Fukushima Daiichi nuclear power plant
Following the accident that occurred on 11 March 2011 at the Fukushima Daiichi nuclear power plant, operated by TEPCO, a number of initiatives were rapidly intro-duced to enable complementary assessments to be conducted in the light of the events affecting the Fukushima Daiichi plant, based on experience feedback from the accident.
84. Vulcan Experimental Nuclear System.
85. Fuel element containing an absorbing element.
86. Critical assembly similar to EOLE (see section 5.2).
The European Council meeting on 24 and 25 March 2011 demanded that the various EU Member States conduct these complementary safety assessments, known as“stress tests”, on their nuclear power plants. A specification was drawn up for this purpose, based on proposals byWENRA87.
Complementary safety assessments were therefore simultaneously carried out in the different EU countries based on similar specifications, and in some cases extended–as was the case in France and Belgium– to other types of nuclear facility or even to other issues88. These complementary safety assessments could therefore be conducted not only on nuclear power reactors, but also on research reactors, fuel cycle facilities, etc.
Some of the first lessons learned from theaccident at the Fukushima Daiichi nuclear power plantwere indeed generic in nature. In particular they concerned the robustness of facilities with respect to the extreme hazards that can affect nuclear facility sites, emergency response, organization and also the role of the safety authorities. Because these issues are also relevant to research reactors and fuel cycle facilities, many coun-tries have included these in their list of facilities to undergo complementary safety assessments, with priorities relevant to the risks they present (inventory of radioactive materials, age, proximity to residential areas, etc.).
The complementary safety assessments carried out in the EU Member States have generally looked at:
– the possibility of extreme hazards occurring that exceed those considered when the facility was designed, leading to station blackout or total loss of cooling, so that additional measures can be identified where necessary to limit the consequences of these accident situations;
– the actual physical conditions of structures, systems and components significant to the safety of each facility and the potential effects of the failure of non-safety class elements on elements significant to safety, should an extreme event occur (hence the need to conduct detailed inspections of the facility);
– the ability of I&C and facility monitoring equipment to provide appropriate information in the accident situations considered by the complementary safety assessments (extreme hazards, loss of power or cooling).
The key features of the complementary safety assessments carried out on research reactors in France are presented insection 10.2.
Complementary safety assessments have also been carried out or are planned in countries other than the EU Member States, prioritised on the basis of the risks pre-sented by each facility.
In June 2011, theIAEA organized a conference at ministerial level. An action plan was put in place by the IAEA to improve nuclear safety worldwide.
87. Western European Nuclear Regulators Association.
88. Issues concerning the contractors used by operators were therefore addressed in France.
In this context, in 2011 theIAEAbegan preparing a procedure for conducting safety reassessments of research reactors, based on the lessons learned from theaccident at the Fukushima Daiichi nuclear power plant. The purpose of this procedure, which was the subject of a final report89published in March 2014, was to obtain consistency be-tween the different approaches of different countries, and to form the basis for any reassessments yet to be carried out. Some of the principles expressed in this IAEA report are explained below.
In this report, the IAEA expressly states firstly that, although the inventory of radioactive material, and consequently the potential hazard associated with research reactors worldwide, is much lower than that for nuclear power plants, generally speak-ing there are aspects that justify safety reassessment in the light of feedback from the accident at the Fukushima Daiichi nuclear power plant:“the majority of research reac-tors worldwide were designed decades ago, and their design requirements are not fully in conformance with IAEA Safety Standard No. NS-R-4. In addition, many research reactors are located near populated areas, and for some of these the leaktightness of their con-finement buildings is inadequate. These issues complicate the management of accidents that result in radioactive releases. In some other cases, the characteristics of the research reactor site and the site area and site vicinity may have changed since the facility was constructed. Not all the above mentioned issues are reflected in the safety analysis for many facilities”. Whether a reassessment is required should be decided on the basis of the potential hazard associated with each research reactor.
In terms of feedback from the accident at the Fukushima Daiichi nuclear power plant, theIAEAdraws particular attention in this report to the role and responsibilities of the safety authorities, which need to be clearly defined for both normal operating conditions and accident situations. Moreover, the safety authorities must have the nec-essary expertise to supervise and review the post-Fukushima reassessments to be car-ried out by operators.
