Chapter 4

GENERAL HOOKE’S LAW

 Contents

GENERAL HOOKE’S LAW ....................................................................................................... 89

4.1 Introduction ................................................................................................................................. 89

4.2 First Law of Thermodynamics, Internal Energy Density and Complementary Internal Energy
Density .................................................................................................................................................... 89
4.2.1 Linear Strain Elasticity and Internal-Energy Density ......................................................... 91
4.2.2 Elasticity and Complementary Internal-Energy Density .................................................... 92
4.3 Anisotropic Elasticity.................................................................................................................. 94
4.4 Isotropic Elasticity ...................................................................................................................... 95
4.4.1 Isotropic and Homogeneous Materials................................................................................ 95
4.4.2 Strain-Energy Density of Isotropic Elastic Materials ......................................................... 96
4.5 Equations of Thermo-elasticity for Isotropic Materials .............................................................. 99
Solved Examples................................................................................................................................... 101
Review Questions ................................................................................................................................. 106
Exercise ................................................................................................................................................. 107
Answers ................................................................................................................................................ 109

88
 CHAPTER 4
GENERAL HOOKE’S LAW

4.1 Introduction
In last two chapters, we presented separate theories for stress and strain. These theories are based
on the concept of a general continuum. Consequently, they are applicable to all continua. In
particular, the theory of stress is based solely on the concept of force and the associated concept
of force per unit area. Similarly, the theory of strain is based on geometrical concepts of
infinitesimal line extensions and rotations between two infinitesimal lines. However, to relate the
stress at a point in a material to the corresponding strain at that point, knowledge of material
properties is required. These properties enter into the stress-strain-temperature relations as material
coefficients. The theoretical basis for these relations is the first law of thermodynamics, but the
material properties themselves must be determined experimentally.

In this chapter, we employ the first law of thermodynamics to derive linear stress-strain-
temperature relations. In addition, certain concepts, such as complementary strain energy, that
have application to nonlinear problems are introduced. These relations and concepts are utilized
in many applications of solid mechanics.

4.2 First Law of Thermodynamics, Internal Energy Density and

Complementary Internal Energy Density
The derivation of load-stress and load-deflection relations requires stress-strain relations that relate
the components of the strain tensor to components of the stress tensor. The form of the stress-strain
relations depends on material behavior. In most of the application, we treat mainly materials that
are isotropic; that is, at any point they have the same properties in all directions.

Stress-strain relations may be derived with the first law of thermodynamics, a precise statement of
the law of conservation of energy. The total amount of internal energy in a system is generally
indeterminate. Hence, only changes of internal energy are measurable. If electromagnetic effects
are disregarded, this law is described as follows:
The work performed on a mechanical system by external forces plus the heat that flows into the
system from the outside equals the increase in internal energy plus the increase in kinetic energy.

Symbolically, the first law of thermodynamics is expressed by the equation

𝛿𝑊 + 𝛿𝐻 = 𝛿𝑈 + 𝛿𝐾 ….. (4.1)
where 𝛿𝑊 is the work performed on the system by external forces, 𝛿𝐻 is the heat that flows into
the system, 𝛿𝑈 is the increase in internal energy, and 𝛿𝐾 is the increase in kinetic energy.

To apply the first law of thermodynamics, we consider a loaded member in equilibrium. The
deflections are assumed to be known. They are specified by known displacement components

89
(𝑢, 𝑣, 𝑤) for each point in the deflected member. We allow each point to undergo infinitesimal
increments (variations) in the displacement components (𝑢, 𝑣, 𝑤) indicated by (𝛿𝑢, 𝛿𝑣, 𝛿𝑤). The
stress components at every point of the member are considered to be unchanged under variations
of the displacements. These displacement variations are arbitrary, except that two or more particles
cannot occupy the same point in space, nor can a single particle occupy more than one position
(the member does not tear). In addition, displacements of certain points in the member may be
specified (e.g., at a fixed support); such specified displacements are referred to as forced boundary
conditions. By Equation (3.4), the variations of the strain components resulting from variations
(𝛿𝑢, 𝛿𝑣, 𝛿𝑤) are
𝜕𝛿𝑢 𝜕𝛿𝑣 𝜕𝛿𝑤
𝛿𝜀𝑥 = ; 𝛿𝜀𝑦 = ; 𝛿𝜀𝑧 =
𝜕𝑥 𝜕𝑦 𝜕𝑧
….. (4.2)
𝜕𝛿𝑣 𝜕𝛿𝑢 𝜕𝛿𝑣 𝜕𝛿𝑤 𝜕𝛿𝑢 𝜕𝛿𝑤
𝛿𝛾𝑥𝑦 = + ; 𝛿𝛾𝑦𝑧 = + ; 𝛿𝛾𝑥𝑧 = +
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑦 𝜕𝑧 𝜕𝑥

To introduce force quantities, consider an arbitrary volume 𝑉 of the deformed member enclosed
by a closed surface 𝑆. We assume that the member is in static equilibrium following the
displacement variations (𝛿𝑢, 𝛿𝑣, 𝛿𝑤). Therefore, the part of the member considered in volume 𝑉
is in equilibrium under the action of surface forces (represented by stress distributions on surface
𝑆) and body forces (represented by distributions of body forces per unit volume 𝐵𝑥 , 𝐵𝑦 and 𝐵𝑧 in
volume 𝑉).

For adiabatic conditions (no net heat flow into 𝛿𝐻 = 0) and static equilibrium (𝛿𝐾 = 0), the first
law of thermodynamics states that, during the displacement variations (𝛿𝑢, 𝛿𝑣, 𝛿𝑤), the variation
in work of the external forces 𝛿𝑊 is equal to the variation of internal energy 𝛿𝑈 for each volume
element. Hence, for 𝑉 we have
𝛿𝑊 = 𝛿𝑈 ….. (4.3)

It is convenient to divide 𝛿𝑊 into two parts: the work of the surface forces 𝛿𝑊𝑆 and the work of
the body forces 𝛿𝑊𝐵 . At point 𝑃 of surface 𝑆, consider an increment of area 𝑑𝑆. The stress vector
𝝈𝑷 acting on 𝑑𝑆 has components 𝜎𝑃𝑥 , 𝜎𝑃𝑦 and 𝜎𝑃𝑧 . The surface force is equal to the product of
these stress components and 𝑑𝑆. The work is equal to the sum of the work of these forces over the
surface 𝑆. Thus,

𝛿𝑊𝑆 = ∫ 𝜎𝑃𝑥 𝛿𝑢𝑑𝑆 + ∫ 𝜎𝑃𝑦 𝛿𝑣𝑑𝑆 + ∫ 𝜎𝑃𝑧 𝛿𝑤𝑑𝑆

𝑆 𝑆 𝑆

= ∫ (𝜎𝑥𝑥 𝑙 + 𝜎𝑦𝑥 𝑚 + 𝜎𝑧𝑥 𝑛)𝛿𝑢𝑑𝑆 + ∫ (𝜎𝑥𝑦 𝑙 + 𝜎𝑦𝑦 𝑚 + 𝜎𝑧𝑦 𝑛)𝛿𝑣𝑑𝑆

𝑆 𝑆

+ ∫ (𝜎𝑥𝑧 𝑙 + 𝜎𝑦𝑧 𝑚 + 𝜎𝑧𝑧 𝑛)𝛿𝑤𝑑𝑆

𝑆 ….. (4.4)

90
For a volume element 𝑑𝑉 in volume 𝑉, the body forces are given by products of 𝑑𝑉 and the body
force components per unit volume (𝐵𝑥 , 𝐵𝑦 , 𝐵𝑧 ). The work 𝛿𝑊𝐵 , of the body forces that act
throughout 𝑉 is

𝛿𝑊𝐵 = ∫ (𝐵𝑥 𝛿𝑢 + 𝐵𝑦 𝛿𝑣 + 𝐵𝑧 𝛿𝑤)𝑑𝑉 ….. (4.5)

𝑉

The variation of work 𝛿𝑊 of the external forces that act on volume 𝑉 with surface 𝑆 is equal to
the sum of 𝛿𝑊𝑆 and 𝛿𝑊𝐵 . The surface integral in Equation (4.4) may be converted into a volume
integral by use of the divergence theorem. Thus,
𝛿𝑊 = 𝛿𝑊𝑆 + 𝛿𝑊𝐵
𝜕 𝜕
=∫ [ (𝜎𝑥𝑥 𝛿𝑢 + 𝜎𝑥𝑦 𝛿𝑣 + 𝜎𝑥𝑧 𝛿𝑤) + (𝜎 𝛿𝑢 + 𝜎𝑦𝑦 𝛿𝑣 + 𝜎𝑦𝑧 𝛿𝑤)
𝑉 𝜕𝑥 𝜕𝑦 𝑦𝑥
𝜕
+ (𝜎𝑧𝑥 𝛿𝑢 + 𝜎𝑧𝑦 𝛿𝑣 + 𝜎𝑧𝑧 𝛿𝑤)
𝜕𝑧
+ (𝐵𝑥 𝛿𝑢 + 𝐵𝑦 𝛿𝑣 + 𝐵𝑧 𝛿𝑤)] 𝑑𝑉 ….. (4.6)

