NASA/TM–2019–220275

An Analytical Method to Calculate Effective Elastic

Properties of General Multifunctional Honeycomb
Cores in Sandwich Composites

Erik Sather and Thiagaraja Krishnamurthy

Langley Research Center, Hampton, Virginia

April 2019
 NASA STI Program. . . in Profile

Since its founding, NASA has been dedicated • CONFERENCE PUBLICATION.

to the advancement of aeronautics and space Collected papers from scientific and
science. The NASA scientific and technical technical conferences, symposia, seminars,
information (STI) program plays a key part or other meetings sponsored or
in helping NASA maintain this important co-sponsored by NASA.
role.
• SPECIAL PUBLICATION. Scientific,
The NASA STI Program operates under the technical, or historical information from
auspices of the Agency Chief Information NASA programs, projects, and missions,
Officer. It collects, organizes, provides for often concerned with subjects having
archiving, and disseminates NASA’s STI. substantial public interest.
The NASA STI Program provides access to
the NASA Aeronautics and Space Database • TECHNICAL TRANSLATION. English-
and its public interface, the NASA Technical language translations of foreign scientific
Report Server, thus providing one of the and technical material pertinent to
largest collection of aeronautical and space NASA’s mission.
science STI in the world. Results are
published in both non-NASA channels and Specialized services also include organizing
by NASA in the NASA STI Report Series, and publishing research results, distributing
which includes the following report types: specialized research announcements and
feeds, providing information desk and
• TECHNICAL PUBLICATION. Reports of personal search support, and enabling data
completed research or a major significant exchange services.
phase of research that present the results
of NASA programs and include extensive For more information about the NASA STI
data or theoretical analysis. Includes Program, see the following:
compilations of significant scientific and
technical data and information deemed to • Access the NASA STI program home page
be of continuing reference value. NASA at http://www.sti.nasa.gov
counterpart of peer-reviewed formal
professional papers, but having less • E-mail your question to
stringent limitations on manuscript length [email protected]
and extent of graphic presentations.
• Phone the NASA STI Information Desk at
• TECHNICAL MEMORANDUM. 757-864-9658
Scientific and technical findings that are
• Write to:
preliminary or of specialized interest, e.g.,
NASA STI Information Desk
quick release reports, working papers, and
Mail Stop 148
bibliographies that contain minimal
NASA Langley Research Center
annotation. Does not contain extensive
Hampton, VA 23681-2199
analysis.

• CONTRACTOR REPORT. Scientific and

technical findings by NASA-sponsored
contractors and grantees.
NASA/TM–2019–220275

An Analytical Method to Calculate Effective Elastic

Properties of General Multifunctional Honeycomb
Cores in Sandwich Composites
Erik Saether and Thiagarajan Krishnamurthy
Langley Research Center, Hampton, Virginia

National Aeronautics and

Space Administration
Langley Research Center
Hampton, Virginia 23681-2199

April 2019
The use of trademarks or names of manufacturers in this report is for accurate reporting and does not
constitute an offical endorsement, either expressed or implied, of such products or manufacturers by the
National Aeronautics and Space Administration.

Available from:

NASA STI Program / Mail Stop 148

NASA Langley Research Center
Hampton, VA 23681-2199
Fax: 757-864-6500
 Table of Contents

Abstract .........................................................................................................................1
1. Introduction ...................................................................................................................1
2. The unit cell approach for a honeycomb core structure ................................................5
3. Analytical expressions for effective material properties of honeycomb
structures .......................................................................................................................6
4. FEM analysis of a unit cell ............................................................................................8
4.1 In-plane analysis of a unit cell using a 1-D beam model ........................................9
4.2 In-plane analysis of a unit cell honeycomb structure with 2-D shell elements ....10
4.3 In-plane analysis of a unit cell honeycomb structure with 3-D solid elements ....11
4.4 General 3-D analysis of a unit cell for a complete set of elastic moduli ...............12
5. Equivalent isotropic single layer material properties to represent a
multi-layered honeycomb wall ...................................................................................14
6. Comparison of equivalent single layer isotropic properties for layered
honeycomb structures using a finite element analysis of a unit cell ............................16
7. Reference solution for sandwich honeycomb structures with layered core-walls ......21
8. Homogenized finite element models used to assess effective core properties ...........22
9. Comparison of maximum center deflection with reference solutions ........................24
9.1 Deflection comparison of a three-layered isotropic [Nomex/Nomex/Nomex]
Laminate-1 core in a Sandwich plate with reference and homogenized
solutions ................................................................................................................24
9.2 Deflection comparison of a symmetric three-layer isotropic [Al/Cu/Al]
Laminate-2 core in a sandwich plate with reference and homogenized
solutions ................................................................................................................25
9.3 Deflection comparison of an unsymmetrical three-layer isotropic
[Al/Epoxy/Cu] Laminate-3 core in a sandwich plate with reference
and homogenized solutions ...................................................................................25
10. Summary .....................................................................................................................27
11. References ....................................................................................................................28
 Abstract

Sandwich composite structures are ideal configurations in which to incorporate additional

functionality beyond load carrying capabilities. The inner core-walls can be layered to incorporate
other functions such as power storage for a battery. In this work we investigate an assemblage of
analytical tools to compute effective properties that allow complex layered core architectures to
be homogenized into a single continuum layer. This provides a great increase in computational
efficiency to numerically simulate the structural response of multifunctional sandwich structures
under applied loads. We present a coupled analytical method including an extensive numerical
verification of the accuracy of this method.

1. Introduction
Multifunctional structures seek to maximize operational efficiency by using materials that can
perform several functions simultaneously. Sandwich composites are ideally suited for
incorporating additional functions beyond load carrying capabilities. These composite structures
are composed of two faceplates separated by a core material to increase bending stiffness and
provide a light, stiff structure. A representative honeycomb core geometry is shown in Figure 1
and a depiction of a sandwich structure showing the attached faceplates in Figure 2.

Figure 1. A representative honeycomb core configuration.

Various multifunctional applications for composites have been discussed in the literature, such as
piezoelectric actuators, self-healing, sensing, and battery functions [1-4]. In this study, we will
focus on potential core functionality as a battery. Structural battery materials are those that can
carry mechanical loads while also storing electrical energy. Because these core geometries possess
an open architecture of repeating cells, the cell walls can be layered with a suitable choice of
materials to function as electrodes and electrolytes as required in a battery. The general battery
configuration and the scope of the analysis performed in the present investigation was developed

1
in the NASA project entitled Multifunctional Structures for High Energy Lightweight Load-
bearing Storage (M-Shells) [5].

Face Plate

Honeycomb
core

Face Plate

Figure 2. A schematic of a typical sandwich structure.

