An Explanation of Why the

Elastic Matrix is Symmetric

Dr. Lonny L. Thompson
November 11, 2024

    







σ11 






 C11

C12 C13 C14 C15 C16 





ε11 





 
σ22  C12 C22 C23 C24 C25 C26 ε22

 
   

 
   


 
  


 
   

 
σ33  C13 C23 C33 C34 C35 C36 ε33

 
   

 
   

   

= 










σ 23







 C



14 C24 C34 C44 C45 C46 






2ε23 
σ  C C25 C35 C45 C55 C56 2ε13 

 
   
13 15

 
 

 
  

 
  
 σ12  C16 C26 C36 C46 C56 C66 2ε12 

 
   


  

 

Swipe for the next page

Contents

1 Why is the Elastic Matrix Symmetric? 2

1.1 Tensor Form of Constitutive Equations . . . 3
1.2 Physical Interpretation of Stress Components 4
1.3 Symmetry of the Stress Tensor . . . . . . . 7
1.4 Minor and Major Symmetry Requirements
for the Stiffness Tensor . . . . . . . . . . . . 10
1.5 Constitutive Equations in Voigt Notation . . 16
1.6 Anisotropic Materials . . . . . . . . . . . . . 19
1.7 Orthotropic Symmetry . . . . . . . . . . . . 20
1.8 Transversely isotropic . . . . . . . . . . . . 22
1.9 Isotropic materials . . . . . . . . . . . . . . 24
1.10 Orthotropic Material Behavior for Plane Strain 26
1.11 Isotropic Material Behavior for Plane Strain 31
1.12 Plane Stress Assumption . . . . . . . . . . . 33
1.13 Conclusions . . . . . . . . . . . . . . . . . . 34

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 1 of 37

1 Why is the Elastic Matrix Symmetric?

Why is the Elastic (Stiffness) Matrix relating indepen-

dent stress and strain components for linear elastic
materials symmetric?
• An important property of the stiffness matrix for finite
element analysis (FEA) for stress analysis is that the ma-
trix is symmetric. This requires the elastic matrix [C]
relating independent stress and strain components for
linear elastic materials to be symmetric. The question
then arises: Why is the Elastic Matrix Symmetric?
• Due to major symmetry and minor symmetry due to
the symmetry of stress and strain tensors of the elastic
tensor, the corresponding stiffness matrix [C]T = [C] is
symmetric, such that the transpose operator does not
change the matrix and the rows and columns can be
interchanged with no change CJI = CIJ .
• The matrix form {σ} = [C]{ε} of the stiffness tensor con-
stitutive equations σij = Cijkl εkl allows for standard ma-
trix algebra operations for linear transformations and
matrix implementation in computational methods such
as the finite element method (FEM).

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 2 of 37

1.1 Tensor Form of Constitutive Equations
Let’s start with the most general elastic tensor in three
dimensions
• The general 4th-order stiffness (elastic) tensor Cijkl re-
lates strain to stress tensors in a linear elastic material
through the nine constitutive equations:
3 3
σij = Cijkl εkl , i = 1, 2, 3, and j = 1, 2, 3.
X X

k=1 l=1

σij : components of the 2nd-order, symmetric Cauchy

stress tensor,
εkl : components of the 2nd-order linearized strain
tensor (symmetric part of linear displacement gradi-
ent tensor), and
Cijkl : components of the 4th-order elastic (stiffness)
tensor.

• The indices i, j, k, and l can each take values of 1, 2,

or 3 in three dimensions (or 1 and 2 in two dimensions)
corresponding to the rectangular spatial coordinates:

x = x1 ,
y = x2
z = x3

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 3 of 37

 1.2 Physical Interpretation of Stress Com-
ponents
1. Normal Stresses: (σ11, σ22, σ33) represent forces act-
ing perpendicular to the respective planes within the
material. Positive values indicate tension, while neg-
ative values indicate compression.
2. Shear Stresses: (σ12, σ13, σ23) represent forces act-
ing parallel to the respective planes within the ma-
terial. Shear stresses tend to deform the material by
sliding layers over one another.