Using Equation (4.2) with the differential equation of equilibrium, Equation (4.6) reduces to

𝛿𝑊 = ∫ (𝜎𝑥𝑥 𝛿𝜀𝑥 + 𝜎𝑦 𝛿𝜀𝑦 + 𝜎𝑧 𝛿𝜀𝑧 + 2𝜎𝑥𝑦 𝛿𝜀𝑥𝑦 + 2𝜎𝑦𝑧 𝛿𝜀𝑦𝑧 + 2𝜎𝑧𝑥 𝛿𝜀𝑧𝑥 )𝑑𝑉 ….. (4.7)
𝑉

The internal energy 𝑈 for volume 𝑉 is expressed in terms of the internal energy per unit volume,
that is, in terms of the internal-energy density 𝑈0 . Thus,

𝑈 = ∫ 𝑈0 𝑑𝑉 ….. (4.8)
𝑉

and the variation of internal energy becomes

𝛿𝑈 = ∫ 𝛿𝑈0 𝑑𝑉 ….. (4.9)

𝑉

Substitution of Equations (4.7) and (4.9) into Equation (4.3) gives the variation of the internal-
energy density 𝛿𝑈0 in terms of the stress components and the variation in strain components. Thus,
𝛿𝑈0 = 𝜎𝑥 𝛿𝜀𝑥 + 𝜎𝑦 𝛿𝜀𝑦 + 𝜎𝑧 𝛿𝜀𝑧 + 2𝜎𝑥𝑦 𝛿𝜀𝑥𝑦 + 2𝜎𝑦𝑧 𝛿𝜀𝑦𝑧 + 2𝜎𝑧𝑥 𝛿𝜀𝑧𝑥 ….. (4.10)

4.2.1 Linear Strain Elasticity and Internal-Energy Density

The strain-energy density 𝑈0 is a function of certain variables; we need to determine these
variables. For elastic material behavior, the total internal energy 𝑈 in a loaded member is equal to
the potential energy of the internal forces (called the elastic strain energy). Each stress component
is related to the strain components; therefore, the internal-energy density 𝑈0 at a given point in the
member can be expressed in terms of the six components of the strain tensor. If the material is

91
nonhomogeneous (has different properties at different points in the member), the function 𝑈0
depends on location (𝑥, 𝑦, 𝑧) in the member as well. The strain-energy density 𝑈0 also depends on
the temperature 𝑇.

Since the strain-energy density function 𝑈0 generally depends on the strain components, the
coordinates, and the temperature, we may express it as function of these variables. Thus,
𝑈0 = 𝑈(𝜀𝑥 , 𝜀𝑦 , 𝜀𝑧 , 𝜀𝑥𝑦 , 𝜀𝑦𝑧 , 𝜀𝑧𝑥 , 𝑥, 𝑦, 𝑧, 𝑇) ….. (4.11)

Then, if the displacements (𝑢, 𝑣, 𝑤) undergo a variation (𝛿𝑢, 𝛿𝑣, 𝛿𝑤), the strain components take
variations 𝛿𝜀𝑥 , 𝛿𝜀𝑦 , 𝛿𝜀𝑧 , 𝛿𝜀𝑥𝑦 , 𝛿𝜀𝑦𝑧 and 𝛿𝜀𝑧𝑥 and the function 𝑈0 takes on the variation
𝜕𝑈0 𝜕𝑈0 𝜕𝑈0 𝜕𝑈0 𝜕𝑈0 𝜕𝑈0
𝛿𝑈0 = 𝛿𝜀𝑥 + 𝛿𝜀𝑦 + 𝛿𝜀𝑧 + 𝛿𝜀𝑥𝑦 + 𝛿𝜀𝑦𝑧 + 𝛿𝜀 ….. (4.12)
𝜕𝜀𝑥 𝜕𝜀𝑦 𝜕𝜀𝑧 𝜕𝜀𝑥𝑦 𝜕𝜀𝑦𝑧 𝜕𝜀𝑧𝑥 𝑧𝑥

Therefore, since Equations (4.10) and (4.12) are valid for arbitrary variations (𝛿𝑢, 𝛿𝑣, 𝛿𝑤),
comparison yields for rectangular coordinate axes (𝑥, 𝑦, 𝑧)
𝜕𝑈0 𝜕𝑈0 𝜕𝑈0
𝜎𝑥 = ; 𝜎𝑦 = ; 𝜎𝑧 =
𝜕𝜀𝑥 𝜕𝜀𝑦 𝜕𝜀𝑧
….. (4.13)
1 𝜕𝑈0 1 𝜕𝑈0 1 𝜕𝑈0
𝜎𝑥𝑦 = ; 𝜎𝑦𝑧 = ; 𝜎𝑧𝑥 =
2 𝜕𝜀𝑥𝑦 2 𝜕𝜀𝑦𝑧 2 𝜕𝜀𝑧𝑥

4.2.2 Elasticity and Complementary Internal-Energy Density

In many members of engineering structures, there may be one dominant component of the stress
tensor; call it 𝜎. This situation may arise in axially loaded members, simple columns, beams, or
torsional members. Then the strain-energy density 𝑈0 [Equation (4.11)] depends mainly on the
associated strain component 𝜀; consequently, for a given temperature 𝑇, 𝜎 depends mainly on 𝜀.

By Equation (4.13), 𝜎 = 𝑑𝑈0 /𝑑𝜀 and, therefore, 𝑈0 = ∫ 𝜎𝑑𝜀. It follows that 𝑈0 is represented by
the area under the stress-strain diagram (Figure 4.1). The rectangular area (0,0), (0, 𝜀), (𝜎, 𝜀),
(𝜎, 0) is represented by the product 𝜎𝜀. Hence, this area is given by
𝜎𝜀 = 𝑈0 + 𝐶0 ….. (4.14)
where 𝐶0 is called the complementary internal-energy density or complementary strain energy
density. 𝐶0 is represented by the area above the stress-strain curve and below the horizontal line
from (𝜎, 0), (𝜎, 𝜀). Hence, by Figure 4.1,

𝐶0 = ∫ 𝜀𝑑𝜎 ….. (4.15)

or,
𝑑𝐶0
𝜀= ….. (4.16)
𝑑𝜎

92
 Figure 4.1: Strain Energy Densities

This graphical interpretation of the complementary strain energy is applicable only for the case of
a single nonzero component of stress. However, it can be generalized for several nonzero
components of stress as follows. We assume that Equation (4.13) may be integrated to obtain the
strain components as functions of the stress components. Thus, we obtain
𝜀𝑥 = 𝑓1 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
𝜀𝑦 = 𝑓2 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
𝜀𝑧 = 𝑓3 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
….. (4.17)
𝜀𝑥𝑦 = 𝑓4 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
𝜀𝑦𝑧 = 𝑓5 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
𝜀𝑧𝑥 = 𝑓6 (𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 )
where 𝑓1 , 𝑓2 , …. 𝑓6 denote functions of the stress components. Substitution of Equation (4.17)
into Equation (4.11) yields 𝑈0 as a function of the six stress components. Then direct extension of
Equation (4.14) yields
𝐶0 = −𝑈0 + 𝜎𝑥 𝜀𝑥 + 𝜎𝑦 𝜀𝑦 + 𝜎𝑧 𝜀𝑧 + 2𝜎𝑥𝑦 𝜀𝑥𝑦 + 2𝜎𝑦𝑧 𝜀𝑦𝑧 + 2𝜎𝑧𝑥 𝜀𝑧𝑥 ….. (4.18)