A core is typically configured as an array of cells which can assume many different geometrical
architectures such as circular or polygonal cross sections [6]. Here we will focus on hexagonal
honeycomb configurations that are sized to maintain a large bending stiffness in the sandwich
panel. These cores are additionally assumed to include battery functionality. Figure 3 shows the
idealized battery as a three-layer configuration with electrodes and electrolyte incorporated into
the walls of the honeycomb.

Figure 3. Walls of a honeycomb serving as a battery with materials that

can function as electrodes and electrolyte.

2
Composite structures are typically analyzed using the finite element method (FEM). Composite
sandwich structures, however, present a modeling challenge because they consist of a core region
composed of a large number of cells that also have a complex geometry. While certain analyses -
such as simulating sequential failure propagation within the core - would require a detailed finite
element modeling of a small representative region of the core to determine the material response
in damage propagation, other simulations of sandwich composite behavior involving a complete
panel would require modeling a large number of core cells making an explicit simulation
computationally expensive. Therefore, some homogenization method is needed to replace the core
with a simplified solid layer with effective (equivalent) material properties. Many research efforts
have been published that present various analytical methods for determining effective mechanical
properties of homogenized core geometries that can avoid explicit modeling of the honeycomb
geometry [7-11]. In general, the effective material properties for honeycomb structures are
obtained from analyzing a unit cell representing a repeating element in a honeycomb structure.
The most prevalent unit cell model for effective property determination is the 1-D isotropic beam
analysis approach of Gibson et al. [7]. Gibson assumes that the linear-elastic response of the
honeycomb deformations and resulting core properties depend only on bending of the core cell
walls. Other deformation modes have been investigated that include stretching and shear
deformation of the cell walls [8]. Many other approaches have been published [6] to estimate the
effective material properties of honeycomb structures using the finite element method. For
multifunctional honeycomb sandwich structures with built in battery functionality, it is necessary
to compute equivalent elastic properties of core-walls composed of an arbitrary number of layers
possessing different material properties. In particular, the electrolyte layers can have moduli that
are orders of magnitude less than the materials used for the electrodes. There is a need to develop
analytical expressions to predict the homogenized effective material properties for multifunctional
honeycomb sandwich cores. These analytical expressions can then be used for rapid prototyping
of multifunctional sandwich composites during the design phase to size the core for the intended
service loads and for projected energy requirements.

Figure 4 illustrates the homogenization approach undertaken in this work. This approach is
performed in two steps: (1) determine a single effective modulus for the multi-layered wall used
in the core; and (2), use Gibson’s approach to obtain equivalent elastic properties of the full core
configurations such that they can be used in a uniform solid material representation. The
homogenized core is then utilized to simplify finite element modeling of a sandwich core. This
procedure is depicted in Figure 5.

Thus, the two objectives sought in this investigation are, first, to enhance the Gibson analysis by
extending it to multi-layered isotropic walls, and second, to validate the methodology by
comparing the analytical results with reference FEM simulations in which the multi-layered
honeycomb geometry is explicitly modeled. All analyses are assumed to involve small
deformations and exhibit an exclusively linear elastic response.

3
 Honeycomb core architecture Equivalent homogenized core

Multifunctional layered core-wall

L a y e r 1 E 11 ,  112

Step 1: Determine Step 2: Determine

L a y e r 2 E 12 ,  122
effective wall modulus effective core modulus
L a y e r 3 E 13 ,  132 E e ff  f E 1
1 , E 12 , E 13  E c o r e  g  E e ff 
Figure 4. Overview of homogenization methodology sought in the current investigation.

Complex core geometry Equivalent shell model

Honeycomb detail Detail of wall Homogenized core layer

z
w
x

Figure 5. Simplified modeling of complex core geometry.

4
This report is organized into several sections: Section 2 defines unit cell models for the honeycomb
structure used in the present investigation. Next, in Section 3, Gibson’s analytical solutions for
the effective in-plane elastic core properties are presented. In Section 4, the validity of the
analytical approach is examined by using 1-D, 2-D, and 3-D finite element simulations to predict
the elastic properties. The calculation of an equivalent modulus for multi-layered core-walls is
contained in Section 5. This equivalent modulus will be used in the Gibson equations to simulate
the elastic behavior of core-walls that are representative of multifunctional battery construction.
In Section 6, various layer configurations are identified and the enhanced Gibson equations are
used to compute effective moduli of the core. These moduli are validated by comparing them to
elastic properties computed through unit cell finite element simulations. This is followed, in
Section 7, by the development of reference solutions for several sandwich composite
configurations with different layered cores. Homogenized 2-D shell and 3-D solid element models
are developed in Section 8 to illustrate the use and accuracy of homogenizing sandwich cores that
allow simplified finite element models to be used for rapid simulations. These models incorporate
the derived effective core properties and, in Section 9, their maximum center deflection is
compared to the reference solutions to verify the overall procedure for modeling multifunctional
sandwich composite structures. Finally, a discussion is presented in Section 10 that focuses on the
major issues examined in this report.

2. The unit cell approach for a honeycomb core structure

The effective homogenized material properties for a honeycomb structure are derived from
analyzing the unit cell (repeating element) of the honeycomb. The geometry of the unit cell
depends upon the method used to fabricate the honeycomb core. In this report, the honeycomb
structure is assumed to be constructed from corrugated sheet-based technology [6]. The corrugated
sheets are joined together by an adhesive to form the honeycomb geometry. The thin layer of
adhesive material bonding the cells walls is considered to add negligible stiffness. In addition, the
adhesive used in the manufacturing process can produce fillets where the honeycomb walls join,
but the influence of a fillet on stiffness of the unit cell representation is assumed to have only a
second order contribution and is neglected. However, with other geometries, the effect of fillet
radius can be important in the failure modes exhibited by core cell structures under applied loads
[6] and will be addressed in future work. In addition, at the interface at which the core is bonded
to the faceplates, various fillets or adhesive layers can be formed but are assumed to be of
secondary influence compared to the large stiffness of the faceplates [11].

The unit cell selected for this study is shown in Figures 6 and 7. The cell size ( d ) , angle ( ) ,
thickness (t ) and width (w) completely define any honeycomb geometry of the unit cell. The length
of the sides of the hexagonal honeycomb core ( L ) can be determined from d and as shown in
Figure 6. For a regular hexagonal honeycomb structure, the angle  equals 30 degrees.

The definition of the width, w, of the honeycomb geometry is shown in Figure 7. The width of the
unit cells represents the separation of the faceplates.

5
 unit cell
t L, t
L, t L, t
L / 2,2t d
L / 2, 2t

y L, t  L, t
x L, t
  30 degrees
L d ‐Cell size
L ‐Length of the side d
L

t ‐Thickness of the wall 2cos( ) 
Figure 6. Unit cell derived as a repeating substructure of the core.

