Continuum Mechanics

F
F
x x
x
x
x
x
Continuum Mecanics SS 2013
Prof. Dr. Ulrich Schwarz
Universitt Heidelberg, Institut fr theoretische Physik
Tel.: 06221-54-9431
EMail: Ulrich.Schwarz@bioquant.uni-heidelberg.de
Homepage: http://www.thphys.uni-heidelberg.de/~biophys/
Latest Update: July 15, 2013
Address comments and suggestions to gmarschmann@web.de
Contents
1. Introduction 2
2. Linear viscoelasticity in 1d 4
2.1. Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2. Elastic response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.3. Viscous response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.4. Maxwell model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.5. Laplace transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.6. Kelvin-Voigt model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.7. Standard linear model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.8. Boltzmanns Teory of Linear Viscoelasticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.9. Complex modulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3. Distributed forces in 1d 19
3.1. Continuum equation for an elastic bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2. Elastic chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4. Elasticity theory in 3d 25
4.1. Material and spatial temporal derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.2. Te displacement vector eld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.3. Te strain tensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.4. Te stress tensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.5. Linear elasticity theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.6. Non-linear elasticity theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5. Applications of LET 46
5.1. Reminder on isotropic LET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
5.2. Pure Compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.3. Pure shear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.4. Uni-axial stretch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.5. Biaxial strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.6. Elastic cube under its own weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.7. Torsion of a bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.8. Contact of two elastic spheres (Hertz solution 1881) . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.9. Compatibility conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.10. Bending of a plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.11. Bending of a rod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2
1 Contents
6. The Finite Element Method (FEM) 64
6.1. Classication of PDEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
6.2. Te weak form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.3. Shape functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
A. Overview of tensors in elasticity theory 66
CHAPTER 1
Introduction
Continuum Mechanics describes the movement of mater on a suciently large length scale ( 100 nm) such that
it can be considered to be continuously distributed, rather than consisting of many discrete atoms, molecules or
larger particles (like grains of sand).
Te appropriate mathematical concept is that of continuous elds, for example:
(r, t) mass density eld
v(r, t) velocity eld
u(r, t) deformation eld
Terefore continuum mechanics can be considered to be a classical eld theory (like electrodynamics). It can be
combined with other fundamental parts of physics, e.g. thermodynamics (if temperature, heat, work and entropy
are important) or electrodynamics (if electric and/or magnetic elds exist). Important applications are engineering,
materials science (including biomaterials), environmental sciences and astrophysics. A very modern branch is
'active systems', where movement results from local energy sources.
Te mechanics of continuous bodies branches into a few major areas as depicted in g.1.2. In this lecture we will
take a rather unconventional approach and begin with the theory of viscoelasticity in one dimension, describing
the ow of complex uids and combining the two worlds of solid mechanics and uid dynamics into one. Only
then we will turn to the three-dimensional theory, rst for elastic solids and then for uids.
F
F
x x
x
x
x
x
(A)
F
d
F
g
(B)
a
a
htp://en.wikipedia.org/wiki/File:Stokes_sphere.svg
Figure 1.1.: (A) By deforming an elastic body, in general all points will change their position. By mapping the
original onto the deformed shape, one obtains the deformation eld u
e
= x
e
x
e
(B) Velocity eld of ow around a sphere in the context of Stoke's Law
2
3
Hookean Solid
Newtonian Fluid
Solid Mechanics Fluid Dynamics
Continuum Mechanics
Imperfections
Elasticity Theory
Rheology
Hydrodynamics Gas Dynamics
defects
fracture
plasticity
processing and
stability of
materials
linear
vs.
non-linear
shells & beams
elastic waves
buckling
polymer melts &
solutions, blood, liquid
crystals, glaciers, earth,
etc.
flow of complex fluids: flow of simple liquids:
variable density &
temperature
combustion engines,
star formation,
supersonic flights
Maxwell,
Kelvin,
Boltzmann
aerodynamics (cars
& planes),
flow through porous
media (ground water),
atmosphere &
oceans
Navier Equation
Navier-Stokes-Equation
Figure 1.2.: Substructure of Continuum Mechanics
CHAPTER 2
Linear viscoelasticity in 1d
2.1. Motivation
Continuum Mechanics is a eld theory for vectors and tensors of higher rank. However, before we develop the
three dimensional theory, we rst consider its 1d (scalar) version. In particular we introduce the concepts of vis-
coelasticity and the complex modulus.
A typical experiment in this context would be the mechanical test of a ber. In a relaxation experiment, one pre-
scribes the deformation and records the force (the alternative would be a creep experiment where one prescribes the
force and records the deformation). A typical result of a relaxation experiment looks like this:
l l
F
F
Figure 2.1.: Fiber
Stretching a ber in one direction (g.2.1): We neglect the compression in the other di-
rection and denote the time-dependent extension of the elastic material, the deformation,
as l = l l
0
.
Fig. 2.2 depicts the loading protocol (input) as a series of stretches. In a perfect elastic
system the force will follow exactly the stretch prole (g. 2.3) whereas in a real system,
the asymmetry between loading and unloading leads to a hysteresis cycle. Te energy
W =
_
Fdl ,= 0, corresponding to the area of the rectangle, is dissipated as heat and thus
presents a possibility to distinguish between elastic (W = 0) and viscoelastic systems.
l
time t
Figure 2.2.: Input
time t
force F
Figure 2.3.: Output
force F
l
Figure 2.4.: Hysteresis cycle
4
5 2.2 Elastic response
F F
k
Figure 2.5.: Te response of an elastic spring, characterized by its spring constant k, to an applied force F is in linear
approximation given by Hooke's Law.
2.2. Elastic response
As the simplest model consider a spring with spring constant k (g.2.5).
Te simplest possible mechanical response to a force is an elastic one: F = k (l l
o
).
We dene:
stretc =
l
l
0
=
l
0
+ l
l
0
(2.1)
strain
1
=
l
l
0
(2.2)
Equation 2.2 can be used to dene the 1d modulus C which characterizes material properties of the ber:
F = k l = (k l
0
) C (2.3)
In a 3d stretch experiment, a force F applied over an area A stretches the material from length l to l.
A linear elastic response implies:
F
A
= E
l
l
(2.4)
where stress
F
A
and strain act as cause and eect
= E (2.5)
where E is the so called Young's modulus or rigidity of the material. From here we nd C = E A.
Equation 2.5 might be recognized as Hooke's law conjuring up the image of macroscopic deformation as the result
of the stretching of a large set of microscopic springs corresponding to the elastic elements within the material.
A dimensional analysis of the quantities in question reveals the following:
[] = 1 (2.6)
[C] = [F] = N, (2.7)
[] =
N
m
2
= Pa (2.8)
[E] = [] = Pa (2.9)
For solid materials the Young's modulus is typically in the range of GPa, much larger than for sof mater, e.g. cells
with rigidity in the order of 10 kPa.
Stretching the relationship between stress and strain a bit further, we can rewrite Equation 2.5 as
F = (E A) = C = (k l
0
) (2.10)
k =
E A
l
0
(2.11)
where k is considered to be the 'spring constant' of the material.
1
Linear elasticity theory (LET) is an expansion in small
6 2.3 Viscous response
l
F
F
A
(A)
F
C
(B)
Figure 2.6.: (A) Elastic material of length l and cross-sectional area A is stretched by a force F which will result in
a deformation of the material.
(B) Unlike viscoelastic material (compare g. 2.4) an elastic material shows no hysteresis and does not
dissipate energy.
2.3. Viscous response
Most biological materials showviscoelastic behaviour. In this section we will cast this behaviour in a one-dimensional
mathematical format.
For a viscous element, force results from movement:
F = C
1
l
dl
dt
(2.12)
where C
is the damping coecient and