By Equations (4.17) and (4.18), the complementary energy density 𝐶0 may be expressed in terms
of the six stress components. Hence, differentiating Equation (4.18) with respect to 𝜎𝑥 , noting by
the chain rule of differentiation that
𝜕𝑈0 𝜕𝑈0 𝜕𝜀𝑥 𝜕𝑈0 𝜕𝜀𝑦 𝜕𝑈0 𝜕𝜀𝑧 𝜕𝑈0 𝜕𝜀𝑥𝑦 𝜕𝑈0 𝜕𝜀𝑦𝑧 𝜕𝑈0 𝜕𝜀𝑧𝑥
= + + + + + ….. (4.19)
𝜕𝜎𝑥 𝜕𝜀𝑥 𝜕𝜎𝑥 𝜕𝜀𝑦 𝜕𝜎𝑥 𝜕𝜀𝑧 𝜕𝜎𝑥 𝜕𝜀𝑥𝑦 𝜕𝜎𝑥 𝜕𝜀𝑦𝑧 𝜕𝜎𝑥 𝜕𝜀𝑧𝑥 𝜕𝜎𝑥
and employing Equation (4.13), we find
𝜕𝐶0
𝜀𝑥 = ….. (4.20)
𝜕𝜎𝑥

Similarly, taking derivatives of Equation (4.18) with respect to the other stress components
(𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 ), we obtain the generalization of Equation (4.16):

93
 𝜕𝐶0 𝜕𝐶0 𝜕𝐶0
𝜀𝑥 = ; 𝜀𝑦 = ; 𝜀𝑧 =
𝜕𝜎𝑥 𝜕𝜎𝑦 𝜕𝜎𝑧
….. (4.21)
1 𝜕𝐶0 1 𝜕𝐶0 1 𝜕𝐶0
𝜀𝑥𝑦 = ; 𝜀𝑦𝑧 = ; 𝜀𝑧𝑥 =
2 𝜕𝜎𝑥𝑦 2 𝜕𝜎𝑦𝑧 2 𝜕𝜎𝑧𝑥

Because of their relationship to Equation (4.13), Equation (4.21) are said to be conjugate to
Equation (4.13). Equation (4.21) are known also as the Legendre transform of Equation (4.13).

4.3 Anisotropic Elasticity

In the one-dimensional case, for a linear elastic material the stress 𝜎 is proportional to the strain 𝜀;
that is, 𝜎 = 𝐸𝜀 where the proportionality factor 𝐸 is called the modulus of elasticity. The modulus
of elasticity is a property of the material. Thus, for the one-dimensional case, only one material
property is required to relate stress and strain for linear elastic behavior. The relation 𝜎 = 𝐸𝜀 is
known as Hooke’s law. More generally, in the three-dimensional case, Hooke’s law asserts that
each of the stress components is a linear function of the components of the strain tensor; that is
(with 𝛾𝑥𝑦 , 𝛾𝑦𝑧 , 𝛾𝑧𝑥 ),
𝜎𝑥 = 𝐶11 𝜀𝑥 + 𝐶12 𝜀𝑦 + 𝐶13 𝜀𝑧 + 𝐶14 𝛾𝑥𝑦 + 𝐶15 𝛾𝑦𝑧 + 𝐶16 𝛾𝑧𝑥
𝜎𝑦 = 𝐶21 𝜀𝑥 + 𝐶22 𝜀𝑦 + 𝐶23 𝜀𝑧 + 𝐶24 𝛾𝑥𝑦 + 𝐶25 𝛾𝑦𝑧 + 𝐶26 𝛾𝑧𝑥
𝜎𝑧 = 𝐶31 𝜀𝑥 + 𝐶32 𝜀𝑦 + 𝐶33 𝜀𝑧 + 𝐶34 𝛾𝑥𝑦 + 𝐶35 𝛾𝑦𝑧 + 𝐶36 𝛾𝑧𝑥
….. (4.22)
𝜎𝑥𝑦 = 𝐶41 𝜀𝑥 + 𝐶42 𝜀𝑦 + 𝐶43 𝜀𝑧 + 𝐶44 𝛾𝑥𝑦 + 𝐶45 𝛾𝑦𝑧 + 𝐶46 𝛾𝑧𝑥
𝜎𝑦𝑧 = 𝐶51 𝜀𝑥 + 𝐶52 𝜀𝑦 + 𝐶53 𝜀𝑧 + 𝐶54 𝛾𝑥𝑦 + 𝐶55 𝛾𝑦𝑧 + 𝐶56 𝛾𝑧𝑥
𝜎𝑧𝑥 = 𝐶61 𝜀𝑥 + 𝐶62 𝜀𝑦 + 𝐶63 𝜀𝑧 + 𝐶64 𝛾𝑥𝑦 + 𝐶65 𝛾𝑦𝑧 + 𝐶66 𝛾𝑧𝑥

where the 36 coefficients, 𝐶11 , 𝐶12 ..., 𝐶66 , are called elastic coefficients. Materials that exhibit
such stress-strain relations involving a number of independent elastic coefficients are said to be
anisotropic.

In reality, Equation (4.22) is not a law but merely an assumption that is reasonably accurate for
many materials subjected to small strains. For a given temperature, time, and location in the body,
the coefficients 𝐶𝑖𝑗 are constants that are characteristics of the material. Equations (4.13) and (4.22)
yield
𝜕𝑈0
= 𝜎𝑥 = 𝐶11 𝜀𝑥 + 𝐶12 𝜀𝑦 + 𝐶13 𝜀𝑧 + 𝐶14 𝛾𝑥𝑦 + 𝐶15 𝛾𝑦𝑧 + 𝐶16 𝛾𝑧𝑥
𝜕𝜀𝑥
𝜕𝑈0
= 𝜎𝑦 = 𝐶21 𝜀𝑥 + 𝐶22 𝜀𝑦 + 𝐶23 𝜀𝑧 + 𝐶24 𝛾𝑥𝑦 + 𝐶25 𝛾𝑦𝑧 + 𝐶26 𝛾𝑧𝑥 ….. (4.23)
𝜕𝜀𝑦
𝜕𝑈0
= 𝜎𝑧 = 𝐶31 𝜀𝑥 + 𝐶32 𝜀𝑦 + 𝐶33 𝜀𝑧 + 𝐶34 𝛾𝑥𝑦 + 𝐶35 𝛾𝑦𝑧 + 𝐶36 𝛾𝑧𝑥
𝜕𝜀𝑧

94
 𝜕𝑈0
= 𝜎𝑥𝑦 = 𝐶41 𝜀𝑥 + 𝐶42 𝜀𝑦 + 𝐶43 𝜀𝑧 + 𝐶44 𝛾𝑥𝑦 + 𝐶45 𝛾𝑦𝑧 + 𝐶46 𝛾𝑧𝑥
𝜕𝛾𝑥𝑦
𝜕𝑈0
= 𝜎𝑦𝑧 = 𝐶51 𝜀𝑥 + 𝐶52 𝜀𝑦 + 𝐶53 𝜀𝑧 + 𝐶54 𝛾𝑥𝑦 + 𝐶55 𝛾𝑦𝑧 + 𝐶56 𝛾𝑧𝑥
𝜕𝛾𝑦𝑧
𝜕𝑈0
= 𝜎𝑧𝑥 = 𝐶61 𝜀𝑥 + 𝐶62 𝜀𝑦 + 𝐶63 𝜀𝑧 + 𝐶64 𝛾𝑥𝑦 + 𝐶65 𝛾𝑦𝑧 + 𝐶66 𝛾𝑧𝑥
𝜕𝛾𝑧𝑥

Hence, the appropriate differentiations of Equations (4.23) yield

𝜕 2 𝑈0
= 𝐶12 = 𝐶21
𝜕𝜀𝑥 𝜕𝜀𝑦
𝜕 2 𝑈0
= 𝐶13 = 𝐶31
𝜕𝜀𝑥 𝜕𝜀𝑧
……
….. (4.24)
……
𝜕 2 𝑈0
= 𝐶45 = 𝐶54
𝜕𝛾𝑥𝑦 𝜕𝛾𝑦𝑧
𝜕 2 𝑈0
= 𝐶46 = 𝐶64
𝜕𝛾𝑥𝑦 𝜕𝛾𝑧𝑥

These equations show that the elastic coefficients 𝐶𝑖𝑗 = 𝐶𝑗𝑖 are symmetrical in the subscripts 𝑖, 𝑗.
Therefore, there are only 21 distinct 𝐶′𝑠. In other words, the general anisotropic linear elastic
material has 21 elastic coefficients. In view of the preceding relation, the strain-energy density of
a general anisotropic material is by integration of Equation (4.23)
1 1 1
𝑈0 = 𝐶11 𝜀𝑥 2 + 𝐶12 𝜀𝑥 𝜀𝑦 + ⋯ … + 𝐶16 𝜀𝑥 𝛾𝑧𝑥
2 2 2
1 1 1
+ 𝐶12 𝜀𝑥 𝜀𝑦 + 𝐶22 𝜀𝑦 2 + ⋯ … + 𝐶26 𝜀𝑦 𝛾𝑧𝑥
2 2 2 ….. (4.25)
+⋯…
1 1 1
+ 𝐶16 𝜀𝑥 𝛾𝑧𝑥 + 𝐶26 𝜀𝑦 𝛾𝑧𝑥 + ⋯ … + 𝐶66 𝛾𝑧𝑥 2
2 2 2

4.4 Isotropic Elasticity

4.4.1 Isotropic and Homogeneous Materials

If the constituents of the material of a solid member are distributed sufficiently randomly, any part
of the member will display essentially the same material properties in all directions. If a solid
member is composed of such randomly oriented constituents, it is said to be isotropic. Accordingly,

95
if a material is isotropic, its physical properties at a point are invariant under a rotation of axes. A
material is said to be elastically isotropic if its characteristic elastic coefficients are invariant under
any rotation of coordinates.

