• Aucun résultat trouvé

Effective properties of architectured materials

N/A
N/A
Protected

Academic year: 2021

Partager "Effective properties of architectured materials"

Copied!
2
0
0

Texte intégral

(1)

HAL Id: hal-01713384

https://hal.archives-ouvertes.fr/hal-01713384

Submitted on 20 Feb 2018

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Effective properties of architectured materials

Justin Dirrenberger, Samuel Forest, Dominique Jeulin, Christophe Colin, Jean-Dominique Bartout, Mathieu Faessel, François Willot

To cite this version:

Justin Dirrenberger, Samuel Forest, Dominique Jeulin, Christophe Colin, Jean-Dominique Bartout, et al.. Effective properties of architectured materials. 1st International School on Architectured Materials, May 2011, Grenoble, France. 2011. �hal-01713384�

(2)

EFFECTIVE PROPERTIES

OF ARCHITECTURED MATERIALS

J. Dirrenberger

1

, S. Forest

1

, D. Jeulin

1,2

, C. Colin

1

, J.D. Bartout

1

, M. Faessel

2

, F. Willot

2

1

Centre des Matériaux, MINES-ParisTech, CNRS UMR 7633, BP 87, 91 003 Evry Cedex, France

2

Centre de Morphologie Mathématique, MINES-ParisTech, 35, rue St-Honoré, 77 305 Fontainebleau, France

Introduction

The MANSART project, for "architectured sandwich materials", aims at developing innovative architectured materials, exhibiting improved functional properties in

comparison with traditional composites. Our contribution is focused on modeling and simulation of architectured microstructures, prediction of their e

ffective properties

using numerical homogenization implemented with finite elements (Z-set software). Rapid prototyping using selective laser melting allows us to fabricate samples and

validate the numerical prediction. Two types of morphologies have been identified: randomly distributed entangled Poisson’s fibres and auxetic periodic lattices. Effective

properties of Poisson’s fibres is investigated although estimation is challenging since statistical representative volume element (RVE) size is a priori unknown, due to

an infinite integral range as shown in

[Jeulin, 1989]

. Scaling laws for variances of properties are computed. Auxetics are architectured materials exhibiting a negative

Poisson’s ratio. Mechanical behavior of such materials is investigated for both elastic and plastic domains. 2D and 3D lattices are proposed for industrial applications.

Keywords: Homogenization, architectured materials, auxetics, Poisson’s fibres, RVE, finite elements, mathematical morphology, selective laser melting.

Determination of statistical RVE and e

ffective properties for Poisson’s fibres

– Modeling straight infinite randomly entangled interpenetrating Poisson’s fibres with intensity λ such as, λ = ln(1−VV)

π2R2 with R fibres radius and VV volume fraction of fibres. – Optimization of surface (vtkSim, Yams) and volume meshes (GHS3D) for finite element computation.

– Estimation of RVE size based on method established in [Kanit et al., 2003].

– Simulation of samples with an increasing quantity of fibres; scaling laws are computed.

– Homogenization using mixed, kinetic (KUBC) and static uniform boundary conditions (SUBC).

N = 8

N = 32

N = 71

N = 127

Homogenization and mechanical investigation of auxetic lattice structures

– Modeling of original auxetic periodic cells and examples from the literature.

– Numerical homogenization (FE) using periodic boundary conditions with imposed macroscopic stress or strain. – Analytical representation of elastic moduli and Poisson’s ratio in 3D.

– Design and manufacture of 3D quasi-isotropic auxetic lattice structures.

Auxetic lattice from the literature (hexachiral)

ν = −0.83, VV = 7% [Prall and Lakes, 1997]

Original transverse isotropic chiral structure

ν = −0.25, VV = 7%

[Dirrenberger et al., 2011]

Cubic anti-chiral structure from the literature

ν = f (φ), VV = 7% [Alderson et al., 2010]

Elastic moduli as function of φ for the anti-chiral cell

Original 3D periodic auxe-tic cubic cell (hexatruss)

Elastic moduli as function of φ for the hexatruss cell

εp

ν12

Plastic Poisson’s ratio ν12 vs. cumulative

plas-tic strain εp for a hexachiral cell for an imposed

macroscopic stress.

