|About this Abstract
||2022 TMS Annual Meeting & Exhibition
||Design of 2D-Mechanical Metamaterials with Spinodal Topologies
||Kiara McMillan, Doğacan Öztürk, Pinar Acar
|On-Site Speaker (Planned)
This study investigates the design of a subset of mechanical metamaterials that demonstrate enhanced material performance through their spinodal topologies. They are generated through spinodal decomposition modeled by the Cahn–Hilliard equation. Instead of utilizing the inefficient method of modeling this time-dependent equation, formulation introduced by Kumar and co-workers, which uses Gaussian random fields to generate a special form of spinodal microstructures called spinodoids, is followed. With this approach, 2D-spinodoid images are generated by controlling their topology with orientation parameters. Principal Component Analysis (PCA) is applied to reduce the high dimensionality of this image representation and present an efficient means of characterizing this dataset. Physics-based simulations will be conducted to determine the homogenized mechanical properties as a function of the spinodoid topology. After characterizing the generated spinodoids, an inverse problem with uncertainty will be considered to find the corresponding 2D-spinodoid that provide prescribed values of properties using the PCA representation.
||Characterization, Phase Transformations, Other