If the material properties are identical for every point in a member, the member is said to be
homogeneous. In other words, homogeneity implies that the physical properties of a member are
invariant under a translation. Alternatively, a member whose material properties change from point
to point is said to be nonhomogeneous.

If an elastic member is composed of isotropic materials, the strain-energy density depends only on
the principal strains, since for isotropic materials the elastic coefficients are invariant under
arbitrary rotations.

4.4.2 Strain-Energy Density of Isotropic Elastic Materials

The strain-energy density of an elastic isotropic material depends only on the principal strains (𝜀1 ,
𝜀2 , 𝜀3 ). Accordingly, if the elasticity is linear, Equation (4.25) yields
1 1 1 1 1 1
𝑈0 = 𝐶11 𝜀1 2 + 𝐶12 𝜀1 𝜀2 + 𝐶13 𝜀1 𝜀3 + 𝐶12 𝜀1 𝜀2 + 𝐶22 𝜀2 2 + 𝐶23 𝜀2 𝜀3
2 2 2 2 2 2 ….. (4.26)
1 1 1
+ 𝐶13 𝜀1 𝜀3 + 𝐶23 𝜀2 𝜀3 + 𝐶33 𝜀3 2
2 2 2

By symmetry, the naming of the principal axes is arbitrary. Hence, 𝐶11 = 𝐶22 = 𝐶11 = 𝐶1 , and
𝐶12 = 𝐶23 = 𝐶31 = 𝐶2 . Consequently, Equation (4.26) contains only two distinct coefficients. For
linear elastic isotropic materials, the strain-energy density may be expressed in the form
1
𝑈0 = 𝜆(𝜀1 + 𝜀2 + 𝜀3 )2 + 𝐺(𝜀1 2 + 𝜀2 2 + 𝜀3 2 ) ….. (4.27)
2
where 𝜆 = 𝐶2 and 𝐺 = (𝐶1 − 𝐶2 )/2 are elastic coefficients called Lame’s elastic coefficients. If
the material is homogeneous and temperature is constant everywhere, 𝜆 and 𝐺 are constants at all
points. In terms of the strain invariants Equation (4.27) may be written in the following form:
1 2
𝑈0 = ( 𝜆 + 𝐺) 𝐼1̅ − 2𝐺𝐼2̅ ….. (4.28)
2

Returning to orthogonal curvilinear coordinates (𝑥, 𝑦, 𝑧) and introducing the general definitions of
𝐼1̅ and 𝐼2̅ , we obtain
1 2
𝑈0 = 𝜆(𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 ) + 𝐺(𝜀𝑥 2 + 𝜀𝑦 2 + 𝜀𝑧 2 + 2𝜀𝑥𝑦 2 + 2𝜀𝑦𝑧 2 + 2𝜀𝑧𝑥 2 ) ….. (4.29)
2
where (𝜀𝑥 , 𝜀𝑦 , 𝜀𝑧 , 𝜀𝑥𝑦 , 𝜀𝑦𝑧 , 𝜀𝑧𝑥 ) are strain components relative to orthogonal coordinates (𝑥, 𝑦, 𝑧).
Equations (4.13) and (4.29) now yield Hooke’s law for a linear elastic isotropic material in the
form (for orthogonal curvilinear coordinates 𝑥, 𝑦, 𝑧)
𝜎𝑥 = 𝜆𝑒 + 2𝐺𝜀𝑥 ; 𝜎𝑦 = 𝜆𝑒 + 2𝐺𝜀𝑦 ; 𝜎𝑧 = 𝜆𝑒 + 2𝐺𝜀𝑧
….. (4.30)

96
 𝜎𝑥𝑦 = 2𝐺𝜀𝑥𝑦 ; 𝜎𝑦𝑧 = 2𝐺𝜀𝑦𝑧 ; 𝜎𝑧𝑥 = 2𝐺𝜀𝑧𝑥
where 𝑒 = 𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 = 𝐼1̅ , is the classical small-displacement volumetric strain (also called
cubical strain). Thus, we have shown that for isotropic linear elastic materials, the stress-strain
relations involve only two elastic constants.

By means of Equation (4.30), we find

𝐼1 = (3𝜆 + 2𝐺)𝐼1̅
2
𝐼2 = 𝜆(3𝜆 + 4𝐺)𝐼1̅ + 4𝐺 2 𝐼2̅ ….. (4.31)
3
𝐼3 = 𝜆2 (𝜆 + 2𝐺)𝐼1̅ + 4𝜆𝐺 2 𝐼1̅ 𝐼2̅ + 8𝐺 3 𝐼3̅
which relate the stress invariants 𝐼1 , 𝐼2 , 𝐼3 to the strain invariants 𝐼1̅ , 𝐼2̅ , 𝐼3̅ .

Inverting Equation (4.30), we obtain

1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 − 𝜈𝜎𝑧 )
𝐸 𝑥
1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 − 𝜈𝜎𝑧 )
𝐸
1
𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 + 𝜎𝑧 )
𝐸
….. (4.32)
1 1+𝜈
𝜀𝑥𝑦 = 𝜎𝑥𝑦 = 𝜎𝑥𝑦
2𝐺 𝐸
1 1+𝜈
𝜀𝑦𝑧 = 𝜎𝑦𝑧 = 𝜎𝑦𝑧
2𝐺 𝐸
1 1+𝜈
𝜀𝑧𝑥 = 𝜎𝑧𝑥 = 𝜎𝑧𝑥
2𝐺 𝐸
where
𝐺(3𝜆 + 2𝐺) 𝜆
𝐸= ; 𝜈= ….. (4.33)
𝜆+𝐺 2(𝜆 + 𝐺)
are elastic coefficients called Young’s modulus and Poisson’s ratio, respectively. Also, inverting
Equation (4.33), we obtain the Lame coefficients 𝜆 and 𝐺 in terms of 𝐸 and 𝜈 as
𝜈𝐸 3𝜈𝐾 𝐸
𝜆= = ; 𝐺= ….. (4.34)
(1 + 𝜈)(1 − 2𝜈) (1 + 𝜈) 2(1 + 𝜈)
where
𝐸
𝐾= ….. (4.35)
3(1 − 2𝜈)
is the bulk modulus. The bulk modulus relates the mean stress 𝜎𝑚 = 𝐼1 /3 to the volumetric strain
𝑒 by 𝜎𝑚 = 𝐾𝑒.

97
Alternatively, Equation (4.30) may be written in terms of 𝐸 and 𝜈 as follows:
𝐸
𝜎𝑥 = [(1 − 𝜈)𝜀𝑥 + 𝜈(𝜀𝑦 + 𝜀𝑧 )]
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎𝑦 = [(1 − 𝜈)𝜀𝑦 + 𝜈(𝜀𝑧 + 𝜀𝑥 )]
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎𝑧 = [(1 − 𝜈)𝜀𝑧 + 𝜈(𝜀𝑥 + 𝜀𝑦 )]
(1 + 𝜈)(1 − 2𝜈)
….. (4.36)
𝐸
𝜎𝑥𝑦 = 𝜀
(1 + 𝜈) 𝑥𝑦
𝐸
𝜎𝑦𝑧 = 𝜀
(1 + 𝜈) 𝑦𝑧
𝐸
𝜎𝑧𝑥 = 𝜀
(1 + 𝜈) 𝑧𝑥

For the case of plane stress, 𝜎𝑧 = 𝜎𝑦𝑧 = 𝜎𝑧𝑥 = 0, Equations (4.36) reduce to
𝐸
𝜎𝑥 = (𝜀 + 𝜈𝜀𝑦 )
1 − 𝜈2 𝑥
𝐸
𝜎𝑦 = (𝜈𝜀𝑥 + 𝜀𝑦 ) ….. (4.37)
1 − 𝜈2
𝐸
𝜎𝑥𝑦 = 𝜀
1 + 𝜈 𝑥𝑦

