Application of the SERPENT code to neutronic analyses of the MYRRHA core: a sensitivity and uncertainty approach

In 80 years of nuclear reactor history two types of reactors were constructed, one leading to energy production and the other one for scientific research. At the moment, the operational nuclear reactors are represented by 440 nuclear power plants and 220 research reactors, therefore it can be easily deducted the high relevance that the scientific community reserves for research reactors. In the early days of the nuclear history, research reactors were used to product fissile isotopes for weapon applications. Nowadays they have several utilities in the medical field, as the production of radionuclides for cancer treatments, and in the industrial field for the recycling of nuclear waste and material testing applications. The Multipurpose hYbrid Research Reactor for High-tech Applications (MYRRHA) designed at SCK-CEN in Belgium is one of the promising project in this areas. This reactor is committed to fulfill the criteria of the future GEN-IV reactor type for what concerns safety, proliferation resistance and sustainability. The innovative features of such reactor are the mixed oxide fuel (MOX), the lead-bismuth eutectic (LBE) as coolant and spallation target, and the possibility to operate as critical reactor and subcritical system driven by a spallation neutron source of a proton accelerator, mode called Accelerator Driven System (ADS). In this thesis, all the analyses were performed on the MYRRHA version 1.6 while operating in critical mode, since the main safety requirements concern the criticality operation. The Monte Carlo code chosen for the evaluation of such safety parameters was SERPENT, a modern code with several interesting features. This code was applied in particular for several void injection analyses due to different accidental scenarios leading to distinct reactivity effects. In addition, sensitivity calculations were conducted in order to prove the SERPENT capabilities and to assess which neutron induced reaction (and from which nuclide) has the greatest effect on different parameters, as the multiplication factor and the kinetic parameters. The simulations highlight a reactivity enhancement due to the void presence, an increase affected either by the volume of the void and the spatial location of the injection. In order to understand the physics behind the reactivity increase, local and global analyses were performed. Local analyses such as flux and reaction rates tallying for different reactions and nuclides. Global analyses such as sensitivity calculations, in which the nominal condition and the void injection condition leading to the highest reactivity increase was analyzed. Moreover, sensitivity analyses were performed to estimate the uncertainties in the main reactor parameters and to identify nuclear data shortcomings. This thesis compares the SERPENT results for MYRRHA basic parameters respect to the validated MCNP code, investigates and benchmarks the SERPENT capabilities for sensitivity calculations and analyses different void accidental scenarios. For the nature of such calculations, they can be used as an addition to previous safety results obtained with other codes for the same MYRRHA model and for the future ones. They can be also used as new tools for analyzing void injections thanks to the easy SERPENT geometry modeling routine and for sensitivity analyses, especially on kinetic parameters without the needs of core homogenization.