A comparison of dose rate calculations for a spent fuel storage cask by using MCNP and SAS4

This paper presents a detailed comparison of the surface dose rate calculations for the NAC-UMS spent fuel storage cask by using MCNP and SAS4 computer codes. Their accuracy and computation efficiencies are compared. For such a real world deep penetration and streaming problem, effective variance reduction techniques are indispensable for a Monte Carlo simulation to obtain results of small statistic errors within reasonable computing time. The TORT-coupled MCNP calculation based on the CADIS methodology has been used in this study. The main differences between MCNP and SAS4 calculations are the underlying cross-section libraries and the adjoint functions used for variance reduction in Monte Carlo simulations. The cross-section libraries and their formats should be the root cause for some significant discrepancies between the MCNP and SAS4 results. In addition, limited by the 1D adjoint biasing scheme, SAS4 is inefficient in calculating the dose rates near inlet/outlet apertures. Considering all the computer time spent and the statistical errors of results obtained, the overall computation efficiency by using the TORT-coupled MCNP is better than SAS4 in the shielding calculations of spent fuel storage casks. More specifically, although the SAS4 efficiency is better when the cask side calculation is the only concern, the TORT-coupled MCNP technique is more efficient for the gamma-ray transport in cask top configurations and almost all the vent-streaming problems.