For the case of plane strain, 𝜀𝑧 = 𝜀𝑦𝑧 = 𝜀𝑧𝑥 = 0, Equations (4.36) reduce to
𝐸
𝜎𝑥 = [(1 − 𝜈)𝜀𝑥 + 𝜈𝜀𝑦 ]
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎𝑦 = [𝜈𝜀𝑥 + (1 − 𝜈)𝜀𝑦 ]
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎𝑧 = [𝜈(𝜀𝑥 + 𝜀𝑦 )] ….. (4.38)
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎𝑥𝑦 = 𝜀
(1 + 𝜈) 𝑥𝑦
𝜎𝑦𝑧 = 0
𝜎𝑧𝑥 = 0

Substitution of Equation (4.32) into Equation (4.23) yields the strain-energy density 𝑈, in terms of
stress quantities. Thus, we obtain

98
 1
𝑈0 = [𝜎 2 + 𝜎𝑦 2 + 𝜎𝑧 2 − 2𝜈(𝜎𝑥 𝜎𝑦 + 𝜎𝑦 𝜎𝑧 + 𝜎𝑧 𝜎𝑥 ) + 2(1 + 𝜈)(𝜎𝑥𝑦 2 + 𝜎𝑦𝑧 2 + 𝜎𝑧𝑥 2 )]
2𝐸 𝑥
1
= [𝐼1 2 − 2(1 + 𝜈)𝐼2 ] ….. (4.39)
2𝐸

If the (𝑥, 𝑦, 𝑧) axes are directed along the principal axes of strain, then 𝜀𝑥𝑦 = 𝜀𝑦𝑧 = 𝜀𝑧𝑥 = 0.
Hence, by Equation (4.38), 𝜎𝑥𝑦 = 𝜎𝑦𝑧 = 𝜎𝑧𝑥 = 0. Therefore, the (𝑥, 𝑦, 𝑧) axes must also lie along
the principal axes of stress. Consequently, for an isotropic material, the principal axes of stress are
coincident with the principal axes of strain. When we deal with isotropic materials, no distinction
need be made between principal axes of stress and principal axes of strain. Such axes are called
simply principal axes.

4.5 Equations of Thermo-elasticity for Isotropic Materials

Consider an unconstrained member made of an isotropic elastic material in an arbitrary zero
configuration. Let the uniform temperature of the member be increased by a small amount Δ𝑇.
Experimental observation has shown that, for a homogeneous and isotropic material, all
infinitesimal line elements in the volume undergo equal expansions. Furthermore, all line elements
maintain their initial directions. Therefore, the strain components resulting from the temperature
change Δ𝑇 are, with respect to rectangular Cartesian coordinates (𝑥, 𝑦, 𝑧),
𝜀′𝑥 = 𝜀′𝑦 = 𝜀′𝑧 = 𝛼Δ𝑇 ; 𝜀′𝑥𝑦 = 𝜀′𝑦𝑧 = 𝜀′𝑧𝑥 = 0 ….. (4.40)
where 𝛼 denotes the coefficient of thermal expansion of the material.

Now let the member be subjected to forces that induce stresses 𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 , 𝜎𝑥𝑦 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 at point 0
in the member. Accordingly, if 𝜀𝑥 , 𝜀𝑦 , 𝜀𝑧 , 𝜀𝑥𝑦 , 𝜀𝑦𝑧 , 𝜀𝑧𝑥 denote the strain components at point 0 after
the application of the forces, the change in strain produced by the forces is represented by the
equations
𝜀′′𝑥 = 𝜀𝑥 − 𝛼Δ𝑇 ; 𝜀′′𝑦 = 𝜀𝑦 − 𝛼Δ𝑇 ; 𝜀′′𝑧 = 𝜀𝑧 − 𝛼Δ𝑇
….. (4.41)
𝜀′′𝑦𝑥 = 𝜀𝑥𝑦 ; 𝜀′′𝑦𝑧 = 𝜀𝑦𝑧 ; 𝜀′′𝑧𝑥 = 𝜀𝑧𝑥

In general, Δ𝑇 may depend on the location of point 0 and time 𝑡. Hence Δ𝑇 = Δ𝑇(𝑥, 𝑦, 𝑧).
Substitution of by Equation (4.41) into by Equation (4.30) yields
𝜎𝑥 = 𝜆𝑒 + 2𝐺𝜀𝑥 − 𝑐Δ𝑇
𝜎𝑦 = 𝜆𝑒 + 2𝐺𝜀𝑦 − 𝑐Δ𝑇
….. (4.42)
𝜎𝑧 = 𝜆𝑒 + 2𝐺𝜀𝑧 − 𝑐Δ𝑇
𝜎𝑥𝑦 = 2𝐺𝜀𝑥𝑦 ; 𝜎𝑦𝑧 = 2𝐺𝜀𝑦𝑧 ; 𝜎𝑧𝑥 = 2𝐺𝜀𝑧𝑥
where
𝐸𝛼
𝑐 = (3𝜆 + 2𝐺)𝛼 = ….. (4.43)
1 − 2𝜈

99
Similarly, substitution of Equations (4.42) into Equations (4.32) yields
1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 − 𝜈𝜎𝑧 ) + 𝛼Δ𝑇
𝐸 𝑥
1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 − 𝜈𝜎𝑧 ) + 𝛼Δ𝑇
𝐸
1
𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 + 𝜎𝑧 ) + 𝛼Δ𝑇
𝐸
….. (4.44)
1 1+𝜈
𝜀𝑥𝑦 = 𝜎𝑥𝑦 = 𝜎𝑥𝑦
2𝐺 𝐸
1 1+𝜈
𝜀𝑦𝑧 = 𝜎𝑦𝑧 = 𝜎𝑦𝑧
2𝐺 𝐸
1 1+𝜈
𝜀𝑧𝑥 = 𝜎𝑧𝑥 = 𝜎𝑧𝑥
2𝐺 𝐸

Finally, substituting of Equations (4.44) into of Equations (4.28) or (4.29), we find that
1 2 3
𝑈0 = ( 𝜆 + 𝐺) 𝐼1̅ − 2𝐺𝐼2̅ − 𝑐𝐼1̅ Δ𝑇 + 𝑐𝛼(Δ𝑇)2 ….. (4.45)
2 2

In terms of the strain components, we obtain

1 2
𝑈0 = 𝜆(𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 ) + 𝐺(𝜀𝑥 2 + 𝜀𝑦 2 + 𝜀𝑧 2 + 2𝜀𝑥𝑦 2 + 2𝜀𝑦𝑧 2 + 2𝜀𝑧𝑥 2 )
2 ….. (4.46)
3
− 𝑐(𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 )Δ𝑇 + 𝑐𝛼(Δ𝑇)2
2

Equations (4.42) and (4.44) are the basic stress-strain relations of classical thermos-elasticity for
isotropic materials. For temperature changes Δ𝑇, the strain-energy density is modified by a
temperature-dependent term that is proportional to the volumetric strain 𝑒 = 𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 and by
a term proportional to (Δ𝑇)2 [Equations (𝟒. 𝟒𝟓) and (𝟒. 𝟒𝟔)].

We find by Equations (𝟒. 𝟒𝟒) and (𝟒. 𝟒𝟔)

1
𝑈0 = [𝐼1 2 − 2(1 + 𝜈)𝐼2 ] ….. (4.47)
2𝐸
and
1
𝑈0 = [𝜎 2 + 𝜎𝑦 2 + 𝜎𝑧 2 − 2𝜈(𝜎𝑥 𝜎𝑦 + 𝜎𝑦 𝜎𝑧 + 𝜎𝑧 𝜎𝑥 )
2𝐸 𝑥 ….. (4.48)
+ 2(1 + 𝜈)(𝜎𝑥𝑦 2 + 𝜎𝑦𝑧 2 + 𝜎𝑧𝑥 2 )]
in terms of stress components. Equation (4.48) does not contain Δ𝑇 explicitly. However, the
temperature distribution may affect the stresses. Note that Equations (4.47) and (4.48) are identical
to the results in Equation (4.39).

100
Solved Examples
Example 4.1
A wing of an airplane is subjected to a test in bending, and the principal strains are measured
at several points on the wing surface as 𝜺𝟏 = 𝟎. 𝟎𝟎𝟐 and 𝜺𝟐 = −𝟎. 𝟎𝟎𝟐. The wing material is
aluminum alloy 𝟕𝟎𝟕𝟓 𝑻𝟔 (𝑬 = 𝟕𝟐 𝑮𝑷𝒂 and 𝝂 = 𝟎. 𝟑𝟑). Determine the corresponding
principal stresses and the third principal strain.

