«Item 7b Severe Accidents Related Issues Preliminary Monitoring Report Report to the Federal Ministry of Agriculture, Forestry, Environment and Water ...»
Was credit taken for corium cooling with thin layer of water?
8.6.6 What is the composition and What are the dimensions of the concrete, including reinforcing material, in the containment floor between Elevation +10.8 meters and the floor level outside the reactor cavity?
8.6.7 What is the composition and what are the dimensions of the concrete, including reinforcing material, in the containment in the reactor cavity floor between Elevation +13.2 meters and the top of the reactor cavity floor?
8.6.7a What is the composition and what are the dimensions of the concrete, including reinforcing material, in the containment in the cylinder wall between Elevation +10.8 meters and the top of the cylinder wall?
8.6.7b What is the composition and what are the dimensions of the concrete, including reinforcing material, in the containment for the interior walls in the reactor cavity in the area above Elevation +13.2 meters?
8.6.8 There is a heavy steel door between the reactor cavity and the lower part of the containment. Do you envision operating Temelín with this door normally open?
If not, what physical conditions (e.g., pressure, temperature, etc.) are required for force open the reactor cavity door if it is closed at the start of an accident and on what basis these conditions were determined (e.g., engineering judgement, stress calculations, etc.)?
8.6.9 Are there any means of opening the reactor cavity door without entering the containment?
If not, under what conditions, if any, to do the Temelín SAMGs foresee a containment entry for the purpose of manual opening of the reactor cavity door?
What are the instrumentation, alarm, phenomenological, or other cues to the operators to accomplish this action, and how much time is available from the occurrence of these cues in order to accomplish this action?
From the time of entry into the outer-most door into the containment, until the action is completed and exit from the outer-most door from the containment is completed, how much time is required?
8.6.10 How do the Temelín SAMGs consider actions after RPV melt-through to protect the local compartments in that area?
What guidance is available on actions that should be taken if this is not successful?
116 ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues State-of-the-art requirements and practices The problems of molten corium-concrete interaction were not given high attention in the design of NPPs of the vintage similar to Temelín (early 1980s), and in the reviews of NPP safety even in recent years. For example, in the IAEA TECDOC on Generic issues for NPPs with light water reactors [TECDOC 1044] the problem of basemat protection is not mentioned at all, even in the issue of “containment integrity during severe accidents”, which deals mostly with hydrogen control measures and containment venting. Core melt-concrete interaction is mentioned there, but only in the context of hydrogen generation which can lead to combustible gas mixture. It is only in the design of NPPs built in the last decade that this issue has been addressed.
The recent WENRA pilot study on harmonization of nuclear safety in WENRA countries in the chapter on ”Instrumentation and hardware provisions for SAM” includes the goal that ”Means shall be provided to prevent containment melt-through“, as one of the reference levels that reflect the highest quartile of existing national requirements [WENRA 03]. The study has indicated however, that ”none of the WENRA countries totally complies with the reference levels“, and adds that when ”harmonisation measures are not judged to be reasonably practicable, an extended analysis of possible compensatory measures is recommended“.
The resolution of the problem has not been yet reached due among other reasons to inherent difficulties in experimental simulation of high temperature chemical reactions of molten corium with internal heat sources in large-scale installations. In the past decades much experimental work was done to learn the relationships influencing MCCI processes. The aspects studied on various experimental stands were the composition and temperature of molten corium, composition of concrete and influence of rebars in the basemat, and possibilities of top cooling of corium with water. In the US these experiments were used to validate the code CORCON [ERI 93], which is presently used as the module for MCCI calculations in MELCOR, and in European Union the validation was done for WECHSL code, used as MCCI calculating module in ESCADRE code system [TACIS 02].
Among the open questions is the issue of whether covering the basemat with water will be enough to stop corium-concrete interaction and eventual concrete penetration, or not. The experiments performed so far suggest that after corium quenching with water a thin oxide crust develops which prevents further cooling of the remaining liquid phase of corium. However, taking into account large dimensions of the corium pool it is expected that the crust will break, improving cooling of the molten corium layers beneath.
The experiments performed within the ECOSTAR programme at FZR Karlsruhe showed that the development of an oxidic crust prevents effective cooling by top flooding [Steinwarz 01, EUR20281]. Also in a large-scale test COMETPC-H4 the melt formed a surface crust under the water and prevented fragmentation and cooling of the liquid melt underneath. Bottom cooling seems to be promising, however it is technically not feasible to be installed in existing NPPs and therefore not envisaged. Large-scale 2-D experiments are planned within EU research programmes, but their results have not been published yet.
Thus, according to the present day state-of-the-art, it cannot be demonstrated that the strategy of covering reactor cavity with water and corium spreading will be sufficient to prevent basemat penetration.
Current plant status The main strategy of Temelín NPP to prevent basemat penetration consists in filling up the reactor cavity with water before the RPV failure and opening the cavity door to facilitate corium spreading over an area of about 100 m2. This is in accordance with recommendations of EU TSO [EUR 20163] that the spreading area should be at least 0,02 m2/MW of reactor thermal power, which in the case of Temelín would yield an area of 60 m2.
ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues 117 Both Czech calculations and PN7 analysis showed that corium spreading would significantly slow down the rate of basemat penetration. In the case of LB LOCA the time to basemat failure – without taking account of the water layer – was evaluated in PN7 as 48 hours without spreading and 74 hours with spreading over 100 m2. Similar values were obtained in Czech calculations [Pazdera 03].
