Sensitivity Analysis Applied to Mechanistic Calculation for Effective Diffusivity of Ag through the SiC Layer in TRISO modeled with BISON

Many advanced reactor designs plan to utilize TRistructural ISOtropic (TRISO) fuel such as high-temperature gas-cooled reactors (HTGRs) and fluoride salt-cooled reactors (FHRs). The quantification of the diffusional release of fission products from TRISO fuel particles is crucial to ensure safe operations for these advanced reactors. While the silicon carbide (SiC) layer serves as the main fission product barrier, silver (Ag) release has been observed from intact TRISO particles and is the focus of this work. This study consists of a systematic global sensitivity analysis to determine best-estimate parameters on a mechanistic mesoscale effective Ag diffusion coefficient (D_eff) in SiC that was determined by leveraging atomistic silver (Ag) diffusivity calculations at the mesoscale to be a function of temperature and microstructure variables. The Sobol variance decomposition method was used, and the Sobol indices were computed to determine the impact of each parameter of the effective diffusivity on diffusional Ag release predictions. The determined best-estimate parameters were then utilized to determine Ag release prediction uncertainties associated with calculated temperature and microstructural length measurements to further optimize BISON predictions. Finally, the optimized values are employed with computationally determined normal and transient histories representative of the generic fluoride-salt cooled reactor (gFHR) to provide Ag release predictions for this reactor design.