Solution
The principal stresses can be expressed in terms of the principal strains as
𝐸
𝜎1 = [(1 − 𝜈)𝜀1 + 𝜈(𝜀2 + 𝜀3 )] … (a)
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎3 = [(1 − 𝜈)𝜀2 + 𝜈(𝜀1 + 𝜀3 )] … (b)
(1 + 𝜈)(1 − 2𝜈)
𝐸
𝜎3 = [(1 − 𝜈)𝜀3 + 𝜈(𝜀1 + 𝜀2 )] … (c)
(1 + 𝜈)(1 − 2𝜈)

As we know that, for an element undergoing bending, 𝜎3 = 0. Then Equation (c) gives
𝜈
𝜀3 = − (𝜀 + 𝜀2 ) … (d)
(1 − 𝜈) 1

Substituting 𝜀3 from Equation (d) into Equation (a), we get

𝐸 𝜈2
𝜎1 = [(1 − 𝜈)𝜀1 + 𝜈𝜀2 − (𝜀 + 𝜀2 )]
(1 + 𝜈)(1 − 2𝜈) (1 − 𝜈) 1

𝐸 𝜈2 𝜈2
= [{(1 − 𝜈) − } 𝜀1 + {𝜈 − }𝜀 ]
(1 + 𝜈)(1 − 2𝜈) (1 − 𝜈) (1 − 𝜈) 2

𝐸
∴ 𝜎1 = (𝜀 + 𝜈𝜀2 ) … (e)
(1 − 𝜈 2 ) 1

Substituting 𝜀3 from Equation (d) into Equation (b), we get

𝐸 𝜈2
𝜎2 = [(1 − 𝜈)𝜀2 + 𝜈𝜀1 − (𝜀 + 𝜀2 )]
(1 + 𝜈)(1 − 2𝜈) (1 − 𝜈) 1

𝐸 𝜈2 𝜈2
= [{𝜈 − } 𝜀1 + {(1 − 𝜈) − }𝜀 ]
(1 + 𝜈)(1 − 2𝜈) (1 − 𝜈) (1 − 𝜈) 2

𝐸
∴ 𝜎2 = (𝜈𝜀1 + 𝜀2 ) … (f)
(1 − 𝜈 2 )

101
Substituting 𝐸 = 72 𝐺𝑃𝑎, 𝜈 = 0.33, 𝜀1 = 0.002 and 𝜀2 = −0.002 into Equations (e) and (f)
𝜎1 = 108.27 𝑀𝑃𝑎 and 𝜎2 = −108.27 𝑀𝑃𝑎

Similarly, substituting 𝜀1 = 0.002 and 𝜀2 = −0.002 into Equation (d) and (f)
𝜈 0.33
𝜀3 = − (𝜀1 + 𝜀2 ) = − (0.002 − 0.002) = 0
(1 − 𝜈) (1 − 0.33)

Example 4.2
A square plate in the side of a ship with 𝟖𝟎𝟎 𝒎𝒎 sides parallel to the 𝒙 and 𝒚 axes has a
uniform thickness 𝒉 = 𝟏𝟎 𝒎𝒎 and is made of an isotropic steel (𝑬 = 𝟐𝟎𝟎 𝑮𝑷𝒂 and 𝝂 =
𝟎. 𝟐𝟗). The plate is subjected to a uniform state of stress. If 𝝈𝒛 = 𝝈𝒚𝒛 = 𝝈𝒛𝒙 = 𝟎 (plane stress),
𝝈𝒙 = 𝝈𝟏 = 𝟓𝟎𝟎 𝑴𝑷𝒂, and 𝜺𝒚 = 𝟎 for the plate, determine 𝝈𝒚 = 𝝈𝟐 and the final dimensions
of the plate, assuming linearly elastic conditions.

Solution
The strains can be expressed in terms of the stresses as
1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 ) … (a)
𝐸 𝑥
1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 ) … (b)
𝐸
1
𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 ) … (c)
𝐸

Substituting 𝜀𝑦 = 0 and 𝜎𝑥 = 500 𝑀𝑃𝑎 into Equation (b), we get

−𝜈𝜎𝑥 + 𝜎𝑦 = 0
∴ 𝜎𝑦 = 𝜈𝜎𝑥 = 0.29 × 500 = 145 𝑀𝑃𝑎

Then using Equations (a) and (c), we get

1 1
𝜀𝑥 = (𝜎𝑥 − 𝜈𝜎𝑦 ) = 9
(500 − 0.29 × 145) × 106 = 0.002290
𝐸 200 × 10
and
1 1
∴ 𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 ) = (−0.29 × 500 − 0.29 × 145) × 106 = −0.000935
𝐸 200 × 109

Hence the final dimensions of the plate can be determined as

𝐿′ = (1 + 𝜀𝑥 )𝐿 = (1 + 0.002290) × 800 = 801.83 𝑚𝑚
𝑊′ = (1 + 𝜀𝑦 )𝑊 = (1 + 0) × 800 = 800 𝑚𝑚
𝐻′ = (1 − 𝜀𝑧 )𝐻 = (1 − 0.000935) × 800 = 9.99 𝑚𝑚

102
Example 4.3
The nonzero stress components at a point in a steel plate (𝑬 = 𝟐𝟎𝟎 𝑮𝑷𝒂 and 𝝂 = 𝟎. 𝟐𝟗) are
𝝈𝒙 = 𝟖𝟎 𝑴𝑷𝒂, 𝝈𝒚 = 𝟏𝟐𝟎 𝑴𝑷𝒂, and 𝝈𝒙𝒚 = 𝟓𝟎 𝑴𝑷𝒂. Determine the principal strains.

Solution
The nonzero strain components can be expressed in terms of the stresses as
1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 − 𝜈𝜎𝑧 ) … (a)
𝐸 𝑥
1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 − 𝜈𝜎𝑧 ) … (b)
𝐸
1
𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 + 𝜎𝑧 ) … (c)
𝐸
1 1+𝜈
𝜀𝑥𝑦 = 𝜎𝑥𝑦 = 𝜎𝑥𝑦 … (d)
2𝐺 𝐸

Substituting 𝐸 = 200 𝐺𝑃𝑎, 𝜈 = 0.29, 𝜎𝑥 = 80 𝑀𝑃𝑎, 𝜎𝑦 = 120 𝑀𝑃𝑎, and 𝜎𝑥𝑦 = 50 𝑀𝑃𝑎, we get
1
𝜀𝑥 = (80 − 0.29 × 120 − 0.29 × 0) = 0.000226
200 × 109
1
𝜀𝑦 = (−0.29 × 80 + 120 − 0.29 × 0) = 0.000484
200 × 109
1
𝜀𝑧 = (−0.29 × 80 − 0.29 × 120 + 0) = −0.00029
200 × 109
1 + 0.29
𝜀𝑥𝑦 = × 50 = 0.0003225
200 × 109

Hence, the strain tensor for the given problem can be expressed as
0.000226 0.0003225 0
[𝜀] = [0.0003225 0.000484 0 ]
0 0 −0.00029

Then the characteristic equation for the principal strain can be written as
0.000226 − 𝜀 0.0003225 0
| 0.0003225 0.000484 − 𝜀 0 |=0
0 0 −0.00029 − 𝜀
or, (𝜀 + 0.00029)[(0.000226 − 𝜀)(0.000484 − 𝜀) − (0.0003225)2 ] = 0
or, (𝜀 + 0.00029)(𝜀 2 − 0.000710𝜀 + 2.5284 × 108 ) = 0
or, (𝜀 + 0.00029)(𝜀 − 0.0000077)(𝜀 − 0.000702) = 0
∴ 𝜀1 = −0.00029; 𝜀2 = 0.0000077 and 𝜀2 = 0.000702

103
Example 4.4
An airplane wing spar (Figure E4.4) is made of an aluminum alloy (𝑬 = 𝟕𝟐 𝑮𝑷𝒂 and 𝝂 =
𝟎. 𝟑𝟑), and it has a square cross section perpendicular to the plane of the figure. Stress
components 𝝈𝒙 and 𝝈𝒚 are uniformly distributed as shown.

Figure E4.4

(a) If 𝝈𝒙 = 𝟐𝟎𝟎 𝑴𝑷𝒂, determine the magnitude of 𝝈𝒚 so that the dimension 𝒃 = 𝟐𝟎 𝒎𝒎

does not change under the load.
(b) Determine the amount by which the dimension 𝒂 changes.
(c) Determine the change in the cross-sectional area of the spar.

Solution
(a)
The dimension 𝑏 = 20 𝑚𝑚 does not change under the load means that there is no strain along 𝑦
axis, i.e., 𝜀𝑦 = 0.