Figure 7. The unit cell width dimension

In this report, all finite element models have a honeycomb cell size of d = 4.8 mm.

3. Analytical expressions for effective material properties of honeycomb

structures

The most widely used analytical expressions for the effective material properties of a honeycomb
core were derived by Gibson [7]. These analytical expressions were derived assuming the walls of
the honeycomb unit cell deform solely due to bending of the inclined walls. The effective in-plane
moduli are given by

6
 3
t
Ec   Cos  
Exx  L
(1)
1  Sin    Sin 2  
3
t 
Ec   1  Sin   
E yy    3
L
(2)
Cos  
Cos 2  
 xy  (3)
1  Sin    Sin  
 t  1  Sin  
3

Gxy  Ec   (4)
 L  3Cos  

where Ec is the elastic modulus of the wall. The out-of-plane elastic properties derived using
Gibson’s assumptions are presented in Reference [12] and are represented by

Ezz  E (5)
c c
where  c is the relative density of the equivalent core and

Exx
vxz   c (6)
Ezz
Eyy
vyz   c (7)
Ezz
 t  Cos  
G xz  G yz  Gc   (8)
 L  1  Sin   

where c and Gc are the Poisson’s ratio and shear modulus of the wall. The Gibson’s Equations
(1) - (8) are valid only for a uniformly isotropic wall material.

Constraint effects on the effective in situ core properties are an important consideration. At the
boundary where the core cells are bonded to the face plates, bending in the honeycomb cells is
constrained and the core deformation is limited to the stretching of the face plates. Effective in-
plane moduli for core deformation solely due to cell wall stretching modes, as shown in Figure 8,
have been derived by Masters and Evans [8] and are given by Equations 9 – 12 for the in-plane
properties. This effect would be expected to diminish within a region extending out from this
interface. In this region the effective core moduli should transition from stretching-dominated
elastic properties to bending-dominated elastic properties. Assuming a wall modulus, Ec, of unity,
Table 1 shows a comparison of calculated effective moduli considering separately bending or
stretching deformation modes for t/L ratios between 0.01 and 0.1. The lower magnitude modulus
in bending indicates a naturally preferred deformation mode from an energy standpoint.
Constraining the honeycomb cell to only deform through stretching can have a large effect on the

7
local core properties, and various aspects of this effect have been investigated in References [18]
and [19]. For the remainder of this report, all deformations of the core cells are assumed to be
unconstrained and consist only of bending in the core walls; however, a study of the effect of
deformation mode transitioning on in situ core properties is planned for future work.

y

y
x x

x
y
(a) (b)
y–directional loading x–directional loading
Figure 8. Stretching modes in a honeycomb core unit cell.

 tE 
E xx   c 
1  Sin    (9)

 L  Cos   2  Sin 2   
 tE  1
E yy   c  (10)
 L  Cos   1  Sin   

Et Cos     Sin   

Gxy   c  (11)

 L  Cos    1  Sin    Sin   
2

Sin   1  Sin   
 xy   (12)
2  Sin 2  

Table 1. Comparison of effective moduli due to flexural (Gibson) and stretching (Masters)
deformations.
t/L Flexural Exx,Eyy Stretching Exx,Eyy
0.010 2.31E-6 7.70E-3
0.033 7.94E-5 2.50E-2
0.055 3.84E-4 4.23E-2
0.078 1.07E-3 5.97E-2
0.100 2.31E-3 7.70E-2

8
4. FEM analysis of a unit cell

Numerical simulations of the hexagonal unit cell for the honeycomb core were performed using
FEM with appropriate periodic boundary conditions. The FEM solution simulates all pertinent
deformation modes including bending, stretching and shear, and is a reference solution for the
analytical Gibson solution which assumes only bending deformations. The Gibson formula for the
in-plane moduli is independent of the core width. To evaluate the assumptions of the Gibson
analysis, three different models were analyzed to account for 1-D, 2-D and 3-D stress states in the
walls of the unit cell. These models are:

(a) 1-D beam model

(b) 2-D shell model
(c) 3-D solid model

The results from these studies are compared with the analytical results of Gibson et al. [7] which
are based solely on the 1-D elasticity of beam bending. Linear elastic isotropic properties were
used with t = 0.052 mm, Ec = 3150 MPa, Gc = 1125 MPa and cas used in Reference [18].
With these values, Gibson’s formula predicts the in-plane moduli of the core to be: Exx = Eyy =
0.04806 MPa.

4.1 In-plane analysis of a unit cell using a 1-D beam model
The unit cell beam model shown in Figure 9 was analyzed using ABAQUS [16] with B21 linear
beam elements. The width was fixed at 1.0 mm and the effective in-plane moduli Exx and Eyy were
calculated. In this model, the cell width effects the moment of inertia of the cross section but cannot
alter the 1-D stress state. The results from FEM unit cell analysis are compared with analytical
results from Gibson in Table 2. The beam FEM model predicts effective properties within one
percent of Gibson’s analytical model and shows that any axial deformation is negligible.

Figure 9. Unit cell beam FEM model

9
Table 2. Effective core moduli for a beam model: Comparison of Gibson (Exx = Eyy = 0.04806
MPa) and numerical FEM results.

Exx (MPa) E yy (MPa)

% difference of Gibson % difference of Gibson
FEM FEM
solution from FEM solution from FEM
0.0476 0.96 0.0477 0.96

4.2 In-plane analysis of a unit cell honeycomb structure with 2-D shell elements
The FEM for the unit-cell shell model is shown in Figure 10. The shell model was analyzed for
different honeycomb widths ranging from 0.125 mm to 10 mm. Simulations were performed using
ABAQUS with 4-node S4R reduced integration shell elements. The results from FEM analyses
are compared with the analytical solution of Gibson [7] in Table 3. For widths greater than 5 mm,
the maximum difference between the FEM and the Gibson’s analytical solution is around 15
percent, indicating a departure of a 1-D beam solution from a 2-D stress state obtained using shell
elements. With increasing width, this maximum difference is approached asymptotically and,
therefore, is bounded.

Figure 10. Unit cell shell FEM model

10
Table 3. Effective core moduli for a shell model: Comparison of Gibson (Exx = Eyy = 0.04806 MPa)
and numerical FEM results.

