Large Strain Finite Element Analysis of Spinodal Shell Structures

Abstract

Nanoarchitected materials combine architecture properties with nano-scale size effects such as increased tensile strength, but often suffer from localized failure and low mechanical efficiency. Architectures derived from spinodal decomposition have promising material properties due to their continuous doubly curved surfaces with minimal stress concentrations. In this work, finite element analysis was used to model spinodal shell structures, allowing the study of the architecture properties and inspection into the interior of the structure to allow a better understanding of the nanomaterial behavior. Isotropic, lamellar, columnar, gradient, and bioinspired conch architectures were simulated up to 0.5 strain at relative densities between 0.001 and 0.01, in order to match the conditions of nanoindenter tests of the same structures. The simulations provided results close to the experiments in elastic stiffness and plateau stress. Each of the architectures scales elastic modulus and yield strength proportional to relative density squared. The experiments showed extensive localization effects, especially localized buckling at the bottom layer and localized fracture in the top layer, but the simulations showed only localized auxetic behavior in the middle of the structures. Strain localization was observed in the gradient structures, as the changing anisotropy encouraged strain to localize to each layer, but this strain did not cause localized failure. The stress state within the structures showed stress concentrations occur in areas of connections between layers, but the structures may have additional load paths that can be activated when the highly stressed areas buckle or fracture, preventing catastrophic failure and enabling recovery. This work shows the benefits of the spinodal architecture in eliminating localized deformation, controlling strain patterns through anisotropic and gradient architecture designs.

Description

Thesis (Master's)--University of Washington, 2023

Citation

DOI