The strains can be expressed in terms of the stresses as

1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 ) … (a)
𝐸 𝑥
1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 ) … (b)
𝐸
1
𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 ) … (c)
𝐸

Substituting 𝜀𝑦 = 0, 𝜈 = 0.33 and 𝜎𝑥 = 200 𝑀𝑃𝑎 into Equation (b), we get

−𝜈𝜎𝑥 + 𝜎𝑦 = 0
∴ 𝜎𝑦 = 𝜈𝜎𝑥 = 0.33 × 200 = 66 𝑀𝑃𝑎

(b)
Then using Equations (a) and (c), we get

104
 1 1
𝜀𝑥 = (𝜎𝑥 − 𝜈𝜎𝑦 ) = (200 − 0.33 × 66) × 106 = 0.002475
𝐸 72 × 109
and
1 1
∴ 𝜀𝑧 = (−𝜈𝜎𝑥 − 𝜈𝜎𝑦 ) = (−0.33 × 300 − 0.33 × 66) × 106 = −0.001219
𝐸 72 × 109

The change in dimension 𝑎 can be determined as

∆𝑎 = 𝜀𝑥 𝑎 = 0.002475𝑎

(c)
The change in dimension along 𝑧 axis can be determined as
∆𝑐 = 𝜀𝑧 × 200 = −0.001219 × 20 = −0.2438 𝑚𝑚

The change in cross-sectional area of the spar can be determined as

∆𝐴 = 𝑏 × ∆𝑐 = 200 × (−0.2438) = −48.77 𝑚𝑚2

Example 4.5

A brass sheet 𝟐𝟎 𝒎𝒎 × 𝟑𝟎 𝒎𝒎 × 𝟐 𝒎𝒎 is clamped in a very rigid frame whose coefficient

of thermal expansion is almost zero, Figure E4.5. Given that the temperature drops by
𝟏𝟎𝟎𝟎 𝑪, calculate the resulting stresses in the sheet. For brass, 𝑬 = 𝟏𝟐𝟎 𝑮𝑷𝒂, 𝝂 = 𝟎. 𝟑𝟑 and
𝜶 = 𝟏𝟔 × 𝟏𝟎−𝟔 𝟎 𝑪−𝟏 .

Figure E4.5
Solution
The strain components 𝜀𝑥 and 𝜀𝑦 for a thermos-elasticity problem can be expressed in terms of the
stresses 𝜎𝑥 and 𝜎𝑦 as
1
𝜀𝑥 = (𝜎 − 𝜈𝜎𝑦 ) + 𝛼Δ𝑇 … (a)
𝐸 𝑥

105
 1
𝜀𝑦 = (−𝜈𝜎𝑥 + 𝜎𝑦 ) + 𝛼Δ𝑇 … (b)
𝐸

The given sheet is rigidly fixed, i.e., 𝜀𝑥 = 𝜀𝑦 = 0. Substituting these into Equations (a) and (b),
we get
(𝜎𝑥 − 0.33 × 𝜎𝑦 ) + 120 × 109 × 16 × 10−6 × 200 = 0
∴ 𝜎𝑥 − 0.33 × 𝜎𝑦 + 192 × 106 = 0 … (c)
and
(−0.33 × 𝜎𝑥 + 𝜎𝑦 ) + 120 × 109 × 16 × 10−6 × 200 = 0
∴ −0.33 × 𝜎𝑥 + 𝜎𝑦 + 192 × 106 = 0 … (d)

Solving Equations (c) and (d) for 𝜎𝑥 and 𝜎𝑦 , we get

𝜎𝑥 = −286.57 𝑀𝑃𝑎 and 𝜎𝑦 = −286.57 𝑀𝑃𝑎

Review Questions
1. How can you apply first law of thermodynamics to determine internal energy density?
2. What are the work due to surface forces and body forces?
3. Write down an expression for work in terms of stress and strain components.
4. Write down the expressions for stress components in terms of the strain-energy density
function 𝑈0 .
5. Define complementary internal energy density.
6. Write down the expressions for strain components in terms of complementary internal energy
density 𝐶0 .
7. What do you mean by anisotropic elasticity? Write and expression for the strain-energy density
function 𝑈0 for an anisotropic solid.
8. What is the difference between isotropy and homogeneity?
9. Write an expression for the strain-energy density function 𝑈0 for an isotropic solid in terms of
principal strains.
10. Write an expression for the strain-energy density function 𝑈0 for an isotropic solid in terms of
general strain components.
11. Write down the expressions for strain components in terms of stress components and Lame’s
constants 𝜆 and 𝐺.
12. Write down the expressions for strain components in terms of stress components and elastic
constants 𝐸 and 𝜈.
13. Write down the expressions for stress components in terms of strain components and elastic
constants 𝐸 and 𝜈.
14. Write down the expressions for strain components in terms of stress components and elastic
constants 𝐸 and 𝜈 for a plane stress problem.
15. Write down the expressions for stress components in terms of strain components and elastic
constants 𝐸 and 𝜈 for a plane strain problem.

106
16. Write down the expressions for stress components in terms of strain components and Lame
constants 𝜆 and 𝐺 for a thermo-elasticity problem.
17. Write down the expressions for strain components in terms of stress components and elastic
constants 𝐸 and 𝜈 for a thermo-elasticity problem.
18. Write down the expression for the strain-energy density function 𝑈0 for an isotropic solid in
terms of general strain components for a thermo-elasticity problem.

Exercise
1. Compute Lame’s coefficients 𝜆 and 𝐺 for
(a) steel having 𝐸 = 207 𝐺𝑃𝑎 and 𝜈 = 0.3.
(b) concrete having 𝐸 = 28 𝐺𝑃𝑎 and 𝜈 = 0.2.
2. Table P4.2 lists principal strains that have been calculated for several points in a test of a
machine part made of 𝐴𝐼𝑆𝐼 − 3140 steel (𝐸 = 200 𝐺𝑃𝑎 and 𝜈 = 0.29). Determine the
corresponding principal stresses.
Table P4.2
Strain Point 𝟏 Point 𝟐 Point 𝟑 Point 𝟒 Point 𝟓
𝜺𝟏 0.008 0.006 −0.007 0.004 0.009
𝜺𝟐 −0.002 −0.003 −0.008 −0.005 0.002
𝜺𝟑 0 0 0 0 0
3. A wing of an airplane is subjected to a test in bending, and the principal strains are measured
at several points on the wing surface (Table P4.3). The wing material is aluminum alloy
7075 𝑇6 (𝐸 = 72 𝐺𝑃𝑎 and 𝜈 = 0.33). Determine the corresponding principal stresses and the
third principal strain.
Table P4.3
Strain Point 𝟏 Point 𝟐 Point 𝟑 Point 𝟒
𝜺𝟏 −0.004 0.008 0.006 −0.005
𝜺𝟐 −0.006 0.002 0.002 −0.008
4. A square plate in the side of a ship with 800 𝑚𝑚 sides parallel to the 𝑥 and 𝑦 axes has a
uniform thickness ℎ = 10 𝑚𝑚 and is made of an isotropic steel (𝐸 = 200 𝐺𝑃𝑎 and 𝜈 = 0.29).
The plate is subjected to a uniform state of stress. If 𝜎𝑧 = 𝜎𝑦𝑧 = 𝜎𝑧𝑥 = 0 (plane stress), 𝜎𝑥 =
𝜎1 = 500 𝑀𝑃𝑎, and 𝜀𝑥 = 2𝜀𝑦 for the plate, determine 𝜎𝑦 = 𝜎2 and the final dimensions of the
plate, assuming linearly elastic conditions.
5. A square plate in the side of a ship with 800 𝑚𝑚 sides parallel to the 𝑥 and 𝑦 axes has a
uniform thickness ℎ = 10 𝑚𝑚 and is made of an isotropic steel (𝐸 = 200 𝐺𝑃𝑎 and 𝜈 = 0.29).
The plate is subjected to plane of strain (𝜀𝑧 = 𝜀𝑦𝑧 = 𝜀𝑧𝑥 = 0). If 𝜎𝑥 = 𝜎1 = 500 𝑀𝑃𝑎, 𝜀𝑥 =
2𝜀𝑦 , determine the magnitudes 𝜎𝑦 = 𝜎2 and 𝜎𝑧 = 𝜎3 , assuming linearly elastic conditions.
6. For an isotropic elastic medium subjected to a hydrostatic state of stress, 𝜎𝑥 = 𝜎𝑦 = 𝜎𝑧 = −𝑝,
where 𝑝 denotes pressure. Show that for this state of stress 𝑝 = −𝐾𝑒, where 𝐾 =

