Table 1.
Selected gamma-ray emission lines for the detection and identification of high explosives and CWA [5, 16, 17].
Table 2.
Elemental ratio H/N for some common high explosives [18].
Fig 1.
Cross-section drawings of the optimized geometries of the proposed instruments for a case with (a) neutron-sensitive detectors (“shell-point” design) and (b) detectors that are not affected by the presence of a large neutron flux (“shell-side design”). See Table 3 for the list of numbered parts and Table 4 for their dimensions.
Table 3.
List of volumes represented in GEANT4 for the instruments under study, along with their material composition at the start of the optimization studies and after completion of the study.
Table 4.
List of volumes represented in GEANT4 for the instruments under study, along with their final dimension at the end of the study, for the “shell-point” design.
Fig 2.
GEANT4 model of the instrument.
GEANT4 cutaway image of the initial geometry used to study neutron flux characteristics in the unknown-object cavity. The shell-point design is used, with the detector and artillery shell omitted. The neutron scoring plane is shown in solid red.
Fig 3.
Neutron energy distributions in the unknown-object cavity with varying moderator thicknesses.
Effect of the thickness of the polyethylene moderator on the energy distribution of (a) direct fast neutrons, (b) integrated fast neutrons, (c) direct thermal neutrons and (d) integrated thermal neutrons in the unknown-object cavity. A carbon elastic scattering resonance can be observed just below 2.1 MeV. In (b), the broad, asymmetrical peak between 1.8 and 2 MeV is due to neutrons reflected by the graphite shield located around the unknown-object cavity, shown as part #14 in Fig 1.
Fig 4.
Neutron time distributions in the unknown-object cavity with varying moderator thicknesses.
Effect of the thickness of the polyethylene moderator on the time distribution of (a) direct fast neutrons, (b) integrated fast neutrons, (c) direct thermal neutrons and (d-e) integrated thermal neutrons in the unknown-object cavity, shown for two different time ranges.
Fig 5.
Energy distribution of the integrated thermal neutrons at the centre of the unknown-object cavity for different moderation scenarios.
Two distributions are the result of simulations where the moderator volume is made up of polyethylene, with (red) and without (green) an artillery shell in the cavity. The other two distributions use Teflon in the moderator volume, with (brown) and without (blue) an artillery shell in the cavity.
Fig 6.
Neutron-activated gamma-ray emission spectrum of a TNT-filled iron artillery shell.
The spectrum is obtained using the instrument configuration shown in Fig 1(a), after optimisation of the neutron flux. Some of the highest-intensity peaks are labelled with their isotopic source.
Fig 7.
Energy spectrum of gamma rays incident on the active detector material in the initial design of the shell-point instrument configuration.
The total incident spectrum is shown in blue, while gamma rays from the artillery shell and TNT payload are shown in red. The most important peaks for identification of explosives, H and N, are labelled, as well as other large peaks from iron in the shell and copper in the neutron generator target.
Fig 8.
Number of gamma rays incident on the detector originating from each simulated volume that contributes significantly to the hydrogen gamma ray background, as a function of gamma ray energy.
The histograms are drawn in a stack such that each histogram presents the contribution of a given volume on top of all volumes drawn below it.
Fig 9.
Number of gamma rays (a) emitted by the artillery shell and TNT payload and (b) incident on the detector after material substitutions in the instrument design.
Fig 10.
Number of gamma rays emitted by the artillery shell and TNT payload after updating material choices in the instrument design and after adding lead shielding.
Fig 11.
Energy spectrum of the gamma rays incident on the detector in the fully optimized instrument design.
The total incident spectrum is shown in blue, while gamma rays from the artillery shell and TNT payload are shown in red. The spectra for the shell-point configuration of Fig 1(a) are shown in (a) while the spectra for the shell-side configuration of Fig 1(b) are shown in (b).
Fig 12.
Hydrogen signal fraction SFH as a function of the start time of the thermal-neutron only timing window.
The shell-point instrument design is shown in (a) and the shell-side instrument design is shown in (b). The error bars correspond to the 1σ statistical uncertainty resulting from the size of the Monte Carlo simulated data samples.
Fig 13.
Energy spectrum of gamma rays incident on the detector for the time windows t <1 μs, t <30 μs and t >50 μs.
The fully optimized instrument design in the shell-side configuration is used. The total gamma ray spectrum is shown in (a), while gamma rays from the artillery shell and TNT payload are shown in (b).
Fig 14.
Energy spectrum of the gamma rays incident on the detector in the fully optimized instrument design in the thermal-only time window.
The total incident spectrum is shown in blue, while gamma rays from the artillery shell and TNT payload are shown in red. Spectra for the shell-point configuration of Fig 1(a) are shown in (a) and spectra for the shell-side configuration of Fig 1(b) are shown in (b).
Fig 15.
Spectrum of the energy deposited by all incident particles in a 2×3 inch LaBr3 detector in the fully optimized instrument design in the thermal-only time window.
The simulated energy deposits are smeared using the energy resolution characteristics of [24]. Spectra for the shell-point configuration of Fig 1(a) are shown in (a) and spectra for the shell-side configuration of Fig 1(b) are shown in (b). The inset in (b) shows the peak from the hydrogen 2.22 thermal capture line. The result of a Gaussian peak fit over linear background is shown in red.
Table 5.
Value of SFH in simulated event samples with the shell-point instrument design at various stages of the design optimization process.