Exx (MPa) Eyy (MPa)

Width % difference of % difference of
Mm FEM Gibson solution FEM Gibson solution
from FEM from FEM
0.125 0.0477 0.6 0.0477 0.6
0.25 0.0478 0.6 0.0478 0.6
0.5 0.0491 2.2 0.0491 2.2
1 0.0513 6.7 0.0512 6.6
5 0.0546 14.0 0.0546 14.0
10 0.0550 15.0 0.0550 15.0

4.3 In-plane analysis of a unit cell honeycomb structure with 3-D solid
elements

The unit cell model was analyzed using ABAQUS 8-node solid continuum elements and is shown
in Figure 11. The ABAQUS C3D8I element incorporates incompatible modes and full integration.
Solid elements can represent a complete 3-D state of stress which was not expected in these
simulations due to the small thickness-to-length ratio of the honeycomb walls. Here a similar
parametric study as made in Section 4.2 was performed with the widths of the unit cell model
varied from 0.125 mm to 10 mm.

Number of 3D brick elements = 6720

Number of Nodes = 10004

Figure 11. Unit cell 3-D solid FEM model.

11
The 3-D elastic solid elements prediction of effective moduli are compared with Gibson’s
analytical predictions in Table 4. For widths greater than 5 mm, the difference between the FEM
and the Gibson’s analytical solutions asymptotically approach approximately18 percent.

Table 4. Comparison of Gibson (Exx = Eyy = 0.04806 MPa) and numerical results for effective
core moduli using a solid FEM model.

Exx (MPa) E yy (MPa)

Width (mm) % difference of % difference of

FEM Gibson solution FEM Gibson solution
from FEM from FEM
0.125 0.0484 1.0 0.0483 1.0
0.25 0.0498 3.8 0.0496 3.3
0.5 0.0526 9.6 0.0519 8.0
1 0.0547 14.0 0.0540 13.0
5 0.0566 18.0 0.0558 16.0
10 0.0568 18.0 0.0561 17.0

From the analyses using both 2-D shell and 3-D solid models, it is clear that the unit-cell FEM
model predicts the effective modulus of the material to agree with Gibson’s analysis for small
widths but asymptotically increases to an 18 percent difference for larger widths. A comparison of
results from the shell and solid element models indicates that 2-D shell elements are sufficient to
represent the stress state of honeycomb cores. This will be further corroborated in Section 8 where
cores with different layer properties are investigated.

4.4 General 3-D analysis of a unit cell for a complete set of elastic moduli

A final set of computations was performed to obtain the six in-plane and out-of-plane moduli of a
unit cell model with various cell widths. These equivalent moduli are presented in Table 5. When
the calculated Gibson values for these moduli are normalized by the FEM predictions, the
normalized moduli for Exx and Eyy, and for Ezz and Gyz follow the same curve and have been
perturbed slightly to be visible in Figure 12.

12
 Table 5. Complete set of equivalent elastic moduli of a honeycomb core as a function of
cell width, w.
Elastic Numerical FEM, w = Analytical
Component 0.125 mm 0.25 mm 2.00 mm 14.0 mm 24.0 mm Gibson
Exx (MPa) 0.04899 0.04831 0.0532 0.0552 0.0553 0.0481
Eyy (MPa) 0.04899 0.04831 0.0532 0.0552 0.0553 0.0481
Ezz (MPa) 91.0000 90.9727 91.013 91.003 90.003 91.000
Gxy (MPa) 0.01493 0.01576 0.0175 0.0173 0.0173 0.0120
Gxz (MPa) 10.564 10.565 10.564 10.564 10.563 12.188
Gyz (MPa) 12.205 12.205 12.205 12.199 12.197 12.188

Figure 12. The six in-plane and out-of-plane elastic moduli for a unit hexagonal core cell.

In Figure 12, the in-plane moduli, Exx, Eyy and Gxy, show a dependence on cell width when the
width becomes lower than 15.0. This is presumed to be due to the stress state transitioning from a
1-D beam bending stress state to a 2-D stress state. This is corroborated by the results in Tables 3
and 4 that show an asymptotic convergence to in-plane moduli after a core width of approximately
5 mm. The other out-of-plane elastic moduli appear to be relatively independent of the cell width
and are effectively constant for the dimensions investigated.

13
5. Equivalent isotropic single layer material properties to represent a
multi-layered honeycomb wall

In order to rapidly assess design variations and sizing requirements of configurations with multi-
layered cell walls, the multi-layer wall must be replaced with a single equivalent isotropic layer.
To store energy as a battery, the honeycomb walls consist of three distinct materials to function as
an anode, a cathode, and an electrolyte. The modulus of the equivalent wall layer is then used
within the analytical expressions derived by Gibson et al. [7] to allow multi-layered cores to be
modeled as a simple homogeneous material. In this section, classical lamination theory (CLT) is
used to compute equivalent single-layer in-plane elastic properties of the laminated walls.

For a multi-layered wall, effective in-plane properties can be computed via CLT using the laminate
A, B and D matrices [13,14].

 A11 A12 A16 B11 B12 B16 

A A22 A26 B21 B22 B26 
 21
  A     A61
B  A26 A66 B61 B62 B66 
 T   (13)
 B   D   B11 B21 B61 D11 D12 D16 
 B12 B22 B62 D21 D22 D26 
 
 B16 B26 B66 D61 D62 D66 
where
Aij    Qij  zk  z k 1
 
n
i, j  1, 2, 6 (14)
k 1 k

Bij    Qij  zk2  zk21  

1 n
i, j  1, 2, 6 (15)
2 k 1 k

Dij 
1 n

3 k 1
 Qij  zk3  zk31
  i, j  1, 2,6 (16)
k

Where Qij are the reduced stiffnesses of the kth layer, zk is the location of the layer interface through
the thickness of the core-wall, and n is the number of layers in the laminate. The constitutive
matrix relating strains and curvatures with force and moment resultants for the laminate is given
by

 o   a b N 
  T   (17)
   b d   M 
where
1
a b  A B
bT 
d   BT D 
(18)


The equivalent in-plane material properties for a layered wall obtained by CLT are given by

14
 1
Exx 
a11t
1
E yy 
a22t
(19)
a
 xy   12
a11
1
Gxy 
a66t

Chen and Chan [15] derive a set of modified in-plane moduli that account for induced shear and
bending deformation in the equivalent single layer. These moduli are defined by elements of a P
matrix defined as:

 P   a   b d  b
1 T
(20)
such that

1
Exx 
 P162 
 11
P  t
 P66 
1
E yy 
 P262 
 P22  t
 P66 
P13 P26 (21)
P12 
P66
 xy  
P2
P11  16
P66
1
Gxy 
 
1

 P66   P16 P22  2 P26 P16 P26 P11
2 2
 t
 1 
where
1  P11 P22  P122 (22)

Averaged properties defined for an equivalent single layer using CLT in Equation (19) or by the
Chen-Chan Equations (21) are then used in the Gibson Equations (1) through (8) to obtain the
effective homogenized material properties of a honeycomb core possessing a layered wall. It
should be noted that if the laminate is balanced such that the coupling B matrix is null, the Chen-
Chan Equations and the CLT Equations yield identical results. The homogenized effective core
properties obtained using Gibson’s equations together with the averaged single layer properties
are validated using a unit cell finite element analysis in the following section.