By understanding these components, engineers can an-

alyze how materials respond to different types of load-
ing conditions, such as tension, compression, shear, bend-
ing, and torsion, and design structures to withstand these
loads safely.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 4 of 37

 Stress Component Indexes i and j

x2
σ22
σ21
σ23 σ12
σ32
σ11
σ31σ13
σ33 x1

x3
Stress components on volume element – in the limit, this element
shrinks to a point.

To grasp the physical meaning of the surface and direc-

tion of these stress components, let’s break down the
notation σij
face direction
1. First Subscript i: Direction of the normal to the per-
pendicular face surface on which the stress acts.
2. Second Subscript j: Direction of the stress compo-
nent acting on that plane. This indicates the direc-
tion in which the internal force is applied.
LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 5 of 37
 Interpretation of Stress Components σij
In a three-dimensional space, stress is represented by a
second-order tensor (or matrix) with nine components:
 
σ11 σ12 σ13


σ = σ21 σ22 σ23 , σij
 



σ31 σ32 σ33
 
face direction

x2

σ22
σ32
σ
σ21
σ23σ 12
31 σ33
σ13 σ32
σ11 σ11
σ31 σ13
σ12 σ23 x1
σ33
σ21

x3 σ22

Index convention for σij components represents the jth component

of the surface force acting on the surface whose normal is parallel to
the xi-axis. All components shown are positive.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 6 of 37

 1.3 Symmetry of the Stress Tensor
Within classic elasticity theory, the stress tensor is sym-
metric, meaning that σij = σji.
x2
σ22

σ12
σ23 σ12
σ23
σ11
σ13σ13
σ33 x1

x3
Stress components acting on surfaces of a volume element.
When the volume limits to zero, the element shrinks to a point.

If the solid is in equilibrium, within classical elasticity

theory, the stress tensor must be symmetric.

This symmetry reflects the fact that shear stresses on
perpendicular planes must be equal in magnitude but
act in opposite directions to maintain rotational equi-

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 7 of 37

 librium. Therefore:

σ12 = σ21, σ13 = σ31, σ23 = σ32

Consider the components of stress σ12 and σ21. σ12

and the same σ12 on the opposite side of the volume
element create a couple that would act to produce
an angular rotation. σ21 is the only couple that can
balance σ12. Therefore, σ12 = σ21, and with the same
argument for the other two shear pairs, then σij = σji
for equilibrium.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 8 of 37

  
σ11 σ12 σ13


σ = σ12 σ22 σ23
 



σ13 σ23 σ33
 

Normal Stress Components on Diagonal

(σ11, σ22, σ33) represent stress components acting per-

pendicular to the respective planes within the material.
Positive values indicate tension, while negative values
indicate compression. Normal stresses tend to elon-
gate or contract the material along the respective axis
x1, x2, and x3.

Shear Stresses on Off-Diagonals

(σ12, σ13, σ23) represent stress components acting paral-

lel to the respective planes within the material. Shear
stresses tend to deform the material by distortion.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 9 of 37

1.4 Minor and Major Symmetry Requirements
for the Stiffness Tensor
The stiffness tensor Cijkl exhibits two types of symmetry:
Minor and Major.
1. Minor Symmetry: Symmetry within the the first and
last pairs of indices ij and kl.

C(ij)kl = C(ji)kl and Cij(kl) = Cij(lk)

This symmetry arises because the stress and linear strain

tensors themselves are symmetric:

σij = σji and εkl = εlk

2. Major Symmetry: Symmetry between the first and

second pairs pairs of indices ij and kl.

Cijkl = Cklij

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 10 of 37

How do we show the stiffness tensor has ma-
jor symmetry?
1. One way of showing major symmetry is the require-
ment that the scalar strain energy density quadratic
function in stress or strain components is the same (in-
variant) regardless of the path taken to reach a partic-
ular strain state.