107
 𝐸/[3(1 − 2𝜈)] is the bulk modulus and 𝑒 = 𝜀𝑥 + 𝜀𝑦 + 𝜀𝑧 is the classical small-displacement
cubical strain (also called the volumetric strain).
7. A triaxial state of principal stress acts on the faces of a unit cube of soil. Show that these
stresses will not produce a volume change if 𝜈 = 0.5. Assume soil is a linearly elastic isotropic
material. If 𝜈 ≠ 0.5, show that the condition necessary for the volume to remain unchanged is
for 𝜎1 + 𝜎2 + 𝜎3 = 0.
8. An airplane wing is made of an isotropic linearly elastic aluminum alloy (𝐸 = 72 𝐺𝑃𝑎 and
𝜈 = 0.33). Consider a point in the free surface of the wing that is tangent to the (𝑥, 𝑦) plane.
If 𝜎𝑥 = 250 𝑀𝑃𝑎, 𝜎𝑦 = −50 𝑀𝑃𝑎, and 𝜎𝑥𝑦 = −150 𝑀𝑃𝑎, determine the directions for strain
gages at that point to measure two of the principal strains. What are the magnitudes of these
principal strains?
9. A bearing made of isotropic bronze (𝐸 = 82.6 𝐺𝑃𝑎 and 𝜈 = 0.35) is subjected to a state of
plane strain (𝜀𝑧 = 𝜀𝑦𝑧 = 𝜀𝑧𝑥 = 0). Determine 𝜎𝑧 , 𝜀𝑥 , 𝜀𝑦 and 𝛾𝑥𝑦 if 𝜎𝑥 = 90 𝑀𝑃𝑎, 𝜎𝑦 =
−50 𝑀𝑃𝑎, and 𝜎𝑥𝑦 = 70 𝑀𝑃𝑎.
10. A steel plate ((𝐸 = 200 𝐺𝑃𝑎 and 𝜈 = 0.29) is subjected to a state of plane stress (𝜎𝑥 =
−80 𝑀𝑃𝑎, 𝜎𝑦 = 100 𝑀𝑃𝑎, and 𝜎𝑥𝑦 = 50 𝑀𝑃𝑎). Determine the principal stresses and
principal strains.
11. A cubical element is subjected to the following state of stress.
𝜎𝑥 = 100 𝑀𝑃𝑎, 𝜎𝑦 = −20 𝑀𝑃𝑎, 𝜎𝑧 = −40 𝑀𝑃𝑎 and 𝜎𝑥𝑦 = 𝜎𝑦𝑧 = 𝜎𝑧𝑥 = 0
Assuming the material to be homogeneous and isotropic, determine the principal shear strains
and the octahedral shear strain, if 𝐸 = 200 𝐺𝑃𝑎 and 𝜈 = 0.25.
12. A steel element (𝐸 = 207 𝐺𝑃𝑎 and 𝐺 = 80 𝐺𝑃𝑎) is subjected to the following state of strain:
0.001 0 −0.002
[𝜀𝑖𝑗 ] = [ 0 −0.003 0.0003 ]
−0.002 0.0003 0
Determine the stress matrix.
13. A rubber cube shown in Figure P4.13 is inserted in a cavity of the same form and size in a
steel block and the top of the cube is pressed by a steel block with a pressure of 𝑝 pascals.
Considering the steel to be absolutely hard and assuming that there is no friction between steel
and rubber, find
(a) the pressure of rubber against the box walls, and
(b) the extremum shear stresses in rubber.
14. A thin rubber sheet is enclosed between two fixed hard steel plates (see Figure P4.14). Friction
between the rubber and steel faces is negligible. If the rubber plate is subjected to stresses 𝜎𝑥
and 𝜎𝑦 as shown, determine the strains 𝜀𝑥 and 𝜀𝑦 and also the stress 𝜎𝑧 .
15. The nonzero strain components at a critical point in an aluminum spar of an airplane (𝐸 =
72 𝐺𝑃𝑎 and 𝜈 = 0.33) are measured on a free surface as 𝜀𝑥 = 0.002, 𝜀𝑥 = 0.001 and 𝜀𝑥𝑦 =
0.001.
(a) Determine the corresponding nonzero stress components.

108
 (b) A design criterion for the spar is that the maximum shear stress cannot exceed 𝜏𝑚𝑎𝑥 =
70 𝑀𝑃𝑎. Is this condition satisfied for the measured strain state?

Figure P4.13 Figure P4.14

16. On the free surface of a bearing made of commercial bronze (𝐸 = 110 𝐺𝑃𝑎 and 𝜈 = 0.35),
the principal strains are determined to be 𝜀1 = 0.0015 and 𝜀2 = 0.0005. A design criterion for
the bearing is that the maximum tensile stress not exceed 200 𝑀𝑃𝑎. Is this criterion satisfied
for the given strain state
17. Calculate the volumetric change of the metal block shown in Figure P4.17 subjected to
uniform pressure 𝑝 = 160 𝑀𝑃𝑎 acting on all faces. Use 𝐸 = 210 𝐺𝑃𝑎 and 𝜈 = 0.3.
18. The aluminum [𝐸 = 70 𝐺𝑃𝑎 and 𝛼 = 22 × 10−6 0 𝐶 −1 ] block shown in Figure 4.18 rests on
a smooth, horizontal surface. When the body is subjected to a temperature change of ∆𝑇 =
+200 , determine
(a) The thermal strains 𝜀𝑇𝑥 , 𝜀𝑇𝑦 and 𝜀𝑇𝑧 .
(b) The deformations in the coordinate directions 𝑥, 𝑦 and 𝑧,
(c) The shearing strain 𝛾𝑥𝑦 .

Figure P4.17 Figure P4.18

Answers
1. 119.42 𝐺𝑃𝑎, 79.62 𝐺𝑃𝑎; 7.78 𝐺𝑃𝑎, 11.67 𝐺𝑃𝑎
2. 1882.62 𝑀𝑃𝑎, 332.23 𝑀𝑃𝑎, 642.30 𝑀𝑃𝑎; 1251.39 𝑀𝑃𝑎, −143.97 𝑀𝑃𝑎, 321.15 𝑀𝑃𝑎;
−2691.03 𝑀𝑃𝑎, −2846.07 𝑀𝑃𝑎, −1605.75 𝑀𝑃𝑎;

109
 513.11 𝑀𝑃𝑎, −882.25 𝑀𝑃𝑎, −107.05 𝑀𝑃𝑎;
2572.91 𝑀𝑃𝑎, 1487.63 𝑀𝑃𝑎, 1177.55 𝑀𝑃𝑎
3. −483.18 𝑀𝑃𝑎, −591.45 𝑀𝑃𝑎, 0.004925;
699.7 𝑀𝑃𝑎, 374.9 𝑀𝑃𝑎, −0.0049;
538.1 𝑀𝑃𝑎, 321.6 𝑀𝑃𝑎, −0.0039;
−617.3 𝑀𝑃𝑎, −779.7 𝑀𝑃𝑎, 0.0064;
4. 801.59 𝑚𝑚, 800.79 𝑚𝑚, 9.98 𝑚𝑚 5. 377.19 𝑀𝑃𝑎, 254.39 𝑀𝑃𝑎
8. 0.3827𝒊 + 0.9239𝒋, −0.002988; 0.9239𝒊 − 0.3827𝒋, 0.004849
9. 14 𝑀𝑃𝑎, 0.001242, −0.001046, 0.002288
10. −92.96 𝑀𝑃𝑎, 112.96 𝑀𝑃𝑎; −0.000629, 0.000699, −0.000029
11. 0.000125, −0.000875, 0.00075; 0.000773
−67.79 0 −160
12. [ 0 −707.88 24 ] 𝑀𝑃𝑎
−160 24 −227.88
𝜈𝑝 𝜈𝑝 1 − 2𝜈
13. − ,− ; 𝑝
1−𝜈 1−𝜈 1−𝜈
1+𝜈 1+𝜈
14. 𝜈(𝜎𝑥 + 𝜎𝑦 ), [(1 − 𝜈)𝜎𝑥 − 𝜈𝜎𝑦 ], [(1 − 𝜈)𝜎𝑦 − 𝜈𝜎𝑥 ]
𝐸 𝐸
15. 188.26 𝑀𝑃𝑎, 134.13 𝑀𝑃𝑎, 54.14 𝑀𝑃𝑎; Satisfied
16. 209.97 𝑀𝑃𝑎, 128.49 𝑀𝑃𝑎; Not satisfied
17. 0.002743 𝑚3
18. 450 × 10−6 , 450 × 10−6 , 450 × 10−6 ; 0.0450 𝑚𝑚, 0.0225 𝑚𝑚, 0.0338 𝑚𝑚; 0

110