εp map for an imposed macroscopic strain E11 = 0.1 of an

auxetic microstructure from [Milton, 1991]

Direct fabrication using selective laser melting and experimental validation

– Direct fabrication from alloy or polymer powders – Compressive, 3 and 4-points bending tests

– Image correlation measurements

– Comparison between FEM and experiments

– Available process for production of micro-architectured materials in the industry (aerospace, automotive, biomedical, etc.)

316L steel hexachiral lattice sample made with SLM

4-points bending test of PA 12 re-entrant honeycombs

Conclusions and prospects

– Modeling and homogenization tools, in association with the set of goals for the MANSART project, resulted in the development of architectured materials likely to replace aluminium honeycombs currently serving as core for sandwich composites intensively used in the aerospace industry. The non-linear static and dynamic mechanical behavior of such materials has yet to be characterized using FEM and experiments; indentation and acoustic damping tests will be conducted in order to bring out interesting functional properties for auxetics, due to their high shear modulus. Quasi-isotropic 3D auxetic lattices shall be developed.

– The study of Poisson’s fibres highlighted a scaling law for the volume fraction variance proportional to V32. Efforts are being made to obtain scaling laws concerning elastic moduli, simulations with more fibres are needed. Statistical convergence is slower; for the same precision, Poisson’s fibres’ RVE is larger than for a classical Boolean variety. Future work on these microstructures will include: static uniform boundary conditions (SUBC) calculations, larger simulation size, mesh rationalization, domain decomposition, thermal homogenization and Fast Fourier Transform computations.

[Alderson et al., 2010] Alderson A. et al. (2010). Composites Science and Technology, vol. 70 (7), pp 1042–1048.

[Dirrenberger et al., 2011] Dirrenberger J., Forest S., Jeulin D., Colin C. (2011). ICM11, Procedia Engineering, in press.

[Jeulin, 1989] Jeulin, D. (1989). Modèles de Fonctions Aléatoires Multivariables, Ecole des Mines de Paris, Centre de Géostatistique.

[Kanit et al., 2003] Kanit T., Forest S., Galliet I., Mounoury V. and Jeulin D. (2003). International Journal of Solids and Structures, vol. 40, pp 3647–3679.

[Milton, 1992] Milton G.W. (1992). Journal of the Mechanics and Physics of Solids, vol. 40 (5), pp 1105–1137.

[Prall and Lakes, 1997] Prall D. and Lakes R. S. (1997). International Journal of Mechanical Sciences, vol. 39 n◦ 3, pp 305–314.

Contact :

justin.dirrenberger@mines-paristech.fr

CENTRE DES MATERIAUX

Références

Documents relatifs

l’élève   Non ­ verbal Organisation ­ structuration Guide ­ modèle Demande à l’élève de refaire le trait (101) Psychomotrice Se déplace sur la droite de l’élève

Table I shows ATL, the average differences between the transmission losses measured in the A0 room condition and those measured in the A l , A2, and A3 conditions using the

bitals of titanium atoms give rise to a threefold degen- erate band[10, 11]. The degeneracy is lifted by a com- bination of spin-orbit coupling and tetragonal distortion. Recently,

In that section, we investigate the particular case of our simple 2×2 games, the implications of a simple idea: replace players by populations, and probabilities in mixed strategies

Title Page Abstract Introduction Conclusions References Tables Figures J I J I Back Close Full Screen / Esc. Print Version

Plasma and RBC fatty acids relative and quan- titative levels (DHA, EPA, total n-3 PUFA and n-6 PUFA) and choline concentrations were compared between rats fed the

Several issues related to cementitious materials are then addressed, such as the high volume fractions of inclusions needed either for the initial suspension of cement, or to model

4.2 Evolution of the mean of D1 depending on the coarse graining length for the square lattice and standard deviation... Table