Potential influence of core/concrete interactions on PWR containment pressurization

For certain postulated accidents beyond the present design basis for LWRs it has been shown that the potential exists to fail the containment building as a result of extensive core debris-concrete interactions. The MARCH code was developed at BCL to analyse the response of a LWR to core meltdown accidents. MARCH has been found to be nonconservative with regard to predicting core-concrete interactions in the following areas: (1) energy is not conserved, i.e. heat transfer from the top of the molten core is lost. It is not added to the atmosphere or used to heat-up and degas concrete. (2) MARCH uses INTER to model core-concrete interactions. INTER models only a molten pool attacking concrete. The core materials rapidly freeze; this limits the applicability of the model. This paper removes the above nonconservation in MARCH to more realistically determine the potential for containment pressurization and basemat penetration resulting from extensive core-concrete interactions. Large amounts of combustible gases are also released during core-concrete interactions; however this paper is primarily concerned with the partial pressure exerted by these gases on containment. It therefore considers accidents without the containment heat removal systems (CHRS) operating so that burning of the combustibles is prevented by lowmore » oxygen concentration.« less