The analysis of water sources available in the plant and of the possibilities of flooding the reactor cavity has been made and the corresponding curve showing the achievable level of water has been established. It has been shown that the level of water can maximally reach up to the bottom of the RPV, but not to cool down the molten corium inside the RPV. Czech specialists estimate that the presence of water at the moment of RPV break would significantly reduce the initial MCCI rate [Sykora 03]. However, no details of the calculations or of underlying relationships considering the influence of water and slowing down or stopping the MCCI have been provided.
The Czech side provided a document indicating the composition of basemat concrete after the Prague Workshop. It was stated that the base concrete is the same in the layer below the reactor cavity and outside, below the equipment room. It is characterized by the absence of carbon compounds, which significantly reduces the amount of gases produced due to MCCI.
The rates of concrete penetration in vertical and horizontal direction are calculated by CORCON module, which is a part of the MELCOR code system, validated against several experiments [ERI 93]. The calculations of basemat penetration made initially by PN7 team were repeated for updated concrete composition after the Prague Workshop [Sart 03b] and provided results comparable to those quoted by Czech side for the concrete penetration without influence of overlaying water layer. However, no stopping of concrete penetration was found under any conditions.
The technical problems related to spreading of corium are to be resolved by the Temelín NPP within the preparation of SAMG implementation in the plant. In order to ensure opening of the cavity door, a remote control device will be installed and the door is to be opened early in the accident sequence, much before the moment of RPV failure. Two movable barriers will be installed in the neighbouring containment room to prevent molten corium flowing to the equipment hatch (which would fail the containment integrity) and to protect the area from missiles that can be thrown from the hermetic door after RPV break.
An analysis of molten corium progression in the instrumentation channels has been performed and it was shown that the corium would quickly freeze and would not be able to flow down the channels [Kujal 03]. In addition, special protection of the channels is planned. Thus the general features of the strategy to prevent basemat penetration are defined and look feasible.
In answer to questions, NPP Temelín experts acknowledged that there is a possibility of improving leak tightness of the rooms below the containment basemat and sealing them off in case of a severe accident involving basemat melt through. The provision of leak tightness of rooms adjacent to reactor containment is a recognized measure, similar to the measures recommended already in Rasmussen report [WASH 1400] where it was required that any possible leakages bypassing the containment (V-sequences) should be captured in closed rooms and not released directly to atmosphere. This strategy however, has not been studied in Temelín in detail yet. Temelín NPP stated that it prefers to concentrate on prevention, and
this prevention includes:
- Prevention of DBAs from developing into BDBAs by following so EOPs,
- Prevention of RPV failure by RCS depressurization and water injection,
- Prevention of basemat penetration by covering the floor of reactor cavity with water and corium spreading.
The publications of Czech specialists include statements that the basemat integrity can be kept even after RPV break, for example in Ref. [Sykora 01], there is a statement that with delayed ECCS recovery the depth of concrete ablation after 300 000 seconds (~ 3,5 days) will 118 ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues be only 1,4 m. However, in view of the uncertainties connected with the top water-cooling of molten corium pools the issue cannot be judged closed.
The situation in Temelín is more critical than in a typical PWR NPP, because the basemat is situated about 13 m above the ground, and in case of its penetration the containment atmosphere will get in direct contact with the outside atmosphere. This can result in air mixing with the hydrogen accumulated inside the containment, which could not burn in view of inerted atmosphere, but will be able to burn or even detonate once it gets in contact with the free air, either outside or inside the containment. The analyses of the consequences of such a burn are not known so far, but should be a part of emergency planning.
Evaluation While the measures already implemented and being planned by Temelín NPP go in the right direction, the PN7 team considers that they do not assure protection of the basemat if RPV failure occurs. The probability of RPV failure is small, as shown by recent PSA, but it exists.
After RPV failure there are no visible means of stopping the basemat penetration completely, and the planned measures assure only slowing down of the process. Therefore, additional measures aimed at improving leak tightness of rooms below the containment basemat should be considered.
Inasmuch as the SAMG strategies include a capability for filtered venting, the potential for using filtered venting in connection with mitigating basemat failure could be considered. Use of filtered venting before basemat failure would reduce the containment pressure, so that once the basemat fails it would be less likely that the reactor building structure would fail due to overpressure. In addition, venting would reduce the amount of hydrogen in the containment atmosphere; also making it less likely that hydrogen combustion in the reactor building after basemat failure would result in reactor building structural failure. This approach could help preserve the reactor building as an independent fission product barrier (or at least as a function of hold up and attenuation of the source term before release to the environment).
Further monitoring should cover the implementation of technical means for:
- Timely opening the rector cavity door before the RPV break,
- Installation of removable shielding walls to restrict molten corium pool area and protect the containment liner and other barriers against molten corium and missiles hazards,
- Protection of containment penetrations and reactor cavity instrumentation channels against molten corium penetration.
The monitoring should also include the new set of analyses to be performed for Temelín NPP scheduled for the end of 2003, in particular the evaluation of possibilities to stop the MCCI, the estimates of possibilities of hydrogen burn in case of basemat failure and resulting revolatilization of fission products deposited inside the containment, and the effectiveness of possible reduction of leakages through the rooms under the containment basemat.
The hazards due to radioactive releases in case of basemat melt-through are much smaller than in the case of early containment break, as discussed in section 5.6 below. As shown in TACIS programme, the mass of radioactive aerosols still suspended in the containment atmosphere is dramatically smaller than the mass released from the core to the containment and available for release in case of an early containment failure. The radiological hazards are correspondingly reduced.