Atomic scale simulation of gaseous and volatile fission products with application to LWR and GEN-IV fuels

Caglak Emre


Dubus Alain, (ULB),

SCK•CEN Mentor

Govers Kevin
+32 14 33 30 93

Expert group

Fuel Materials

PhD started


Short project description

Nuclear fuel performance is highly affected by the behaviour of gaseous (Xe, Kr) and volatile (Cs, I, Te) fission products, particularly at elevated burnups, where large amounts of them can potentially be relocated in the fuel rod, leading to different penalizing phenomena such as fuel swelling, overpressure in the rod or corrosion of the cladding (Cs, I); all potentially leading to clad failure. In addition, during accidental conditions, these elements are the first ones which would pass the first confinement barrier. The importance of gaseous and volatile fission product release was the motivation for large efforts to understand, both experimentally and theoretically, the different steps of the process.

Extensions to higher burnups, together with the growing interest in novel types of fuels such as inert matrix fuels envisaged for the transmutation of minor actinides, make that one is still looking for a permanently better modeling, based on a physical understanding and description of all stages of the release mechanism.


Experimental determinations of the diffusion coefficient exist for most gaseous and volatile fission products in different stoichiometric regimes, but suffer from large uncertainties, because of the difficulty to perform separate effect studies. Therefore, these data do not yet enable to determine the effective migration mechanism, even less under in-pile conditions.

Computer simulations nowadays offer an elegant way to circumvent these lacks by simulating the trajectories of atoms at the atomic scale, for a well-defined system. Such investigations are already available or are in progress for bulk diffusion of different gaseous and volatile elements, using both empirical interatomic potentials and first-principle calculations



The objective of the research proposed would be to investigate the behaviour of gaseous and volatile fission products at the atomic scale in the nuclear fuel using computer simulations. We propose to extend the scope of previous studies to their behaviour along dislocation lines and at grain boundaries. Parameters of importance are :

- the influence on the stability of the dislocation or grain boundary;

- the mobility of these elements, in order to better predict their release from the matrix and their chemical interactions with the fuel and the cladding;

- the swelling induced, being it as solute element in the matrix or solid or gaseous precipitate.


This will be considered in a first time in uranium dioxide (model compound for which more experimental data are available) for fission gases (Xe, He) and volatile products (Cs, I, Te). For the last ones, additional difficulties can potentially be encountered, since it is known e.g. that Cs could modify the valence state of U atoms from +IV to +VI.

The objective of the work performed in uranium dioxide double. On the one hand, it will generate the necessary inputs for multi-scale modeling of the fuels (up to fuel performance codes); on the other hand it will provide the basis for an extension of the study to other GEN-IV fuels, considering both oxide fuels (MOX, IMF-CERCER); as such as, on the long term, other types of compounds such as carbides and nitrides based fuels – (U,Pu)C, (U,Pu)N.


The research will be done using empirical interatomic potentials simulation techniques, which involve both classical molecular dynamics simulations, as such as advanced hyperdynamics simulations (temperature accelerated dynamics, biased potentials dynamics).


This project will fit into two EURATOM /Horizon 2020 research programmes, namely INFORMS, which will provide a large international framework for the proposed research.