On the compressive strength of open-cell metal foams with Kelvin and random cell structures

Two families of finite element models of anisotropic, aluminum alloy, open-cell foams are developed and their predictions of elastic properties and compressive strength are evaluated by direct comparison to experimental results. In the first family of models, the foams are idealized as anisotropic Kelvin cells loaded in the <100> direction and in the second family more realistic models, based on Surface Evolver simulations of random soap froth with N³ cells are constructed. In both cases the ligaments are straight but have nonuniform cross sectional area distributions that resemble those of the foams tested. The ligaments are modeled as shear deformable beams with elasto-plastic material behavior. The calculated compressive response starts with a linearly elastic regime. At higher stress levels, inelastic action causes a gradual reduction of the stiffness that eventually leads to a stress maximum, which represents the strength of the material. The periodicity of the Kelvin cell enables calculation of the compressive response up to the limit stress with just a single fully periodic characteristic cell. Beyond the limit stress, deformation localizes along the principal diagonals of the microstructure. Consequently beyond the limit stress the response is evaluated using finite size 3-D domains that allow the localization to develop. The random models consist of 3-D domains of 216, 512 or 1000 cells with periodicity conditions on the compressed ends but free on the sides. The compressive response is also characterized by a limit load instability but now the localization is disorganized resembling that observed in experiments. The foam elastic moduli and strengths obtained from both families of models are generally in very good agreement with the corresponding measurements. The random foam models yield 5–10% stiffer elastic moduli and slightly higher strengths than the Kelvin cell models. Necessary requirements for this high performance of the models are accurate representation of the material distribution in the ligaments and correct modeling of the nonlinear stress–strain response of the aluminum base material.     

Compressive response of open-cell foams. Part I: Morphology and elastic properties

This study is concerned with the understanding and modeling of the compressive response of open cell foams. The response starts with a nearly linear elastic regime which terminates into a limit load followed by an extensive load plateau. The plateau, which is responsible for the excellent energy absorption capacity of foams, is followed by a second stiff branch. Results from polyester urethane open cell foams with relative densities of about 0.025 are used to illustrate this behavior using experiments coupled with several levels of modeling. The experiments include characterization of the microstructure and the properties of the base material and measurement of the compressive response of the foams of various cell sizes.A sequence of models for predicting the complete response of such foam is developed. The foam is idealized to be periodic using the space-filling Kelvin cell assigned the major geometric characteristics found in the foams tested. The cells are elongated in the rise direction, the ligaments are assumed to be straight, to have Plateau border cross-sections and nonuniform cross-sectional area distribution. The ligaments are modeled as shear-deformable extensional beams and the base material is assumed to be linearly elastic. Prediction of the initial elastic moduli are addressed in Part I. Closed form expressions for the material constants are presented as well as results using a FE model of the characteristic cell. Comparison between measurements and predictions is very favorable. The paper finishes with results from a limited parametric study of the elastic moduli. The results demonstrate that inclusion of the geometric complexities mentioned above is essential for successful prediction of the moduli of such foams. The nonlinear parts of the response including the foam crushing behavior are addressed in Part II.     

On the crushing response of random open-cell foams

This study examines the effect of randomness of the cellular microstructure on the calculated compressive response of a class of open-cell aluminum alloy foams. The foams are modeled using realistic random soap froth with N³ cells generated using the Surface Evolver software. The ligaments are made straight but with non-uniform cross sectional area distributions that mimic those of the physical foams. The models are also assigned the density and anisotropy measured. The ligaments are modeled as shear deformable beams with the elasto-plastic material behavior of the Al-alloy. The microstructure is discretized with finite elements using LS-DYNA, which allows for beam-to-beam contact on the outer surface of the ligaments. 10³ cell domains compressed between rigid planes are shown to reproduce the measured compressive responses in both the rise and transverse directions. This includes the complete response from the initial elastic regime, through “yielding,” the extended stress plateau, to densification. More importantly, localized bands of crushed cells that develop and gradually spread throughout the domain resemble closely experimental observations made using X-ray tomography. This is a major improvement over previous models that idealized monodispersed foams as periodic Kelvin cells, and should allow modeling of polydisperse foams. The contact algorithm, friction between ligaments, and generally the discretization play crucial roles in the accuracy of the calculation as well as their numerical stability.     

