Validation and benchmarking of calculation methods for photon and neutron transport at cask configurations

The reliability of calculation tools to evaluate and calculate dose rates appearing behind multi-layered shields is important with regard to the certification of transport and storage casks. Actual benchmark databases like SINBAD do not offer such configurations because they were developed for reactor and accelerator purposes. Due to this, a bench-mark-suite based on own experiments that contain dose rates measured in different distances and levels from a transport and storage cask and on a public benchmark to validate Monte-Carlo-transport-codes has been developed. The analysed and summarised experiments include a ⁶⁰Co point-source located in a cylindrical cask, a ²⁵²Cf line-source shielded by iron and polyethylene (PE) and a bare ²⁵²Cf source moderated by PE in a concrete-labyrinth with different inserted shielding materials to quantify neutron streaming effects on measured dose rates. In detail not only MCNP^TM (version 5.1.6) but also MAVRIC, included in the SCALE 6.1 package, have been compared for photon and neutron transport. Aiming at low deviations between calculation and measurement requires precise source term specification and exact measurements of the dose rates which have been evaluated carefully including known uncertainties. In MAVRIC different source-descriptions with respect to the group-structure of the nuclear data library are analysed for the calculation of gamma dose rates because the energy lines of ⁶⁰Co can only be modelled in groups. In total the comparison shows that MCNP^TM fits very wall to the measurements within up to two standard deviations and that MAVRIC behaves similarly under the prerequisite that the source-model can be optimized. (author)