EN | FR

Three research areas



Three complementary research areas

The three research areas under the Bataille Act, later supplemented by the 2006 Act, offered a wide range of management methods. The research areas were intended to be complementary rather than competing. Christian Bataille, the promoter of the 1991 Act, did not want any one area to be 'put on a pedestal'.

For example, although there is no question that large disposal facilities will have to be built, transmutation, which can be used to destroy some of the particularly troublesome waste, could considerably reduce the volume of waste that ultimately needs to be buried. The safety of a geological disposal site will only be enhanced by better conditioning, and sensibly designed storage will make it possible to benefit from technological progress.

Three research areas
The three research areas defined by the 1991 Bataille Act aim to solve the radioactive waste issues for which solutions have still not been found and to offer an array of management methods. The three areas are complementary. The first is engaged in exploring the potential for separating and destroying certain radioactive atoms, and the second is looking at the deep geological disposal of high-level waste, which will be inevitable. The third is aiming to improve the conditioning of waste and to enable it to be stored for long periods of time.
IN2P3

Area 1: Partitioning and transmutation

Research on partitioning looks at the methods for sorting nuclear waste. The intention is to go beyond the partitioning carried out at present, which merely separates the uranium and plutonium from the fission products and minor actinides. For example, the research will look at how to partition a fission product like caesium so that it can be conditioned separately.

Partitioning is particularly useful if it is followed by transmutation. It would then be possible to eliminate radioelements such as the minor actinides, which represent less than one thousandth of the mass of spent fuel but are very long-lived. Transmuting actinides into radioactive products with a short half-life, or even into stable elements, takes nuclear reactions and a return to the reactor. But what kinds of reactors are needed to achieve transmutation that is efficient enough?

Area 2: Deep geological disposal

Can final waste, for which we currently have no other use, be buried deep underground without posing any risks? In what kind of soils can this type of disposal be considered? How can we be sure a disposal facility of this kind will continue to function for long periods of time? For how long should a disposal facility be reversible? If by chance our descendants discovered a way of using the waste, or making it disappear, could we be sure they were able to remove it from the disposal facility?

Area 3: Conditioning and storage

Do good conditioning solutions already exist for multiple types of waste? Can they be improved, made stronger, adapted to the types of radioactive atoms that need to be immobilized?

Do good conditioning solutions already exist for multiple types of waste? Can they be improved, made stronger, adapted to the types of radioactive atoms that need to be immobilized?

Spent fuel and vitrified waste are already stored at facilities designed to last for decades. Could the life of these facilities be extended for periods of up to 300 years? Should these storage facilities be surface or subsurface facilities? How can we ensure that storage facilities that precede but are not designed to replace disposal facilities are monitored and looked after for such long periods of time?

Go to page in French