Dynamic crushing of aluminum foams: Part II – Analysis

Part II of this study uses micromechanically accurate foam models to simulate and study the dynamic crushing of open-cell foams. The model starts as random soap froth generated using the Surface Evolver software to mimic the microstructure of the foams tested. The linear edges of the cellular microstructure are “dressed” with appropriate distributions of solid to match those of ligaments in the actual foams and their relative density. The ligaments are modeled as shear-deformable beams with variable cross sections discretized with beam elements in LS-DYNA, while the Al-alloy is modeled as a finitely deforming elastic–plastic material. The numerical contact algorithm of the code is used to model ligament contact and limit localized cell crushing. The quasi-static and all dynamic crushing experiments in Part I are simulated numerically. The models are shown to reproduce all aspects of the crushing behavior including the formation and evolution of nearly planar shocks, the force acting at the two ends, the shock front velocity, the strain in the crushed material behind the shock, and the energy absorbed.     

On the crushing of aluminum open-cell foams: Part II analysis

Part II of this study is concerned with the modeling of all aspects of the compressive response and crushing of the open-cell Al foam studied in Part I. The foam microstructure is modeled using the regular cell of Kelvin with cell anisotropy and ligament geometry established by X-ray tomography. The ligaments are modeled as shear-deformable beams and the material is elastoplastic calibrated to the properties of the Al alloy base material. It is demonstrated that the initiation stress of measured responses is associated with a limit load instability that results from plastification of foam ligaments due to combined bending and axial compression. The periodicity of the Kelvin cell enables calculation of the initial elastic properties as well as the initiation stress with just a single fully periodic characteristic cell. The crushing response is evaluated by considering finite size 3D domains that allow localized deformation to develop. Localization is in the form of shear buckling that develops along the principal diagonals of the Kelvin cell foam. Localized crushing is arrested by contact between the ligaments of the buckled cells. Contact is approximated by limiting the amount a cell can collapse in the direction of the applied load. This arrests local collapse and causes it to spread to neighboring material at a nearly constant stress level as in the experiments. The stress picks up when the whole domain has crushed. Although the calculated collapse patterns differed from the more random ones observed in the experiments, the calculated force–displacement responses match very well the experimental ones in all aspects.     

Modeling and prediction of bulk properties of open-cell carbon foam

The emerging ultralightweight material, carbon foam, was modeled with three-dimensional microstructures to develop a basic understanding in correlating microstructural configuration with bulk performance of open-cell foam materials. Because of the randomness and complexity of the microstructure of the carbon foam, representative cell ligaments were first characterized in detail at the microstructural level. The salient microstructural characteristics (or properties) were then correlated with the bulk properties through the present model. In order to implement the varying anisotropic nature of material properties in the foam ligaments, we made an attempt to use a finite element method to implement such variation along the ligaments as well as at a nodal point where the ligaments meet. The model was expected to provide a basis for establishing a process-property relationship and optimizing foam properties.The present model yielded a fairly reasonable prediction of the effective bulk properties of the foams. We observed that the effective elastic properties of the foams were dominated by the bending mode associated with shear deformation. The effective Young's modulus of the foam was strongly influenced by the ligament moduli, but was not influenced by the ligament Poisson's ratio. The effective Poisson's ratio of the foam was practically independent of the ligament Young's modulus, but dependent on the ligament Poisson's ratio. The effective Young's modulus of the carbon foam was dependent more on the transverse Young's modulus and the shear moduli of the foam ligaments, but less significantly on the ligament longitudinal Young's modulus. A parametric study indicated that the effective Young's modulus was significantly improved by increasing the solid modulus in the middle of the foam ligaments, but nearly invariant with that at the nodal point where the ligaments meet. Therefore, appropriate processing schemes toward improving the transverse and shear properties of the foam ligaments in the middle section of the ligaments rather than at the nodal points are highly desirable for enhancing the bulk moduli of the carbon foam.     

