«Item 7b Severe Accidents Related Issues Preliminary Monitoring Report Report to the Federal Ministry of Agriculture, Forestry, Environment and Water ...»
Experiments have been conducted on the outcome of such conditions as part of the review process for the Combustion Engineering System 80+ design certification process. The experimental setup included the presence of hydrogen igniters, which ensured constant availability of hydrogen ignition sources. Eight tests were conducted under well-mixed conditions and three under stratification conditions. The tests were conducted with sufficient hydrogen to represent an average concentration of 13,5% under dry conditions. In the eleven tests, no detonations were observed. All tests ended in multiple deflagrations or a single deflagration, which in no case was threatening to containment integrity [Blanchat 97]. While there are differences in geometry between the experimental setup and the WWER 1000 design, the experiments suggest that detonations are not very likely outcomes when a steam-inerted, hydrogen-containing atmosphere is subjected to containment spray. These experiments lend credence to the modeling of hydrogen combustion under steam-inerted, containment spray recovery conditions as a deflagration or series of deflagrations, depending on the conditions in the various containment compartments.
Calculations of this situation with MELCOR within PN7 (for a LOFW rather than station blackout, however) suggest that the resulting pressure increase from deflagration under rapidly deinerting conditions are below or slightly above the design pressure for the containment (that is, less than 0,49 MPa), and therefore should not pose a threat to containment integrity (the minimum containment failure pressure, corresponding to the 5th percentile of the failure pressure curve, is 0,8 MPa).
4.2.3 Other Accident Sequences
Besides blackout, other accident sequences involving simultaneous beyond design basis failures in NPP operating systems and safety systems can result in loss of heat removal capability from the core. Usually considered BDBAs are SB LOCAs with simultaneous loss of ECCS and accidents initiated by loss of all feed water to steam generators (LOFW), including auxiliary feed water system and emergency feed water systems (EFWS).
According to recent French estimates, the cases with maximum hydrogen releases are those for SB LOCA (25 to 75 mm diameter break) with CSS available but with no Safety Injection, and LOFW with CSS but no Safety Injection. Within PN7 project both these types of scenarios were calculated, taking into account both in-vessel and ex-vessel phases and considering several possible SB LOCA break sizes. In all transient sequences excluding LB LOCA the question of effective depressurization of the RCS is of primary importance, just as it is for blackout sequences. For checking the plant ability to depressurize the RCS before the RPV break, the sequences of SB LOCA were chosen to have the size of the break at the lower limit of break spectrum, namely 15 mm.
ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues 89 22.214.171.124 Total Loss of Feed Water scenarios The purpose of the analysis of a total loss of FW was to explore SAM action – primary depressurization to mitigate and if possible – prevent core degradation. In case of LOFW there are two possibilities of RCS depressurization – by opening PRZ PORV or by using emergency gas removal (EGR) system. Both systems are safety graded and operable from the control room. The opening of the PORV will result in a rapid increase of pressure in the relief tank, failure of the rupture disks and LOCA to the containment, which is not acceptable for an operational transient. On the other hand, the gases removed by EGR are directed to the gas purification system, so they remain contained and do not contaminate the containment atmosphere. Therefore, using the EGR system is the better solution.
AM procedure investigated was a primary depressurization with the gas evacuation (emergency gas removal) lines from the top of the RPV and from the pressurizer. The scenario was run with availability of all ECCS systems. The criterion for initiation of the SAM procedure was core exit temperature exceeding 650 °C. The analysis assumed total loss of FW (both normal and emergency) and thus – loss of heat sink. Make-up system was also assumed to be lost. All 3 HPIS trains, 4 SITs and all 3 LPIS trains were available. To establish water injection to the primary system it is necessary to achieve depressurization down to the HPIS injection pressure.
Three cases were analyzed:
- Scenario with depressurization using BRU-A as specified in EOP (0,5 h after the accident), then depressurization with gas evacuation system (RPV+PZR) and PZR PORV,
- Scenarion without opening BRU-A, depressurization with gas evacuation system (RPV+PZR) and PZR PORV,
- Scenario with loss of all active ECCS, no depressurization.
