Multiscale Modeling of Mechanical Deformation in Chemically Complex Alloys

Abstract: In my presentation, I will give an overview of three primary areas that have been my focal research interests at NOMATEN CoE: i) crystal and amorphous plasticity, ii) transport properties of high-entropy alloys (HEAs), and iii) micro-structural informatics. In i), my research has employed statistical physics to unravel the microscopic basis of plasticity based on the collective dynamics of shear transformation zones in amorphous solids as well as dislocations mechanics in crystalline metals. Within the context of HEAs, my focus has been on the role of chemical complexities (i.e. local disorder/ordering) investigating their impact on alloy strengths. In ii), I have explored the sluggish diffusion of defects in HEAs and its impact on thermo-mechanical properties. In the area of micro-structural informatics in iii), I have utilized the power of machine learning (ML) and graph neural networks GNNs to infer (hardness) properties solely based on the (surface) microstructural information. Building upon these achievements, we are currently expanding the scope of the above studies by i) employing ML to identify relevant microstructural metrics for predicting bulk plastic properties in bulk metallic glasses as well as HEAs within the microstructure-property paradigm, ii) utilizing machine-learned interatomic potentials for accelerated material discovery, and iii) extending the GNN’s capabilities to infer microstructural signatures and defects based on micro/nano mechanical response (as input data) in different metallic systems and distinct alloy compositions.

Speaker Bio: Kamran Karimi is a computational materials physicist working at the National Centre for Nuclear Research, Otwock Poland.