The main objective of the reassessment is“to evaluate the robustness of the existing reactor protection, in terms of design features and procedures, against the impact of ex-treme events, with an emphasis on fulfilment of the basic safety functions”. A reassess-ment should consist of:
– a review of the design basis of the reactor facility (taking account of the experimental devices and associated equipment), as described in the safety analysis report;
– a study of events that are beyond the design basis of the facility90, which can be initiated by extreme initiating events, in order to assess their potential impact on the basic safety functions and the adequacy of existing measures to mitigate the consequences of accidents, in order to identify the safety improvements needed from both a technical and an organizational point of view.
89. IAEA report entitled“Safety Reassessment for Research Reactors in the Light of the Accident at the Fukushima Daiichi Nuclear Power Plant”‒Safety Reports Series No. 80‒2014.
90. Beyond design basis accidents.
These reassessments must refer to the current status of the facility as built and as operated (maintenance carried out, modifications made, etc.), the most unfavourable permitted operating conditions, including core configurations, and current and planned experimental devices.
Safety reassessments should consider the possibility of the simultaneous occurrence of more than one external hazard, as well as events that could occur as a result of this.
On the basis of these reassessments, additional measures to prevent or mitigate the consequences of accidents should be defined and implemented if necessary.
Reassessment of the site should look at changes in the site characteristics since the facility was built. This includes changes in the distribution of workers on the site and in the surrounding population, changes made to other facilities within the site area, includ-ing changes of use, changes in local transportation routes, changes in local land use and changes in hydrology and topography. Accidents that could occur simultaneously at dif-ferent facilities should be considered.
The potential impact of extreme hazards on access to the reactor site for operating or on-site response personnel, and the availability of off-site response organizations and response personnel, should also be reassessed. In other words, the emergency response measures must be assessed to make sure they are adequate and can be implemented, including if an extreme hazard occurs that affects several facilities simultaneously.
Reassessments also provide an opportunity to verify:
– the existence of an adequate chain of command for response to an emergency and of procedures and means for effective communication during an emergency;
– preparedness of the on-site response teams and off-site response organizations to manage effectively an emergency affecting several facilities on a single site simultaneously;
– the availability of emergency equipment and the performance of periodic checks on that equipment;
– site accessibility for off-site response teams and the availability of the necessary logistical support.
At a conference held by theIAEAin November 2015, various research reactor oper-ators (e.g. the operoper-ators of the IRR1 reactor91in Israel and the SAFARI-1 reactor in South Africa) presented the actions plans that they had proposed to their respective safety regulators following the safety reassessments conducted on the basis of the IAEA’s report mentioned above or the recommendations ofENSREG(European Nuclear Safety Regulators Group92) in the case of the stress tests.
Generally speaking, and by way of illustration, through the installation of new equipment that can withstand earthquakes associated with the sites, including safety margins, or through the modification of existing equipment to improve this earthquake 91. Israel Research Reactor-1.
92. A European Commission consultative group of independent experts.
resistance, the reassessments have resulted in proposals to improve the safety of reac-tors such as:
– seismic detection linked to the reactor protection system, causing an automatic reactor scram in the event of an earthquake;
– an extra system to shut down the chain reaction (injection of a soluble neutron poison, etc.);
– an emergency power supply in addition to the existing power supplies (mobile generator or backup battery), addition of easily accessible external connections;
– additional means of emergency cooling, fire service connections, core spray systems;
– strengthening of the containment vessel to improve its resistance to external natural hazards;
– improvements to emergency ventilation systems and theirfiltration systems;
– reinforcement of the means provided for effective emergency response, creation of off-site emergency control rooms with feedback of the information necessary to monitor the facility, etc.
Most of these measures had already been implemented at research reactors in France, at the time of safety reviews, or had been reinforced or supplemented during the stress tests carried out following theaccident at the Fukushima Daiichi nuclear power plant(this is discussed insection 10.2).
Other proposed improvements resulting from the reassessments concern the safety culture, organizational aspects, and training and qualification programmes of operating personnel.
Schedules have been drawn up for the implementation of these proposed improve-ments.
In conclusion, the complementary safety assessments performed on research reac-tors on the basis of feedback from theaccident at the Fukushima Daiichi nuclear power plantwill help to improve defence in depth, including as regards emergency response.
Peer review of the results of this work has been conducted under the auspices of the IAEA, at various technical meetings.
