The impossible accident happened three times

Illustré par :


Before the Fukushima disaster, there was Chernobyl, 25 years before. But the first catastrophic accident in the history of civil nuclear power took place a few years earlier in the United States, the country where the nuclear industry was born.

It was March 28, 1979, at four in the morning. For an unknown reason, the feed pumps for the secondary cooling system of the No. 2 reactor at the Three Mile Island nuclear power plant in Pennsylvania shut down(1). Backup pumps should normally start automatically. They do not do so as a result of human error during the last maintenance operation. The resulting rise in temperature of the primary circuit water increases the pressure, which causes the pressurizer pressure relief valve to open and the reactor and turbine to shut down automatically. All this took 8 seconds.

The pressure relief valve must be closed again once the pressure has returned to normal. Then a new technical malfunction occurs: the indicator light indicates a closed valve. The pressure continued to drop in the primary circuit, which emptied through the valve that had remained open, releasing 120 m³ of highly radioactive steam and water into the reactor containment. The heart starts to melt. The operator, believing the pressurizer to be full, wrongly concludes that the primary circuit is also full, and manually stops the safety injection circuit!

4 minutes 38 seconds have elapsed. The primary circuit empties directly into the containment. The core continues to melt and the fuel cladding alloy (zirconium based) reacts with water to produce large amounts of hydrogen, which is trapped in the upper parts of the reactor.

At 6:45 a.m., nearly three hours after the start of the accident, the radioactivity detection alarms went off. The plant manager, in a panic, declared a state of emergency and alerted local authorities, the Governor of the State of Pennsylvania and the Atomic Energy Commission (AEC) that the event could have serious radiological consequences for the general public. At 10 a.m., an emergency meeting was held in which, in addition to those in charge of the plant, political representatives from different levels, including the President of the United States, participated. The decision to evacuate the local population is taken as soon as the explosion of the hydrogen bubble formed in the enclosure is a plausible hypothesis.

Attempts by the operators to refill the primary circuit with water finally succeeded in the afternoon; the fuel was gradually cooled. For five days, plant operators and NRC (Nuclear Regulatory Committee) experts sought to understand what was happening and to predict the outcome of the crisis. The reactor was finally brought back under control, but the actual course and effects of the accident were not elucidated until much later.

In 1979, U.S. nuclear safety officials rejected the possibility of a core meltdown, which did occur. It was only six years later, thanks to a probe sent into the reactor vessel, that we had proof: 50% of the corium(2) had melted and 20% had sunk to the bottom of the vessel. It didn’t go through the tank but it didn’t take much…

The official version of events is that the radioactive releases were very limited and had no impact on the local population. In reality, the balance sheet is far from negligible.

  • In the first hours, there was a massive release (several million billion Bq) of radioactive gases (iodine, xenon and krypton) which were not measured, given the ambient disarray;
  • Periodic releases of radioactive gases took place during the following months (large quantities of Krypton 85);
  • The large volumes of contaminated water present in the enclosure and the cooling system (respectively 2 million liters and 350,000 liters with an average contamination rate of 4 billion Bq per liter) were released over time.

It is therefore incorrect to classify the accident at level 5 of the INES scale, as the international control bodies have done. Official forecasts of the health consequences concluded that the impact of the releases caused by the accident was not significant. However, in contrast, a study covering the period 1979–1998, published in 2003(3), showed a significant increase in the rate of breast cancer and diseases of lymphatic and hematopoietic tissues in relation to the assessed level of radiation exposure in residents near Three Mile Island.

Admittedly, the TMI accident was neither a human nor an ecological disaster like Chernobyl or Fukushima, but it revealed the vulnerability of a power generation system that had been considered infallible until then.

A reading of the events shows that a very serious accident, which could have been catastrophic, occurred due to the unexpected and therefore unforeseen combination of multiple failures. Some phenomena that appeared during the accident had never been considered. One would look in vain in the safety reports on American or European PWR reactors prior to 1979 for any trace of the possible formation of a hydrogen bubble. Although the reaction between the zirconium in the fuel cladding and steam was well known, the physical conditions necessary to produce a significant amount of hydrogen were not considered to be achievable. The facts largely refuted this hypothesis: in the first hours of the accident, the quantity of hydrogen produced was sufficient for the NRC experts to consider the possibility of an explosion.

