A Computational Approach to Modeling Nature-Inspired Structural Ceramics
The multi-scale approach permits the computation of the failure probability of any structure under any mechanical load, solely based on considerations of the local micro-scale structure and its failure properties.
Members of the Mechanical Behavior of Materials Program established a novel computational modeling approach to predict the statistical failure of nature-inspired ceramics at multiple length-scales, coupling micro-scale criteria to macroscopic behavior.
Significance and Impact
Material properties can be created at specific microstructural levels, but must be designed with regard to their macroscopic distribution. The new model permits physically-based coupling of length-sales, bridging the micro- and macroscales.
Natural materials generally comprise both soft and hard constituents, often with meager mechanical properties. Yet through the creation of composite structures with complex hierarchical architectures over multiple length-scales, these materials can display spectacular structural performance. Such natural design is providing inspiration to the development of new synthetic materials, in the anticipation of generating unprecedented combinations of properties. One example is biologically-inspired lightweight ceramic structures modeled on the architecture of mollusk shells.
However, theoretical prediction of these ceramic structures’ properties proves difficult, as the structures fail in a very statistical and size-dependent manner. In principle, to quantitatively characterize their fracture properties, it is necessary to perform large numbers of experiments to measure the distribution of strengths down to low failure probabilities on every potential application scale: it has not been possible to link properties measured at one length scale to another. Here we establish a novel computational modeling approach aimed at bridging the statistical failure properties of bioinspired ceramics, at this point from the microscale to macroscopic samples. Although this approach is in its infancy, we believe that it has the potential to reliably model statistically the complex size-dependent fracture properties of these exciting new structural materials.