The ability of C/C composite materials to maintain high mechanical properties at extremely high temperatures (exceeding 3000 K), combined with their low densities, justifies their use in extreme conditions, particularly in aerospace applications. Predicting the behavior of different constituents (fibers, matrices) from small scales is a fundamental step in developing virtual materials at the composite or component scale. Given their highly anisotropic properties, reliable, physics-based modeling at small scales is crucial.

The recent development of a digital synthesis method for dense anisotropic carbons (PolyGranular Image Guided Atomistic Reconstruction, PG-IGAR) has enabled the creation of a database of anisotropic pyrocarbon (pyC) matrix microstructures and the calculation of their elastic properties. Recently, the equations of state for these microstructures have been calculated to define valid anisotropic elasticity under pressure. The postdoctoral goal is to utilize established atomic-scale numerical models implemented in a mesoscopic finite element code to simulate the behavior of these materials under various mechanical loads and compare the model results with microscopic-scale simulations and micro-mechanical experiments.

An initial task will involve using a nonlinear hyper-elasticity model for large deformations implemented in the Lagrangian code Coddex, developed at CEA DAM DIF. To validate the behavior law, simulations of mechanical tests (mesoscopic and molecular dynamics) will be conducted on small-scale systems at deformation rates accessible through molecular dynamics. The emergence of irreversible deformation mechanisms within atomic-scale simulations may lead to a modification of the model, for example, by adding an effective plasticity. Subsequently, experimental-scale simulations will be performed with the Coddex code to compare the model's results with tensile-compression experiments on micro-composites and nano-indentatio