Initial cycling of BRU-A till 30 minutes depletes significantly the SGs secondary inventory (to 1 m below the initial level). The forced opening of BRU-A between 30 minutes and 60 minutes reduces very fast the SG secondary coolant level by another 1 m in about 8÷10 minutes. Thereafter, SGs do not influence the primary pressure response due to loss of the secondary inventory and loss of primary-to-secondary heat transfer.
Shortly after starting the depressurization with BRU-A (at time 0,53 h) the primary pressure is reduced for a short period of time below the shut-off head of the HPIS pumps and there is injection for a short period of about 480 seconds.
Thereafter the primary pressure increases due to the loss of heat sink. At time 2,13 h the active core starts to uncover. Core fully uncovers in 865 s and stays uncovered for almost 1300 s. The operator starts primary depressurization by using Emergency Gas Removal system and PORV at 2,36 h. At time 2,55 h as a result of the primary depressurization by the operator the HPIS starts steady delivery of water to the primary system.
Peak cladding temperature reaches 1540 K at time 2,6 h after the accident - fuel heatup is sufficiently high to fail the cladding. Gap release starts at time 2,58 h after the accident and fuel cladding tightness is lost in all fuel elements.
In 2200 s the primary pressure goes down to 60 bar - SITs start injection of water to the primary system at 2,84 h. Injection from SITs goes for about 540 s. At around 2,51 h and 2,73 h there are observed spikes in the primary pressure due to the sharp evaporation and increase of the steam in the primary system by the heated fuel in the reactor core and at the same time - injection of water to the primary system. At time 276 h the core is covered again and the fuel is cooled down.
The total amount of oxidized Zr is 121,6 kg, representing 0,454 % of the total mass of zirconium in the corewith production of 5,4 kg of hydrogen. After depressurization of the primary system the injection of water by the HPIS brings the reactor into a stable cooled down state and the accident is successfully managed.
90 ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues In the case without BRU-A opening the sequence is similar but the timings are slightly different and the amount of Zr oxidized is smaller. Peak cladding temperature reaches 1400 K at time 3,54 h after the accident - about 1 h later than for the case with opening of BRU-A. The heatup of the fuel is sufficiently high to fail the cladding - gap release starts at time 3,48 h after the accident. However, the fuel does not melt and the releases of fission products from fuel pellets are small.
H2 generation for this case is 1,41 kg as opposed to 5,38 kg of hydrogen for the case with the opening of BRU-A. The peak of the fuel temperature is about 50 minutes earlier and the peak value is 140 K higher for the case with opening of BRU-A. The peak fuel temperature for the case with BRU-A is 1540 K, and for the case without opening of BRU-A – 1400 K.The calculations showed that in case of entering SAMGs at core exit temperature exceeding 650 °C and using EGR and PORV the accident can be successfully managed without core melt.
The third case calculated was that of LOFW with loss of all active ECCS systems, and no SAM actions. In such a case the sequence leads to core melt and RPV bottom head failure under full pressure. Total H2 generated in-vessel is 544 kg, bottom head is penetrated at 6,8 h, and total corium mass ejected is 147 t.
For this case – LOFW without SAM actions – two sets of calculations were performed, the first for generic data on basemat concrete composition, the second for updated concrete composition, which was made known to the Austrian side after the Prague Workshop. The difference consisted mostly in the fact that the concrete used in Temelín for the basemat has no carbon content and so would not produce noncondensible gases CO and CO2 during MCCI. Moreover, reduced PAR capacity was assumed, taken according to the latest data of the PAR producer.
The results for the updated concrete composition showed that in the case without SAM actions the upper layer of serpentinite concrete would be penetrated after 16 hours since the beginning of the accident with the average axial penetration rate of 7,6 cm/h, and the amount of hydrogen generated due to MCCI in serpentinite concrete would be about 850 kg. After that the molten corium would attack the lower base concrete layer. The rate of axial penetration would be 11,8 cm/h and the concrete basemat would be penetrated within about 42 hours after the start of the accident. The amount of hydrogen generated due to MCCI during base concrete penetration would be about 1100 kg.
