Effects of correlations in uncertainties of total cross section and elastic angular distribution for a deep penetration of 14-MeV neutrons in Cu

Uncertainty in neutron reaction rates after penetrating a thick copper benchmark assembly was estimated based on two different kinds of Total Monte Carlo methods under random sampling methodology. Five hundred random nuclear data files were generated for Cu-63 by Bayesian Monte Carlo method by perturbing the underlying nuclear model parameters. In the first method, these files were used directly to generate processed libraries, preserving all the correlations among different physical quantities. In the second method, the random files generated in the first method were used, but the angular distributions of elastic scattering were kept fixed to those of the non-perturbed nominal one. It was found that the two methods have the same neutron reaction rate after 608 mm penetration of copper. However, the uncertainty of the first method was smaller than that of the second method. It shows that the correlation between angular distribution of elastic scattering at 0 degrees and total cross section, which stems from Wick's inequality, affects uncertainty of the calculated neutron reaction rate. It could be concluded that the uncertainty obtained by using the covariance files given in the ENDF-6 format may not give correct results for the uncertainty of neutron penetration calculations.