Jacques Repussard: knowledge, independence, proximity
AtomInfo.Ru, PUBLISHED May 15, 2012
Jacques Repussard, Director General of IRSN, France, answers questions from the on-line journal AtomInfo.Ru.
Jacques Repussard, credit ©Antoine Devouard/IRSN
Measurements are neutral
Mr Repussard, we know that IRSN is celebrating its 10th anniversary this year. Could you tell us, and our readers, how you see IRSN today and how you envisage its future over the coming years? In our opinion, IRSN is a unique structure in the nuclear sector, not only in France but also in the rest of the world, which is why your Institute's activities generate such a lot of interest.
Doubtless you know that the idea to create IRSN was initially political rather than mainly scientific or technical. Before the creation of IRSN, most of the Institute's knowledge base already existed within the Protection and Nuclear Safety Institute, part of the French Atomic Energy Commission.
The French parliament and government took the political decision, ten years ago, to split off the Institute and transform it into a new public entity that is independent of operators and designers of nuclear facilities.
The Institute is placed under the supervision of the French ministries of the environment, economy, research, defence and health - AtomInfo.Ru note.
The initial idea was that, in order to take its decisions within the framework of procedures for authorising nuclear facilities provided for by the law, ASN, the French Nuclear Safety Authority, should be able to rely on the expert opinion of an independent technical body. This happened ten years ago and IRSN therefore became an independent structure.
We obtained a status that allows us to plan, fund and conduct research in the safety field. In addition, this situation allows us to work in an independent manner, including with regard to contract work we receive from the nuclear industry.
Commissions from industry or instead ASN?
Commissions from industry, whether in France or abroad, only represent a small part of our activity compared to commissions from public authorities. Obviously, when we work in this capacity, we are obliged to follow a strict deontology code and a set of ethical standards and guiding principles. For example, for facilities located in France, we do not accept commissions which would lead us to contribute to the establishment of safety assessments for regulatory purposes.
However, the fact that we are legally authorised to work on industry commissions, and not only on requests from ASN , allows our specialists to strengthen their skills by working directly with sites (reactors and other facilities), thereby improving their knowledge in their respective fields.
But does that not lead to conflicts of interests?
In fact, no. The work we do for our clients acts as a complement to the work we do for the French safety authority. What's more, our clients are not uniquely firms and organisations from the French nuclear industry, but also safety authorities from other counties and their TSOs. We collaborate extensively at an international level.
R&D provides us with the necessary knowledge, while contract work allows us to put into practice the knowledge acquired.
IRSN's founding charter also empowers us with the mission of informing the public of our activities and our results. Which is why, immediately after the Fukushima-Daiichi accident, we released information notes on the events on our website.
At the same time, all of our activities need to be bounded within a very strict framework. First and foremost we have to give recommendations to the safety authority, which is our key mission.
What does all this lead us to? For example, in our research work, we do not allow ourselves to participate in activities developing new 4th generation reactor technologies. In other words, the design of the reactor is not our job. On the other hand, we conduct and will continue to conduct research in the safety field, on all the safety systems of reactor projects that are being developed by other organisations. This will allow us at a later stage to conduct critical analyses of future reactor projects on the basis of the most advanced scientific knowledge.
In our contracts, we do not allow ourselves to participate in work involving the drafting of safety assessments. But, before this step, we can carry out studies and research following commissions, from EDF for example, in scientific areas of common interest, with regard to safety or radiation protection.
We can also carry out contract measurements on behalf of operators. These measurements are neutral. They are neither an evaluation nor a judgment.
When we work for the nuclear industry abroad, we always act in full agreement with the country's safety authority.
In our public information actions, we do not reveal commercial secrets and we do not disclose sensitive information, particularly with regard to specific points on which decisions have not yet been taken.
Thus, IRSN is developing thanks to its status, which makes it possible to take into account all the aspects alluded to above and to make progress. We operate in accordance with internal regulations that strictly define what actions are possible and what actions would run contrary to our ethics.
