Comparisons between RFSP and MCNP for modeling pressure tube heavy water reactor cores with thorium-based fuels. (October 2018)

• Record Type: Journal Article
• Authors: Yan, Huiping; Dugal, Cliff; Colton, Ashlea V.; Bromley, Blair P.; Golesorkhi, S.
• Is Part Of: Annals of nuclear energy. Volume 120(2018)
• Page Start: 642
• Page End: 655
• Publication Date: 2018-10
• Subjects: PT-HWR Pressure Tube Heavy Water Reactor -- LC Lattice Concept -- CVR Coolant Void Reactivity -- MCNP® Monte Carlo N-Particle transport code -- NRMSD Normalized Root Mean Square Deviation -- RFSP Reactor Fueling Simulation Program -- WIMS-AECL 2D multi-group neutron transport code developed by AECL at Chalk River Laboratories; Reactor core physics -- Reactor lattice physics -- Heavy water reactors -- Code-to-code comparisons -- Advanced fuels -- Thorium; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2018.06.036
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: Deterministic physics code RFSP was compared against MCNP for PT-HWR core models. Fuels contains mixtures of ThO2 with 235 UO2, PuO2, LEUO2, or 233 UO2 . The core neutron multiplication factors (keff ) and CVR agree well for all cases. Cores with higher-burnup fuels show larger differences in bundle power distributions. Cores with higher-burnup fuels exhibit "double-hump" axial power profiles. Abstract: Thorium-based fuels are recognized to hold significant promise as an option for achieving a long-term, sustainable nuclear fuel cycle and energy security. Pressure tube heavy water reactors (PT-HWRs) are well suited to exploit the energy potential of thorium. Deterministic reactor physics codes are often used in exploratory studies to evaluate the performance and operational characteristics of various fuel bundle, lattice and core concepts with thorium based fuels in PT-HWRs. Because of the approximations inherent in deterministic codes, they are often considered less accurate than stochastic codes. In order to enhance confidence in deterministic code-based predictions, these codes are often benchmarked against stochastic codes, when experimental data is not available for code validation. Code-to-code comparisons of core physics calculations were made between the deterministic reactor physics toolset WIMS-AECL/WIMS-Utilities/RFSP and the stochastic neutron transport code MCNP for a series of core configurations with mixed oxide fuels containing thorium in PT-HWRs. The core neutron multiplication factors (keff ) appear to have a difference (RFSP-MCNP) ranging between −2.4 mk and +4.0 mk. The MCNP full-core calculations confirm that that thorium-based fuels have a lower coolant void reactivity (CVR), ranging from +8.3 mk to +11.3 mk (versus 14 mk for NU fuel). The core cases with NU fuel have a small difference (RFSP-MCNP) in peak bundle power (ranging between −0.13% and 0.65%). Cores with LEU at 1.2 wt% 235 U/U, Pu/Th, and LEU/Th (LEU at 5 wt% 235 U/U) fuel have higher differences in peak bundle power (ranging between −7% and −12%). All these core cases have peak channel power differences between −2.1% and −8.9%. Core with 233 U fuel has the smallest peak bundle difference (−0.05%) and smallest peak channel differences (−0.58%) which represent the best agreement between MCNP and RFSP simulations. The performed code-to-code comparisons have demonstrated that the core physics parameters estimated by RFSP calculations are consistent with MCNP simulations, especially for fuel where the main fissile component are 235 U-based and 233 U-based fuel.