• Scalar Strain Energy defined by an inner product of

stress with strain tensors is invariant to coordinate
transformations.
1 X3 X 3
W = εij σij
2 i=1 j=1
1 X3 X 3 X 3 X 3
= εij Cijkl εkl
2 i=1 j=1 k=1 l=1
1 X3 X 3 X 3 X 3
= εkl Cklij εij
2 k=1 l=1 i=1 j=1

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 11 of 37

2. Another way to show major symmetry from strain en-
ergy density, quadratic in the strain components, is to
consider partial derivatives with respect to strain com-
ponents.

• Stress as the derivative of strain energy with respect

to strain and stiffness as the partial derivative of stress
with respect to strain.
• The first partial derivative of W with respect to εij
gives the stress components, while the second par-
tial derivative gives the 4th-order stiffness tensor:
∂ 2W
Cijkl =
∂εij ∂εkl

• Since the mixed partial derivatives of a scalar func-

tion are symmetric, this implies the major symmetry
between the first and last pairs of indices:
∂ 2W ∂ 2W
Cijkl = = = Cklij
∂εij ∂εkl ∂εkl ∂εij

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 12 of 37

3. Another way to define the major symmetry of a 4th-
order tensor is that it has the property that when tak-
ing the tensor inner product of two arbitrary 2nd-order
tensors, say Aij and Bij , switching the order of the
product does not change the resulting scalar value of
the products.

scalar = Aij Cijkl Bkl = Bij Cijkl Akl

• In the above, summations are implied on the repeated

indices, which is the standard convention first used
by Einstein and is known as Eistein’s summation con-
vention.
• This equality requires major symmetry, such that the
first two and last two indices must be interchange-
able for the tensor to be symmetric. That is,

C(ij)(kl) = C(kl)(ij)

• A short-hand notion for tensor inner products is to

use a colon notation for the summation of a pair of
indices. For the symmetry condition, we write

scalar = A : C : B = B : C : A

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 13 of 37

How do these major and minor symmetries
reduce the number of independent compo-
nents of the elastic tensor and form a sym-
metric elastic matrix?
• The minor and major symmetries reduce the number of
independent components in Cijkl from 34 = 81 to 21 in
the general 3D case for an anisotropic material.
• For materials with additional symmetries, like isotropic
or orthotropic materials, the number of independent
components is further reduced.
• For anisotropic materials, the elastic stiffness matrix gov-
erns the relationship between stress and strain, often
represented as a 4th-order tensor Cijkl or in matrix form
with coefficients Cmn, where m and n are indices that
map the 3D indices ij and kl into a 6 × 6 matrix.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 14 of 37

Professor Voigt to the rescue
• In Voigt notation, the symmetric stress and strain ten-
sors are represented using a single index instead of
two, simplifying the matrix representation of the stiff-
ness tensor.
• The mapping from the tensor indices ij or kl to the sin-
gle Voigt index I or J is as follows:
11 → 1
22 → 2
33 → 3
23, 32 → 4
13, 31 → 5
12, 21 → 6

• Using this mapping, the stiffness tensor components

Cijkl can be written as a 6 × 6 matrix [C] with coefficients
CIJ where I and J correspond to the indices mapped
from ij and kl respectively.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 15 of 37

1.5 Constitutive Equations in Voigt Notation
{σ} = [C]{ε}
    







σ11 






 C11

C12 C13 C14 C15 C16 





ε11 





 
σ22  C12 C22 C23 C24 C25 C26 ε22

 
   

 
   


 
  


 
  




σ33  C13 C23 C33 C34 C35 C36 ε33

 
   

 
  




   
= 










σ 23







 C



14 C24 C34 C44 C45 C46 






2ε23 
σ13   C15 C25 C35 C45 C55 C56 2ε13 

 
   

 

 

 
 σ12  C16 C26 C36 C46 C56 C66 2ε12 

 
   


  

 

• From the major symmetry in the elasticity tensor Cijkl =

Cklij , then the corresponding CIJ = CJI the elements in
the matrix [C] below the main diagonal are the same as
the corresponding elements above the main diagonal,
allowing us to replace the elements below the diagonal
with those above it, reducing the 6 × 6 = 36 elements
to 21.
• Due to both the major symmetry and minor symmetry
due to the symmetry of stress and strain tensors, the
stiffness matrix [C]T = [C] is symmetric, such that the
transpose operator does not change the matrix and the
rows and columns can be interchanged with no change
CJI = CIJ .