The dismay of operators, official experts and politicians became clear to international public opinion: it became obvious to careful observers that nuclear power plants could be the site of catastrophic accidents. In the United States, the Three Mile Island accident was a real disaster for the nuclear industry: no more orders for new reactors were placed; many projects were cancelled and construction sites were considerably delayed because of new requirements from the regulatory authorities. All this contributed to making the nuclear choice financially untenable.

To date, the 99 nuclear units in operation in the United States were all ordered before 1979 and their average age is 35.6 years; 33 are over 40 years old, 35 are between 35 and 40 years old, and 30 are between 21 and 30 years old. Only one is under 20 years old.

It was not until 2013 that four new sites were able to start, with connection to the network planned for 2019 and 2020. Another construction site has been underway since 1972 (!), the Watts Bar- 2 reactor, abandoned for many years and reactivated in 2014.(4)


The Chernobyl disaster took place on April 26, 1986. It affected a reactor of the RBMK line, developed in the USSR since the 1950s. From a safety point of view, certain characteristics clearly differentiate it from the pressurized water reactor (PWR).

Boris Semenov, one of the Soviet safety experts and Deputy Director General of the IAEA in 1983, insisted on the low risk involved with such a reactor:  » a serious accident due to loss of coolant is practically impossible . In 1985, the head of the Chernobyl power plant, Nikolai Formin, said:  » The huge reactor is housed in a concrete silo and is equipped with environmental protection devices. Even if the unbelievable were to happen, the control and safety systems would shut down the reactor within seconds… « (5)

We see it. The confidence and serenity of Soviet engineers was identical to that displayed by their Western colleagues before Three Mile Island.

It is true that the safety of the RBMK reactor is designed to counteract the maximum accident envisaged. As for PWR reactors, it is the one resulting from a loss of cooling following the appearance of a breach in the water-steam circuits.

The question is no longer whether a nuclear disaster can happen in Europe, but when it will happen

In Chernobyl, the scenario was completely different from that of Three Mile Island. To counter the same reference accident, the Soviets planned modular containment systems, with the fuel assemblies placed in nearly 1,700 tubes of independent strength. In case of heating, the possible core melting concerns only a limited number of assemblies. The criticism, which has been widely propagated in the West, about the lack of containment is therefore not really founded, knowing that, as for the PWRs, it is the reference accident, that of a loss of coolant, which determines the design basis of the RBMK, and not an explosion considered impossible, which did in fact occur.

It was not a failure of the primary circuit that caused the accident at Chernobyl, but rather an « excursion (sudden rise) in reactivity linked to an unstable characteristic specific to the RBMK, which prohibits prolonged operation of the reactor at low power (less than 800 Mwth).(6) It is the continuation of an experiment intended to improve safety, which is paradoxically at the origin of the accident. The operating team, while trying to stabilize the operation of the reactor at low power, caused a sudden increase in reactivity leading to a first explosion, quickly followed by a second (hydrogen explosion). This double explosion could only have occurred as a result of a series of violations of basic operating rules. The first explosion pulverized the fuel and created the conditions for the second, non-nuclear, but larger explosion that took place a few seconds later. A considerable amount of radioactive material was released, not comparable to Three Mile Island.

At Chernobyl in 1986, the Soviet authorities did everything possible to cover up the extent of the disaster and prevent correct information from being disseminated. Orders were given by political leaders to the medical community to deny any link between certain pathologies and radiation exposure, especially among the 600,000 liquidators. Some Western governments were not left out; the French government and the country’s nuclear safety officials sent the message of total protection of the national territory against any radiation risk. The European cacophony manifested itself in Belgium through inconsistent decisions regarding the confinement of livestock: the Walloon Region considered the risk of contamination of grasslands to be negligible, while the federal government recommended that cattle be kept in the barn for a few days to avoid ingesting radioactive iodine. Over the years, international agencies have deliberately minimized the health consequences of the disaster.