15
6. Comparison of equivalent single layer isotropic properties for layered
honeycomb structures using a finite element analysis of a unit cell

The equivalent single layer properties defined using CLT Equations (19) or the modified Chen-
Chan Equations (21) have to be verified before using Gibson’s equations (1) to (8) to approximate
the entire core. In this section, the equivalent single layer properties are validated using finite
element analysis of a honeycomb unit cell. For finite element simulation, the single layer walls of
the unit cell defined in Figures (6) and (7) are replaced by a three-layered laminate, as shown in
Figure 13.

Figure 13. Honeycomb unit cell geometry with a three-layered laminated wall

Three different laminates were considered for the three-layered honeycomb wall. The lay-ups were
selected to have an increasing difference in elastic moduli and to transition from a symmetric to
an unbalanced laminate.

Laminate-1 consists of three isotropic layers with properties characteristic of lightweight Nomex
core material. These properties are listed in Table 6.

Table 6. Laminate-1 multi-layer core-wall for [Nomex/Nomex/Nomex].

Layer Thickness Modulus Poisson’s Ratio

Nomex Paper 0.01733 mm 3150 MPa 0.40
Nomex Paper 0.01733 mm 3150 MPa 0.40
Nomex Paper 0.01733 mm 3150 MPa 0.40

Laminate-2 consists of aluminum and copper layers. The properties for Laminate-2 are given in
Table 7.

16
 Table 7. Laminate-2 multi-layer core-wall for [Al/Cu/Al].

Layer Thickness Modulus Poisson’s Ratio

Aluminum 0.01733 mm 70,300 MPa 0.330
Copper 0.01733 mm 110,000 MPa 0.343
Aluminum 0.01733 mm 70,300 MPa 0.330

Laminate-3 consists of Aluminum, Epoxy and Copper layers. The properties for Laminate-3 are
given in Table 8. Laminate-3 is selected to represent realistic battery wall properties. The outer
layers represent cathode and anode materials while the middle epoxy layer represents the separator
to prevent contact between the electrodes.

Table 8. Laminate-3 multi-layer battery wall for [Al/Epoxy/Cu].

Layer Thickness Modulus Poisson’s Ratio

Aluminum 0.05 mm 70,300 MPa 0.330
Epoxy 0.20 mm 4237 MPa 0.450
Copper 0.05 mm 110,000 MPa 0.343

Laminate-1 and Laminate-2 possess a balanced assemblage of layers and the B matrix is identically
zero. As stated above, this results in the CLT and Chen-Chan in-plane effective moduli being
equal. However, for Laminate-3, the layer properties are not balanced and the computed wall
modulus for the CLT and Chen-Chan in-plane moduli are different. The computed equivalent
single layer isotropic wall moduli for the three laminates considered are given in Tables 9 to 11.

Table 9. Laminate-1: Equivalent single layer isotropic wall modulus

Effective Core In-plane Moduli using In-plane Moduli using

Wall Properties CLT relations Chen-Chan relations
Exx, Eyy 3150.0 MPa 3150.0 MPa
Gxy 1125.0 MPa 1125.0 MPa
xy 0.4 0.4

Table 10. Laminate-2: Equivalent single layer isotropic wall modulus

Effective Core In-plane Moduli using In-plane Moduli using
Wall Properties CLT relations Chen-Chan relations
Exx, Eyy 83537.2 MPa 83537.2 MPa
Gxy 31270.0 MPa 31270.0 MPa
xy 0.336 0.336

17
 Table 11. Laminate-3: Equivalent single layer isotropic wall modulus

Effective Core In-plane Moduli using In-plane Moduli using

Wall Properties CLT relations Chen-Chan relations
Exx, Eyy 31,507.5 MPa 32,916.3 MPa
Gxy 11,698.6 MPa 12,204.3 MPa
xy 0.347 0.349

The homogenized effective elastic properties for the unit cell are obtained using FEM analysis
modeling the walls of the unit cell as a three-layered composite. The width of the modeled cell
was 14.5 mm. The unit cell finite element model was created using 8064 4-node shell elements
(S4R in ABAQUS) and 8193 nodes. For each of the 9 elastic constants a different set of loads and
support conditions are applied to an FEM model and the response simulated. For example, the
displacement boundary conditions applied to the unit cell to compute the effective in-plane Exx
and E yy moduli are shown in Figure 14. In this approach, the strains are determined from the
applied displacements divided by the model length in the x or y direction, and the stresses are
obtained from the recovered reaction forces due to the imposed loading divided by the area over
which the reaction forces are obtained. The 1-D moduli then become a simple calculation of the
stress divided by the strain along these axes. A succinct enumeration of the boundary conditions
used for each set of boundary conditions is contained in Reference 12 and are not elaborated here.

Figure 14. Applied displacements boundary conditions to impose unit cell strains

18
The estimated homogenized effective material properties for the unit cell from FEM simulations
are compared with the calculated effective core elastic properties from the Gibson’s equations
(using conventional CLT and the Chen-Chan modified properties as presented in Section 5).
Because the Chen-Chan equations are identical to the CLT equations for balanced laminates, only
one set of results is shown for Laminate-1 and Laminate-2, whereas for Laminate-3, which is
nonsymmetric, two separate tables are shown for these two sets of effective properties. These
results are presented in Tables 12 - 15. The percent difference provides a measure of how well the
effective wall properties used in Gibson’s equations are representing the elastic response of the
core-walls.

Table 12. Laminate-1: Effective honeycomb core elastic properties obtained from FEM and the
Gibson equations modified using the Chen-Chan relations.

Property FEM Gibson % Difference

Exx MPa 0.05518 0.04806 12.90
Eyy MPa 0.05516 0.04806 12.87
Ezz MPa 91.00 91.00 0.0031
xy 0.9985 1.00 –0.15
xz 2.425E-4 2.112E-4 12.91
yz 2.424E-4 2.112E-4 12.90
Gxy MPa 0.01680 0.01201 28.51
Gxz MPa 10.43 12.19 -16.87
Gyz MPa 12.20 12.19 0.082

Table 13. Laminate-2: Effective honeycomb core elastic properties obtained from FEM and the
Gibson equations modified using the Chen-Chan relations.