基于三维细观力学模型的开孔泡沫弹性性能预测

泡沫材料的宏观力学性能主要取决于基体材料的力学特性及其微细观结构特征,基于细观力学模型的分析方法是泡沫材料力学性能研究的重要途径。文中基于Matlab语言和Abaqus软件构建了描述中等孔隙率开孔弹性泡沫材料微结构特征的三维随机分布球形泡孔模型,并采用有限元方法对弹性泡沫压缩变形进行了模拟,并计算给出了不同孔隙率弹性泡沫材料弹性模量、剪切模量、体积模量以及泊松比的分布,建立了相应的唯象表达式。与理论模型及测试结果的比较表明,本文基于三维随机泡孔模型模拟结果构建的唯象表达式能够对弹性泡沫材料的弹性力学性能给出很好的预测。     

Modelling open cell-foams based on the Weaire–Phelan unit cell with a minimal surface energy approach

The complex architecture of open cell foams has most often been described by Kelvin cell models. It has been shown that the accuracy to predict the elastic properties of open cell foams increases with an increasing level of detail and resemblance to real foam microstructures. However, the Kelvin cell does not possess pentagonal faces which are the most abundant within real open cell foams. Therefore this study focuses on the use of the Weaire–Phelan unit cell to model the elastic properties of an open cell polyurethane foam. Optical and scanning electron microscopy were used to characterise the architecture of the open cell foam. Surface Evolver software was used to minimize the surface energy and introduce the typical architectural characteristics of the open cell foam to the FE-model. The E-modulus and Poisson coefficient of the Kelvin and Weaire–Phelan cell show a similar behaviour as a function of density. The Weaire–Phelan cell predicts however a higher dependency of the shear modulus on the density. When the influence of the elongation of the cells in the rise direction of the foam and the uncertainty of the solid material properties of the polyurethane is taken into account, a good accuracy of the Kelvin cell and Weaire–Phelan structure based FE-models versus experimental compression tests is found.     

On the stability of Kelvin cell foams under compressive loads

It has been previously shown that the nonlinearity exhibited in the compressive response of open cell foams is governed by cell ligament buckling. Significant insight into this behavior can be gained by idealizing such foams as periodic, space-filling Kelvin cells assigned several of the geometric characteristics of actual foams. The cells are elongated in the rise direction; the ligaments are assumed to be straight, to have Plateau border cross sections, and nonuniform cross sectional area distribution. The mechanical response of such foams can be established using models of a characteristic cell assigned appropriate periodicity conditions. The ligaments are modeled as shear deformable beams. The periodicity of this microstructure allows the use of Bloch wave theory to conduct the search for the critical state efficiently. The method tailored to the present microstructure is outlined. It is subsequently used to establish the critical states for uniaxial and a set of triaxial loadings. A rich variety of buckling modes are identified which are affected by the anisotropy and the mutliaxiality of the applied loads. Under some loadings the critical modes have long wavelengths which are shown to lead to unstable postbuckling behavior involving localization. Under other loading conditions the modes are either local to the characteristic cell or involve an assemblage of a few such cells. For the cases analyzed local modes were found to have a stable postbuckling response.     

Three-dimensional modeling of the mechanical property of linearly elastic open cell foams

Three-dimensional Voronoi models are developed to investigate the mechanical behavior of linearly elastic open cell foams. Dependence of the Young’s modulus, Poisson’s ratio and bulk modulus of the foams on the relative density is evaluated through finite element analysis. Obtained results show that in the low density regime the Young’s modulus and bulk modulus of random Voronoi foams can be well represented by those of Kelvin foams, and are sensitive to the geometric imperfections inherent in the microstructure of foams. In contrast, the compressive plateau stress of the foams is less sensitive to the imperfections. Failure surface of the foams subject to multi-axial compression is determined and is found to comply with the maximum compressive principal stress criterion, consistent with available experimental observations on polymer foams. Numerical results also show that elastic buckling of cell edges at microscopic level is the dominant mechanism responsible for the compressive failure of elastic open cell foams.     

Effects of cell shape and strut cross-sectional area variations on the elastic properties of three-dimensional open-cell foams

The Voronoi tessellation technique and the finite element (FE) method are utilized to investigate the microstructure-property relations of three-dimensional (3-D) cellular solids (foams) that have irregular cell shapes and non-uniform strut cross-sectional areas (SCSAs). Perturbations are introduced to a regular packing of seeds to generate a spatially periodic Voronoi diagram with different degrees of cell shape irregularity (amplitude a), and to the constant SCSA to generate a uniform distribution of SCSAs with different degrees of SCSA non-uniformity (amplitude b). Twenty FE models are constructed, based on the Voronoi diagrams for twenty foam samples (specimens) having the same pair of a and b, to obtain the mean values and standard deviations of the elastic properties. Spatially periodic boundary conditions are applied to each specimen. The simulation results indicate that for low-density imperfect foams, the elastic moduli increase as cell shapes become more irregular, but decrease as SCSAs get less uniform. When the relative density (R) increases, the elastic moduli of imperfect foams increase substantially, while the Poisson's ratios decrease moderately. The effect of the interaction between the two types of imperfections on foam elastic properties appears to be weak. In addition, it is found that the strut cross-sectional shape has a significant effect on the foam properties. Also, the elastic response of foams with the cell shape and SCSA imperfections appears to be isotropic regardless of changes in a, b and R and the strut cross-sectional shape.     