Part 2
Research reactors in France
Chapter 5 Evolution of the French research reactor “fleet”
5.1. The diversity and complementarity of French research reactors
In theIAEA’sResearch Reactor Database (RRDB), 42 reactors in France are identified as being research reactors93(including those no longer in operation, theJules Horowitz reactor (JHR)which is under construction, and the research reactors at defence-related facilities94).
General de Gaulle created the French Atomic Energy Commission (CEA95) by decree in 1945, giving it responsibility for directing and coordinating the development of appli-cations for the fission of the uranium atom nucleus. In this context, a team led by Lew Kowarski started up the first French research reactor in 1948, the ZOÉ atomic pile built at the CEA centre at Fontenay-aux-Roses (figure 5.1). The core of this reactor, consist-ing of uranium oxide-based fuel elements (1,950 kg) sittconsist-ing in heavy water (5 tonnes) in an aluminium tank surrounded by a 90 cm-thick graphite wall, stood within a 1.5 metre-thick concrete containment wall designed to absorb the different types of ionizing radiation emitted by the nuclear reactions in the core. The ZOÉ reactor was used up to a power of 150 kW to study the behaviour of materials under irradiation, and at 93. This database gives the full list of French research reactors. See also the CEA publication entitled
“Research Nuclear Reactors”, a Nuclear Energy Division monograph‒2012, or the publication (in French)“Les réacteurs de recherche”by Francis Merchie, Encyclopédie de l’énergie, 2015.
94. This publication does not cover research reactors used for defence-related purposes.
95. Which would later become the Alternative Energies and Atomic Energy Commission.
low power to characterize the neutronic properties of the materials used in atomic piles at the time (worldwide).
In the 1950s, some ten research reactors were commissioned in France. Having no industrial enrichment capability of its own at the time, France set about improving knowledge of the nuclear data for reactors using natural uranium. The AQUILON reac-tor at Saclay was designed for this purpose. This reacreac-tor and the ALIZÉ reacreac-tor (also at Saclay) were then used to support the design of the on-board reactors in the first French nuclear-powered submarines. The PROSERPINE reactor, also at Saclay, was used specif-ically for studying “homogeneous96” reactors, using plutonium in solution as fissile material. PROSERPINE was light water-moderated. It was used to compare the neutron characteristics of two fundamental fissile elements: plutonium-239 and uranium-235.
In parallel during the 1950s, reactors were built for material testing and technolog-ical research. Thus the EL2 then the EL3 reactors were commissioned at Saclay for the purpose of producing artificial radioisotopes for studying the behaviour under irradia-tion of materials of structures used in reactors.
Towards the end of the 1950s, it became apparent that there was a need for better knowledge of the fundamental neutronics parameters involved in nuclear reactor core physics. In response to this need in particular, the MINERVE reactor was designed and commissioned in 1959 at theCEA centre at Fontenay-aux-Roses.
Figure 5.1. View of ZOÉ, France’sfirst“atomic pile”. CEA historic archives. © CEA/Documentation service.
96. The fuel in a homogeneous reactor is in liquid form (nitrate or sulphate).
A further twenty or so research reactors were commissioned in the 1960s. The development of the nuclear energy industry was already in full swing at the time but there were limited computing resources available. The use of critical assemblies or mockups97and material testing reactors appeared necessary to complete the acquisi-tion of knowledge and data to support the industrial development of nuclear energy.
France was attempting to develop the GCR98(gas-cooled, graphite-moderated) reactor type using natural uranium as fuel. The MARIUS research reactor (commissioned in 1960 on the Marcoule site, then transferred in the mid-1960s to theCEAsite at Cadar-ache) and the CESAR research reactor (commissioned in 1964 at CadarCadar-ache) were used in the early 1960s to conduct studies for the nuclear energy industry.
The use of fast neutron reactors was also explored at this time, especially for the purpose of using the plutonium resulting from the operation of the GCR reactors.
The development of SFRs led in particular to the construction:
– of the HARMONIE reactor at Cadarache, whichfirst went critical in 1965 and was used mainly to determine the neutron characteristics of radiological protection materials (the lateral neutron shielding around the core in fast neutron reactors);
– the MASURCA99critical assembly, also situated at Cadarache and commissioned in 1966, which was used to study neutronics and, much later, to conduct research into the transmutation of the actinides present in highly radioactive nuclear waste.
