In order to increase the capability to detect material with a low signal-to-noise ratio, such as Highly Enriched Uranium (HEU), researchers at Texas A & M are developing a new neutron detector concept that is direction sensitive. This capability would lower the detection threshold of HEU and allow for faster response time and recovery. Exploration of correlated events also led to the examination of using a direction sensitive neutron detector to help eliminate background sources. A detector concept for this system was developed which is theoretically capable of measuring incident angular fluxes on the detector (that is, the detector provides counts as a function of neutron position, energy, and angle).

Angular dependence is generated based on the Doppler broadening effect on the 477 keV gamma ray produced in flight following an (n, alpha) reaction in 10B. This reaction produces an alpha-particle and 7Li nucleus. The 7Li nucleus is in an excited state 94% of the time. It will decay to the ground state by emitting a 477 keV gamma ray. This gamma ray will also be affected by the kinetic energy (and the direction) of the Li particle from which it originated. If this 477 keV gamma ray can be detected and the sum of the original kinetic energies of the Li and alpha particle be measured, then the directional plane the original neutron came from can be unfolded.

Project Updates:

The boron carbide half-cylinder detector system analyzed last quarter was further divided into four equal pieces. This allows for the number of counts from each piece to be processed to give directionality information to determine the location of the source. The signal received from the enriched uranium is still lower than the signal from background, so various methods are being used to try and compensate for this. One such method is to insert polyethylene slabs between the boron carbide to improve fast neutron detection. Research is also being performed in the area of super low signal analysis techniques to help improve the capability of the detector. A Geant4 simulation of the P-N, silicon-boron carbide semiconductor was finished. The simulation was given as best case conditions for the semiconductor energy deposit: 100% pure, 0 Kelvin , and single crystal. The activated region is based on the purest commercially available silicon and assumed boron carbide of similar purity. The material properties of boron carbide were based on those when the crystal is at 300 Kelvin. Given a stream of monoenergetic 5MeV neutrons, the average energy deposit in the activated region was 80.3% of the energy created in a boron reaction. For a 0.025eV stream of neutrons, the average rose to 94.7%. Given more realistic conditions, such as outside temperatures, cross contamination, multiphase, and multicrystal boron carbide, the design was considered possible for detecting thermal neutrons, but it did not have the required characteristics for unscattered neutron detection. Work has begun on trying to find a new design that will have what is needed for unscattered neutron detection. The most significant issue with the design is that energetic ions travel too far in the low density of boron carbide. Therefore either the density or the activated region has to increase. Other boron rich semiconducting materials are being investigated. Other solutions that have a much lower density of boron but have desirable semiconducting properties are also being considered, such as super saturated boron in silicon.

After modeling the existing border monitor in MCNP, various minor aspects of the detector were altered in attempts to increase efficiency. This included changing the location and size of the BF3 tubes and replacing the BF3 entirely. Research done this quarter suggests the best candidate for a solid-state semiconducting material is Boron Carbide. An evaluation of the relative efficiency of such a detector is being completed. Additionally, modifications to the overall design of the system are being modeled to assess whether the traditional slab-type system is most efficient for directionality measurements. Half Cylinder and Elliptical systems are currently being modeled for comparison.

The existing border monitoring device has been modeled in MCNP. A mockup simulation is currently being performed in which a car has been modeled with an HEU source in the trunk. Neutron and gamma count rates are then being computed using various types of detector designs. Modifications to the designs include exchanging moderator material, adding more BF-3 tubes, and modifying the size of the tubes themselves.

One of the requirements for determining neutron direction is knowing the energy of the products of the 10B(n,)7Li reaction. Since high energy resolution is desired, a semiconductor detector is the primary choice. However, infusing the required amount of boron into a semiconductor is difficult. Several ideas for achieving the required level of boron infusion have been researched during this reporting period, and three merit continued study. The first approach is to diffuse a large amount of boron into silicon, anneal it to create a boron precipitate that doesn't affect the silicon crystal structure, and then use the resulting material to construct a silicon detector. The second method is to use a semiconductor material, such as gallium arsenide in which boron is not a dopant, and infuse as much boron as possible. The third approach is to use a semiconducting material made out of boron, such as boron carbide. While all three approaches are still on the table, the third option is the one that looks most feasible. The basic concept is to grow polycrystalline enriched boron carbide on top of a highly n-doped silicon wafer and then grow a polycrystalline highly p-doped silicon layer on top of the boron carbide.