I personally report to the relevant ministries on the running of the Institute. For example, after the Fukushima-Daiichi accident, the results of our studies were communicated to these ministries and also to the Prime Minister's office, which contributed to defining the orientations taken by the government during the crisis.
The Fukushima crisis
Our next question concerns the Fukushima accident. We have to admit that while the accident was actually taking place, particularly over the first few days, we listened attentively to the information coming from France. We even trusted this information more than the information coming from the IAEA. What was IRSN's role during Fukushima?
There was a conference on Fukushima in Vienna in March at which I was one of the co- chairpersons. TEPCO presented new details on the sequence of events involved in the accident.
At present however, we still do not yet have sufficient knowledge of the sequence of events or the exact times at which successive decisions were taken during the accident. Were these decisions right or wrong, were they taken at the right time or too late, and at what exact instant? None of these points are sufficiently clear at the moment. To date we still have scant precise information on the current condition of the reactors themselves.
I'll tell you how IRSN operated, which enabled us to carry out our analysis and our evaluation. Quite frankly, we did not have specific information, we worked on the data transmitted by TEPCO.
Transmitted via the IAEA or coming from other sources?
Yes, and also the data from TEPCO's website. We were lucky in that several Japanese scientists, who were working in Paris at the time, saw that we were monitoring the accident and were trying to understand what was happening. They spontaneously offered us their help for translations.
Our information came from TEPCO, NISA and the American NRC. They sent us the plans of the power plant, which is important. We do not have BWRs (boiling water reactors) in France and so in order to establish our prognosis we needed to have access to a minimum amount of information on this technology.
For two or three weeks we worked in real time, exactly as in a national crisis exercise. In France, these exercises are organised on a regular basis, once a month.
Even before the Fukushima accident, our engineers were accustomed to working with minimum initial information to establish an in-depth prognosis of the condition of the defence barriers (vessel, primary circuit and confinement). We established a diagnosis and gave a prognosis on the increase in pressure, on what could happen to the temperature and on what could happen over the coming hours.
Obviously, we organised conferences between the experts in different fields - for example, experts of reactor systems or calculation codes such as CATHARE (thermohydraulic code for the analysis of accidental situations and the evaluation of safety).
Experts always have computational tools, which make it possible to obtain in a short period of time the first rough evaluations of the parameters of interest to us. We also have other groups of experts, particularly for the environment, who take into account meteorological data and can calculate the volume and the direction of the radioactive plume.
This data was acquired and as early as the 21st March we were able to publish a first estimation of the release. If you recall, we stated at the time that the rough estimation of the Fukushima releases would be around 10% of that of Chernobyl.
The French were the first to state that the Fukushima accident should be at level 6 of the INES scale.
Yes, France did indeed provide this estimation, but not officially. May I remind you that the official indication of the level of seriousness of an accident concerning a nuclear facility can only be given by the government of the country in which the accident has occurred - and by no-one else.
The INES scale was specially created to be able to inform the public on what has occurred. If every expert starts giving his or her own estimation and level of seriousness, it could quite obviously create confusion in the population.
It goes without saying that IRSN did not indicate what INES scale should be attributed to the Fukushima accident. But the Japanese, who officially announced INES-7, were absolutely right, given the importance of the releases into the environment.
At the very moment when France was evoking INES-6, in Russia it was continued to hope to remain at INES-4. We were impressed that you and your colleagues were able to find your way amongst the events that occurred in Fukushima under conditions that were, to say the least, confused.
You're right. But you know, when the accident was taking place and while there still had not been any significant release, there was always a hope that the accident would remain at level 4.
Do you consider it is necessary to change anything in the information exchange system in the event of a serious accident? We are not accusing our Japanese colleagues, but you have to admit that, over the first few days following the accident, the information provided by TEPCO was not very precise. And the Americans, by giving their recommendation of 50 miles, made a mistake in their estimation of the time it would take for the storage pool of the 4th unit to run dry. In short, there were errors. Do modifications need to be made to the information exchange system to avoid such errors in the future?
It's an important question, to which I'll give several answers.
