Computational Thermodynamics—An Invaluable Tool for Predicting the Thermochemical Behavior of Nuclear Materials in All State

Fuel and structural materials in advanced nuclear reactors are often complex systems containing numerous chemical elements. Under operating conditions, many fission products form in the fuel whose chemical composition changes with burnup. The impact on the thermodynamic and thermal-physical properties of the fuel has to be predicted. Furthermore the chemical interaction with the surrounding structural materials in normal and off-normal conditions has to be investigated too. In this frame, Computational thermodynamics (CT) approach constitutes an essential step in the multi-scale modelling of materials to optimize the synthesis process, and simulate the high temperature behavior of these materials in various environments. The overall methodology to develop thermodynamic databases using the CALPHAD method, will be described, and how the Gibbs energy parameters of the phases can be assessed using both experimental measurements and atomic scale methods (Density Functional Theory, DFT and Quasi-Harmonic Approximation, QHA). Recent progress on the determination of the uncertainties on the thermodynamic parameters and how it propagates when calculating thermodynamic and phase diagram data using a Bayesian approach will be reported. To exemplify the capability of such computational tool, calculations on the chemistry of irradiated oxide fuels using the advanced TAF-ID (Thermodynamics Advanced Fuels – International Database) database will be shown [1]. Finally examples on how the TAF-ID database can be used with the Open Calphad software to take into account the thermochemistry of nuclear fuels in multi-physics codes will be presented.<br/>[1] C. Guéneau <i>et al.</i>, ‘TAF-ID: An international thermodynamic database for nuclear fuels applications’, <i>Calphad</i>, 72 (2020) 102212