Monte Carlo analysis of KRITZ-2 critical benchmarks on the reactivity temperature coefficient using ENDF/B-VII.1 and JENDL-4.0 nuclear data libraries

Abstract

A set of KRITZ-2 experiments light water moderated lattices with uranium oxide and mixed-oxide fuel rods, at room and elevated temperatures, performed in the early 1970’s have been assessed. Using the MCNP6.1 code with the most recent cross section libraries: JENDL-4 and ENDF/B-VII.1, the critical experiments KRITZ: 2-1, KRITZ: 2-13, and KRITZ: 2-19 achieved in the Sweden reactor KRITZ were analyzed. We have used the ENDF/B-VII.1 data provided with the MCNP6.1.1 version in ACE format and the Makxsf utility to handle the data in the specific temperatures not available in the MCNP6.1.1 original data. The JENDL-4 evaluations were processed using NJOY99 (update 364) to the temperatures of interest. The detailed comparisons of the calculated and measured (Benchmark, 2005) effective multiplication factors and pin power distributions for UO2 and MOX fuelled cores presented in this work demonstrate a good agreement between calculation and measurements. The maximum deviation of the calculation from the experimental data for keff, is 0.58% (absolute value) obtained for the KRITZ 2:1 at 248.5°C using ENDF/B-VII.1 data.To investigate better the influence of cross sections differences on the reactivity and temperature coefficient, we break down the infinite multiplication factor into its components using a pin cell model. Using this simple model we evaluated the temperature effect on the infinite multiplication factor and the effect on its components.We have also analyzed the temperature effect on the leakage by calculating the non leakage probability and the associated temperature coefficient. The overestimation of the calculated non-leakage probability values using JENDL-4 relatively to the calculated one using ENDF/B-VII.1 for all configurations and all temperatures can be explained by the overestimation of the absorptions in the JENDL-4 library especially in the resonance of the heavy nuclides. We can note that this discrepancy between the libraries decreases in hot conditions.For the reactivity temperature coefficient, our analysis has shown that the tendency of a negative error (overestimation by calculation of the absolute value of the RTC) usually observed when analyzing similar UO2 and MOX LWR lattices is confirmed. However, the level of the calculation error has been reduced significantly by using Monte Carlo modeling associated with the most recent nuclear data libraries.