We have been calibrating rem meters and spectrometers with monoenergetic neutrons from 200 keV to 4.5 MeV by bombarding Li and Be with protons produced by the Positive Ion Accelerator (PIA) at the Naval Surface Warfare Center, Carderock Division (NSWCCD), in Bethesda, Maryland. Now we are bombarding Be with alpha particles, extending the energy range upward to 10 MeV, and moderating the neutrons to extend the range downward to thermal energy (0.025 eV). The alpha Be reaction is the same one in common neutron sources such as PuBe and AmBe, so we can produce the same spectrum. Dose rates are yet to be determined, but a rem meter indicated that we may expect at least a few hundred mrem per hour at a distance of 1 m from the Be target. Contamination of the beam with scattered radiation and gamma rays is expected to be lower than is observed with isotopic neutron sources.
Unlike the proton reactions, the alpha-Be reaction results in broader energy spectra, even for thin Be targets, on the order of 12.5 microns. The reaction adds about 5.7 MeV to the alpha energy, so alphas of 9 MeV produce neutron energies up to 14.7 MeV. At 9 MeV, the alpha particle loses roughly 1.5 MeV in our thinnest possible Be foil. So there is a consequent spread in neutron energy that is 3 to 10 times larger than the spread in neutron energy that is produced typically in either the Li-7 or Be-9 proton-induced reactions for neutron production. The residual C-12 nucleus formed in the reaction has a much higher probability of being in an excited state than, for example, the Be-7 nucleus from the proton-Li reaction, so very few neutrons are emitted with the full excitation energy, i.e., the projectile's incident kinetic energy and the Q-value for the reaction (5.7 MeV). The most likely neutron emitted has an energy of 0.3 to 0.7 of the full excitation energy for a given alpha energy. Also, high-energy gamma rays are released in the subsequent decay to the C-12 ground state. These gammas range from 4.4 to 10 MeV. However, the 4.4 MeV-gamma predominates. When the alpha energy is reduced enough, the possibility of populating the highest energy levels of C-12 is eliminated, which cleans up both the neutron spectra and the gamma background. Initial tests indicated that the gamma dose at 1 m from the Be target remains on the order of 5% of the fast neutron dose equivalent. This value is practically identical to the observed gamma doses from the proton-induced reactions on Li or Be. So the 4.4 MeV-gamma ray contribution appears to be insignificant in terms of total dose compared to the capture gammas emanating from the borated polyethylene shielding on the beamline. The Be-9-alpha reaction can be used as a "surrogate" for either a PuBe or AmBe source of neutrons, so our calibrations may be compared with common neutron sources.
Thermal neutrons are produced by moderation of neutrons with water, heavy water, or graphite. Light water can produce a spectrum not unlike that outside a light-water reactor. D2O and carbon have low thermal neutron capture cross sections, resulting in a larger flux of thermal neutrons than light water. Tests with and without a thermal neutron absorber, such as cadmium or boron, isolate the thermal response from the response to the moderated spectrum.
Dr. David Gilliam, head of Neutron Standards at the National Institute of Standards and Technology (NIST), came on May 1 and May 2 to measure the PIA fluence with standard fission chambers. The beams are more pure than those from isotopic sources. Thermal neutron shields and measurements of detector response versus distance isolate the response to the beam from the response to scattered radiation, which is of lower energy. These corrections are small compared to those used with isotopic neutron sources, such as PuBe, AmBe, and Cf-252, because isotopes send radiation in all directions, while the beam sends most radiation along its axis.