Behind the ITER Fusion Energy Project, Mountains of Toxic Metals and Radioactive Waste (2/3)

by Celia Izoard, Reporterre, June 17, 2021

Derrière le projet Iter, des montagnes de métaux toxiques et de déchets radioactifs

Translated by Dennis Riches

The original publication cites many links to French language sources. These have been removed in the English translation.

Part 2 of a three-part series: ITER, the Reality Behind the Promises of Nuclear Fusion

Presented as a “clean” project that will contribute to the fight against climate change, the experimental ITER fusion reactor nevertheless requires a quantity of polluting and carcinogenic metals, and it will produce a great deal of radioactive waste. ITERwants to be the showcase of nuclear fusion, whose qualities, according to its promoters, surpass those of the fission reactors used in conventional nuclear power plants. This investigation looks at the heart of a massive project that will have disastrous health and environmental consequences.

Part 1: The Future ITER Nuclear Reactor: A Titanic and Energy-Intensive Project

Part 3: The Abyss at ITER Fails to Discourage Other Nuclear Fusion Projects

“If I come to work every morning, it is because I believe that there is no alternative to nuclear fusion to fight climate change. It will be necessary to power the entire fleet of electric vehicles. Wind and solar will not be enough,” explains Laban Coblentz, ITER’s director of communications, to Reporterre from the glass office that dominates the gigantic construction site of the experimental reactor located in the Bouches-du-Rhône. In theory, a nuclear fusion reaction does not produce greenhouse gases. “The main by-product is helium, a non-toxic inert gas,” according to the ITER Organization. But in the most optimistic scenario where the nuclear fusion reaction is controlled, a functioning fusion-powered electricity sector could not see the light of day until after 2070.[1] To replace fossil fuels and hope to contain the rise in temperatures before the end of the century, it would be too late.

Apart from this problem, can we consider that ITER will contribute 100% to the fight against climate change, a claim that Europe used recently to justify allocating 6.6 billion euros to the project?[2]

On the ITER site in Cadarache, the construction of about forty monumental buildings has already required the excavation of 3 million cubic meters of earth, producing 150,000 cubic meters of concrete, and the installation of a THT line and a 4-hectare substation. To cool the reactor’s superconducting magnets, Air Liquide built the world’s largest cryogenic plant on site, powered by helium (produced from methane) imported from Qatar. To perform the calculations necessary to parameterize the fusion reaction, ITER uses supercomputers that, from 2035, will generate 2.2 petabytes of data every day, the equivalent of 20,000 hard drives of consumer computers. The energy consumed by these drives will be all the more energy-intensive as they will require two daily backups. But the most worrying aspect of the environmental accounting of fusion energy is the unprecedented quantities of metals needed for such a reactor.

Consider a little-known metal, niobium. The principle of the fusion reaction that will take place in the ITER tokamak—a kind of magnetic bottle—requires confining a plasma raised to more than 150 million degrees Celsius by means of gigantic magnetic fields. To produce them, 10,000 tons of superconducting magnets, the largest ever designed, are on their way to ITER. Their coils are made of two alloys of precious metals, niobium-titanium and niobium-tin.

Niobium mine in Brazil

“The exceptional size of ITER’s magnets […] has disrupted the global superconducting market,” enthuses the Organization in its magazine ITER Mag. The machine will use more than one-fifth of the world’s annual production of niobium-titanium. As for niobium tin, its production […] had to be multiplied by six in order to meet ITER’s needs alone. In total, ITER will use nearly 450 tons of niobium. And it is only an experimental reactor. We are far short of an actual fusion energy sector that would consume many more metals. What would all this mean in practice?

Although the ITER Organization has not been able to tell us where its metals originate, we know that niobium is the cherished metal of Jair Bolsonaro, the far-right president of Brazil, who considers it “more important than oil”[3]. In fact, 85% of the niobium extracted in the world comes from two Brazilian mines. In the state of Minas Gerais, the Companhia Brasileira de Metalurgia e Mineração (CBMM) is the world’s leading producer. Rodrigo de Castro Amédée Péret has denounced the company, saying, “It has been contaminating the groundwater of the Araxá basin for at least thirty-six years.” He has been working with the group Franciscan Action for Ecology and Solidarity. As a result of contamination of water with barium, chromium, lead, vanadium and uranium, many families have suffered from various types of diseases, such as cancer, and kidney and cardiovascular diseases. About 200 families living in the area had to leave their homes after the contamination was discovered. Through the Barreiro Residents’ Association, these families were able to win a court ruling that the city government must provide mineral water to the residents who still live there.”[4]

To meet the growing demand for niobium[5], the Brazilian President is in the process of setting up an exceptional procedure allowing the deposits of the Rio Negro basin to be exploited in the near future, in a nature reserve in the Amazon rainforest where twenty-three indigenous tribes live, including the Yanomami[6]. The fusion energy solution already requires the destruction of the Amazonian forest and the expropriation of its last inhabitants.

