Thursday, 21 January 2021Speaker: Prof. Dennis KochmannInstitute of Mechanical Systems – ETH Zürich Spinodoids – cellular architectures with intriguing properties |
Abstract
Tailoring the architecture of cellular materials – from random foams to periodic structures based on trusses, plates,and shells – has resulted in a variety of lightweight architected materials with beneficial mechanical properties. While periodic networks offer advantages such as a simple effective property extraction, they come with significant shortcomings including the sensitivity to symmetry-breaking defects, challenges and limitations arising from spatial unit cell variations, and non-scalable fabrication techniques. So-called spinodal architectures, by contrast, are non-periodic and show little sensitivity to defects, and they evolve naturally during diffusion-driven self-assembly processes. Inspired by spinodal architectures found in nature, we introduce spinodoid topologies as an efficient theoretical description of “spinodal-like” structures. Spinodoids follow a simple mathematical parametrization and have intriguing mechanical properties such as optimal stiffness scaling with density as well as a superb resilience due to their curvature distribution. They further cover an enormous anisotropic property space in a seamless fashion, which enables spatially graded structures with locally optimized mechanical properties and multiscale topology optimization of engineering structures. Moreover, we present experimental prototypes of spinodoid architectures and demonstrate their mechanical resilience. Finally, we address the inverse challenge of identifying a microstructural architecture from the huge design space to achieve a sought combination of macroscale properties. We demonstrate that a data- driven approach based on the integration of two neural networks for the forward- and inverse-problems renders the challenge well-posed and reliably and accurately generates foam-type cellular metamaterials with as-designed 3D anisotropy and density in a spatially uniform or functionally graded fashion. As a particular example, we highlight the suitability of this new approach for the generation of bio-mimetic bone replacements.
Biography
Dennis Kochmann received his education at Ruhr-University Bochum and at the University of Wisconsin- Madison. He was a postdoc and Fulbright fellow at Wisconsin and a Feodor Lynen fellow at Caltech, before joining the Aerospace Department at Caltech as Assistant Professor in 2011. From 2016 to 2019 he was Professor of Aerospace at Caltech. Since April 2017 he has been Professor of Mechanics and Materials at ETH Zürich, where he served as Head of the Institute of Mechanical Systems and is currently Deputy Head of the Department of Mechanical and Process Engineering. His research focuses on the link between structure and properties of a variety of materials and metamaterials, which includes the development of theoretical, computational and experimental methods to bridge across scales from nano to macro. His research has been recognized by, among others, the Bureau Prize in Solid Mechanics form IUTAM, the Richard von Mises Prize by GAMM, an NSF CAREER Award, the T.J.R. Hughes Young Investigator Award by the ASME, and an ERC Consolidator Grant.
Notes
by Benjamin Dieu and Chloé Giraudet- Spinodoids are architectured materials which allow to mimic natural materials such as bones, which have non periodic spatial variations.
- In contrast to thruss and plates, spinodoids have the advantage that they do not have corners and therefore avoid stress concentrations.
- Numerically, spinodoids can be created thanks to mixing and demixing two materials through diffusion-driven phase separation. This way is computationally expensive, but the spinodoids can also be created through a sum of well chosen Gaussian Random Fields (GRFs) without solving any equation.
- Spinodoids can be anisotropic, and the anisotropy is tunable. Lamellae or column-like structures can be easily created by tuning a small set of parameters used for the GRFs generation. It is also possible to create materials that goes smoothly from a lamellae end to a columnar end.
- One of the main challenge in architectured material is to get a topology given a desired stifness. Usually the problem is ill-posed because there are many topologies that can correspond to a given striffness distribution. With two neural network, it is possible to train them to obtain a topology from a given stiffness, resulting in a fast and reliable method.
- There are many application for the spinodoids. Here are presented bones like structures in ceramics. The deformation level achieved in the experiments shows an impressive resilience of the material, when avoiding stress concentrations.
Suggested readings
- S. Kumar, S. Tan, L. Zheng and D. M. Kochmann
Inverse-designed spinodoid metamaterials
npj Computational Materials 6, 73, 2020. - C. M. Portela, A. Vidyasagar, S. Krödel, T. Weissenbach, D. W. Yee, J. R. Greer, and D. M. Kochmann
Extreme mechanical resilience of self-assembled nanolabyrinthine materials
PNAS 117, 5686-5693, 2020. - D. M. Kochmann, J. B. Hopkins and L. Valdevit
Multiscale modeling and optimization of the mechanics of hierarchical metamaterials
MRS Bulletin 44, 773-781, 2019. - A. Vidyasagar, S. Krödel and D. M. Kochmann
Microstructural patterns with tunable mechanical anisotropy obtained by simulating anisotropic spinodal decomposition
Proc. R. Soc. 474, 20180535, 2018. - J. Gao, Z. Luo, H. li and L. Gao
Topology optimization for multiscale design of porous composites with multi-domain microstructures
Computer Methods in Applied Mechanics and Engineering 334, 454-476, 2019. - J. W. Cahn
On spinodal decomposition
Acta Metallurgica 9, 795-801, 1961.
LMS seminars on related topics
- 17/12/2020, Pr. Fengwen Wang, Architected materials using topology optimization
- 17/12/2020, Dr. Emmanuel Siéfert, Inflating to shape: from soft architectured elastomers to patterned fabric sheets