The French Approach to Nuclear Waste

Declan Butler

[This article is adapted from “France digs deeper for nuclear waste” published in Nature 466(7308), 804-805 (12 August 2010). We are grateful to Dr. Butler and Nature for permission to edit it. The original article can be found at]

The American physics community will be acutely aware of the on-again, off-again long-term nuclear waste repository proposed for Yucca Mountain, which now looks to be abandoned after two decades of work and more than $10 billion in investment (P&S July 2008, January 2009, April 2009, April 2010). One of the main problems is that the selection of Yucca Mountain by Congress in 1987 was, from the outset, a political rather than a scientific choice. Efforts by the United States government to find a site have been stymied by opposition from individual states, where representatives and constituents are uneasy about having a nuclear dump in their backyard. In contrast, the Canadian government established a Nuclear Waste management Organization (NWMO) to develop an approach involving significant public input for the long-term management of their nuclear fuel (P&S, October 2009). Similarly, European countries have taken a more scientific and stepwise approach to locating sites, which has engendered greater public confidence; for example, Sweden and Finland involved local communities in decisions from the outset, which has increased acceptability [1]. As a result, Finland and Sweden plan to open deep geological repositories in about 2020-2025, whereas Germany hopes to open its own long-term repository in 2035. Several smaller European countries have banded together to form a European Repository Development Organization to work on the concept of a shared facility.

In this article, I describe the French approach to management of long-lived nuclear waste management. France generates about 80% of its electricity from 58 nuclear power plants, and is a world leader in the technology. Nuclear power enjoys staunch cross-party support in the country; anti-nuclear groups enjoy little public clout.

The Bure Laboratory

Half a kilometer beneath rolling wheat fields outside the small town of Bure in the Lorraine region of northeast France, the country is preparing to dispose of its radioactive waste. In a €1-billion (US$1.3 billion) underground laboratory, the French National Radioactive Waste Management Agency (ANDRA) is testing the soundness of the rock and the technologies to contain the waste. ANDRA scientists are convinced that the rock formations can safely house highly radioactive waste, and plan an industrial-scale facility that would open deep below a site nearby in 2025. The surface-level footprint of the site will occupy about 30 square kilometers; the underground repository itself will be smaller. It would be among the world’s first geological repositories for high-and medium-level long lived nuclear waste, and the largest. Patrick Landais, a geologist and scientific director of ANDRA, questions Yucca’s geological suitability not least, because of due to nearby volcanoes, and says that there are far better geological sites in the US than Yucca Mountain.

The Bure lab, authorized by the French government in 1999, has largely established the geological suitability of the area, and its findings have been endorsed by international experts. Now it is shifting into high gear, spending €100 million a year on research to pin down exactly how waste would be stored at the planned repository. ANDRA must present a blueprint for the repository to the government in 2014; if approved by the French National Assembly in 2016, construction would begin the following year. The assembly will then consider licensing the facility to open in 2025. Once completed, the repository would store all of the existing 2,300 cubic meters of high level and 42,000 cubic meters of medium-level long-lived radioactive waste which has been generated by France’s nuclear power stations as well as new waste created over at least the next 20 years. The existing waste is currently being stored at temporary sites in La Hague, Marcoule and Cadarache. The economic incentives that the facility offers to the Bure region have been welcomed by local officials, and there has been little effective resistance to the facility. Mobilizing public opinion to oppose the repository is difficult because the majority of the French are indifferent to nuclear power issues. Greenpeace France’s nuclear campaign does not oppose geological storage research, but has expressed concern that plans to seal the repository after a century of use would make it almost impossible to deal with a subsequent problem in the facility.


Subterranean experimental gallery of the Bure laboratory. Copyright 2006, Eric Sutre. Other photos of the facility can be found at php?id=itemmenu_phototheque_196_tout_1_1

The lab itself contains no radioactive waste, and never will. Instead, researchers at Bure are focusing on testing the rock and prototype waste-containment strategies. Almost all of the research results are analyzed remotely. Once scientists have installed their experiments, the output of instruments lining the tunnels is transmitted via the Internet to ANDRA’s own researchers, along with 80 collaborating labs at other research agencies and universities in France and other European countries involved in the project. A remote data-access system presents users with a three-dimensional representation of the galleries in which one can zoom in on any tunnel to find an experiment and pull up its data output and graphs in real time. Despite much remote access to the data, dozens of scientists and engineers still descend into the facility every day. This author had the opportunity to make the descent, accompanied by Dr. Landais.


