Shutdown dose rate benchmarking using modern particle transport codes

• Published in: Nuclear fusion 2020-05, Vol.60 (5), p.56024
• Main Authors: Eade, T., Colling, B., Naish, J., Packer, L.W., Valentine, A.
• Format: Article
• Language: English
• Subjects: FNG; ITER; MCR2S; neutronics; OpenMC; Serpent 2; shutdown dose
• ISSN: 0029-5515; 1741-4326
• DOI: 10.1088/1741-4326/ab8181

Shutdown dose rate calculations provide an essential input to the design and research of fusion power plant technology. They allow the estimation of dose to personnel and equipment during planned and unplanned maintenance. The mesh coupled rigorous 2 step (MCR2S) methodology used at Culham Centre for Fusion Energy (CCFE) was originally developed to link the MCNP particle transport code and the FISPACT-II inventory code. As new particle transport codes are developed there is a strong motivation to move towards a code agnostic approach. This paper details the integration of MCR2S with the FISPACT-II API and two other transport codes, Serpent 2 and OpenMC. Two benchmarks, the FNG shutdown dose rate experimental benchmark and the ITER computational benchmark, have been performed and compared to results produced with MCNP. In general, the results show that MCNP, Serpent 2 and OpenMC give shutdown dose rate results similar to the FNG experiment for both experimental campaigns. However, all codes appeared to slightly overestimate the dose rates for Campaign 1 (all results had a C/E between 1 and 1.5) and underestimate the dose rates for Campaign 2 (all results had a C/E between 0.6 and 1). Differences were seen between OpenMC and MCNP for the ITER port plug benchmark, where the lack of variance reduction in OpenMC meant that the neutron flux estimates at the rear of the model were not converged. This led to differences of up to 13% in the shutdown dose rates. It was shown that Serpent 2 and MCNP, where variance reduction was used, gave shutdown dose rates within 3% of each other. Although some areas for development of the Serpent 2 and OpenMC transport codes have been highlighted, overall the comparisons give confidence that the implementation of these two transport codes into the MCR2S work-flow has been carried out successfully.