Kelvin结构开孔泡沫材料的弹性性能研究

采用一修正的十四面体结构模型(Kelvin结构模型)对开孔泡沫金属的弹性性能进行研究,对低密度开孔泡沫材料表现出不可压的特性进行了分析。该模型考虑作用在泡沫筋条上的弯矩、剪力和轴向力,以及轴向力的平衡。修正模型的数值计算结果与实验结果及其他模型的结果进行了对比,结果表明修正模型计算的杨氏模量比原有模型的略有提高,筋条截面为星形的修正模型计算的结果与实验比较符合。在密度等同的条件下,筋条截面惯性矩越大的开孔泡沫材料,其弹性模量也越大,而泊松比则越小。Kelvin结构的开孔泡沫材料的泊松比随相对密度的减小而趋于0.5。     

On the crushing of aluminum open-cell foams: Part I. Experiments

This two-part study is concerned with the understanding and modeling of the compressive response of open-cell metallic foams. Part I presents experimental results from Al-6101-T6 foams of three different cell sizes with relative densities of about 8%. X-ray tomography is first used to characterize the geometry of the microstructure. The cells are irregular polyhedra of nearly uniform size that are somewhat elongated in one direction. The ligaments are nearly straight with convex, three-sided cross-sections and variable area distribution along their length. Foam specimens were compressed at slow displacement rates along the rise and transverse directions and the evolution of crushing in the specimens was monitored using X-ray tomography. In both directions, the response is initially nearly linear, terminating into a limit load that is followed by an extensive load plateau. At an average strain of about 55% the load increases monotonically again due to densification. The limit load is caused by plastification due to combined compression and bending of the ligaments. Beyond this point, cells start to buckle and collapse locally, forming bands that cover the full cross-section of the specimen. Contact of the collapsing cells arrests local deformation triggering collapse in neighboring cells. In this manner, crushing gradually spreads throughout the specimen and when this is achieved the load required for further deformation starts to rise. The initial elastic modulus, the stresses at the limit load and the plateau and the extent of the plateau have been measured as a function of relative density for both directions. The stress–displacement response in the transverse direction is generally somewhat lower than in the rise direction but the prevalent events were found to be similar in the two directions.     

The high strain mechanical response of the wet Kelvin model for open-cell foams

Surface Evolver software was used to create the three-dimensional geometry of a Kelvin open-cell foam, to simulate that of polyurethane flexible foams. Finite Element Analysis (FEA) with 3D elements was used to model large compressive deformation in the [0 0 1] and [1 1 1] directions, using cyclic boundary conditions when necessary, treating the polyurethane as an elastic or elastic–plastic material. The predicted foam Young’s moduli in the [0 0 1] direction are double those of foams with uniform Plateau border cross-section edges, for the same foam density and material properties. For compression in the [1 1 1] direction, the normalized Young’s modulus increases from 0.9 to 1.1 with foam relative density, and the predicted stress–strain relationship can have a plateau, even for a linearly-elastic polymer. As the foam density increases, the predicted effects of material plasticity become larger. For foam of relative density 0.028, edge-to-edge contact is predicted to occur at a 66% strain for [1 1 1] direction compression. The foam is predicted to contract laterally when the [1 1 1] direction compressive strain exceeds 25%.     

Effects of microstructure on uniaxial strength asymmetry of open-cell foams

Based on the elongated Kelvin model, the effect of microstructure on the uniaxial strength asymmetry of open-cell foams is investigated. The results indicate that this asymmetry depends on the relative density, the solid material, the cell morphology, and the strut geometry of open-cell foams. Even though the solid material has the same tensile and compressive strength, the tensile and compressive strength of open-cell foams with asymmetrical sectional struts are still different. In addition, with the increasing degree of anisotropy, the uniaxial strength as well as the strength asymmetry increases in the rise direction but reduces in the transverse direction. Moreover, the plastic collapse ratio between two directions is verified to depend mainly on the cell morphology. The predicted results are compared with Gibson and Ashby''s theoretical results as well as the experimental data reported in the literature, which validates that the elongated Kelvin model is accurate in explaining the strength asymmetry presented in realistic open-cell foams.     

