Estimation of radiation conditions in the ITER electron cyclotron upper launcher with state-of-the-art simulation techniques

Abstract

• Nuclear analyses in support of ECUL port plug design: computation of nuclear heating to the BSM during plasma operation, and mitigation of dose rate during maintenance. • Significant mitigation of dose rate achieved by shielding options located within the port plug itself; further reductions only possible by optimization of the surrounding environment. • Pioneering application of the MCNP conformal unstructured mesh capability for both geometry description and result tallying for computation of the nuclear heating in the BSM. • The UM succeeds in accelerating geometry preparation and result post-processing, and in removing unphysical results inherent to MCNP’s conventional approach. The ITER Electron Cyclotron Upper Launcher (ECUL) port plug is undergoing final design evolutions towards manufacturing, which include engineering efforts to ascertain and manage the radiation conditions during operation and maintenance. Two important aspects of these efforts are discussed here. First, design and analysis activities undertaken to ameliorate the residual radiation field (shut-down dose rate, SDDR), present during maintenance due to activation of components by plasma neutrons, are presented. Several shielding options were proposed and analysed. It was concluded that very significant improvement is achieved by those located within the port plug itself, and that further necessary reductions are only possible by considering changes in the surrounding environment. Second, the nuclear heating profile in the critical blanket shield module (BSM) component was computed. For this part of the work, a pioneering application of the novel MCNP conformal unstructured mesh capability (UM) for both geometry description and results tallying was made, in addition to the conventional approach of MCNP native geometry plus superimposed structured mesh. Different aspects of the results and functionality of both approaches were compared in a variety of conditions. It is concluded that, whilst the UM succeeds in accelerating geometry preparation and post-processing and in removing unphysical results inherent to the conventional approach, the latter still provides a more computationally efficient representation leading to uniform precision. These may be limiting factors for large analysis models such as those needed for ITER systems.