Another essential metal for nuclear fusion is beryllium. It is refractory, a good thermal conductor, and ultra-resistant. It will be used to cover the walls of the vacuum chamber of the ITER tokamak—the surface that will be closest to the thermonuclear plasma. Alain Bécoulet, physicist and chief engineer of ITER[ 7] acknowledged in 2019 that Beryllium is not sustainable. And for good reason. This metal is on the shortlist of the most toxic natural elements in the world, alongside arsenic and mercury. Beryllium acts as a carcinogenic poison, as described in a 2011 report by the Bureau of Geological and Mining Research (BRGM), and can remain detectable in urine for up to ten years after exposure. Even inhaled in minute doses, it causes two serious diseases, berylliosis and lung cancer.

In the majority of its uses, in electronics, it is used in amounts measured in grams. The ITER reactor will consume 12 tons! While global beryllium production is estimated at more than 300 tons per year, “it is envisaged that several hundred tons of this metal will be used “for future fusion tokamaks that could emerge after ITER,” worries a team of Russian scientists who are wondering how to recycle this beryllium once it has been irradiated in a tokamak.[8]

The extraction and refining of beryllium is already a thorny problem, solely because of its extreme toxicity. Added to this is the fact that most deposits also contain uranium. One can imagine the dangerous nature of the mountains of tailings stored near the mine sites. ITER’s beryllium will be extracted in the United States, Russia and China. China’s main mine, Koktokay No. 3, whose existence was only recently made public[9], is located in Xinjiang, the Uyghur autonomous region subjected by Beijing to ongoing human rights violations.

Russian—and perhaps Chinese—beryllium is refined in the Ulba metallurgical plant in the far east of Kazakhstan. In contrast to the conditions of beryllium extraction in Uyghur, some data exist on the sanitary status of the city of Ust-Kamenogorsk. According to a team of Kazakhstani researchers, the cumulative discharges from this industrial basin have led to a situation of “environmental crisis.” They report an “increasing incidence of cancers” and “respiratory diseases”, noting that “the amount of highly polluting components such as beryllium” present in the city’s air “is not measured”.[10]

Bombarded with neutrons, the beryllium cover of the ITER tokamak will disintegrate rapidly. The lifetime of this metal in a fusion reactor would be five to ten years[11]. It will not only be necessary to replace its modules regularly, but also to evacuate beryllium dust after each experiment. “This dust have a lot of defects,” points out Joëlle Elbez-Uzan, director of safety at ITER. First, it’s beryllium. In addition, it will be heavily irradiated. Finally, it is pyrophoric [ignites easily on contact with air at normal temperatures].” A whole filtration system is provided in the ventilation of the installations to prevent the spread of this dust. “It will be evacuated by giant fully automatic vacuum cleaners, heated in an oven and then fixed in cement dies whose composition is being studied to find a material that will prevent explosions. They will then be put in leak-proof drums to store them with other radioactive wastes.”

Radioactive waste? One of the promises of nuclear fusion was precisely that it would not create such waste. “One of the big advantages of this nuclear sector,” Alain Bécoulet said in 2020 on France Culture, “is that we do not have radioactive waste products either at the beginning or the end of the cycle. Let us review these arguments. No radioactive products as inputs? This is said only at this stage when we are gambling on the future. The planned reaction at ITER requires radioactive tritium, which will come from Canada’s heavy water nuclear fission reactors.