Property FEM Gibson % Difference

Exx MPa 1.177 1.275 -8.32
Eyy MPa 1.194 1.275 -6.78
Ezz MPa 2413.0 2413.0 -0.0002
xy 0.9994 1.00 -0.06
xz 1.614E-4 1.748E-4 -8.30
yz 1.637E-4 1.748E-4 -6.78
Gxy MPa 0.3928 0.3186 18.89
Gxz MPa 356.3 338.8 4.91
Gyz MPa 339.1 338.8E 0.09

19
Table 14. Laminate-3: Effective honeycomb core elastic properties obtained from FEM and the
Gibson equations modified using the Chen-Chan relations.

Property FEM Gibson % Difference

Exx MPa 97.92 96.43 1.52
Eyy MPa 97.55 96.43 1.14
Ezz MPa 5466.0 5486.0 -0.35
xy 0.9630 1.00 -3.84
xz 6.253E-3 6.135E-3 1.88
yz 6.229E-3 6.135E-3 1.49
Gxy MPa 26.35 24.11 8.44
Gxz MPa 816.7 762.8 6.60
Gyz MPa 757.8 762.8 -0.66

Table 15. Laminate-3: Effective honeycomb core elastic properties obtained from FEM and the
Gibson equations modified using CLT approximations.

Property FEM Gibson % Difference

Exx MPa 97.92 92.31 5.73
Eyy MPa 97.55 92.31 5.37
Ezz MPa 5466.0 5251.0 3.93
xy 0.9630 1.00 -3.80
xz 6.253E-3 6.100E-3 2.44
yz 6.229E-3 6.100E-3 2.07
Gxy MPa 26.35 23.08 12.41
Gxz MPa 816.7 731.2 10.47
Gyz MPa 757.8 731.2 3.51

The Gibson equations that are modified by the Chen-Chan approximations to obtain effective core
properties presented in Tables 12 through 14, demonstrated a reasonably close agreement with the
same quantities obtained through finite element simulations. It should be noted that, in Sections
4.2 and 4.3, it was shown that there is a departure of the Gibson results from the FEM predictions
with increasing width of the cell due to the 1-D beam solutions not fully accounting for the 2-D
stress state in deflections, and possibly due to deformation modes other than bending that are not
included in Gibson’s analysis. These modeling effects will influence the deviation of the elastic
properties using Gibson’s equations from the FEM predictions. In the three laminates examined,
the least deviation is seen in the in-plane normal moduli which varied between 1.1% and 12.9%.
The Poisson ratios showed a similar range of deviations for the different laminates, a trend that
was also observed by Sorohan et al. [2]. In general, the shear moduli showed a larger deviation,
which was also noted by Penado [17] in a study of effective moduli determination in which it was
observed that a discrepancy in the calculation of the in-plane shear modulus, Gxy, could be
explained by its relatively small magnitude compared to the out-of-plane shear moduli, which were
1-3 orders of magnitude greater and less susceptible to numerical precision issues.

20
7. Reference solution for sandwich honeycomb structures with layered core-
walls
A high-fidelity FEM of a sandwich structure consisting of a 10 x 10 unit cell honeycomb core and
two faceplates was developed as shown in Figure 15. The model contained a complete
representation of the core architecture with multi-layered cell walls. The deflection response of
this model under uniform pressure on the upper face plate provided reference solutions to compare
with simulations using homogenized effective properties of the core. These homogenized cores
simplify the finite element modeling which, in turn, minimize the computational cost and the
overall simulation times.

Figure 15. Sandwich composite configuration used to obtain the reference

Three reference solutions were generated with the three laminates described in Sections 5 and 6.
The materials used for faceplates and the core for the three reference solutions are given in Table
16. Each model consists of 661,600 4-node S4R elements and 644,462 nodes.

Table 16. Materials and thickness of faceplates and cores used in the reference solutions.

Reference Solution Lower Faceplate Honeycomb Core Upper Faceplate

1 Nomex, w=0.25mm Laminate-1, w=14.0mm Nomex, w=0.25mm
2 Steel, w=0.25mm Laminate-2, w=14.0mm Steel, w=0.25mm
3 Aluminum, w=0.25 mm Laminate-3, w=14.0mm Aluminum, w=0.25mm

The loading and boundary conditions are shown schematically in Figure 16. The nodes on the
vertical faces that extend up from the outer edges of the model were assigned clamped boundary
conditions, and a uniform normal pressure of 1.0 MPa was applied over the upper face.

21
 Figure 16. Schematic of applied surface loads and edge boundary conditions.

High fidelity finite element analyses of the three reference sandwich panels were performed. The
maximum center deflection obtained from the analyses are given in Table 17.

Table 17. Maximum Center deflection for the three reference solutions

Reference Solution Maximum center deflection (mm)

1 -1.4758
2 -.04594
3 -.02768

These reference solutions will be used to verify the center deflection of the homogenized models
in the next section.

8. Homogenized finite element models used to assess effective core properties

A 2-D shell model and a 3-D solid model of the sandwich structure were created to assess the
accuracy of replacing the honeycomb core configurations with an effective homogenized layer to
improve computational efficiency. The shell model is composed of 7,104 4-node S4R elements
and 7,280 nodes, while the solid model is composed of 478,080 8-node C3D8R elements and
502,169 nodes. These models are depicted in Figures 17 and 18.

22
 Figure 17. 2-D shell model with a homogenized core.

Figure 18. 3-D solid model with a homogenized core.

The homogenized 2-D shell model and 3-D solid models were analyzed using the same boundary
conditions and loading used in the reference solution as depicted in Figure 16. The maximum
displacements obtained using the homogenized core properties obtained for the three laminates
described in Section 5 (Tables 9 to 12) are compared with the three reference solutions in the next
section.

23
9. Comparison of maximum center deflection with reference solutions
This section completes the investigation into the coupled methodology to compute effective elastic
properties of honeycomb cores with multi-layered walls. The reference problem selected in Section
7 represents a complex multi-cell model of a sandwich panel with realistic load conditions. The
simplified FEM models developed in Section 8 contain homogenized representations of the central
core and will necessarily contain various modeling aspects inherent to finite element simulations
that may introduce differences in deflections predicted by the reference model and by the
simplified homogenized models. Two sets of comparisons were made, one consisting of finite
element predictions of deflections using FEM-derived effective moduli - which are considered the
most accurate - and the second, consisting of finite element predictions using Gibson’s
approximations to the effective moduli of the core layer. It was assumed that the calculated
deviations of the homogenized model deformations from the reference solution deformations
represents inherent modeling differences between the different FEM representations, and the
deviation of the same homogenized model using Gibson’s effective moduli as the core material
properties would yield a different percent deviation with the difference between the two due solely
to the approximations inherent to Gibson’s analytical method. It is shown in the following
subsections that the difference between the homogenized models with the reference solutions is
small, less than a maximum of 11.5%, but, more importantly, the difference between the
homogenized models using effective FEM-derived core properties and Gibson’s effective moduli
were very small, less than two percent. This good agreement may be partly explained by the close
agreement of the FEM-derived and the Gibson-derived effective modulus Ezz for the out-of-plane
properties as shown in Tables 11 – 13.