D
1
l
dl
dt
(2.13)
is called the rate of deformation. Recalling the stretch parameter =
l
l
0
(Equation 2.1) and the relation = 1 +
we can write
D =
1
l
dl
dt
=
1
d
dt
=
1
1 +
(1 ) (2.14)
for small strain 1 such that
D =
1
l
0
dl
dt
(2.15)
and
F = C
D C
(2.16)
To perform the experiment correctly for all strain values one has to implement a constant deformation rate D =
const:
1
l
dl
dt
=
d ln l
dt
= D = const (2.17)
with solution
l = l
0
e
Dt
(2.18)
subject to the initial conditon l(t = 0) = l
0
, meaning that the endpoint has to be displaced exponentially in time
in order to maintain a constant deformation rate.
7 2.4 Maxwell model
F F
l
(A)
F
(B)
Figure 2.7.: (A) A dashpot is a damping device which resists motion via friction and serves as the mechanical equiv-
alent of a viscous bre.
(B) Response curve
F F
C
C
Figure 2.8.: Maxwell model: Dashpot and spring in series
2.4. Maxwell model
Te Maxwell model is the simplest spring-and-dashpot model for a viscoelastic uid (it ows on long time scales).
Te single elements of the model are given by
F
s
= C
s
, F
d
= C