Estimates provided jointly by the World Health Organization (WHO) and the International Atomic Energy Agency (IAEA), published in 2005, put the number of deaths among liquidators at around 50 and the number of cancer deaths at 4,000 in the three countries most affected by the radioactive fallout, namely Belarus, Ukraine and Russia. More recently, the same authorities have finally conceded that a few thousand cases of thyroid cancer have affected children in the most contaminated regions (while insisting that thyroid cancer is most often curable!). According to the WHO, the accident caused less than 10,000 deaths in total, and the number of patients did not exceed 200,000. These estimates are essentially based on the internationally accepted risk model of the International Commission on Protection against Ionizing Radiation, a model that is contradicted by many facts and challenged by field studies conducted by Russian, Ukrainian and Belarusian scientists, studies that are totally ignored by the various international bodies under the pretext of publications mainly in Russian.

In December 2011, this ridiculous justification was swept aside thanks to the New York Academy of Sciences publishing the bulk of the work in English.(7) The analysis of data for Belarus, Ukraine and the region of Russia closest to Chernobyl shows, among other things, that since the disaster :

- the general morbidity of children has increased significantly in Belarus;

- the phenomenon of early aging is clearly manifested: the biological age of people living permanently in the contaminated territories of Ukraine exceeds the real age by 7–9 years;

- the syndrome of premature aging is characteristic of the 600,000 liquidators who built the Chernobyl sarcophagus; many diseases appear in them 10 to 15 years earlier than in the general population;

- Genetic damage is clearly measurable by the widespread detection of chromosomal aberrations. The genetic consequences of the disaster will reach hundreds of millions of people;

- Experts estimate that nearly 1,500,000 people are at risk of thyroid disease, with cancer being the most serious form of thyroid disease;

- In Belarus, the incidence of all cancerous diseases increased by 40% between 1990 and 2000.

It should be noted that in Western Europe, some regions are still contaminated by radioactivity, particularly in northern Scandinavia, Germany, Scotland and Poland, to the point that some products originating from these regions remain unfit for consumption (game, fish, mushrooms), according to the European Commission.

In terms of health, the work of Martin Tondel et al.(8) on cancer in northern Sweden showed a significant 11% increase in the cancer rate for a Cesium 137 contamination of 100kBq/m².


In the hours following the loss of control of the Fukushima nuclear reactors in the wake of a devastating earthquake and tsunami, reports from Tepco, the plant’s owner and operator, the Japanese safety agency and the government all suggested that while the situation was serious, the disaster was still preventable. Initially assessed at level 4 on the INES scale of severity of nuclear accidents, then at level 5 (as at Three Mile Island), the accident was, a few weeks later, reassessed at level 7 (as at Chernobyl). It was difficult to deny what has become obvious to everyone: Fukushima is a disaster.

Three nuclear reactors in meltdown, four fuel deactivation pools made extremely dangerous, hundreds of thousands of people evacuated, thousands of hectares contaminated, two million people whose health is threatened, mountains of radioactive waste to be stored for centuries to come, permanent discharge of contaminated water into the sea…

Four years later, the reassuring official speeches of the Japanese government and the near disappearance in the Western media of recent information on the aftermath of the disaster may lead us to believe that the situation is back to normal and under control. This is not the case. Certainly, the removal of fuel assemblies from the pool of reactor No. 4 was completed in December 2014 finally ruling out the threat of a collapse whose consequences would have been apocalyptic. However, there is still much to be done in terms of technical control of the site, but it is also necessary to begin (in 2019?) the removal of the fuel present in the pools of the other reactors and above all to begin the removal of the molten fuel from reactors 1, 2 and 3. 30 to 40 years are planned for this.

According to the IRSN (French Institute for Radiation Protection and Nuclear Safety), the announced delays are to be considered as orders of magnitude,  » knowing that important operations of in-depth characterization of the state of the installations as well as research work are still to be carried out « . This diplomatic and complacent language reveals the situation: we are nowhere.

But beyond these enormous management problems, to be implemented under difficult safety conditions (in particular for the thousands of workers, subjected to significant radiation doses), it is the suffering of the populations that must be taken into consideration. Above all, there is the permanent concern for health due to exposure to radioactivity. Thyroid pathologies are at the forefront of concerns. This is all the more legitimate since the precedent of Chernobyl and the absence of preventive protection measures in the hours following the first radioactive releases must make us fear an escalation. The global situation leaves little room for doubt; the human, ecological and economic catastrophe is still ongoing, no matter what the « Japanese officials » say.

