D-Cell alkaline batteries. The multiplicity electronics provides 512 time gates. System software operates on an associated ruggedized handheld computer included with the instrument. In addition to software control, the system may be controlled manually from the front panel. The instrument weights 57 lbs and may be carried on a shoulder strap (Fig 6).