d
(2.19)
for spring and dashpot respectively. Te strains add up to
=
s
+
d
(2.20)
implying
=
s
+
d
. (2.21)
Te forces in the spring and the dashpot are the same, hence the overall strain rate can be writen as
=
s
+
d
=
1
C
F +
1
C
F (2.22)
We consider a relaxation experiment, that is the strain is given and the force F has to be calculated:
Multiplying Equation 2.22 with C and rearranging we get an ordinary dierential equation (ODE) in F:
F +
C
C
F = C (2.23)
Introducing the relaxation time
C
C
we obtain
F +
1
F = C (2.24)
Te general solution of the ODE is given by the homogeneous solution and one particular solution F = F
h
+ F
p
.
Te homogeneous solution to
F
h
+
1
F
h
= 0 (2.25)
8 2.4 Maxwell model
is given by
F
h
= A
1
e
, A
1
= const (2.26)
and one particular solution to the inhomogeneous equation 2.24 by
F
p
= A(t) e
(2.27)
Substitution into Equation 2.24 leads to an expression for A:
dA
dt
= C e
t
(2.28)
hence
A = C
t
_
0
dt
e
t
(t
) (2.29)
subject to the condition that for t < 0 the strain rate vanishes 0.
Likewise imposing F = 0 for t < 0 leads to A
1
= 0 and collecting all the pieces we arrive at the integral solution
F(t) = C
t
_
0
dt
(tt
(t
) (2.30)
Example: 1) Strain ramp
For a strain depending linearly on time (compare g. 2.9) the strain rate is constant = const r. Equation
2.30 then reduces to
F(t) = C r e
t
_
0
dt
e
t
(2.31)
= C r e
[e
t
1] (2.32)
= C
r (1 e
) (2.33)
In the case of short times only the spring is extended and the response is linear and elastic (compare g. 2.9)
F = C
r
t
= Crt (2.34)
For long times we have a constant and viscous response
F = C
r, (2.35)
the spring has a constant extension and the force is dominated by the dashpot.
Example: 2) Relaxation experiment
We keep constant starting at time t
. Ten the strain rate is zero for t > t
= 0 (2.36)
and the force is given by
F = F
(tt
(2.37)
Te spring relaxes back to zero, no energy is stored but dissipated as heat.
9 2.5 Laplace transformation
t
t*
(A)
r
t
t*
F
(B)
ramp relaxation
Figure 2.9.: (A) Strain as a function of time and (B) the force response for the spring-dashpot model.
2.5. Laplace transformation
For the Maxwell model we had to solve the ODE
F +
1
F = C = f(t) (2.38)
for t 0 and with known initial condition F(0). Tis can be nicely done with Laplace transforms.
Denition 1. Let f(t) be a function dened for t 0:
f(s) = /[f(t)] =
_
0
dt f(t) e
st
, s C
Te bac transforms
f(t) =
1
2i
c+i
_
ci
ds

f(s) e
st
(2.39)
follow from complex analysis and are tabulated in many books.
We now show how to solve the ODE (Equation 2.38):
F +BF = f(t) (2.40)

s
F(s) F
0
+B

F(s) = /[f(t)] =

f(s) (2.41)

F(s) =
f(s) +F
0
s +B
(2.42)
Taking the strain ramp as an example from above:
f(t) = C = C r = const, B =
1
, F
0
= 0 (2.43)

f(s) =
C r
s
(2.44)