Rather than recognizing the seriousness of the problems, the government persists in its attitude of denial. In order for Tokyo to host the 2020 Olympic Games, Prime Minister Shinzo Abe did not hesitate to lie brazenly that the situation was under control and that the impacts of the accident were limited to the plant site. The political impact of the disaster is not limited to this event. In December 2013, the parliament passed a law prohibiting the disclosure of so-called « sensitive » information covered by the state secret. The latest annual Reporters Without Borders press freedom index has dropped Japan to 59th place. Before March 11, 2011, Japan was ranked 11th.

The aim is to save Japan’s image in the eyes of international opinion and, with the complicity of the international bodies in charge of nuclear safety and health (IAEA and WHO), to pass on the message that it is perfectly possible to live in a contaminated area after a serious nuclear accident. In Belarus, after the Chernobyl accident, the Ethos program, financed by the European Union, aimed to show how to live in a contaminated area and thus help the population to come to terms with an unacceptable situation. The same program was implemented in Fukushima …


The three catastrophic accidents that put nuclear reactors out of control and caused a radioactive core meltdown were impossible accidents. They took place in a scenario not envisaged by the safety authorities. Each time, simultaneous or successive failures and unthinkable human errors have led to a situation rejected by experts for whom totally improbable is synonymous with impossible.

Impossible was the formation of a hydrogen bubble that could explode at Three Mile Island. Impossible was the nuclear excursion in a RBMK reactor and the explosion of a reactor in Chernobyl. Impossible was a magnitude 9 earthquake, followed by a tsunami and the shutdown of everything that ensured the safe operation of three reactors in Fukushima.

After Three Mile Island, an effective disinformation campaign was able to reassure a complacent public opinion: all things considered, it was not so serious; moreover, there were no victims and the radioactive discharges were insignificant. To hell with the alarmists!

After Chernobyl, blaming the obsolete Soviet technology and the lack of seriousness of the operators could easily reassure Westerners and give credence to the idea that such a disaster was unthinkable here.

After Fukushima, in a country at the cutting edge of modernity, what can we say but invoke a nasty fatality and the exceptional and unthinkable character in other places of a tsunami of such magnitude.

Unfortunately for the believers, blinded by their technical faith, the facts are indisputable: 50 years of nuclear power production in the world have seen three impossible accidents and five radioactive core meltdowns (three in Fukushima). The question is no longer whether a nuclear disaster can happen in Europe, but when it will happen.

Either one accepts this bleak prospect and prepares to live in a contaminated area for decades. Either we program the shutdown of all the nuclear reactors in operation, starting with the most dangerous ones, i.e. those located in seismic zones and in densely populated industrial areas, such as Doel and Tihange.

Paul Lannoye

Notes et références
  1. Ce réacteur était du type PWR (à eau pressurisée) comme ceux de Doel, Tihange et Chooz.
  2. Le corium est le magma résultant de la fusion des métaux du cœur et de l’uranium combustible.
  3. E.O.Talbott et al., Long term follow-up of the Residents of the Three Mile Island Accident area : 1979–1998 ;Environmental Health Perspectives ; Vol 111 n° 3 ; mars 2003.
  4. Mycle Schneider et Antony Froggatt, « World nuclear industry statut report 2015 », juin 2015.
  5. Les deux citations sont de Jean-Pierre Pharabod et Jean-Paul Shapira, tous deux ingénieurs et spécialistes du nucléaire, dans leur ouvrage « Les jeux de l’atome et du hasard », Calmann-Lévy, 1988.
  6. La puissance thermique au cœur du réacteur n’est pas la puissance électrique ; les 2/3 de la puissance sont perdues sous forme de chaleur.
  7. A.Yablokov, V. Nesterenko et A. Nesterenko : Chernobyl ; « Consequences of the catastrophe for people and the environment » ; Annals of the New York Academy of Sciences ; Vol 1181, déc. 2011.
  8. Martin Tondel et al., « Increase of total cancer incidence in North Sweden due to the Tchernobyl accident? »Journal of Epidemiol.Community Health, n° 58, 2004.

Espace membre

Member area