Verification of depletion capability of OpenMC using VERA depletion benchmark. (1st June 2022)

• Record Type: Journal Article
• Authors: Yu, Jiankai; Forget, Benoit
• Is Part Of: Annals of nuclear energy. Volume 170(2022)
• Publication Date: 2022-06-01
• Subjects: OpenMC -- VERA -- Depletion Chain -- ENDF/B-VIII.0; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2022.108973
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: Depletion capability of OpenMC is verified by using VERA depletion benchmark. K-effective and detailed number densities are compared and discussed in depletion analysis. The impacts on depletion from simplified nuclides transmutation chain, and different versions of data libraries are discussed. The advances of high-order integrators are testified based on Gadolinium-bearing fuel depletion. Abstract: Depletion capability has been implemented in the open-source state-of-the-art Monte Carlo particle transport code, OpenMC, via the Python application programming interface (API) and C API. The Python API provides the main depletion functionality including the depletion matrix solver based on the Chebyshev rational approximation method, and high-fidelity numerical integration algorithms for flux and nuclide densities updates. OpenMC serves as the data provider for problem dependent transmutation reactions needed to form the depletion matrix. The data exchange between the OpenMC solver and the depletion solver is controlled by the C API to facilitate in-memory transfers. This paper firstly demonstrates the accuracy of this implementation by comparing results of VERA depletion benchmark against the well-verified Monte Carlo depletion code, Serpent2, including both multiplication factor and nuclide densities of interest. Additionally, a simplified depletion chain based on that of the CASL project is evaluated by comparing its accuracy and efficiency with a full depletion chain based on ENDF/B-VII.1 library. The benefits of high-order numerical integration algorithms have also been verified on selected benchmarks with and without burnable absorbers. Finally, the impact of the ENDF/B-VIII.0 data library on select benchmarks was also evaluated.