F(s) =
C r
s (s +B)
(2.45)
F(t) = C r t(1 e
) (2.46)
1
0
2
.
5
L
a
p
l
a
c
e
t
r
a
n
s
f
o
r
m
a
t
i
o
n
Table 2.1.: Some examples of forward Laplace transformations
f(t)

f(s)
1
_
0
dt 1e
st
=
1
s
e
at
_
0
dt e
(sa)t
=
1
sa
e
at
e
bt 1
sa

1
sb
=
ab
(sa)(sb)
t
_
0
dt te
st
=
_
t
s
e
st
0
+
_
0
dt
1
s
e
st
=
1
s
2
t f(t)
d
ds
_
0
dt f(t)e
st
=
d
ds
f(s)
t
n
_
0
dt e
st
t t
n1
=
d
ds
/[t
n1
] = (1)
n1 d
n1
ds
n1
/[t] =
n!
s
n+1
f
(t)
_
0
dt e
st df
dt
=
_
e
st
f(t)
0
+s
_
0
dt e
st
f(t) = f(0) +s
f(s)
convolution integral:
_
y(t t
)x(t
)dt
f(t)
_
0
dt e
st
_
dt
y(t t
)x(t
) = /[x(t)] /[y(t)], t = t
+t
11 2.6 Kelvin-Voigt model

F F
C
C
Figure 2.10.: Spring-dashpot in parallel is a Kelvin-Voigt arrangement
2.6. Kelvin-Voigt model
Te Kelvin-Voigt model as another example of combined viscoelastic behaviour is the simplest model for a viscoelas-
tic solid (it does not ow on long time scales).
In analogy to an electric circuit, the forces add up so that the total force equals the sum of the forces due to the
elastic spring F
s
and the viscous damper F
d
:
F = F
s
+F
d
= C +C
(2.47)
Te natural way to analyse this situation is a creep experiment where the force is prescribed. Following the math-
ematical treatment of the Maxwell modell, the constitutive ODE for the Kelvin-Voigt model is given by
+
1
=
F
C
linear ODE for (2.48)

Like equation 2.30, but with and F exchanged, the solution is
(t) =
1
C
t
_
0
dt
e
(tt
)/
F(t
) (2.49)
where we assume that at t = 0 we start to pull on the setup. In the case of a force jump to a constant value F
0
at
t = 0, the creep function is given by
J(t) =
(t)
F
0
=
1
C
(1 e
) (2.50)
Te strain response is such that it initially increases and then plateaus which is called creep (see g. 2.11).
Looking at the limits we have for t :
=
F
0
t
C
a linear viscous response (2.51)

and for t :
= F
0
C a constant elastic response (2.52)
12 2.7 Standard linear model
F
t
F
0
(A)
t
(B)
0
F/C
elastic
viscous
Figure 2.11.: (A) Te simple Heaviside step function is ofen used as a loading protocol for viscoelastic models.
(B) Te behaviour of the Maxwell model is reversed, showing that the response not only depends on
scaling but also on the arrangements of elements in the setup.
2.7. Standard linear model
For the standard linear model (see cartoon in g. 2.12) we now combine the Maxwell and Kelvin-Voigt models. For
the linear elements the force has to be the same, whereas in parallel the forces add up. Hence we get the following
relations:
F = F
1
+F
2
(2.53)
= C
1
(
d
) +C
2
(2.54)

d
=
F
1
C
=
F F
2
C
=
F C
2
(2.55)
Inserting (2.54) into (2.55) and thus eliminating F
1
, F
2
and
d
and again introducing the relaxation time (equation
2.4)

F = C
1
(
d
) +C
2
(2.56)
= (C
1
+C
2
) C
1
(
F C
2

C
) (2.57)
we nally obtain the constitutive equation for the standard linear model:

F +F = (C
1
+C
2
) +C
2
(2.58)
By taking away the upper branch, i.e. C
2
0 we regain the Maxwell model and in the limit C
1
the elastic
elements become innitely rigid, leading back to the Kelvin-Voigt model:
C
2
0