You evoked the recommendation of "50 miles". 50 miles is perhaps too great a distance, but 20 km may not be enough.
For a short period - of around two days - there were very great fears regarding the condition of the storage pools. The water level was dropping and everyone was aware that the accident could take on a completely different aspect. You yourself, your colleagues from AtomInfo and other nuclear engineers should easily understand what a total loss of water from a storage pool that contains a large quantity of spent fuel would mean!
The American approach was perhaps too conservative, but that's understandable.
What needs to be changed at the crisis management level? There's no simple answer. The Japanese have been heavily criticised. Very heavily! But let's be completely honest here. If, God forbid, such an accident were to occur in France or in Russia, how much information from expert appraisals would be transmitted abroad in real time?
I can't give a straight answer to this question, and I don't think you can either because the main efforts will quite naturally concern national management of the crisis.
It's a complex subject. I think greater preparation for accidents is necessary. The internal conclusions that we have arrived at within IRSN are that we need to add to our organisation an "international relations" crisis group, which would be in permanent contact with the IAEA, would exchange information with other crisis centres and would have a role of providing information. It would discuss the various calculation results underway with GRS colleagues or other counterparts.
But such a group does not exist for the moment. And if something were to happen tomorrow morning, the problem would quite clearly be apparent.
We are going to resolve it, but to do so we need time. We need to train specialists further, we need to equip additional crisis centres and we need to establish robust communications systems with foreign organisations.
What do we have as things stand? We are fully equipped to establish a situational diagnosis and prognosis. We can calculate the source term and we can calculate the direction of the discharge plume, taking into account weather forecasts, and estimate the doses likely to be received by the population or the radiological consequences for farming produce.
We have a permanent collaborative agreement with Meteo France, which provides us with detailed weather forecasts. We also have mobile measurement equipment, which we can send anywhere in France to measure, on the spot, actual contamination levels. These resources exist and are fully operational.
Contamination of air, fine, but do you know how to calculate the dispersion of contamination in the sea? Everyone knows how to calculate (dispersion) in the air, but not (in the) sea - as far as we know.
You're right! There's an obvious lack of knowledge in this area. Fortunately, there are few inhabitants of the sea itself! The problem exists for the food industry - the need to calculate the contamination of fish and other elements in the food chain. This could be the subject of an international code development project to provide answers to these questions.
At present, we can model the dispersion of radionuclides in water and we have done this for Fukushima. But we do not know how to accurately model the transfer of radionuclides in the marine food chain.
It's an interesting subject. Before Fukushima, there was the Flexblue project in France. It's perhaps not the best time to recall this but it is close to the subject. Does IRSN plan to carry out this type of study linked to Flexblue?
You know the principles of the Flexblue project, which is a French low power modular reactor. IRSN is participating in this project with respect to its safety concepts.
The project is therefore ongoing?
Yes, work is continuing and IRSN is working on contracts in the fields of safety and safety requirements for this type of reactor.
Obviously, it would be quite complicated to build a reactor of this type in France at the moment. Technical problems can be resolved, whereas problems linked to public acceptance are quite another thing. The project is ongoing, come what may.
Codes and experiments
We'd like to return to R&D. One of the particular features of your Institute is that it pays a great deal of attention to R&D and a very high level of interest in it. Could we touch on this question and cite some projects. For example, we have heard talk about SARGEN IV.
We are carrying out research programmes on the safety of facilities and on radiation protection. Among all the research work carried out within IRSN, serious accident modelling studies, which have been more or less halted in other countries, are the most remarkable and provide results.
We have the CABRI programme, which has allowed us to study accidents involving the sudden insertion of reactivity into the fuel, and the PHEBUS research programme in Cadarache, which enables the fusion of the core to be modelled in quite a realistic manner.
With fissile material?
In other words, you obtained corium?
I would go even further. We went as far as simulating the rupture of the reaction vessel to observe the release of all the fission products in a leak tight casing, which represents the reactor containment vessel.
