Severe reactor pressure vessel failure

In the design of pressure vessels safety is the primary consideration, especially for nuclear reactor pressure vessels, due the potential impact of a possible severe accident. In general however, the design is a compromise between consideration of economics and safety. The possible risks of a given mode of failure and its consequences are balanced against the effort required for its prevention the resulting design should achieve an adequate standard of safety at minimum cost. 

The purpose of the present work was to investigate accident sequaices in a typical BWR with MARK-I containment which may lead to the heatup, failure, and subsequent melting, of the steam separator/dryer complex inside the reactor pressure vessel. The analysis was performed using the APRIL.MOD3 severe accident computer code. In addition, scoping calculations were performed to assess the effect of heat radiation to the upper structures on the heatup and melting of these structures, and to verify the APRIL results. 

Since gross failure of the reactor pressure vessel would result in severe core damage, special consideration should be given to ensuring that there is an extremely low probability of such a failure. To design the vessel in accordance with established codes and safety standards is one of the approaches to making such a failure extremely unlikely. 

To reduce the risk of container failure, the pressure vessels are equipped with several safety features. These can include an effective self-venting system where unforeseen overpressure is released by a quick open-resealing step, or the use of safety disks which rupture when their pressure limit is reached. The small vials (0.2-20 mL) of some monomode reactors are protected by the pressure limit (20 bar) of the caps used, which is significantly lower than the operating limit of the vials themselves (40-50 bar). 

A severe flow blockage >90% of the channel flow area is required to cause pressure tube failure due to overheating. A single channel event leading to channel failure also requires analysis of the pressure transient within the calandria, to show that the calandria vessel itself remains intact, that the SDS devices within it can still perform their fimction, and the break does not propagate by causing failures of other reactor channels. 

The emphasis in the VBER-150 design is on incorporation of the inherent safety features to ensure passive reactor shutdown, to limit pressure, temperature, coolant heating rate and energy release in accidents, to reduce the scope of failures leading to depressurization of the primary circuit, to reduce the outflow rate, and to maintain the reactor vessel integrity in severe accidents. 

The risk posed by severe LWR accidents is considered to be dominated by transient and small-break loss-of-coolant accident sequences in which the core is uncovered only after a prolonged boiloff of reactor coolant. The discussions presented in this module presume, for the most part, that the reactor vessel is pressurized. However, the potential for temperature-induced failures of the reactor coolant system pressure boundary is addressed. In addition, the discussion presumes that reactor shutdown (scram) successfully terminates the fission process, so that decay heat drives the core-damage process. Most of the processes discussed in the context of pressurized, decay-heat driven accidents would exist in unpressurized and/or ATWS sequences as well although such sequences would differ in timing, rates and extent of core heating and oxidation, thermal-hydraulic conditions including the presence of water in the lower plenum, and other factors. 

The Mark-I liner attack problem is well known and it will not be elaborated here. Voy briefly, it is concerned with the possibility that the molten corium released from the reactor vessel (in an unmitigated severe accident in a BWR with a Mark-I pressure siq>ptession containment) will come in contact and cause a breach in die containment liner. This liner is the containment pressure boundary, and such a breach would constitute an "early containment failure. The situation is illustrated in Figure 1. The important phenomenology is sketched in Figure 2, which also notes the k terminology enq>loyed in such discussions. The object is to detomine the likelihood of such a liner breach (conditional on die occurrence of an unmitigated severe accident), and especially to consider tte effect of flooding (by wat ) of the drywell floor. 

---