Nuclear power energy is a low carbon system for the electricity production that can contribute to address climate global warming in addition of the use of renewable energy. However, safety is a key issue for nuclear power plant operation due to the potential radiological consequences of an accident with regard to the environment and population. The Fukushima event in 2011 revived the interest for exploring new fuels that could tolerate a severe accident in a reactor core for a longer time than the current UO2/Zr alloy system while maintaining fuel performance during normal operation. Accident-tolerant fuels (ATFs) would enhance resistance to stress corrosion cracking due to pellet-clad interaction and increase the retention of gaseous fission products (FP), in order to avoid as much as possible radioactivity release and minimize the environmental impact of an accident. This double goal may be achieved by adding dopants like Cr2O3 to the UO2 fuel, whose main effect is to increase notably the fuel grain size. The gain of Cr2O3-doped UO2 fuel on the fission gas retention was demonstrated during power ramp tests. However, no data are available on the release of chemically active FP, especially those susceptible to be released in the environment during an accident. Three fission products, caesium, iodine and molybdenum are of particular interest. Cs and I are volatile FP that have a major radiological impact in case of release of their isotopes 137Cs and 131I. Molybdenum itself is not a dosing FP in case of accident, but its behaviour in the event of release strongly affects the chemistry of caesium and consequently that of iodine. In addition, it has a strong impact on the fuel chemistry by acting as an oxidation buffer. In this context, the BENEFICIA project proposes to study the mobility and chemistry of these three key FP in Cr2O3-doped UO2 fuel, in view to evaluate the impact of the Cr dopant on the radioactivity release in case of accident. This project gathers five teams (IP2I – coordinator, IJCLab, IRSN, CEA, FRAMATOME) including the major actors in charge of the design and safety evaluation of nuclear reactors. First, we will study the effect of Cr doping on the mechanisms of FP thermal diffusion in UO2. First, we will seek the influence of the oxygen partial pressure (pO2) and UO2 grain size on the diffusion coefficients. To do so, we will couple experimental (ion implantation -that mimic the FP production in the fuel- and SIMS (Secondary Ion Mass Spectrometry) analysis) and theoretical (DFT, molecular dynamics calculation using SMTB-Q potentials) approaches. The local environment of Cr atoms in UO2 will be investigated by coupling Rutherford Backscattering Spectrometry in channelling mode in Cr-doped single crystals and X-ray Absorption Spectroscopy (XAS). The second question concerns the effect of Cr on the FP speciation in UO2 at high temperature and as function of pO2. Indeed, the FP release during a severe accident is intimately linked with their speciation before the accident and its change as a function of temperature and oxygen potential during the accident. The synthesis of bulk UO2 compounds doped with FP of interest will be carried out. These so-called SIMFUEL will be heated under a controlled atmosphere, which will be followed by the identification of the species present before and after thermal sequence coupling Scanning Electron microscopy, µ-probe, SIMS and XAS. Finally, the experimental results (diffusion coefficients, speciation) will be implemented in safety simulation codes currently developed by IRSN (MFPR-F) and CEA (MARGARET) aiming to model the fission products behaviour in the fuel in accidental conditions, and in thermodynamic databases (MEPHISTA, TAF-ID). A comparison between MFPR-F and MARGARET will be carried out on the prediction of the release of FPs under accidental conditions and to identify possible areas for improvement in the modelling.