Abstract
Loading requirements for dry cask storage of spent nuclear fuel are driven primarily by decay heat capacity limitations, which themselves are determined through recommended limits on peak cladding temperature within the cask. This study examines the relative sensitivity of peak material temperatures within the cask to parameters that influence both the stored fuel residual decay heat as well as heat removal mechanisms. These parameters include the detailed reactor operating history parameters (e.g., characteristics beyond burnup, enrichment, and cooling time, such as the presence of burnable poisons and soluble boron) as well as factors that influence heat removal, including non-dominant processes (such as conduction from the fuel basket to the canister and radiation within canister) and ambient environmental conditions (such as ambient air temperature and solar insolation). By examining the factors that drive heat removal from the cask alongside well-understood parameters such as the relationship of fuel discharge burnup to decay heat, it is therefore possible to make a contextual analysis of the most important parameters to evaluation peak material temperatures within the cask.
The goal of this analysis is to afford modelers the ability to develop best-estimate thermal models useful for material degradation studies, as opposed to more conservative bounding analyses used for safety and licensing studies, which are primarily intended illustrate that recommended temperature limits are not exceeded. In contrast, best-estimate thermal models are necessary for material degradation studies, such as evaluation of potential phenomena such as hydride reorientation and ductile-to-brittle transition of the clad material, both of which as temperature-dependent phenomena.
The cask parameters that have the greatest impact on cask temperatures play a role in convective heat transfer in the cask annulus and within the canister basket. These parameters are the ambient air temperature, canister pressure, and assembly decay heat profiles with measured sensitivity coefficients of 0.500, -0.284, and 0.077, respectively. The sensitivity of peak cladding temperature to reactor operating conditions was determined to be inconsequential in the cooling time spans investigated, excluding burnup which had the greatest impact with a sensitivity coefficient of 0.909.
Meanwhile, while the the specific power history (i.e., distribution of power / burnup by cycle) plays an important role in determining short-term decay heat post-discharge (i.e., within the first 5--7 years), decay heat (and thus peak clad temperature) manifests minimal sensitivity to the specific power history for cooling times over 10 years (and thus is of limited importance for long-term thermal studies). Other fuel history parameters, soluble boron concentrations, presence of burnable absorbers, and fuel and moderator temperature were likewise of comparatively little consequence to peak clad temperature sensitivity.