On entering into the laboratory one is greeted by galleries crammed with scientific instruments. Incessant tannoys (loudspeaker systems) and the din of pneumatic drills and earth borers at work extending the lab, fill the air. The tunnel walls are reinforced with concrete, steel ribs and bolts, but here and there the grey 150-million-year-old Callovo-Oxfordian argillaceous rock (sedimentary rock formed from clay deposits) that would seal the repository is left bare.

Experimental boreholes in the walls carry about 3,500 sensors, which take the pulse of almost every mechanical, chemical and hydrogeological aspect of the rock. The data are fed into models that characterize the rock and also predict its future behavior over periods from decades to more than a million years. The experiments ultimately aim to answer one key question: can France’s most dangerous radioactive wastes be safely contained inside this 150-metre-thick layer of rock? The high-level waste includes the fission products such as cesium-134, cesium-137, strontium-90, and minor actinides such as curium-244 and americium-241. Most nuclear fuel in France is reprocessed to extract useful uranium and plutonium and to concentrate the waste. Although this high-level material comprises just 0.2% of the country’s nuclear waste by volume, it accounts for 95% of its total radioactivity.

The waste is immobilized by blending it into glass in a complex vitrification process which incorporates the radionucleotides in the atomic structure of the glass, a process that was pioneered by the French. The molten glass is poured into stainless steel casks, which are then placed inside steel barrels. Robots in the Bure repository will push these barrels into 70-centimetre-diameter boreholes called alveoli, which are drilled 40 meters horizontally into the walls of the main access tunnels. The medium-level radioactive waste, which comes from used reactor equipment and reagents, would be compressed into circular cakes and piled into steel canisters before being encased in concrete and stored in the tunnels.

Scientists at Bure are already testing the stability of the glass that would be used to immobilize the high-level waste, the rates of corrosion of the stainless steel casks, and the fate of the hydrogen gas that this degradation releases. They are also assessing all the interactions between the glass, the layers of steel and the rock in prototype alveoli. The canisters are designed so that heat from radioactive decay inside does not warm their surface beyond 90°C. Tests using mock-up canisters have shown that prolonged exposure to this temperature does not cause the rock to fissure. Although the volume of high-level waste is much smaller than that of medium-level waste, it will require double the amount of storage space because the hot casks must be spaced out with empty ones to avoid overheating. The scientists are also investigating ways to reduce the volume of waste to be sent to the facility, such as extracting radioactive elements from bulky graphite fuel elements and then concentrating them in order to allow much more medium-level waste to be packed into the repository’s chambers. The repository could eventually operate for at least a century, after which it would be sealed. A few thousand years later, the stainless steel would corrode away until it leaving the vitrified waste, and the rock itself, to provide containment.

The Long Term

ANDRA director Landais warns that rock is not an absolute barrier, as radionuclides would slowly diffuse through it. Of most concern at Bure are radioactive iodide and chloride anions, which are the most mobile in this type of rock. But Landais says that it would take hundreds of thousands of years for them to diffuse to the surface, by which time their low concentrations and lower levels of radioactivity would render any environmental contamination negligible. A more worrying problem is the possibility of a rock fracture, which could lead to radioactive leaks. But the research at Bure has largely confirmed that the layer of rock that would house the repository is homogenous, highly impermeable to water movement, and free from faults and seismic risk. Geologists at Bure are confident that it is a safe, predictable environment for nuclear waste: the rock is 150 million years old, hasn’t budged in the past 20 million years, and won’t in the next, they say. In addition, at the surface, researchers are extensively sampling the air, water and soils in a 250-square kilometer zone around the site to get a comprehensive baseline of environmental data. An observatory, created jointly in April with France’s agricultural research agency, INRA, will monitor this ecosystem for at least a century.

ANDRA researchers are optimistic that their efforts will lead to the opening of a safe, secure, publicly-acceptable repository, and thereby contribute to France’s continued successful program of nuclear-generated power.

[1]See D. Sarewitz, “Politicize me,” Nature 467, 26 (2 September 2010).

Declan Butler is a reporter with Nature magazine.

These contributions have not been peer-refereed. They represent solely the view(s) of the author(s) and not necessarily the view of APS.