F +F = C
1
Maxwell (2.59)
C
1
F = C
+C
2
Kelvin-Voigt (2.60)
F
F
C
C
C
F
F
Figure 2.12.: Te standard linear model with one dashpot and two springs.
13 2.7 Standard linear model
0
(A)
F
t
(C+C)
0
(B)
C
Figure 2.13.: (A) Strain jump.(B) Force response
Equation 2.58 allows for both a relaxation and a creep experiment by specication of the source term.
Example: Strain jump (relaxation experiment, see g. 2.13)
Te relaxation function is given by
F(t) =
0
(C
2
+C
1
e
) (2.61)
Geting the jump at t = 0 is not trivial. We introduce two times t = 0
and t = 0
+
at the lef and right of of t = 0
and rewrite the ODE as:
F(0
) +F(0
+
)
2
+
F(0
+
) F(0
)
t
= C
2
(0
) +(0
+
)
2
+ (C
1
+C
2
)
(0
+
) (0
)
t
(2.62)
Multiplying with t and using the one-sided nature of the function (hence F(0
) = 0 = (0
)) we nd in the
limit t 0
F(0
+
) = (C
1
+C
2
) (0
+
) (2.63)
a nite force jump with magnitude (C
1
+C
2
)
0
.
14 2.8 Boltzmanns Theory of Linear Viscoelasticity
2.8. Boltzmanns Theory of Linear Viscoelasticity
Te spring-and-dashpot models discussed above can be generalized to a class of materials called linear viscoelastic
bodies (Boltzmann 1876). Te basic assumption here is superposition: Individual loading histories add up linearly
to the combined loading history. Terefore, all we need to know is the response to a unit-step perturbation.
Consider a creep experiment where we prescribe the force F = H(t) with Heaviside function H(t).
Te strain in response will follow the force denoted by (t) = J(t) which is called the creep compliance or simply
the creep function. Te superposition principle is graphically shown below:
F
t 0
H(t)
(A)
t 0
(B)
J(t)
Figure 2.14.: Example of a strain response to a prescribed unit-step in force.
t 0
(A)
F
t 0
(B)
Figure 2.15.: Linear superposition of responses.

t
F
F
t'
Figure 2.16.: An arbitrary force linearized.
An arbitrary perturbation can be considered as an innite number of small steps in the force: Te increase in the
force is then given by
F(t
) =
dF(t
)
dt
dt
(2.64)
15 2.9 Complex modulus
and it follows for the strain response that
(t) =

F(t
)dt
J(t t
) (2.65)
Taking the time intervals t
innitesimally small and using the superposition principle, we can add up all the
responses to steps in the force to get an integral expression rst derived by Boltzmann in 1876:
(t) =
t
_
dt
J(t t
)

F(t
) Boltzmann 1876 (2.66)

In the same way one can write for a relaxation experiment:
F(t) =
t
_
dt
G(t t
) (t
) (2.67)
where G(t) is the relaxation function. Obviously G and J must be related to each other
2
.
Again the natural framework for this are Laplace transforms. Te two integral equations become
(s) =

J(s) s

F(s) (2.68)
F(s) =

G(s) s (s) (2.69)

G(s)

F(s) =
1
s
2
(2.70)
t
_
0
dt
J(t t
)G(t
) = t (2.71)
Creep function J and relaxation function G are thus related by an integral equation. In principal it is sucient to
know one of them.
2.9. Complex modulus
So far we have introduced the 1d elastic modulus C and the viscous modulus C
(damping coecient) by
F
s
= C
s
[C] = N (2.72)
F
d
= C

d
[C
] = Ns (2.73)
For composite systems, we have seen that the overall response depends on time scales. It therefore makes sense to
consider harmonic excitations, when loading has the form of a sine or cosine (Fourier analysis):
(t) =
0
cos(t) (2.74)
F(t) =
t
_
dt
G(t t
) (t) Boltzmann linear viscoelasticity (2.75)

=
0
_
ds G(s) (t s), s = t t
(2.76)
=
_
0
ds G(s) (t s) (2.77)
2
G and J act as propagators. In the creep experiment, think of this as creating a perturbation at t
and then propagate it in a linear manner

to the present time and integrate over the past history. G and J are both Green's functions to the ODE of the viscoelastic model of
interest and can be calculated respectively by specifying the source term.
Inserting
(t s) =
0
sin((t s)) (2.78)
=
0
sin(t) cos(s) +
0
cos(t) sin(s) (2.79)
gives
F(t) =
0
cos(t)
. .
in phase, elastic
_
_
_
0
ds G(s) sin(t)
_
_
. .
E
1
(), "storage modulus"

