Comparison of Neutron Data Libraries for the Analysis of the H. B. ROBINSON-2 Pressure Vessel Benchmark with the MCNPX Code

Comparison of Neutron Data Libraries for the Analysis of the H. B. ROBINSON-2 Pressure Vessel Benchmark with the MCNPX Code

The paper refers to an analysis of the `H.B. Robinson-2 Pressure Vessel Benchmark" with the MCNPX code. The benchmark is based on an experiment performed at an operating PWR reactor and includes measurements for both in-vessel and ex-vessel dosimeters.With the aim to enlarge the in-house validation database of the LWR neutron fluence assessments for the Swiss reactors, an MCNPX-based analysis of this benchmark was recently performed, using the general-purpose point-wise library JEFF-3.1. The analysis was done in compliance with the benchmark specifications and using the same or similar assumptions/approximations as employed by the previous participants in order to ensure consistency in the comparison of solutions.A detailed assessment of different libraries was not performed in these past calculations because of the rather poor statistical precision achieved at that time when modeling the high-energy reactions with MCNPX. In the course of recent additional optimization studies the variances of the calculation results have been significantly reduced, allowing thereby comparing the performance of different neutron data libraries for the analysis of this benchmark. This comparison is presented in this paper and includes an assessment of the data libraries JEFF-3.1, JEFF-3.1.1, ENDF/B-VII.0 and IRDF-2002. The differences observed in the results and their potential sources are discussed.

The paper refers to an analysis of the `H.B. Robinson-2 Pressure Vessel Benchmark" with the MCNPX code. The benchmark is based on an experiment performed at an operating PWR reactor and includes measurements for both in-vessel and ex-vessel dosimeters.With the aim to enlarge the in-house validation database of the LWR neutron fluence assessments for the Swiss reactors, an MCNPX-based analysis of this benchmark was recently performed, using the general-purpose point-wise library JEFF-3.1. The analysis was done in compliance with the benchmark specifications and using the same or similar assumptions/approximations as employed by the previous participants in order to ensure consistency in the comparison of solutions.A detailed assessment of different libraries was not performed in these past calculations because of the rather poor statistical precision achieved at that time when modeling the high-energy reactions with MCNPX. In the course of recent additional optimization studies the variances of the calculation results have been significantly reduced, allowing thereby comparing the performance of different neutron data libraries for the analysis of this benchmark. This comparison is presented in this paper and includes an assessment of the data libraries JEFF-3.1, JEFF-3.1.1, ENDF/B-VII.0 and IRDF-2002. The differences observed in the results and their potential sources are discussed.