Characterization of fracture in topology-optimized bioinspired networks

Abstract

Designing strong and robust bio-inspired structures requires an understanding of how function arises from the architecture and geometry of materials found in nature. We draw from trabecular bone, a lightweight bone tissue that exhibits a complex, anisotropic microarchitecture, to generate networked structures using multi-objective topology optimization. Starting from an identical volume, we generate multiple different models by varying the objective weights for compliance, surface area, and stability. We examine the relative effects of these objectives on how resultant models respond to simulated mechanical loading and element failure. We adapt a network-based method developed initially in the context of modeling trabecular bone to describe the topology-optimized structures with a graph theoretical framework, and we use community detection to characterize locations of fracture. This complementary combination of computational methods can provide valuable insights into the strength of bio-inspired structures and mechanisms of fracture.

A graph model to describe the network connectivity of trabecular plates and rods

“…Unlike these previous methods that focused on local measures like trabecular morphology or connectivity density and improved predictions over just using BV/TV, our goal with the PR Network was to emphasize the significance of higher-order connections between these trabecular bone elements in determining mechanical competence. This interrelation has just become a subject of more recent studies and has shown to be Frontiers in Bioengineering and Biotechnology frontiersin.org useful for identifying biomarkers for clinical procedures (Mondal et al, 2019;Nguyen et al, 2019). By evaluating the connectivity with respect to global and local boundary conditions, the developed PR Network increases the influence of channels of high-potential elements, stretching from source to sink.…”

From Brittle to Ductile: Symmetry Breaking in Strut-Based Architected Materials

“…Research aimed at mimicking the microscale features of crystalline structures has been proposed to develop architected materials with higher toughness by leveraging different fracture toughening mechanisms, such as grain boundaries, precipitates, and phases. , Some research aimed at designing graded structures with different relative density distributions within the design domain. − Several methods are also used to create heterostructures, such as applying the Voronoi diagram to the original reference points to rearrange a regular hexagonal honeycomb structure, using multiobjective topology optimization under fixed volume constraint to generate networked structures, stacking two materials with different microstructures, or constructing material foams . However, most of the current approaches for creating asymmetric cellular structures can overly complicate the modeling and manufacturing processes as they often use sophisticated unit cell arrangements or large amounts of input parameters.…”

Edge betweenness centrality as a failure predictor in network models of structurally disordered materials

“…In the context of materials design, it has been proposed that potential failure locations can be identified by correlation to large values of edge betweenness centrality 21 . Recent studies seem to confirm that material failure occurs preferentially at locations, which exhibit large Geodesic (i.e., strictly relying on the shortest path metric) Edge Betweenness Centrality (GEBC) values 22 , 23 . These results demonstrate that assessing failure locations of a system not necessarily requires the calculation of the local loadings, e.g.…”