Investigation and improvement of the solution neutron spectrum around an 241Am-Be based neutron irradiator using the traditional unfolding procedure

Abstract

Neutron spectrum unfolding, from the Bonner spheres measurements, is a challenging work since it requires skills and experience. Furthermore, as the results issued from the unfolding procedure are often sensitive and dependent on many parameters, in particular, to the default spectra used, it is important, therefore, to choose the adequate guess spectrum to initiate the iterative unfolding process of interest. Thus, the main objective of this work is to study the sensitivity effects of the default spectra used in two traditional unfolding codes i.e. MAXED and GRAVEL on the final solution spectrum issued from Am-Be based neutron irradiator located at the Secondary Standard Dosimetry Laboratory (SSDL) of CRNA. To this purpose, a sensitivity analysis was performed taking into account various default spectra provided from different cases: a) the photoneutron spectrum generated around a medical linear accelerator CLINAC 2100C; b) the neutron spectrum issued from Cf-D2O based-source installed at the German institute of metrology (PTB), and c) the OB26 irradiation neutron spectrum obtained using a Bayesian software WinBUGS. Hence, following the unfolding process performed, using MAXED and GRAVEL codes, important dosimetric quantities such as: the total fluence rate, the ambient dose equivalent rate, the mean energy and the spectrum averaged fluence-to-ambient dose equivalent conversion were evaluated for the final neutron spectrum. Overall, it reveals that the unfolded spectra derived by MAXED or GRAVEL, using the default spectra obtained from WinBUGS, are very encouraging and similar to those obtained by the same codes using the Monte Carlo default spectrum calculated by MCNP5.