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 16 of 37

Constructing the 6 × 6 Stiffness Matrix [C]
• Given the mapping of Voigt and the requirement that
the shear strain components are doubled in {ε}, i.e.,
we write {ε} with 2ε23, 2ε13 and 2ε12, the stiffness tensor
Cijkl is converted into the Voigt matrix CIJ = CJI as a
one-to-one mapping:
   


C11 C12 C13 C14 C15 C16   C1111 C1122 C1133 C1123 C1113 C1112 



 C21 C22 C23 C24 C25 C26   C2211 C2222 C2233 C2223 C2213 C2212 


C31 C32 C33 C34 C35 C36   C3311 C3322 C3333 C3323 C3313 C3312
   
 
 = 
 
 
C41 C42 C43 C44 C45 C46   C2311 C2222 C2333 C2323 C2313 C2312
 
 
 
 


 C51 C52 C53 C54 C55 C56   C1311 C1322 C1333 C1323 C1313 C1312 


C61 C62 C63 C64 C65 C66 C1211 C1222 C1233 C1223 C1213 C1212
   

• Using this logical mapping, the first row and column

indices 1, 2, 3, 4, 5, 6 in Voigt notation correspond to the
index pairs (11), (22), (33), (23), (13), (12), respectively.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 17 of 37

Stiffness Matrix and Constitutive Equations
Anisotropic Materials in Voigt Notation
• The 4th-order stiffness tensor Cijkl expressed as a 6 × 6
matrix with mapped indices I and J ranging over the
six independent terms in the strain and stiffness tensors
simplifies the representation and complexity while pre-
serving its full physical meaning.
• The matrix form {σ} = [C]{ε} of the stiffness tensor con-
stitutive equations σij = Cijkl εkl allows for standard ma-
trix algebra operations for linear transformations and
matrix implementation in computational methods such
as the finite element method (FEM).

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 18 of 37

1.6 Anisotropic Materials
For the most general anisotropic material, the stiffness
matrix Cmn is a 6 × 6 symmetric matrix, meaning there are
36 components in total.
    







σ11 






 C11

C12 C13 C14 C15 C16 





ε11 





 
σ22  C12 C22 C23 C24 C25 C26 ε22

 
   

 
   


 
  


 
   

 
σ33  C13 C23 C33 C34 C35 C36 ε33

 
   

 
   

   

= 










σ 23







 C



14 C24 C34 C44 C45 C46 






γ23 
σ  C C25 C35 C45 C55 C56 γ13 

 
   
13 15

 
 

 
  

 
  
 σ12  C16 C26 C36 C46 C56 C66 γ12 

 
   


  

 

Due to symmetries, this reduces to 21 independent elastic

moduli.

• The number of independent elastic constants (moduli)

in this stiffness matrix depends on the symmetry of the
material.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 19 of 37

Special Cases of Symmetry
Reflective planes and rotational symmetries simplify the
stiffness matrix’s form and structure, reducing the number
of independent elastic moduli. Here are some important
symmetry classes in engineering mechanics:

1.7 Orthotropic Symmetry

9 independent elastic constants.
Materials whose properties vary along three orthogonal
directions with three orthogonal (perpendicular) planes
of symmetry, like composites or wood, have different
properties along three mutually perpendicular axes but
still exhibit symmetry in each plane.

