Nuclear fission

9.3.2An Example from German PWR’s

GRS has performed several accident analyses with MELCOR 1.8.6 for shutdown states, part of it with open RPV head. For example, one simulation assumed a station black out in a PWR when RPV and spent fuel pool are connected, and when the decay heat is low. Another simulation assumed failure of heat removal and RPV flooding rather early in the shutdown regime with low (mid loop) coolant inventory in a PWR.
Based on a rather detailed interpretation of the MELCOR analyses the following conclusions have been drawn:

• 
  Availability of water resources can of course stop or at least slow down the core melt progress. For example, water from the spent fuel pool could partially be used for that purpose. However, the final containment pressure will be increased with increasing water resources. Consequently, loads to the containment and to the containment venting system may be higher than in the original analyses for core melt analyses from full power mode. In addition, a better representation of the containment sump and MCCI in the most recent MELCOR analyses seems to indicate higher pressure build-up in the containment than former analyses.
  
• 
  Once the core melt process has begun, the time until relocation into the lower RPV plenum will be a few hours. If it has not been possible to install successful preventive SAM before core melt, it seems rather unlikely to perform successful mitigative SAM in the remaining short time under less favourable conditions.
  
• 
  There is concern that the containment steel shell above the open RPV might be subject to significantly elevated temperature. The MELCOR runs confirmed this; the containment reaches temperature up to 640 K. However the temperature is still low enough for sufficient mechanical strength. Nevertheless it is recommended to precisely evaluate temperature and structural properties above core melt in an open RPV.
  
• 
  It is conceivable that air from above enters the open RPV, changing conditions for core melt. However, the MELCOR analyses showed, first, that the atmosphere above the RPV is almost pure steam (the air has been removed), and secondly that the containment atmosphere hardly moves downward into the hot steaming core. Consequently, the amount of hydrogen produced with open RPV is similar to that with closed RPV, and almost no air-driven oxidation of zirconium has been observed.
  
• 
  With open RPV there is of course a higher release of radionuclides from the core into the containment than with closed RPV. However, later in the accident when the RPV bottom is molten through, the amount of fission products in the containment atmosphere becomes comparable in both cases. Therefore, a significant difference between open and closed RPV head with regard to releases into the environment exists only if there is a very early containment failure, or if the containment isolation has failed.
  

9.3.3An example from Spanish BWR (Mark-III containment)

Like most NPPs, the accidental risk with RPV open in a BWR Mark-III containment is linked to the outage activities. Next are some examples of initiators during this phase:

• 
  drainage into the RHR system with failure of the isolation and loss of the injection systems (initial water level in the top for refuelling activities),
  
• 
  loss of cooling due to power failure without recovery (initial water level to the main steam line).
  

All these scenarios might have core damage in less than 12 h due to the high decay heat power (more than 20 MWt) and the reduced level of water. At any case, the RPV opening is permitted before 24 h since reactor scram due to the capacity and diversity of mitigation systems available, but if the containment were also opening these sequences would be direct contributors to the LERF. Although these initiators have a low probability, to reduce the outside consequences, it is recommended delaying the opening of the containment for refuelling activities, mainly the equipment hatch, until complete the rising level in the vessel.

The progression of the cases in the severe accident phase shows a high concentration of hydrogen in containment before the RPV failure and also an elevated temperature on the containment structures due to high fission products deposition. These elements are a relevant risk for the integrity of the containment if only a fast closure of contention would also be permitted (by procedures). So, the containment mitigation systems capacity would be maintained (ignitors, containment sprays containment heat removal and containment venting) in this phase.
The Mark-III containment is designed with a suppression pool that connects the drywell and wetwell zones and that absorbs most of the core thermal load when RPV is failed. When the RPV is opened, the suppression pool is bypassed and the mitigation systems likes containment sprays might be insufficient to fully absorb this thermal load released. Thus, the most effective severe accident mitigation strategy is one that is performed directly on core damaged into the vessel (i.e. using diesel portable pump) that would require later cooling into the containment (i.e. using CFVS as evaporative cooling). Long times available (more than 5 h) would give credibility to this manual actuation. Sensitivity analyses with the MAAP code show an effective cooling of the core damage, even with a delayed injection at time of failure of the support plate.

9.3.4An example from Swedish NPP’s

In Sweden, the PSA for the low power and shutdown period is performed in the same way as a PSA for the full power operation. The main differences compared to power operation that need to be addressed during the LPSD PSA are:

different operability readiness requirements (availability of systems);

increased risk for disturbances due to operator/human intervention;

a lower residual heat (lower system requirements and more time available for operator actions (recovery of failing systems or use of alternative systems).