9.1 Deflection comparison of a three-layered isotropic [Nomex/Nomex/Nomex] Laminate-1

core in a Sandwich plate with reference and homogenized solutions

The faceplates were assigned isotropic properties corresponding to a Nomex material with
E  3150 MPa,  xy  0.4 and t  0.25 mm. Note that for Laminate-1, the homogenized effective
core properties are the same using both CLT and the Chen-Chan equations due to the vanishing of
the coupling matrix, B. The core in the homogenized model was assigned material properties from
Table 12. The central deflections are listed in Table 18 where the percent deviation from the
reference solution is also presented.

Table 18. Laminate-1 homogeneous model results compared to the reference model solution. The
reference solution calculated a maximum center deflection of  = -1.4758 mm.

Homogeneous Maximum Deviation from Maximum deflection Deviation from

Models deflection using reference using effective reference
effective moduli solution moduli from solution
from FEM analysis Gibson’s equations
2-D Shell -1.5660 mm 6.1% -1.5672 mm 6.1%

3-D Solid -1.5364 mm 4.1% -1.5376 mm 4.2%

24
This analysis shows a good agreement between the simplified homogeneous models compared to
the high-fidelity reference solution for approximating the behavior of the honeycomb core replaced
by effective material properties. Comparing the maximum center deflection of the homogeneous
models using FEM-derived and Gibson-derived effective core properties, the solutions are
virtually identical, differing only by at most 0.1%

9.2 Deflection comparison of a symmetric three-layer isotropic [Al/Cu/Al] Laminate-2 core

in a sandwich plate with reference and homogenized solutions

The faceplates were assigned steel properties with E  1.0 E 6 MPa,  xy  0.3 and t  0.25 mm.
Note that for Laminate-2, the homogenized effective properties are the same using both the CLT
and the Chen-Chan equations. The core in the homogenized model was assigned material
properties from Table 13. The central deflections are listed in Table 19 where the percent deviation
from the reference solution is also presented.

Table 19. Laminate-2 homogeneous model results compared to reference model solution. The
reference solution calculated a maximum center deflection of  = -.04594 mm.

Homogeneous Maximum deflection Deviation from Maximum deflection Deviation from

models using effective moduli reference using effective moduli reference
from FEM analysis solution from Gibson’s equations solution
2-D Shell -.05065 mm 10.2% -.05122 mm 11.5%

3-D Solid -.04950 mm 7.74% -.05007 mm 8.98%

This analysis shows a good agreement between the simplified homogeneous models compared to
the high-fidelity reference solution for approximating the behavior of the honeycomb core replaced
by effective material properties. The maximum percent difference is about 12 percent. Comparison
of the maximum center deflection of the homogeneous models using FEM-derived and Gibson-
derived effective core properties shows that the solutions are virtually identical, differing by at
most 1.3%

9.3 Deflection comparison of a nonsymmetric three-layer isotropic [Al/Epoxy/Cu]

Laminate-3 core in a sandwich plate with reference and homogenized solutions.

The faceplates were assigned Aluminum properties with E  70,300 MPa, xy  0.33 and t  0.25
mm. For Laminate-3, the homogenized effective properties using CLT and the Chen-Chan

25
equations are different. Hence, both the homogenized properties obtained using CLT (Table 15)
and obtained by Chen-Chan (Table 14) were analyzed and compared with the reference solution.

The central deflections obtained using Chen-Chan properties are compared with the corresponding
center deflection from the reference in Table 20 where the deviation from the reference solution is
also presented.

Table 20. Laminate-3 homogeneous model results using Chen-Chan effective wall modulus in
Gibson’s equations. The reference solution calculated a maximum center deflection of
 = -0.027685 mm.

Homogeneous Maximum deflection Deviation from Maximum deflection Deviation from

models using effective moduli reference using effective moduli reference
from FEM analysis solution from Gibson’s equations solution
2-D Shell -.028913 mm 4.4% -.028523 mm 3.9%

3-D Solid -.028327 mm 2.3% -.027921 mm 0.85%

Comparison of the maximum center deflection of the homogeneous models using FEM-derived
and Gibson-derived effective core properties shows that the solutions are virtually identical,
differing by at most 1.45%. The central deflections obtained using conventional CLT properties
are compared with the corresponding center deflection from the reference solution, in Table 21.

Table 21. Increase in error of center deflection using CLT in Laminate-3 for the effective wall
modulus in Gibson’s equations. The reference solution calculated a center deflection
 = -0.027685 mm.

Homogeneous Maximum deflection Deviation from

Models using effective moduli reference
from Gibson’s equations solution
2-D Shell -.025772 mm 8.9%

3-D Solid -.028951 mm 4.6%

Results obtained using homogenized effective properties from the Chen-Chan equations result in
more accurate predictions of the center deflection for Laminate-3, which consists of materials
representative of electrodes and electrolyte that could be used in a battery.

26
10. Summary

The homogenization of complex core architectures is important to decrease the overall

computational requirements in the simulation of multifunctional sandwich composites. This has a
direct impact on the rapid assessment and optimization of design prototypes accounting for
geometric sizing and material applications. A primary objective of this study was to investigate
the accuracy of extending the widely used Gibson’s approach to cellular structures with multi-
layered walls that are designed for multifunctionality.

This work investigated a coupled approach using Chen-Chan’s correction to classical lamination
theory to compute equivalent moduli for multi-layered core-walls for use in the beam bending
solutions of Gibson. The generality of this approach makes it extendable to wall configurations
with an arbitrary number of layers. An equivalent wall modulus and the effective core properties
can be calculated using a small set of closed-form analytical expressions. It was demonstrated that,
as the width of the face plate separation increased, Gibson’s 1-D elasticity solutions departed from
the FEM simulation of the effective core moduli as the 2-D stress state became more pronounced
and that a plate bending solution was more accurate in capturing the cell deformations. This
departure, however, was shown to approach a bounded maximum value as the core width increased
beyond a few multiples of the cell size. For the core widths examined, this maximum departure
was 18% for the in-plane moduli, Exx and Eyy.

For illustration, several multi-layered wall configurations were selected with differing material
properties and symmetry across the midline. It was shown through comparison to unit cell finite
element calculations that the analytically estimated equivalent core properties are obtained with
good accuracy for the different multi-layered wall layers considered.