The calculations covered also evaluation of containment pressure and hydrogen concentration during the accident, assuming division of containment into 11 control volumes with uniform gas concentration in each volume. The pressure peaks in the early phase after the RPV break in the reactor cavity were reaching 0,4 MPa, and in the long term the maximum pressure in the containment was kept at about 0,38 MPa, so under the design strength of the containment. Due to large steam release at the moment of RPV failure the atmosphere in the containment would be inerted. Later on the fraction of steam decreases and at about 16 hours there will be a possibility for hydrogen deflagration in the containment. In the long term the oxygen depletion in the result of hydrogen recombination in PARs will bring the steamhydrogen-air mixture to inerted conditions, but this process is slow due to low capacity of PARs and the depletion threshold is reached after about 35 hours.
In the reactor cavity the concentration of H2 is high, but after penetration of the RPV bottom head a large amount of water from SITs enters the cavity. The evaporation of this water increases the steam concentration up to almost 100 % and this inert condition inside the cavity continues for the first 7 hours of the MCCI. After inertization of the cavity the cavity atmosphere has insufficient O2 for deflagration. Later on, the depletion of oxygen provides for inerted conditions. The peak recombination rate of one PAR is about 0,18 g/s for the H2 and 0,1 g/s for the CO. The H2 mass recombined until 31 hours of the accident in one PAR situated in the optimum position is about 14,3 kg and that of CO – 3,34 kg. Of the total amount of ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues 91 hydrogen produced in-vessel and ex-vessel about 315 kg H2 (11,5 %) are recombined within the time covered by the calculation. The rest of combustible gases stays inside the containment.
The penetration of the bottom head of the RPV creates very high steam concentration also inside the containment compartments. In fact after the onset of MCCI the containment atmosphere is inert. 16÷17 hours after the accident the H2 concentration exceeds the critical value of 10 % where self-deflagration is possible. The steam and gas distribution inside the containment is not fully uniform. There is significant steam and H2 stratification predicted by the MELCOR code. The containment atmosphere is still inert by the high steam concentration. The condensation of the steam on the containment walls slowly decreases the steam concentration and 17 hours after the accident the steam concentration in the containment dome goes down below 55 %. The O2 concentration is sufficient for deflagration (above 5 %).
All this provides conditions for deflagration of the hydrogen. The operation of the recombiners slowly reduces the amount of O2 in the containment. Due to the small capacity of the recombiners the reduction below the minimal 5 %, at which deflagration is still possible, is not earlier than 34,8 hours after the accident.
The steam condensation inside the ECCS compartment is faster. This compartment is not inert during all of the accident - steam concentration is below 55 vol.%. At 15 hours after the accident the H2 concentration in the ECCS compartment reaches 10 vol.%. This creates conditions for deflagrations inside it. During the next 6 hours multiple deflagrations are observed in this compartment. The temperature peaks during the deflagrations are about 680÷690 OC. The pressure oscillations in the containment are very mild because the deflagrations do not propagate to other larger compartments due to the inert conditions in the rest of the containment. These multiple deflagrations slowly deplete the oxygen in the containment. Towards 35 hours after the accident the O2 volumetric concentration in the dome is below 3 vol.%. The steam concentration is about 51,3 vol.% and the H2 concentration is above 13,5 vol.%. Due to the lack of sufficient oxygen there are no more deflagrations.
Although there is no need for spraying because the containment pressure is low, a calculations run was made assuming that the spray system is started to see what would be the impact on the containment integrity if deflagration occurs at this moment.
The spray condenses rapidly the steam in the dome and in the SG boxes, so that the concentration of steam is reduced to 43 vol.%. This results in an increase of the O2 concentration above 5 vol.%. The H2 concentration is also increased reaching 16,3 vol.%. This creates a large deflagration inside the containment, with the pressure peak of 5,40 bar. The temperature reaches 757 OC. After this major deflagration the oxygen is depleted below 2,1 vol.%.
The containment parameters during that deflagration do not exceed the lower limit of the containment strength, so there is no danger of containment early failure.