Finite element micromechanics model of impact compression of closed-cell polymer foams

Finite element analysis, of regular Kelvin foam models with all the material in uniform-thickness faces, was used to predict the compressive impact response of low-density closed-cell polyethylene and polystyrene foams. Cell air compression was analysed, treating cells as surface-based fluid cavities. For a typical 1 mm cell size and 50 s^?1 impact strain rate, the elastic buckling of cell faces, and pop-in shape inversion of some buckled square faces, caused a non-linear stress strain response before yield. Pairs of plastic hinges formed across hexagonal faces, then yield occurred when trios of faces concertinaed. The predicted compressive yield stresses were close to experimental data, for a range of foam densities. Air compression was the hardening mechanism for engineering strains <0.6, with face-to-face contact also contributing for strains >0.7. Predictions of lateral expansion and residual strains after impact were reasonable. There were no significant changes in the predicted behavior at a compressive strain rate of 500 s^?1.     

Virtual improvement of ice cream properties by computational homogenization of microstructures

Optimal shape design of microstructured materials has recently attracted a great deal of attention in materials science. The shape and the topology of the microstructure have a significant impact on the macroscopic properties. This paper presents different computational models of random microstructures, to virtually improve the physical properties of ice cream. Several sensory properties of this heterogeneous material issued from food industry are directly controlled by the elastic and thermal conducting ones. The material effective elastic and thermal conducting properties are obtained through direct large scale numerical simulations. The different formulations address the problem of finding the shape of the representative microstructural element for random heterogeneous media that increase the elastic moduli and thermal conductivity compared to existing products. The computational models are established using finite element method and images of virtual microstructures. In this paper we propose a new model of microstructures. This model is constructed with hexagonal prismatic rods and plates with volume fractions around 0.7 for the hard phase represented by hexagons of ice. A comparison between three two-phase elastic heterogeneous microstructures models is drawn. This illustrates the concept of design of microstructures using computational homogenization tools.     

聚氨酯泡沫塑料压缩杨氏模量的理论预测

通过微分法导出了泡沫塑料剪切模量和体积模量所满足的微分方程组，再利用联系泡沫塑料泊松比和孔隙比的Ｋｅｒｎｅｒ－Ｒｕｓｃｈ经验关系及泡沫塑料弹性常数间满足足的关系，在基体材料不可压缩的假设下，确定了泡沫塑料的杨氏模量。本文针对几种密度的泡沫塑料，分别对它们的杨氏模量进行了理论预测和实验测定，结果表明：理论预测的模量在较高密度下与实验符合的很好，在低密度下也给出相当好的近似值。此外，本文的结果同其他理     

A general tetrakaidecahedron model for open-celled foams

A micro-mechanics model for non-isotropic, open-celled foams is developed using an elongated tetrakaidecahedron (Kelvin model) as the repeating unit cell. Assuming the cell edges possess axial and bending rigidity, the mechanics of deformation of the elongated tetrakaidecahedron lead to a set of equations for the Young’s modulus, Poisson’s ratio and tensile strength of the foam in the principal material directions. These equations are written as a function of the cell edge lengths and cross-section properties, the inclination angle and the strength and stiffness of the solid material. This micro-mechanics model employs an elongated Kelvin model geometry which is more general than that employed by previous authors, as the size and shape of the repeating unit cell are defined by specifying three independent dimensions. As a result, the model accounts for an additional variation in the unit cell shape which is not accounted for in the previous models. The effect of this additional shape parameter on the non-isotropic stiffness and strength behavior is demonstrated and the advantages of this more general micro-mechanics model are illustrated.     

Bounds on the non-local effective elastic properties of composites

The paper deals with the effective linear elastic behaviour of random media subjected to inhomogeneous mean fields. The effective constitutive laws are known to be non-local. Therefore, the effective elastic moduli show dispersion, i.e¹ they depend on the “wave vector” k of the mean field. In this paper the well-known Hashin-Shtrikman bounds (1962) for the Lamé parameters of isotropic multi-phase mixtures are generalized to inhomogeneous mean fields k ≠ 0. The bounds involve two-point correlations of random elastic moduli. In the limit k → ∞ the bounds converge to the exact result. The interest is focussed on composites with cell structures and on binary mixtures. To illustrate the results, numerical evaluations are carried out for a binary cell material composed of nearly spherical grains of equal size.