For a 3D orthotropic material, the stiffness matrix in Voigt

notation is given by:
    







σ11 






 C11

C12 C13 0 0 0 





ε11 





 
σ22  C12 C22 C23 0 0 0 ε22

 
   

 
   


 
  


 
   

 
σ33  C13 C23 C33 0 0 0 ε33

 
   

 
   

   

= 










σ 23







 0



0 0 C44 0 0 






γ23 
σ  0 0 0 0 C55 0 γ13 

 
   
13

 
 

 
  

 
  
 σ12  0 0 0 0 0 C66 γ12 

 
   


 
 



LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 20 of 37

The nine components define the material’s stiffness prop-
erties.

C11, C22, C33, C12, C13, C23, and C44, C55, C66

Orthotropic material behavior often occurs in composites

and metamaterials.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 21 of 37

1.8 Transversely isotropic
These materials have isotropic properties in a plane per-
pendicular to a unique axis
(usually referenced to the z = x3 axis).
The stiffness matrix for a transversely isotropic material
is more restrictive due to its higher symmetry reducing
to only five independent components:
 
C11


C12 C13 0 0 0 

C12 C11 C13 0 0 0
 
 
 
 
C13 C13 C33 0 0 0
 
 
 
[C] = 



0



0 0 C44 0 0 



0

 0 0 0 C44 0 


C11 −C12
 
0 0 0 0 0
 
2
Here, the five components

C11, C12, C13, C33, and C44

define the material’s stiffness properties, with additional

restrictions due to the symmetry expressed by
C11 − C12
C66 =
2
representing the in-plane shear modulus.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 22 of 37

 C11 and C12 represent the in-plane (transverse) stiffnesses
where the material behaves isotropically.
C13 and C33 represent the coupling between the in-plane
strains and the out-of-plane direction, and the stiffness
in the out-of-plane direction, respectively.
C44 represents the shear modulus in planes containing
the unique axis.
C11 −C12
C66 represents the in-plane shear modulus 2 .

The 5 independent elastic constants in the stiffness matrix

for transversely isotropic materials are:
1. C11: Stiffness in the isotropic plane.
2. C12: Stiffness due to in-plane Poisson’s effect.
3. C13: Coupling between in-plane and out-of-plane strains.
4. C33: Stiffness along the unique axis.
5. C44: The shear modulus involving the unique axis.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 23 of 37

1.9 Isotropic materials
These are the simplest case; there are only two inde-
pendent elastic constants.
Isotropic materials have the same mechanical proper-
ties in all directions. Due to this symmetry, the descrip-
tion of the material’s behavior simplifies to only two in-
dependent elastic constants in every direction, reduc-
ing the complexity of the stiffness tensor Cijkl and cor-
responding symmetric matrix [C].
For isotropic materials, [C] must be invariant under rota-
tion. This implies that the first three and last three diag-
onal terms are equal, and it can be shown that relation-
ships exist between the diagonals and off-diagonals re-
duce to a form with only two independent parameters,
typically chosen as

C11 = λ + 2G

and shear modulus (modulus of rigidity)

C44 = G

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 24 of 37

 In this case, the symmetric stiffness matrix for an isotropic
material simplifies to:
 
λ + 2G


λ λ 0 0 0 
λ λ + 2G λ 0 0 0 
 



λ λ λ + 2G 0 0 0 
 


[C] = 
 




0 0 0 G 0 0 




0 0 0 0 G 0 

0 0 0 0 0 G
Here, λ and G (also known as Lamé constants) are related
to Young’s modulus E and Poisson’s ratio ν by:
Eν
λ=
(1 + ν)(1 − 2ν)
E
G=
2(1 + ν)

Bottom Line: Isotropic materials have infinite rotational

and reflective symmetry, reducing to only two indepen-
dent elastic constants.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 25 of 37

1.10 Orthotropic Material Behavior for Plane
Strain
Let’s explore this further and see how the orthogonal planes
of symmetry can be used to reduce the number of inde-
pendent material constants in the case of a two-dimensional
state of plane stress and strain.

• For plane strain, ε13 = ε23 = ε33 = 0.

