Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

An analysis of a hypothetical loss of coolant accident (LOCA) in a pool-type research reactor is presented. The study was implemented for the Israel Research Reactor 1 (IRR-1), which is a 5MW reactor using highly enriched MTR-type fuel plates reflected by Graphite elements. The reactor core is cooled by downward forced flow of light water during normal operation and by upward natural convection flow through a safety flapper valve during shutdown. LOCA in pool-type research reactors may be initiated by various incidents such as ruptures and leakages from pipes and valves in the primary cooling system, ruptures of beam tubes or cracking of the pool wall caused by, e.g., strong earthquakes. Each one of these scenarios results in a rapid drop of the pool water level after reactor SCRAM. If water flow through the break persists, the core could eventually uncover completely and be exposed to the ambient air. The present study analyzes the possibility of passively cooling an exposed reactor core by thermal radiation and natural convection to air. The core uncover time is estimated by conservatively assuming that the LOCA was initiated by a guillotine break of a 10 inch outlet cooling pipe at the bottom of the pool, causing the core to uncover about 20 min after reactor SCRAM. Longer uncover times were used for parametric comparison. Since the Graphite reflector elements surrounded the core are typically solid that do not generate heat, they have the potential to act as a heat sink. The effect of the reflector on the core cooling was studied by comparing the total heat transfer from the core with and without considering the thermal contact between the core and the Graphite reflector elements. It is shown that for an uncover time of 20 min the core could reach its melting point if thermal contact with the Graphite is neglected. On the other hand, considering perfect thermal contact between the core and the Graphite reflector, the core temperature is predicted to remain indefinitely below the clad melting point (580 oC). The decay heat generation rate after reactor shutdown plays an important role in the analysis of LOCA. Several empirical correlations and theoretical models are available for predicting the decay heat after shutdown of a continuously operating power reactor. These correlations could not be simply applied for research reactors that work intermittently. A conservative decay heat generation curve was, therefore, estimated by comparing numerical results obtained by the BGCore computer code with available semi-empirical fitting functions and the ANS 5.1 standard curves. It has been shown that the BGCore computer code predict the decay heat generation rate with a small deviation from the corresponding semi-empirical functions results and the ANS 5.1 standard curves.