Over a decade ago, it was confirmed that detonation nanodiamond (DND) powders reflect very cold neutrons (VCNs) diffusively at any incidence angle and that they reflect cold neutrons quasi-specularly at small incidence angles. In the present publication, we report the results of a study on the effect of particle sizes on the overall efficiency of neutron reflectors made of DNDs. To perform this study, we separated, by centrifugation, the fraction of finer DND nanoparticles (which are referred to as S-DNDs here) from a broad initial size distribution and experimentally and theoretically compared the performance of such a neutron reflector with that from deagglomerated fluorinated DNDs (DF-DNDs). Typical commercially available DNDs with the size of ~4.3 nm are close to the optimum for VCNs with a typical velocity of ~50 m/s, while smaller and larger DNDs are more efficient for faster and slower VCN velocities, respectively. Simulations show that, for a realistic reflector geometry, the replacement of DF-DNDs (a reflector with the best achieved performance) by S-DNDs (with smaller size DNDs) increases the neutron albedo in the velocity range above ~60 m/s. This increase in the albedo results in an increase in the density of faster VCNs in such a reflector cavity of up to ~25% as well as an increase in the upper boundary of the velocities of efficient VCN reflection.
Keywords: Monte Carlo; albedo; detonation nanodiamonds; fluorination; nanopowder; reflectors of slow neutrons; size separation of nanodiamonds.