The RAPSODIE reactor at Cadarache was the first fast neutron research reactor and ran on plutonium100fuel and liquid sodium coolant. Numerous irradiation experiments were conducted in RAPSODIE between 1967 (the year of its first criticality) and 1982 (permanent shutdown in 1983), as part of the development of steel cladding for so-dium-cooled fast neutron reactors (SFRs). Experiments known as“end-of-life testing”, extending as far as the melting of fuel at the core of certain fuel rods, were conducted in 1982 (DISCO and FONDU tests).
TheCABRIreactor, the first French reactor designed specifically for studying acci-dent situations in SFRs (in a sodium loop), was built at Cadarache at the start of the 1960s; the first criticality of this reactor took place in December 1963. Tests were also carried out in the sodium loop to study accident situations in pressurized water reactors (tests known as REP-Na). The SCARABEE reactor, used in the 1980s for tests related to sodium-cooled fast neutron reactors (since shut down and dismantled), shared the main equipment of the CABRI reactor. It had a sodium loop of larger diameter than the one used in the CABRI reactor.
The willingness of the USA to supply highly enriched fuel with uranium-235 meant that it was possible in the 1960s to design higher power reactor cores with more intense 97. Reactors using fuel element arrangements representative of the cores being studied (“assembly”)
and operating at almost zero power (“just critical”).
98. UNGG (Uranium Naturel Graphite-Gaz) in French.
99. Breeding assembly at the Cadarache Research Centre.
100. The RAPSODIE reactor, like subsequent French SFRs, ran on mixed UO2-PuO2fuel; axial and radial“blankets”of uranium-238 (depleted uranium), a fertile material under a fast neutronflux, were also placed around thefissile zone.
neutron fluxes, so the reactors could be used for material testing. In France, three mate-rial testing reactors were designed at this time: the 30 MW PEGASE reactor at Cadarache, the 35 MW SILOE reactor at Grenoble (forced downstream water circula-tion; this reactor was in operation from 1963 to 1997), and the 70 MW OSIRIS reactor (figure 5.2) at Saclay (upstream water circulation; in operation from 1966 to 2015).
These reactors were each accompanied by a critical assembly: PEGGY for PEGASE, SILOETTE (figure 5.2) for SILOE and ISIS for OSIRIS.
Unlike the high flux reactor (RHF101) in Grenoble and the ORPHEE reactor, SILOE was a light water pool-type reactor built for irradiating materials and equipment. The core (figure 5.3) consisted of elements containing fuel enriched to 90% uranium-235.
However, the SILOE reactor also had neutron channels not aimed directly at the core, as well as a beryllium102wall along one of the four sides of the core103. To start with, there were only two radial channels. After the MELUSINE reactor was shut down in 1988, a tangential channel was added to SILOE, which aimed at the beryllium wall through the slice. This brought the number of instruments to six, with two devices per channel (spectrometers, diffraction meters). Despite difficult working conditions for the scientists (high temperatures, confined spaces, etc.), this equipment gave excel-lent service from a scientific point of view and was used to train scientists, especially in powder diffraction and single crystal diffraction, as well as polarised neutron scattering.
In 1969, France took the decision to stop building GCR reactors in favour of US-designed pressurized water reactors. In the decade that followed, with experimental Figure 5.2. Left, Osiris reactor core. View of the submerged neutron radiography installation (2004).
© L. Godart/CEA; right, view of the SILOETTE critical assembly. © CEA.
101. Réacteur à Haut Fluxin French.
102. This material is a neutron source when struck by high energy gamma rays in particular.
103. Source: ARILL (“Association des retraités de l’Institut Laue-Langevin”):“Le réacteur de recherche Siloé”.
needs largely being met, few new research reactors were built compared to the 1950-1970 period.
However, two high flux reactors, with neutron beams for fundamental physics experiments, were then commissioned: the high flux reactor (the 58 MW RHF) at Grenoble, run by the Institut Laue–Langevin (ILL), which went critical in 1971, and the ORPHEE reactor (14 MW) at Saclay, which went critical at the end of 1980.
In 1972, in association withEDF, theCEAset up the Pile Construction Department within the Atomic Piles Division at the CEA. The Department was subsequently hived off and attached to the companyTechnicatome104, which would become the TA branch of
In 1972, in association withEDF, theCEAset up the Pile Construction Department within the Atomic Piles Division at the CEA. The Department was subsequently hived off and attached to the companyTechnicatome104, which would become the TA branch of