These experiments have shown that the values retained for the release of iodine for example in the previous models were too low. This work has made it possible to correct and validate the relevant codes.
We are continuing to study the behaviour of fuel under accidental conditions, for example corium-concrete interactions.
Yes, there is a certain lack of experimental data on this subject.
We are carrying out experiments at a 1 :1 scale within the framework of a French-American programme. We are also conducting studies on iodine, including organic iodine, and on its interaction with cladding materials.
We have a research programme on fires, which are an important risk for nuclear facilities, and studies on criticality accidents. And we are striving, either with our own resources, or more usually in collaboration with our partners, to model all of the risks.
What do you mean by criticality experiments? Criticality experiments have been ongoing for decades.
For example, let's suppose that AREVA develops new technologies linked to the need to reprocess MOX fuel. It will be necessary to construct new plants, which will have new dimensions, employ different materials and there will be new chemical compounds involved.
The safety reports that will be prepared by AREVA to obtain an operating licence will be based, for all of these new features, on existing computations and experiments so as to optimise dimensions, thicknesses, etc.
The experimental data, which we obtain independently from industry, gives us the possibility of having our independent source of data and experimental results, which we can then compare with what is proposed by the designer.
By carrying out expert appraisals we can thus ensure the soundness of the approaches, solutions and computations prepared by the project designer.
Criticality is a known phenomenon, but the behaviour of new materials needs to be modelled, taking into account all new existing data.
I could in fact say that, from the safety point of view, it is necessary to choose the most conservative approach and allow for very big margins. However, if we take our own calculations, which we have confidence in, we can, if necessary, be convinced of not having to impose excessive margins.
And from an economic point of view, it would be much more efficient.
Let's return to calculation codes. You mentioned calculation codes for serious accidents. But these codes are also used by designers, including in their safety evaluations.
Yes, they are used by designers as results of research activities.
A European group known as SARnet exists, in which many European countries are members. Among them are representatives of designers and safety authorities, who can all use this data. We consider that it is necessary to widely disseminate the results of our research work, since they are useful to everyone.
Does your Institute use these same calculation codes?
Yes, we use these codes. We have in IRSN a simulator known as SOFIA (formerly known as SIPA) which enables the normal or accidental operation of French reactors to be modelled. AREVA has the same simulator. It can work under normal operating conditions, or in situations of deviations from normal operating rules, or in accidental situations.
The results of the experiments enable us to enlarge the correct operating limits of these simulators so that they can be used one day to simulate vessel rupture situations. And we would like to go even further. It lets us elaborate accident scenarios, which are then played out during national crisis exercises, and to train our engineers.
In our question, we wanted to place the accent on the fact that the constructor and the designer will hand over to you, and to the safety authority, a project which - from their point of view - will be extraordinarily good and which will ideally meet all safety requirements. If your specialists repeat the calculations with the same codes as those used by the developers, is there not a risk of simply confirming, somewhat mechanically, the affirmations of the project designers?
We are fully aware of this risk and so we carry out checks with other codes. When, for example, we know that AREVA and EDF use certain codes in their calculations, we can check these calculations with other codes, for example MELCOR.
The developers know how to use different codes, but so do we. When the designer uses a set of codes for the preparation of his safety report, we can carry out checks on his work with another set of codes.
The lessons of Fukushima
We spoke about international working groups. But in France EDF has proposed an interesting initiative, the creation of a Nuclear Rapid Action Force. What does IRSN think of this initiative by operators and are you going to take part in the set up of this Action Force?
Since you return to the subject of the lessons of Fukushima, allow me to make a few digressions.
We have already discussed crisis management and the exchange of information. We consider that there are still three important points that could be improved.
Firstly, it is necessary to take into account all external events and perturbations that can add to the course of events during an accident.
For example, in 1999 in the Blayais power plant in France an incident occurred which could have turned out badly. It happened on the evening of the 27 December 1999 as a result of a combination of a heavy storm and a big tide.