0
sin(t)
. .
out of phase, viscous
_
_
_
0
ds G(s) cos(s)
_
_
. .
E
2
() "loss modulus"
(2.80)
=
0
E
1
cos(t)
0
E
2
sin(t) (2.81)
= F
0
cos(t +) (2.82)
with
= arctan(
E
2
E
1
) phase shif (2.83)
F
0
=
0
_
E
2
1
+E
2
2
amplitude (2.84)
We can again look at the elastic and viscous limit
elastic limit: E
2
= 0, = 0, F
0
=
0
E
1
(2.85)
viscous limit: E
1
= 0, =

2
, F
0
=
0
E
2
(2.86)
where in the elastic limit no energy is dissipated and the force in the viscous limit is shifed by half a cycle.
Te amount of work for one loading cycle can be calculated
W =
2
_
0
dt F (2.87)
=
2
_
0
dt [
0
E
1
cos(t)
0
E
2
sin(t)]
0
sin(t) (2.88)
=
2
0
E
2
Tis work is disspated as heat (2.89)
Te work represented by E
1
is released again during unloading (storage modulus), but not the work represented
by E
2
(loss modulus).
In harmonic calculations it is ofen convenient to use comlex numbers. Instead of equation 2.74 write
(t) = Re
_
0
e
i
t
_
(2.90)
Ten the convolution integral becomes
F(t) = Re
__
dt
G(t t
) (t
)
_
(2.91)
= Re
__
ds G(s) (t s)
_
(2.92)
= Re
_
_
_
_
_
i
0
e
it
_
ds G(s)e
is
. .
G
()
_
_
_
_
_
(2.93)
where we recognize G
() as the Fourier transform of G(t).

We then dene the complex modulus as
Denition 2. E
() = iG
() = E
1
() +iE
2
()
F(t) = Re
_
0
e
it
E
()
_
(2.94)
= Re (
0
(cos(t) +i sin(t))(E
1
() +iE
2
())) (2.95)
=
0
E
1
cos(t)
0
E
2
sin(t) (2.96)
thus E
1
and E
2
are the storage and loss modulus as dened above.
With these denitions, the fundamental relation between force and strain
F(t) =
ds G(s) (t s) (2.97)
simply becomes
F
() = G
()i
() (2.98)
= E
()
() (2.99)
In the physics literature the complex modulus E
() is usually denoted as
G() = G
() +i G
() (2.100)
where G
() is the storage and G
() the loss modulus.

G() carries the complete viscoelastic information of a system. It is measured in a rheometer.
A typical result for a polymer melt (liquid) is shown in g. 2.18:
Te polymer is elastic at high frequencies (G
> G
) and viscous at low frequencies (G
> G
) (this corresponds
to the longtime limit).
Here
G
= C
(2.101)
and
G

C
2
2
C
(2.102)
. Equating at the intersection gives an expression for and thus a simple way to extract the relaxation time :
=
C
C
=
1
(2.103)
Perfect elastic networks like PDMS (Polydimethylsiloxan) have a large and constant storage modulus and small loss
modulus.
Te Newtonian liquid water has a constant loss modulus for 10
5
Hz < < 10
5
Hz (viscoelasticity is expected for
10
10
Hz!)
x x x x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
xx x x
x x
x
x
x x x x
x
x x x x x
(a) (b) (c)
Figure 2.17.: Sketches of rotational rheometers: (a) cylindrical, (b) cone and plate, (c) parallel plate
Figure 2.18.: Response of a linear polybutadiene at 25 . Filled symbols are the storage modulus G
and open
symbols are the loss modulus G
(Rubinstein and Colby, Polymer Physics, 2003, p. 293).

