Tunable Energy Absorption in Diatom-inspired Architected Materials Designed for Additive Manufacturing

Abstract

Boosted by additive manufacturing, architected materials have opened new opportunities to extend the perfor-mance of engineering materials. Yet, their development is held back by the intense efforts required to understand their com-plex property-structure-process-performance relationship. Therefore, data-driven biomimetic approaches are becoming in-creasingly popularto unveil suchrelationships. Here we mimic the functionally graded structures found in Coscinodiscus sp.diatom to understand the role of their shapes and define new guidelines for the design of novel architectedhoneycombswithtunable mechanical properties. Finite element simulations, validated on the outcome of a testing campaign performed on3D-printedelastomeric samples, are used to build a dataset for machine learning algorithmtraining. Differentmachinelearningtechniques areused tolinkthe geometric features of the designed biomimetic structuresto theirenergy absorption proper-tiesand,in particular,to the specific absorbed energy divided by the peak force,here used asthe performance index. Theproposed approach leads toa novel design, which featuresaperformanceincrease of 250% w.r.t. conventional honeycombs.