Development and Characterization of a Fast Neutron Imaging Telescope (FNIT) for 1--20 MeV Neutrons From the Sun and Nuclear Material

We discuss the development and complete characterization of a double scatter telescope for 1--20 MeV neutrons intended for applications in solar physics and nuclear security. In high-energy solar physics, detecting the presence of low energy accelerated ions in the low corona is recognized as an important goal. The surest indication of the acceleration of these particles is the detection of low energy (\u3c10 MeV) neutrons. These measurements can only be made in the inner heliosphere due to the finite neutron lifetime and flux divergence as they leave the Sun. Additionally, the field of nuclear security has interest in an instrument that can detect, measure, and locate sources of (\u3c10 MeV) neutrons from nuclear material. Materials of interest, namely uranium and transuranics, emit neutrons via spontaneous or induced fission. Unlike other neutral emission from nuclear material, (e.g. gamma rays), copious and penetrating neutron emission is unique to fissionable material. The FNIT instrument was carefully tailored for both applications with a low energy threshold. A double scatter instrument allows for background rejection techniques to obtain increased sensitivity. A small, modular prototype instrument was constructed at UNH with laboratory calibration completed to tune the pulse height and shape, threshold, and time-of-flight for neutron measurements. Quasi-monoenergetic neutron beams calibrated the prototype over the full energy range and fission neutrons were used to test the response and performance of the instrument. Simulations characterized the instrument energy response and were used to generate response matrices for data inversion. We used zeroth-order Tikhonov regularization de-convolution algorithms to obtain the true neutron source spectrum for a given regularization (smoothing) parameter, lambda. Independent of the binning strategy, lambda is of order 10-6. We find that lambda +/- sigma results in a 2% error in total neutron counts; an error within +/-5-sigma results in a variation of ≤ 30% in total neutron counts. Double scatter imaging, adopted from gamma-ray telescopes, demonstrate source location identification can be obtained. We apply laboratory and simulation information to obtain performance estimates of future instruments near the Sun and in the field