As a final test, a highly refined model of a 10 x 10 cell sandwich composite panel with all cells
explicitly modeled was generated to serve as a reference solution to compare the accuracy of
simplified FEM models incorporating homogenized cores. The panel was clamped along the edges
and subjected to a normal pressure over the upper surface. A simplified 2-D shell element model
and a 3-D solid element model were developed to simulate the sandwich panel with a homogenized
core. For each model, two different simulations were performed and compared to the reference
solution. One used FEM-derived equivalent moduli for the core – which are considered the most
accurate – and one used the analytical moduli based on Gibson’s approach. The difference in the
maximum center deflection between the homogenized model with FEM-derived equivalent moduli
and the reference solution were assumed to be based on various FEM modeling approximations
inherent to the representation of the sandwich panel. The same deviation was calculated using
effective core moduli obtained from Gibson’s analysis. All these departures from the reference
solution were reasonably small, the maximum being 11.5%. However, it was also shown that the
difference between the homogenized models using FEM-derived or Gibson’s effective moduli was
within 2% for the three different laminated walls considered.

In conclusion, the coupled analytical solution for effective core moduli investigated in this study
demonstrated an efficient method to simplify the modeling of multifunctional sandwich composite
cores while maintaining a high accuracy compared to a computationally intensive explicit
simulation of complex core configurations.

27
11. References

1. P. Liu, E. Sherma and A. Jacobsen, “Design and fabrication of multifunctional structural

batteries,” J. Power Sources, Vol. 189, pp. 646-650, 2009.
2. R.F. Gibson, “A review of recent research on mechanics of multifunctional composite
materials and structures,” Comp. Struct., Vol. 92, pp. 2793-2810, 2010.
3. K.J. Narayana and R.G. Burela, “A review of recent research on multifunctional composite
materials and structures with their applications,” Materials Today: Proceedings, Vol. 5, pp.
5580-5590, 2018.
4. L.E. Asp and E.S. Greenhalgh, “Structural power composites,” Comp. Sci. Tech., Vol. 101,
pp. 41-61, 2014.
5. P.L. Loyselle, “Multifunctional Structures for High Energy Lightweight Load-bearing
Storage,” NASA Glenn Research Center, https://ntrs.nasa.gov , Document ID
20180000928.
6. H.E. Soliman, “Mechanical properties of cellular core structures,” Ph.D. Thesis, Virginia
Polytechnic Institute and State University, March 21, 2016.
7. L. J. Gibson et al., “The mechanics of two-dimensional cellular materials,” Proc. R. Soc.
Lond. A, Vol. 382, pp. 25-42, 1982.
8. I.G. Masters and K.E. Evans, “Models for the elastic deformation of honeycombs,” Comp.
Struct. Vol. 35, pp. 403-422, 1996.
9. S. Malek and L. Gibson, “Effective elastic properties of periodic hexagonal honeycombs,”
Mech. Mat., Vol. 91, pp. 226-140, 2015.
10. H.S. Moghaddam, C. Yang, S.R. Keshavanarayana, and A.L. Horner, “Analysis of in-plane
mechanical response of hexagonal honeycomb core: Effect of node bond adhesive and
proposed analytical method,” Proc. American Society for Composites, 32nd Technical
Conference, Oct. 23-25, 2017.
11. S.R.Keshavanarayana, H.Shahverdi, A.Kothare, C.Yang and J.Bingenheimer,”The effect
of node bond adhesive fillet on uniaxial in-plane responses of hexagonal honeycomb core,”
Comp. Struct., Vol. 175, pp. 111-122, 2017.
12. S. Sorohan, M. Sandu, D.M. Constantinesce, and A.G. Sandu, “On the evaluation of
mechanical properties of honeycombs by using finite element analysis,” INCAS Bulletin,
Vol. 7, pp. 135-150, 2015.
13. M. Hajianmaleki and M.S. Qatu, “Mechanics of composite beams,” Ch22 Advances in
Composite Materials – Analysis of Natural and Man-Made Materials, Ed. P. Tesinova,
IntechOpen, 2011.
14. R.M. Jones, Mechanics of Composite Materials, Hemisphere Publishing Corp., 1975.
15. D. J. Chen and W.S. Chan, “Use of composite effective moduli for lumping layers in finite
element analysis,” American Institute for Aeronautics and Astronautics paper AIAA-96-
1568-CP, 1996.
16. Abaqus Analysis User’s Guide. 2018. Abaqus 2018, DS Simulia Corp., Providence, RI,
USA.
17. F.E. Penado, “Effective elastic properties of honeycomb core with fiber-reinforced
composite cells,” Open J. Comp. Mater., Vol. 3, pp. 89-96, 2013.

28
18. M. Tauhiduzzaman and L.A. Carlsson, “Influence of constraints on the effective in-plane
extensional properties of honeycomb core,” Comp. Struct., Vol. 209, pp. 616-624, 2019.
19. J. Hohe and W. Becker, “A refined analysis of the effective elasticity tensor for general
cellular sandwich cores,” Int. J. Solids Struct., Vol. 38, pp. 3689-3717, 2001.

29
 Form Approved
REPORT DOCUMENTATION PAGE OMB No. 0704-0188

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data
sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other
aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information
Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other
provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
1-04-2019 Technical Memorandum
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER

5b. GRANT NUMBER

An Analytical Method to Calculate Effective Elastic Properties of General
Multifunctional Honeycomb Cores in Sandwich Composites
5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

Saether, Erik; Krishnamurthy, Thiagarajan 5e. TASK NUMBER

5f. WORK UNIT NUMBER

975105.09.23
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
REPORT NUMBER

NASA Langley Research Center

Hampton, VA 23681-2199
L-21020
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S)

National Aeronautics and Space Administration NASA

Washington, DC 20546-0001 11. SPONSOR/MONITOR'S REPORT
NUMBER(S)
NASA-TM-2019-220275
12. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified-
Subject Category 24
Availability: NASA STI Program (757) 864-9658

13. SUPPLEMENTARY NOTES

14. ABSTRACT
Sandwich composite structures are ideal configurations in which to incorporate additional functionality beyond load carrying capabilities. The inner core-walls can be
layered to incorporate other functions such as power storage for a battery. In this work we investigate an assemblage of analytical tools to compute effective properties
that allow complex layered core architectures to be homogenized into a single continuum layer. This provides a great increase in computational efficiency to
numerically simulate the structural response of multifunctional sandwich structures under applied loads. We present a coupled analytical method including an extensive
numerical verification of the accuracy of this method.

15. SUBJECT TERMS

Effective moduli; Honeycomb core; Sandwich composites

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
ABSTRACT OF
a. REPORT b. ABSTRACT c. THIS PAGE STI Help Desk (email: [email protected])
PAGES
19b. TELEPHONE NUMBER (Include area code)
U U U UU (757) 864-9658
35
Standard Form 298 (Rev. 8/98)
Prescribed by ANSI Std. Z39.18