The level of water in the sea was high and the strong wind generated very high waves. The site was flooded and there was a loss of external power and the stoppage of some safety systems. This event was qualified as level 2 on the INES scale. Analysis of the event has shown that the project did not take into account all the events that could arise.
What I'm saying is intended first and foremost to countries that are getting ready to develop nuclear power. They will need to devote a lot of effort to taking into account, in the most exhaustive manner possible, environmental risk factors - earthquakes, floods (in particular when the projected site is located near to a body of water), tsunamis or sand storms.
In our opinion, none of these natural events are sufficiently taken into account for the moment in projects. We consider, and it's the result of stress-tests in France, that French nuclear reactors are not well enough protected against a long lasting power cut and a loss of cooling.
It is for this reason that IRSN has proposed a new "hard core" concept concerning safety systems, a concept that has been taken up and presented by the safety authority.
We assume that the design of additional elements such as diesel generators, pumps and alternative cooling sources need to take into account greater margins than the design margins selected for the remainder of the facility.
Will these additional elements be on the site itself or off site?
On the site itself. They include electrical supply sources, cooling and particularly resistant buildings, in which the crisis teams will meet when an accident occurs.
This latter point is very important. There absolutely needs to be areas and buildings on sites that are protected against radiation. Such a building existed in Fukushima, which meant that specialists were able to work efficiently on the site itself.
These additional elements, reinforced buildings, etc., form part of the concept that we call the "hardened safety core".
The hard core is the second point that I would like to discuss. The third point is off site crisis management. This concerns the resources of the operator, IRSN, communications and transport facilities.
We have considered the situation where, if a serious accident were to occur, road access no longer exists. We now have equipment that can be transported by plane or helicopter, for example air sampling devices that can be attached to planes or helicopters and which allow us to take samples at the actual location of the accident.
And what is your opinion of the Nuclear Rapid Action Force?
It's a good idea. But its implementation will be the responsibility of the operator, which is going to organise and oversee this force.
Are you going to check what will be done in this Force? The proposals that we have heard about indicate that this rapid action force will be in a position to take charge of the whole process of combating an accident as of the second day.
I'll tell you what we'd do. In all cases, EDF would have to transmit its emergency plans to the safety authority. When these plans mention a nuclear rapid action force, EDF will need to obtain authorisation for it from the safety authority.
In this case ASN will very certainly ask us to carry out an expert appraisal. At the end of this expert appraisal, I will be in a position to tell you what conclusions have been arrived at.
I would however like to draw your attention to another discussion that is taking place in Europe. Within the European Union, the following question is being posed: If an incident were to occur on a nuclear power plant, for example in Belgium, would the French be in a position to help the Belgian power plant? In such instances, knowledge of the facility would be vital, including with regard to the most trivial aspects such as "will we be able to plug equipment from France into their electrical switchboards?"
The question has also been raised in France as to the usefulness of creating common European reserves in the event of a serious accident in a nuclear power plant?
Do you mean reserves of technical resources? It's not a bad idea, but it needs to be discussed first between European operators.
Yes, obviously. But in Europe relations are always simpler. Stress tests have been the first very successful step in European collaboration.
IRSN's international collaborations form a quite separate part of our questions. Quite recently, we saw in Obninsk a delegation representing IRSN at the conference dedicated to the 50th anniversary of the BFS critical model. The presentations made by your specialists generated a great deal of interest among the participants. We saw with our own eyes that the IRSN delegation was in talks with IPPE representatives. We do not know the content of these talks, but the fact that they are taking place is very encouraging.
There are not that many countries in the world with extensive experience in the nuclear sector. Which is why this collaboration is important.
We have well-established and fruitful collaborations with the United States, Russia, Japan, China and Germany. We also collaborate with other countries on more targeted subjects.
The collaboration with the United States covers numerous aspects, including exchanges of codes and information.
We used to collaborate extensively with the Soviet Union, then the collaboration with your country tailed off, but it is now coming back into force. I know Professor Bolshov well and we have a strategic cooperation agreement with his institute. We have a partnership with the Kourchatov Institute, although the volume of our exchanges has diminished compared to what existed before.
