3D Periodic Cellular Materials with Tailored Symmetry and Implicit Grading☆
Graphical abstract
Introduction
3D periodic cellular materials are an essential ingredient in the manufacture of advanced functional parts. They allow triggering various elastic responses within a part by structuring its internal volume at a small scale. Ideally, a method to model cellular materials would allow for a wide gamut of elastic responses and geometries that could be freely chosen within the part volume.
A central line of research is the design of periodic cellular materials, defined from a representative volume element (or tile) repeated in space. The periodicity simplifies the design and analysis while limiting computational and storage requirements. We consider two types of cellular materials [1]: closed-cell structures which are formed by isolated void cavities (cells), and open-cell structures, formed by connected solid beams (possibly curved).
A key to achieving complex, dynamic behaviors [2] is to allow for free spatial gradation of the structures within the part volume. Most techniques employ a catalog of tiles, pre-determined or optimized, placing different tiles with different elastic responses in different locations. Adjacent tiles will generally not connect properly, requiring applying constraints on the sets of tiles or optimizing after placement (see Section 2).
In this work, we take a distinct view on this problem. Rather than reasoning in terms of a fixed number of tiles, we start from a periodic lattice of nuclei (points) and associate each parametric description of the local structure being produced — a 3D star-shaped set. Each star-shaped set implicitly defines a distance function. Since the star-shaped set is possibly non-convex, we span a wide range of possible distance functions and a more diverse space of cellular solids compared to existing techniques. An efficient computational process grows the structures around each nucleus. Intuitively speaking, the cells centered around the lattice nuclei grow, obeying a law of growth governed by the star-shaped distance while being forbidden to overlap. Once the growth process is complete, we obtain a cellular structure, which can be extracted as either closed or open-cell. If the same star-shaped distance is used at every nucleus, we obtain a periodic structure, from which a single period can act as a traditional periodic tile. If different star-shaped distances or lattices are used, we obtain spatial gradations.
Resorting to a classical Voronoi diagram under the star-shaped distance may lead to Voronoi cells with more than one connected component. Instead, the growth process ensures that each cell is connected by construction. More precisely, we consider a discrete growth process since it is challenging to formulate a continuous one: our growth process is atypical, guided by the growth of a non-convex set. To the best of our knowledge, continuous growth processes have been studied for convex distances (e.g. [3]) but not for star-shaped ones.
Our structures can be prescribed to enforce certain crystallographic geometric symmetries — and hence a given type of elastic response: monoclinic, orthorhombic, trigonal, and so on. The parametric star-shaped distances can also be easily interpolated, making it possible to transition from one geometry to another smoothly. Thus, our approach supports interpolation across different symmetries and topologies without any special treatment. It spans a broad space of possible geometries with controlled symmetries that can be freely combined and spatially graded. Finally, the compact parameterization of the distance and the lattice facilitates exploring the wide variety of possible cellular materials.
We now define the terminology used throughout the paper. A tile, or representative volume element of a structure, is a periodic computational domain. The domain is covered by a regular grid, a set of discrete points each representing the center of a voxel (a cubic cell of the grid centered on the point). We define a nucleus as the starting point of growth of cavities or cells on the grid. We call a lattice the discrete set of nuclei distributed within the tile. The lattice has translational symmetries; therefore, it can be represented by periodic crystals [4]. Finally, a compact set is star-shaped with respect to the origin if for all , the segment is contained in .
Section snippets
Related work
There is a large body of prior work studying periodic cellular materials, especially as 3D printing now enables the fabrication of small-scale internal structures [5]. A key challenge is to infill the interior of an object while grading physical properties. Earlier works focused on density alone, but recent approaches attempt to give finer control over the object’s average elastic behavior.
Method
This section introduces the two fundamental ingredients of our method: the computation of the parametric cellular geometry and the analysis of the connection between parameters, geometric symmetries, and elastic symmetries. We start by introducing in Section 3.1 our computational method to generate parameterized cellular structures with a target symmetry. Then, in Section 3.2, we examine how geometric symmetries derive into symmetries of the elastic behavior.
Results
We now explore the variety of structures and elastic behaviors spanned by our method, focusing on symmetries. Please keep in mind that our method can seamlessly transition between all the shown structures of the same class (open or closed-cell) throughout the result sections.