Perhaps one day, it will no longer be necessary, if the cover of the beryllium vacuum chamber makes it possible to produce tritium within the tokamak itself, from lithium subjected to neutrons. But the French Institute of Radiation Protection and Nuclear Safety (IRSN) considers that “this objective, which is a sine qua non for the industrial operation of tokamak fusion reactors for the production of electricity, is difficult to achieve.”[12]

No radioactive products at the end of the cycle? This is clearly untrue. “During a fusion experiment,” says Michel Claessens, ITER’s former director of communication, in his book ITER, Star of Science. “Barely 2% of the tritium will be consumed. The remaining 98% will spread in pipes and materials. Therefore, tritium contamination of cooling water cannot be avoided.”[13] Tritium absorbed by the walls will have to be constantly recovered to try to reinject it into the reactor, and the rest of this tritium will have to be separated from the cooling water to fix it in matrices, such as beryllium, before storing it. Moreover, it must be borne in mind that the entire 23,000 tons of the tokamak (three times the weight of the Eiffel Tower), will be irradiated during the experiments, so it will itself become a mountain of nuclear waste. There will be so many metals that it will be almost impossible to recycle them.

In fact, the ITER Organization has always—discreetly—specified that the reactor would indeed generate radioactive waste, but “no high-level waste of long duration”—the type of waste that is planned to be stored 500 meters underground for several tens of thousands of years. At least 40,000 tons of waste would have to be stored for fifty years, including irradiated beryllium, which, because of its uranium content, will become medium-level long-lasting nuclear waste[14]. In fusion reactors, “The level of radioactivity per kilogram of waste should be lower than that of fission reactors,” summarizes Daniel Jassby, a physicist emeritus in nuclear fusion at Princeton University, “but their volume and mass would be higher.”[15]

“At this stage,” admits Joëlle Elbez-Uzan, “we cannot say that fusion is a clean and waste-free energy. But the goal is promising.” Faced with ITER’s “considerably high carbon footprint,” Daniel Jassby wonders on the contrary “how the reactor could ever be deemed a success after all the energy expended on it.” Obviously, he thinks it’s not possible.


[1] The ITER organization states, “From 2050, after tests at ITER to be held between 2035 and 2040, a reactor called Demo would have to be constructed, which is the prototype of a fusion reactor that could generate electricity. Only after this could a practical fusion energy sector be built.” Joëlle Elbez-Uzan, a spokesperson for environmental safety at ITER thinks that sector could come between 2080 and 2100.

[2Conseil de l’Union européenne, 12793/20, Bruxelles, 11/11/20, p. 8.

[3“Niobium’s Silent Impact in Brazil,” Caio de Freitas Paes, 5/04/19,

[4CBMM : 36 Anos de Contaminação em Araxá, 28/08/18, Fala Chico.

[5Production increased by 162.8% between 2005 and 2018. Main uses aside from superconductors: reinforcement of steel for bridges, gas and oil pipelines, reactor turbines, aircraft, electricity generating stations, rockets.

[6“Niobium mining in Brazilian Amazon would cause significant forest loss : Study,” Taran Volckhausen, Mongabay, 24/07/20.

[7“Fusion nucléaire : l’énergie à profusion,” La méthode scientifique, France Culture, 12/06/19.

[8] B.N. Kolbasov, V.I. Khripunov, A.Yu. Biryukov, Institut Kurchatov (Moscow), Proceedings of the 11th IEA International Workshop on beryllium technology, Barcelona, 2013, p. 107.

[9] For many decades, this secret mine provided beryllium for Soviet and Chinese nuclear weapons.

[10] T. Alimbaev, B. Omarova, B. Abzhapparova, K. Ilyassova, K. Yermagambetova, Z. Mazhitova, “Environment of East Kazakhstan : state and main directions of optimization,” 2020.

[11] B.N. Kolbasov, V.I. Khripunov, A.Yu. Biryukov, Institut Kurchatov (Moscow), Proceedings of the 11th IEA International Workshop on Beryllium Technology, Barcelona, 2013, p. 104-109.

[12“Réacteurs nucléaires de fusion. Considérations sur les questions de sûreté,” IRSN, 2017, p. 45.

[13Iter, étoile de la science, éd. du Menhir, 2018, p. 198.

[14] M. Claessens, op. Cit., p. 199. B.N. Kolbasov, V.I. Khripunov, A.Yu. Biryukov, Institut Kurchatov (Moscou), Proceedings of the 11th IEA International Workshop on beryllium technology, Barcelona, 2013, p. 104-109. [15Bulletin of Atomic Scientists, 14/02/18. See also L. El-Guebaly, V. Massaut, K. Tobita, L. Cadwallader “Goals, challenges and successes of managing fusion-activated materials,”Fusion Engineering and Design, Volume 83, Issues 7–9, December 2008, p. 928-935.