Hydrogen distribution and Passive Autocatalytic Recombiner (PAR) mitigation in a PWR-KWU containment type

The evaluation of Passive Autocatalytic Recombiners (PARS) performance has been foreseen from the EU stress tests in the framework of a complementary and comprehensive review of the safety of the Nuclear Power Plants (NPPs). The study presented in this work analyses the size, location and number of the PARs to minimise the risk arising from a hydrogen release and its distribution in the containment building during a hypothetical severe accident. A detailed 3D model of a PWR-KWU containment type was used for the simulations. The numerical tool is the GOTHIC 8.1 containment code, which can model certain aspects of the system geometry and behaviour in more detail than typically considered in containment performance analysis. The severe accident scenario chosen is a fast release of hydrogen-steam mixture from hot leg creep rupture during SBO (Station Black-Out) accident. In the first place, the hydrogen preferential pathways and points of hydrogen accumulation were studied and identified starting from the base case scenario without any mitigation measure. Secondly, a configuration of PARs was simulated under the same conditions of the unmitigated case. The number of PARs considered is 40 units distributed all over the containment building. The PAR configuration offered an improvement in the chosen accidental scenario, decreasing the possibility of hydrogen combustion in all the containment compartments at the end of the transient. The analyses showed that this PAR configuration could lead to a reduction between 30-45% of the final hydrogen concentration. The hydrogen combustion risk is decreased with final hydrogen concentration values below the flammability limit (hydrogen concentration below 7%). Nonetheless, the analysis showed the inability of the PARs to recombine in the early stage of the fast release (the first 1-2 min in this sequence), due to their inertia and occurrence of oxygen starvation conditions. (C) 2017 Elsevier Ltd. All rights reserved.