Descriptif du sujet
During a severe accident like a Loss of Coolant Accident (LOCA) in a Nuclear Power
Plant considerable amount of hydrogen and carbon monoxide may be produced due to various chemical reactions inside and outside the reactor vessel. The local or global
concentration of hydrogen inside the containment may increase enough to ignitie and
induce combustion pressure loads which may threaten the reactor containment integrity. Consequently, several European nuclear reactors have been equiped with Passive Autocatalytic Recombiners (PARs) in order to keep the local molar fraction of hydrogen below 10 % to avoid flame acceleration, and the global molar fraction of hydrogen below 8 % to avoid excessive pressure in case of combustion (French criteria for safety).
PARs convert hydrogen and carbon monoxide respectively to steam and carbon dioxide with catalytic surface reactions. The heat release due to the exothermic reactions on the platinum coated catalytic plates develops a self-feeding by upward natural convection inside the chimney of the recombiner, leading to a very efficient recombination of hydrogen. However, high temperatures on the catalytic plates may lead to gas-phase ignition of the mixture close to the PARs. This local ignition may spread outside the PAR-box causing possible damages to reactor containment safety components.
The PhD aims first at identifying the key phenomena leading to gas-phase ignition, and then at avoiding this ignition, typically by limiting the catalyst temperature of the recombiner, while preserving the efficient hydrogen conversion, a sufficient buoyancy-driven flow, and the robustness of the recombiner against catalyst deactivation. This work is done in collaboration between Severe Accidents Department at IRSN in France and Institute for Nuclear Waste and Reactor Safety, Forschungsentrum Juelich (FZJ) in Germany.
The experimental studies to better understand the fundamental mechanisms of gasphase ignition on catalytic surfaces will be carried out in the hydrogen laboratory at
Forschungszentrum Juelich. Simultaneously, design and interpretation calculations will
be made within the REKO-DIREKT code (FZJ) and the SPARK code (IRSN). Furthermore,
means will be proposed to avoid PAR gas-phase ignition, and experimental demonstrations will be made on the new PAR design. A comprehensive report will be submitted at the end of the PhD summarizing all the results.