I'm going to visit Russia this year. I plan to meet the management of Rostechnadzor and SEC/NRS. I would add that I regret that the remit of SEC/NRS, our counterpart organisation, does not extend to research work.
Obviously, in Russia, Rosatom is in charge of many things and deals with numerous safety aspects. But as I have already said, IRSN does not develop actual facilities. We develop and explore in greater detail knowledge with regard to safety. We see a similar approach to our own in the United States. Unfortunately, in Russia it is not as obvious.
Does your Institute collaborate with Ukraine?
I know Mrs Mykolaichuk and Mr Gromov very well. We have an active collaboration and we provide them with our help in terms of expertise.
We have also worked extensively on the Chernobyl site. We are conducting a radioecology research programme on the Chernobyl polygon. We have a representative in Kiev, who even gives lectures at Kiev University.
Do you collaborate with Russia on Chernobyl?
Yes, we are conducting a research programme on the effects of the consequences of the Chernobyl accident on the health of children living in the Bryansk Oblast, on areas with low contamination.
We are studying the effects of chronic exposure to low doses, because there are still a lot of unknowns in the interaction of ionising radiation with living beings.
We agree about low doses. In Russia, some people say that low doses can be good for health.
In France as well there are those who hold this view. Perhaps they're right, perhaps they're wrong.
You know, low doses are good for us, the collaborators of AtomInfo.Ru. We checked.
What can I say? Life began and developed in an environment that was more radioactive than that which exists today on our planet. Perhaps living organisms then had natural defences against the perturbations caused by ionising radiation, but we cannot prove it for the moment.
As regards Chernobyl, IRSN is conducting studies on non-cancerous pathologies. It's a study that we began with the Obninsk Institute MRRC RAMS, through a pilot project that convinced us of the possibility of obtaining interesting results. We are continuing this study in direct collaboration with the Bryansk clinical diagnosis centre.
It's still too early to reveal the results, but they show that the French-Russian collaboration is very fruitful.
A personal challenge
Mr Repussard, a more personal question. How do you feel in your job as Director General of IRSN? It is said that when you were appointed for this job, many people considered that the task was both daunting and difficult.
It was a real challenge for me. Unlike you, I am not an atomic scientist. I've not come from the French Atomic Energy Commission. I'm a graduate of the Ecole Polytechnique and the Ecole Nationale des Ponts et Chaussees and I was appointed to this post by President Chirac, who doubtless considered that I had the necessary skills to succeed.
Before joining IRSN, I was head of INERIS and I spent many years in Brussels. Consequently, I am not unfamiliar with the general technical culture of European countries and this useful experience was taken into account when I was appointed. Obviously, I had to delve much deeper into subjects that I had studied previously in a more superficial manner.
I am very pleased to see that, under my management, IRSN has become an expert body whose expertise and independence are widely recognised.
During the French revolution, our ancestors chose a three word motto "Liberty, equality, fraternity". Our institute has its own motto: "Knowledge, independence, proximity".
Knowledge is indispensable in our work. Independence is also important, because there are conflicts of interest, economic aspects, and I have to be able to sign reports without any pressure being placed on me. And finally we need to be close to all problems, to be close to all players, including journalists, and to provide information and answers to all the questions you ask us.
To illustrate the point about proximity: several months after I joined IRSN a request from Greenpeace arrived addressed to me. I asked the experts to prepare a response. They told me: "Mr Repussard, we're not used to responding to anti-nuclear organisations". To which I replied: "We will not reveal any state or trade secrets, but we will not leave them without any answer".
Mr Repussard, thank you very much for this interview given to the e-journal AtomInfo.Ru.
It is fully belonged now to the Kozloduy NPP.
We want to put it into commercial production at the end of this year.
The company will stop power production at the Tomari nuclear power plant's third reactor.
Hero of the day
Not quite so. The authors of the concept, which was difficult to be realized in practice, turned to a clearer concept of a standing wave reactor (TP-1) that in principle allows finding the solution to the tasks stated for TWRs.