For a two-dimensional orthotropic material, which has
two symmetry planes the material properties are direction-
dependent but exhibit symmetry about the two planes.
This symmetry can be exploited to reduce the number
of independent material constants in the stiffness ma-
trix from six (for a general anisotropic material in 2D) to
four.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 26 of 37

• The 2D in-plane stress-strain relationship for a general
anisotropic material reduces to:
    






σ11 





 C11 C12 C16







ε11 






σ

 C
12 C22 C26

ε22 

=

22 

  

 
  

 σ C16 C26 C66 γ12 

 
  
 
12 
  


σ11, σ22 : the normal stress components.

σ12 : in-plane shear stress component.
ε11, ε22 : normal strain components.
γ12 = 2ε12 : engineering shear strain angle (measured
in nondimensional radians) within the x1 − x2 plane.
• For a material with two orthogonal symmetry planes,
the following simplifications occur:
1. Symmetry about the x1-axis: Reflecting the mate-
rial about the x1-axis should not change the mate-
rial properties. This symmetry implies that C16 = 0
because the shear response should be symmetric,
eliminating any coupling between normal strains ε11
and the shear strain γ12.
2. Symmetry about the x2-axis: Reflecting the mate-
rial about the x2-axis should not change the mate-
rial properties. This symmetry implies that C26 = 0,
as it similarly eliminates any coupling between nor-
mal strains ε22 and the shear strain γ12.
LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 27 of 37
• These simplifications can also be shown rigorously us-
ing reflection matrices about the x1 and x2 axis to trans-
form strain and stress components and confirming that
since σ11 and σ22 remain unchanged, then for arbitrary
strains C16 and C26 must be zero.

Any non-zero values for C16 or C26 would violate the

symmetry of the material, as they would introduce a
coupling between the normal strains and the shear strain
that would change sign under reflection, which contra-
dicts the requirement that the material’s behavior re-
mains unchanged upon reflection about the two orthog-
onal (perpendicular) axis.
The key idea is that the material’s properties must re-
main invariant under reflections about the symmetry planes.
This imposes constraints on the stiffness matrix com-
ponents, forcing certain components to be zero and
reducing the overall number of independent compo-
nents.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 28 of 37

 Given these symmetries, the stiffness (elastic) matrix
for orthotropic material behavior simplifies to:
    






σ11 





 C11 C12

0 





ε11 






σ

 C12 C22

ε22 

= 0

22
 
  

 
  

 σ 0 0 C66 γ12 

 
  
 
12 
  


This matrix contains the following four independent ma-

terial constants:
C11 and C22: Relates σ11 to ε11, and σ22 to ε22, represent-
ing stiffness in the x1 and x2 directions respectively.
C12: Relates σ11 to ε22 and σ22 to ε11, representing the
coupling between strains in orthogonal directions due
to Poisson’s ratio effects.
C66: Relates σ12 to γ12, representing shear stiffness in
the material, decoupled from normal stress and strains.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 29 of 37

 Using the two symmetry planes for a 2D orthotropic
material, the number of independent material constants
in the constitutive (stiffness, elastic) matrix is reduced from
six (for a 2D fully anisotropic material) to four.

The key simplifications are the elimination of the cou-

pling terms C16 and C26, which correspond to shear-
normal coupling terms that would otherwise exist in a
fully anisotropic material. This reduction in indepen-
dent constants simplifies the analysis and modeling of
orthotropic materials.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 30 of 37

1.11 Isotropic Material Behavior for Plane
Strain
Now, let’s see how to reduce the four orthotropic proper-
ties to two for isotropic.

For isotropic materials, the stiffness matrix must be in-

variant under rotation.
This implies that the diagonal terms C11 = C22 are equal
since isotropic material behaves the same in any direc-
tion, and it can be shown that the relationship between
C11 and C12 is C12 = C11 − 2C66, and
C66 (the shear modulus) is related to the other constants
by C66 = C11−C
2
12 .

Substituting the isotropy conditions into the stiffness ma-

trix, the matrix reduces to a form with only two indepen-
dent constants, typically chosen as C11 and C66.
 


