(December 21, 2001) Analysis carried out following an incident at Dampierre in France indicates that when mixed-oxide (MOX) fuel is used, incorrect loading of fuel into a reactor could result in the reactor going critical before it is started up. Workers involved in loading the fuel could then be irradiated in what has been described as a "mini-Tokai Mura" accident (1).
(560.5351) WISE Amsterdam - The fuel loading incident at Dampierre's reactor no. 4 occurred on 2 April 2001. In that incident, 113 fuel elements had been loaded in the wrong position in the reactor. However, subsequent analysis showed that the workers at Dampierre had a lucky escape. If the fuel had been loaded in an even more unfavorable configuration, the result could have been a criticality incident, with the workers involved receiving a significant radiation dose.
This was only realized after analysis by Electricité de France (EdF), carried out on the request of the French nuclear safety authority ASN, which was reported on 2 October 2001 (2,3,4). The incident was then re-classified as Level 2 on the International Nuclear Event Scale (INES), which has 7 levels. It was the second Level 2 incident in France in 2001(5). It is expected to be the joint "winner" of the "prize" for most serious French nuclear incident of the year.
However, neither the ASN press release(2) nor the accompanying "note d'information"(3) mentioned the crucial fact that the reactor was "moxed" - loaded with a mixture of 30% MOX fuel and 70% conventional uranium oxide fuel(6). MOX fuel contains a mixture of oxides of plutonium and depleted uranium. It is already known that the presence of plutonium in this fuel reduces the safety margin when operating the reactor(7). Now it seems that there is a new hazard with MOX during fuel loading - i.e. even before the reactor is started up!
SURPRISE IN JAPAN
News of the Dampierre incident has provoked surprise in Japan. Mr. Keiji Kobayashi, a reactor physicist at the Research Reactor Institute of Kyoto University, said that he had often warned of the dangers of incorrect MOX loading. However, up until now it was mostly assumed that perhaps one or two assemblies might be loaded in the wrong place - not 113 assemblies, as happened at Dampierre. "I don't think that anyone has considered the possibility that more than two-thirds of the assemblies could be loaded in the wrong places, as happened at reactor 4 at Dampierre...I don't think that the Japanese authorities are aware of this problem", he commented. "Such a large-scale incorrect loading of 113 assemblies as that which occurred in France could, depending on the configuration, cause a reactivity accident as EdF has calculated."
Source: Email from Mr. Kobayashi, translated by Satomi Oba, WISE Hiroshima
Contact: WISE Hiroshima at email@example.com
The crucial point is that the amount of MOX in each fuel assembly varies according to the position of the assembly in the core. As the ASN's note says, "The fuel loading layout, which specifies precisely the position of each fuel assembly in the reactor vessel, is calculated before fuel loading begins. It is designed to ensure a homogenous distribution of power in the reactor core." So, for correct functioning of the reactor and proper control of the neutron flux, it is essential to load each assembly in its allocated position.
However, during charging it was noticed while inserting the 135th assembly that 113 of them had been "shifted" by one place due to "human error" and "deficiencies in the supervision of the fuel loading process"(8). A subsequent ASN inspection on 13 June revealed a range of extra faults including lack of organization and actions carried out by unauthorized personnel(3).
Following the incident, the ASN asked EdF to analyze what would have happened if the fuel assemblies had been wrongly loaded in different configurations. Initial results showed that bad configurations could lead to the nuclear chain reaction beginning during loading, while the lid of the reactor vessel was off with workers in the vicinity.
Not only that, but the existing measuring devices would have been unable to detect the rise in power and warn the workers before the fuel went critical. Furthermore, according to a newspaper report(9)), radio communication was forbidden in case it interferes with safety systems, and the nearest telephone was several meters away, so contacting the workers could have been difficult.
The note did not describe what would happen after the chain reaction started. ASN deputy director Jérôme Goellner claimed that the workers would be subjected to "limited but significant" irradiation, "without health risks". However, the possibility of a criticality accident without detectors to warn the operators has led to the scenario being described as a "mini-Tokai Mura"(1).
An obvious question is: why didn't EdF carry out its calculations before loading MOX? Twenty French 900-megawatt PWRs are currently loaded with MOX, and MOX loading is planned for a further eight reactors of this type. Another question: have these calculations been carried out for the reactors in other countries that are loaded with MOX? At the time of writing it seems that the answer is "no".
This new danger from MOX is likely to raise concern in many countries, particularly in Japan, where the Tokai-Mura accident has delayed plans for MOX use(10, 11).
Now that this problem has been identified, it is important that:
- The situation in France must be improved: better detectors that can warn of the onset of a chain reaction, better communication with personnel during fuel loading, better training and supervision.
- Where MOX fuel is used in reactors other than the French 900MW design, utilities need to carry out similar calculations to see if criticality during loading is possible, and if so, they too must install suitable detectors and train personnel to avoid a criticality accident.
- Systems need to be in place to stop criticality during fuel loading if it occurs.
- Belbéoch, B. "Dampierre, un nouveau type d'incident très sérieux avant démarrage du réacteur" [Dampierre, a new type of very serious incident before starting the reactor], in La lettre d'information du Comité Stop Nogent-sur-Seine, July-October 2001.
- ASN press release, 2 October 2001
- ASN "note d'information", 2 October 2001
- WISE News Communique 555.5323, "France: test equipment gets in the way"
- For the first Level 2 incident in France in 2001, see WISE News Communique 548.5270, "France: Regulator says fix those valves now".
- Arrêts de tranche, DIN Orléans, 28 February 2001 (http://www.asn.gouv.fr/regions/orleans/arrets.asp)
- Section 6.2.5, "Behavior of MOX fuel in the reactor" in WISE News Communique special edition 469-470, "The MOX Myth".
- Magnuc, 18 April 2001
- Libération, 3 October 2001
- WISE News Communique 521.5109, "MOX program under discussion after Tokai-Mura accident"
- Responses of Japanese experts to the Dampierre incident are available from Satomi Oba at WISE Hiroshima.
Contact: Bella Belbéoch, Comité Stop Nogent, c/o Nature et Progrès, 49 rue Raspail, 93100 Montreuil, France