BENCHMARK ANALYSIS OF EXPERIMENTS IN FAST CRITICAL ASSEMBLIES USING ACONTINUOUS-ENERGY MONTE-CARLO CODE MVP

Critical experiments in the FCA and ZPPR-9 assemblies were re-analyzed systematically by using a continuous-energy Monte Carlo code MVP with the JENDL-3.2 library. The experimental geometries were simulated alm ost as built and the neutronic parameters were calculated with the phy sics model where all neutron reactions given in the evaluated nuclear data library were treated explicitly As a result, the calculated k(eff ) values underestimated the experimental ones by 0.5-0.8%Delta k/k exc ept for FCA X-2. The calculated results for most reaction rate distrib utions and central reaction rate ratios agree well with the measured o nes within the uncertainties. Furthermore, in order to investigate the cause of the underestimation of the k(eff) values, re-calculations we re performed using the modified JENDL-3.2 library where the inelastic scattering cross sections and the energy-angular distributions for U-2 38 were replaced by those of the ENDF/B-VI library. The k(eff) values increased by 0.4-0.5%Delta k/k for the FCA assemblies. by similar to 0 .8%Delta k/k for ZPPR-9 and the C/E values a ere improved. It is found that the cause of the increase in the k(eff) values is attributed to the energy distribution for secondary neutrons scattered inelastically by U-238.