#2. Bridging scales in the microstructure modeling of nuclear materials


  • Luca Messina, CEA Cadarache, France (luca.messina@cea.fr)
  • Antoine Claisse, Westinghouse AB, Sweden
  • Jaime Marian, UCLA, USA
  • Pascal Bellon, University of Illinois, USA
  • Larry K. Aagesen, Idaho National Laboratory, USA


Nuclear materials are an outstanding example of materials involving multiscale and multiphysics phenomena triggered by irradiation from the atomic level (creation of defects) to macroscopic degradation phenomena. Devising predictive models is crucial to foresee the performance of structural materials and fuel elements in order to anticipate and avoid critical failures. However, this is extremely challenging due to the vast range of length and time scales involved. Physical knowledge must be collected from the lowest scale (starting from electronic-structure calculations) and passed on to the higher scales, up to mechanical models. Ensuring that the collected information is correctly transferred across modeling scales is crucial to ensure the reliability of the final prediction, but such transfer is often subjected to simplifications and approximations. As computational methods and technologies advance, models become more and more sophisticated, allowing for a more efficient transfer of knowledge from one scale to the next one. This is even more relevant when considering the advent of advanced data mining techniques and artificial intelligence, whose use is becoming more and more common in materials science.

To inspire a true multiscale vision and encourage a stronger interaction between researchers working at different scales, this symposium is focused on the investigation of thermo-kinetic properties, phase transformations, and microstructure evolution of nuclear materials, aimed at predicting their behavior during operation. Particular emphasis is on bridging between modeling scales, i.e., how physical information is drawn from the lower ones, and how it can be transferred to the higher ones.

Materials of interest include metallic and ceramic structural materials for fission and fusion applications, nuclear fuels, and cladding materials.

The symposium covers the following topics:
• Electronic structure calculations and molecular dynamics simulations of static and dynamic properties of bulk and defected materials
• Thermodynamic modeling of phase properties at equilibrium and in out-of-equilibrium conditions; kinetic modeling of atomic transport and diffusion phenomena
• Nano and microstructure evolution simulations (Kinetic Monte Carlo, Rate Theory, Cluster Dynamics, Dislocation Dynamics, Phase Field, …) of defect-driven phenomena (phase transformations, solute clustering and precipitation, segregation, surface reactions, swelling, creep), and construction of the underlying parameterization
• Interatomic potentials, data mining algorithms, artificial intelligence and any other model aimed at transferring physical information across modeling scales
• Experimental methods linked with supporting multiscale models to improve the understanding of microstructure evolution under irradiation
• Use of the previous topics to derive mesoscopic models of thermo-mechanical properties of fuel elements and structural components during operation

Confirmed Keynote Speakers

• Marie-France Barthe (CNRS Orléans, France)
• Noam Bernstein (US Naval Research Laboratory, United States)
• Michael Cooper (Los Alamos National Laboratory, United States)
• Guang-Hong Lu (Beihang University, China)
• Ludovic Thuinet (Université de Lille, France)
• Michael Tonks (University of Florida, United States)