CHAPTER 3
Distributed forces in 1d
In the preceding chapter, deformations of 1d structures were represented only by one variable.
In this chapter we address the deformation of 1d structures in more detail, in particular the eect of spatially
distributed deformations and forces.
3.1. Continuum equation for an elastic bar
Consider a bar loaded by a force F at x = L and clamped in space at x = 0 (g. 3.1) . Dierent from Chapter 2 we
now allow the cross sectional area A and the elastic modulus E to be functions of x. We also allow for some force
per volume Q to act in x-direction (e.g. gravity).
Consider a slice of thickness x. Here we have
N(x) = N(x + x) +Q(x)A(x) x (3.1)
where N(x) is the normal force on the slice surface. With q(x) = Q(x)A(x) and in the continuum limit (x 0)
we obtain
dN
dx
+q(x) = 0 (3.2)
If the force per volume Q is zero, the normal force N is constant throughout the bar.
We now introduce a displacement eld u(x). Ten the strain at position x reads
=
u(x + x) u(x)
x
x 0
=
du(x)
dx
(3.3)
[] =
m
m
= 1 (3.4)
F
x
x L
N(x) N(x+x)
Figure 3.1.: Representation of an elastic bar with magnication of the forces acting on a slice of thickness x.
19
20 3.1 Continuum equation for an elastic bar
Strain is related to stress by Hooke's law:
= E (3.5)
where the stress is dened as
=
N
A
(3.6)
Substitution of N = A into equation 3.2 yields a second order dierential equation for the displacement u(x)
q =
dN
dx
=
d
dx
(A) (3.7)
=
d
dx
(EA) (3.8)
=
d
dx
_
C(x)
du
dx
_
(3.9)
where C(x) = E(x) A(x) is the 1d modulus.
d
dx
_
C(x)
du
dx
_
+q(x) = 0 central ODE (3.10)
Example: 1) Homogeneous bar without gravity (see g. 3.2)
A = const, E = const, Q = 0 and the ODE reduces to
C u
= 0 (3.11)
u
= const = a (3.12)
u = a x +b (3.13)
subject to the boundary conditions of suppressed displacement at the origin (3.15) and a force F applied at the free
end of the bar (3.16):
u(x = 0) = 0 (3.14)
N(x = L) = F = C u
(L) (3.15)
hence the solution reads
b = 0, a =
F
C
u =
F
C
x (3.16)
Te stress is then given by
= E u
=
F
A
= const (3.17)
F
x L
Figure 3.2.: A bar of length L, with uniform cross section A and constant Young's modulus E. No volume force is
present.
21 3.1 Continuum equation for an elastic bar
F
x L
(A)
(B)
u
x L
A=const
A(x)
Figure 3.3.: (A) Te Young's modulus E is constant throughout the bar of length L, but the cross section A(x) varies
along the x-axis.
(B) Te displacement along the bar is linear for a homogeneous cross section Awhereas for an increasing
cross section the displacement becomes weaker for larger A.
Example: 2) Increasing cross section (see g. 3.3)
Let
A(x) = A
0
_
1 +
x
3L
_
(3.18)
Integration of
d
dx
_
EA
du
dx
_
= 0 (3.19)
yields
du
dx
=
a
EA
0
(1 +
x
3L
)
(3.20)
Integration again gives
u =
3aL
EA
0
ln
_
1 +
x
3L
_
+b (3.21)
with integration constants a and b to be specied by the boundary conditions:
u(0) = 0 b = 0 (3.22)
EA u
(L) = F u
(L) =
3aL
EA
0
1
/3L
4
/3
=
3a
4EA
0
=
F
EA
0
4
/3
a = F (3.23)
and all in all the displacement reads
u =
3FL
EA
0
ln
_
1 +
x
3L
_

F
EA
0
x for x L, A = const (3.24)
Example: 3) Homogeneous bar with gravity (see g. 3.4)
A = const, E = const, q = const = g A and the central equation becomes
d
dx
_
C
du
dx
_
= q (3.25)
u =
q
2C
x
2
+
a
C
x +b (3.26)
subject to the boundary conditions
u(x = 0) = 0 b = 0 (3.27)
C u
(L) = F = q L +a a = F +qL (3.28)

