I2S-LWR Activation Analysis of Heat Exchangers Using Hybrid Shielding Methodology With SCALE6.1

Abstract

The Integral Inherently Safe Light Water Reactor (I2S-LWR) concept developed by a team lead by Georgia Tech is a novel PWR reactor delivering electric power of 1000 MWe while implementing inherent safety features typically reserved for Generation III+ small modular reactors. The main safety feature is based on an integral primary circuit configuration, bringing together a compact core design with 121 fuel assemblies (FA), control rod drive mechanism (CRDM), 8 primary heat exchangers (PHE), 4 passive decay heat removal systems (DHRS), 8 pumps, and other integral components. A high power density core based on uranium silicide fuel (U3Si2) is selected to achieve high thermal power which is extracted with PHEs placed in the annular region between the barrel and the vessel. The compact and integrated design of I2S-LWR leads to activation of integral components, mainly made from stainless steel, so accurate and precise Monte Carlo (MC) simulations are needed to quantify potential dose rates to personnel during routine maintenance operation. This shielding problem is therefore very challenging, posing a non-trivial neutron flux solution in a phase space. This paper presents the performance of the hybrid shielding methodologies CADIS/FW-CADIS implemented in the MAVRIC sequence of the SCALE6.1 code package. The main objective was to develop a detailed MC shielding model of the I2S-LWR reactor along with effective variance reduction (VR) parameters and to calculate neutron fluence rates inside PHEs. Such results are then utilized to find the neutron activation rate distribution via 60Co generation inside of a stack of microchannel heat exchangers (MCHX), which will be periodically withdrawn for the maintenance. 59Co impurities are the main cause of (n,γ) radiative gamma dose to personnel via neutron activation since 60Co has half-life of 5.27 years and is emitting high energy gamma rays (1.17 MeV and 1.33 MeV). The developed MC model was successfully used to find converged fluxes inside all 8 stacks of PHEs with respect to MC statistics using the FW-CADIS methodology. For that purpose the SN module Denovo, based on forward-adjoint transport theory, was used to find VR parameters (importance map and biased source) to effectively bias MC simulation. Further research is required to account for other activation pathways, i.e. isotopes of iron which may generate 59Co through neutron activation and beta decay.