The basis provided by the existing power operation PSA in terms of FMEA, system analysis, sequence analysis, and the L2 PSA (including PSA model) is a very important input to the LPSD PSA.
Level 1 End States – Consequences

Compared to full power PSA, more consequences are considered in LPSD PSA [24]. Scenarios which may be considered in Shutdown PSAs are:

core damage or fuel overheating (fuel in-core or ex-core in the spent fuel pool),

partial core damage,

physical (mostly mechanical) fuel damage (e.g. from heavy load drops or fuel handling accidents),

boiling (i.e. risk of a higher radiation level on refuelling floor),

ex-core criticality events and related damage,

radioactive releases without core or fuel damage, e.g. tritium release for reactors moderated with heavy water.

In order to determine if the defined consequences (end states) can occur and when they will occur during LPSD more information about plant attributes (characteristics) is needed. The following list provides examples of such attributes (characteristics) that need to be considered when defining the end states (consequences):

POS decay heat level,

Primary circuit coolant inventory,

Water inventory in SFP and other places where the fuel may be located.

A suitable tool that can be used in the identification process of initiating events is a Master Logic Diagram as the example presented in the figure below.

Figure 9.3. General Master Logic Diagram for Overheating of Fuel [45]

The level 2 PSA sequences begin with the end states as defined in Level 1. Any end state having the potential to release the fission products from the fuel in the core or SFP are transferred to Level 2. The L2 PSA assesses the amount, probability and timing of a release of radioactive substances leaking from the nuclear power plant during severe accidents.

Table 9.3. Plant Operating States (POS) for SPSA [45]

ID	Description	Hours
POS4:1	FW is disconnected, AF is started, RC temperature 175 ºC, One RH train in operation, RC is filled	4
POS4:2	PRZ is full, 2 RH trains in operation (second train started at temperature 120 ºC)	2
POS5:1	RC temperature 93 ºC, The operability requirements - less restrictive, The phase covers the lowering of RC level up to DT5*.	30
POS5*:1	Starts when first SRV is dismantled, The RC is now open, The RPV head is dismantled, The reactor cavity is filled.	25
POS6:1	The reactor cavity is filled, Fuel is transported to the fuel pool	45
POS7	The RPV does not contain any fuel, The SGs may be drained	24 / 96 / 240^*
POS6:2	Re-loading, The reactor cavity is filled, Draining the RC level to RPV flange	60
POS5*:2	The RPV head is mounted, Diesel tests are performed (i.e. in next phase all diesels are available), SI logic and full flow test are performed directly before filling of the RCS Note that two different situations exist namely with or without water in the SGs (primary side)	40
POS5*:3	The RCS is filled, The diesels are available, The SI is fully available	8
POS5:2	The RCS is filled, All SRVs are mounted, Warming up	30
POS4:3	RC temperature above 93 ºC, Continued warming, Bubble in PRZ	12

^* The different times indicated for POS7 are due to different types of refuelling outages.
Table 9.3. Shutdown PSA Level 2 – Example of outage period for a BWR in terms of residual heat

Figure 9.3. SPSA Level 2 a Swedish NPP example (POS definitions)

Shutdown PSA Level 2 – Source terms

Usual definition of source terms is based on:

Noble gases;

Volatile fission products CsI and CsOH;

Non-volatile fission products BaO and MoO2;

Rutenium;

• 
  During power operation conditions no large quantities of Ruthenium can be expected due to the high vaporization temperature of the metal.
  
• 
  If however heating of the core occurs in air, oxidation will occur and Ruthenium oxides vaporize at much lower temperature. Ruthenium is considered a volatile nuclide.
  
• 
  A prerequisite for significant amount of Ruthenium oxides to form is that enough air is available. It is most likely that it is necessary with double openings in RC in order for the air to flow through the core.
  

It is however very difficult to calculate the amount of Ruthenium and there is no model in MAAP that takes this into account.

MAAP analysis

For Shutdown L2 PSA a number of MAAP analysis cases have been run [34]. The purpose of MAAP analysis cases are:

• 
  Calculate released amount of fission products during shutdown conditions depending on containment spray availability.
  
• 
  Analyze importance of lower residual heat and different water level in RC during shutdown compared to power operating conditions.
  
• 
  Verify the assumption that release category calculations for power operating conditions can be used also for shutdown conditions.