Conclusions
We have introduced a method for generating cellular materials with a compact and straightforward parameterization covering a broad spectrum of elastic symmetries. We provide a first study that attempts to shed some light on the materials given by our parameterization. Notably, our method can generate periodic tiles or spatial gradations implicitly without requiring to deal with spatially-varying topological or geometrical changes. We have numerically verified the expected symmetry of the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was partly supported by the ANR MuFFin (ANR-17-CE10-0002) and “Région Lorraine and FEDER” . We thank Pierre-Alexandre Hugron and Pierre Bedell for providing their help with 3D printing and making the photos of the results.
References (59)
- et al.
Additive manufacturing (3D printing): A review of materials, methods, applications and challenges
Composites B
(2018) - et al.
Fracture of three-dimensional lattices manufactured by selective laser melting
Int. J. Solids Struct.
(2019) - et al.
The structural performance of the periodic truss
J. Mech. Phys. Solids
(2006) - et al.
The fracture toughness of octet-truss lattices
J. Mech. Phys. Solids
(2017) - et al.
Octet-truss cellular materials for improved mechanical properties and specific energy absorption
Mater. Des.
(2019) - et al.
Periodic truss structures
J. Mech. Phys. Solids
(2016) Tailoring materials with prescribed elastic properties
Mech. Mater.
(1995)Materials with prescribed constitutive parameters: An inverse homogenization problem
Int. J. Solids Struct.
(1994)- et al.
Design of manufacturable 3D extremal elastic microstructure
Mech. Mater.
(2014) - et al.
Multi-scale space-variant FRep cellular structures.
Comput. Aided Des.
(2013)
Parametric design of voronoi-based lattice porous structures
Mater. Des.
Compatibility in microstructural optimization for additive manufacturing
Addit. Manuf.
Concurrent design with connectable graded microstructures
Comput. Methods Appl. Mech. Engrg.
Generalized cowin–mehrabadi theorems and a direct proof that the number of linear elastic symmetries is eight
Int. J. Solids Struct.
Effective properties of composite materials with periodic microstructure: a computational approach
Comput. Methods Appl. Mech. Engrg.
Fourier-based schemes for computing the mechanical response of composites with accurate local fields
C R Méc
A new proof that the number of linear elastic symmetries is eight
J. Mech. Phys. Solids
Numerical technique for the 3D microarchitecture design of elastic composites inspired by crystal symmetries
Comput. Methods Appl. Mech. Engrg.
A numerical method for computing the overall response of nonlinear composites with complex microstructure
Comput. Methods Appl. Mech. Engrg.
Cellular solids: structure and properties
Flexible mechanical metamaterials
Nat Rev Mater
Integration in a dynamical stochastic geometric framework
ESAIM Probab. Stat.
Crystal lattices and crystal symmetry
Design, representations, and processing for additive manufacturing
3D metamaterials
Nat Rev Phys
Design of smart composite materials using topology optimization
Smart Mater. Struct.
Shape optimization by the homogenization method, Vol. 146
Design of graded two-phase microstructures for tailored elasticity gradients
J. Mater. Sci.
Topology optimization of functionally graded cellular materials
J. Mater. Sci.
Cited by (7)
Solid Modelling for Manufacturing: From Voelcker's Boundary Evaluation to Discrete Paradigms
2022, CAD Computer Aided DesignCitation Excerpt :Design of internal microstructures Lattices have also demonstrated their usefulness in weight and used material reduction, as well as in energy absorption applications. In particular, cellular materials [63] may be used to modify the elastic response of printed parts. Other more complex microstructures [64] allow for orientation control of the deformation of the resulting material.
Mechanical and corrosion performance of biodegradable iron porous structures
2024, Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and ApplicationsNeural Metamaterial Networks for Nonlinear Material Design
2023, ACM Transactions on GraphicsRecent Progress on Geometric Microstructure Design: A Survery
2023, Jisuanji Xuebao/Chinese Journal of ComputersNeural Network-Assisted Design: A Study of Multiscale Topology Optimization With Smoothly Graded Cellular Structures
2023, Journal of Mechanical Design
- ☆
This paper has been recommended for acceptance by “Michael Barton, George-Pierre Bonneau & Saigopal Nelaturi”.