C11 C11 − 2C66 0 
[C] =  C11 − 2C66 C11 0 
 

0 0 C66
 

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 31 of 37

 Letting the diagonal terms be
C11 = λ + 2G, C66 = G,
and off-diagonal terms
C12 = C11 − 2C66 = λ,
the stiffness matrix for a 2D plane isotropic material can
be expressed in terms of the Lamé parameters λ and G:
 
 λ + 2G λ 0 

[C] =  λ λ + 2G 0 
 

0 0 G
 

or equivalently expressed in terms of E and ν.

• Thus, for the plane strain assumption: εxz = εyz = εzz =
0, the in-plane constitutive equations for linear elastic
isotropic materials and the elasticity matrix [C] in {σ} =
[C]{ε} is a square symmetric matrix with three rows and
three columns:
    






σ 
xx 




λ + 2G

λ 0 ε 






xx 




{σ} = σyy  =  λ
 
λ + 2G 0 εyy 
  




σ 0 0 G γ

 
  
 

xy xy 
 
  


Eν E
λ= , G=
(1 + ν)(1 − 2ν) 2(1 + ν)

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 32 of 37

1.12 Plane Stress Assumption
• For the plane stress assumption: σ13 = σ23 = σ33 = 0, it
can be shown that the in-plane constitutive equations
for linear elastic isotropic materials and the elasticity
matrix [C] in {σ} = [C]{ε} is a square symmetric ma-
trix with three rows and three columns:
    

σ 
xx  1 ν 0 ε 
 
xx 
E

 
  


 
   

{σ} = σyy  =
 
ν 1

εyy 

0


2
(1 − ν )
 
 
1
−

σ 0 0 (1 ν) γ

 
  


xy xy
 
 



2

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 33 of 37

1.13 Conclusions
• An important property of the stiffness matrix for finite
element analysis (FEA) for stress analysis is that the ma-
trix is symmetric. This requires the elastic matrix [C]
relating independent stress and strain components for
linear elastic materials to be symmetric. The question
then arises: Why is the Elastic Matrix Symmetric?
• The answer lies in the 4th-order elastic tensor trans-
forming components of the strain tensor to components
of the stress tensor.
• Using Voigt notation, engineers map the 4th-order stiff-
ness tensor Cijkl to a 6 × 6 matrix with mapped indices
I and J ranging over the six independent terms in the
strain and stiffness tensors. This matrix representation
of the elastic tensor simplifies the complexity of calcu-
lation while preserving its full physical meaning.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 34 of 37

• The matrix form {σ} = [C]{ε} of the stiffness tensor con-
stitutive equations σij = Cijkl εkl allows for standard ma-
trix algebra operations for linear transformations and
matrix implementation in computational methods such
as the finite element method (FEM).
• Due to both the major symmetry and minor symmetry
due to the symmetry of stress and strain tensors, for
the elastic tensor Cijkl , the stiffness matrix [C]T = [C] is
symmetric, such that the transpose operator does not
change the matrix and the rows and columns can be
interchanged with no change CJI = CIJ .

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 35 of 37

 References
Here are some references (there are many others) that
cover the minor and major symmetries of the 4th-order
stiffness tensor Cijkl as well as material symmetries such
as orthotropic, transversely isotropic, cubic, and isotropic
symmetries.
These books also discuss the constraints on symmetry
reflections and rotations to reduce the number of inde-
pendent moduli:

1. Elasticity: Theory, Applications, and Numerics, by Mar-

tin H. Sadd, 2014, Academic Press, Elsevier.
2. Anisotropic Elasticity: Theory and Applications, by Thomas
C. T. Ting, 1996, Oxford University Press.
3. Theory of Elasticity, by S. P. Timoshenko and J. N. Good-
ier, 1970, McGraw-Hill.

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 36 of 37

I will leave you with a final question
Physically, why do isotropic materials reduce to two
independent material constants, not just one?

LinkedIn � - Dr. Lonny Thompson, Clemson University, Nov. 11, 2024. 37 of 37