22 3.2 Elastic chain
x
(A)
F
L
(B)
u
x L
q=0
Figure 3.4.: (A) For a bar exposed to gravity, the largest elongation occurs at the end, whereas the largest force acts
close to the bounding surface and denes a possible breaking point.
(B) A comparison to the reference case with linear displacement (example 1)) shows an increased slope
at small x. Te boundary condition at x = L forces a crossover to the same slope.
Tus
u =
q
2C
x
2
+
F +qL
C
x (3.29)
For q = 0 this reduces to the linear behaviour of the elastic bar in example 1):
u =
F
C
x (3.30)
Te full solution is depicted in g. 3.4.
3.2. Elastic chain
We now start with a discrete approach and derive the 1d continuum equation for an elastic chain ("coarse-graining"
or "homogenization"). We also include elastic coupling to the ground (g. 3.5).
We have n = 0, 1, ..., N beads at positions x
n
= a n with x
0
= 0 and x
N
= L.
For a given n in the chain the force at equilibrium is
0 = F
N
= k
n+1
(u
n+1
u
n
) k
n
(u
n
u
n1
) K
n
u
n
qa (3.31)
where qa denotes the amount of volume force on bead n.
A Taylor expansion of k
n+1
, u
n+1
and u
n1
for a L, i.e. in
a
L
=
1
N
, and taking u
n
as a reference term gives
0 = F
N
(3.32)
= (k
n
+a
x
k
n
+
a
2
2

2
x
k
n
) (u
n
+a
x
u
n
+
a
2
2

2
x
u
n
u
n
) (3.33)
+k
n
(u
n
(u
n
a
x
u
n
+
a
2
2

2
x
u
n
)) +qa K
n
u
n
(3.34)
= k
n
a
2
2
x
u
n
+a
2
(
x
k
n
) (
x
u
n
) K
n
u
n
+qa +O(a
3
) (3.35)
= k(x)a
2
u
(x) +a
2
k
(x)u
(x) Ku(x) +qa (3.36)

x
(C(x) u
(x)) u(x) +q = 0 (3.37)

with C = E A = k a and =
k
/a.
Tis is the same equation as above where we started with the continuum description in the rst place.
Example: 1) Stretching a bar with elastic foundation, no gravity (see g. 3.6)
......
x L 0
0 1 2 n
N N-1
a
k_n k_(n+1)
K_n
Figure 3.5.: Te beads, separated at a distance a, are laterally connected by springs with spatially varying spring
constants k
n
and atached to the surface with springs with constants K
n
.
Without gravity, q = 0, the constitutive equation becomes, in the case = 0:
Cu
= 0 N = Cu
= F u =
F
C
x (3.38)
like before.
And for ,= 0:
Cu
u = 0 l
2
u
u = 0 (3.39)
where we dene the localization length as l =
_
C
=
_
k
K
a.
u = A e
x
/l
+B e
x
/l
(3.40)
u(0) = 0 B = A (3.41)
u = 2A sinh(
x
/l) (3.42)
C u
(L) = F =
C2A
l
cosh(
L
/l) (3.43)
resulting in
u =
FL
C

sinh(
x
/l)
cosh(
L
/l)
(3.44)
Te force on the foundation F
T
= k u is known as the traction force.
Example: 2) Contracting bar with elastic foundation, no gravity (see g. 3.7)
We now consider that deformation arises not from an external force, but by internal contraction (e.g. thermal
or active contraction). We add force dipoles along the bar:
N = Cu
+P, P = const (3.45)

l
2
u
u = P
= 0 (3.46)
F
x L
(A)
(B)
u
x L
Figure 3.6.: (A) For a bar with elastic foundation, the displacement of string atachments increases with x.
(B) Te displacement is an exponentially increasing function of x.
Tus the displacement has the same solution as before
u = 2A sinh (
x
/l) (3.47)
but the boundary condition is dierent at x = L:
N(L) = 0 (3.48)
C u
(L) +P = 0 (3.49)

P
C
=
2 A(P)
l
cosh (
x
/l) (3.50)
u(x) =
Pl
C

sinh(
x
/l)
cosh(
L
/l)
(3.51)
Tus the deformation vanishes if the contraction is zero P = 0. Note also that the displacement now is to the lef.
x L
(A)
u
x (B) L
Figure 3.7.: (A) A bar with a distribution of contractile forces. (B) Displacement for to a contracting bar.
CHAPTER 4
Elasticity theory in 3d
Up to this point we dealt with one-dimensional equations describing the relationship between stress and strain for
bres and bars. We now turn to the derivation of the fundamental concepts and equations for a 3d theory of elastic
deformations.
4.1. Material and spatial temporal derivatives
In order to describe the deformation and movement of a three-dimensional object in time, we consider how it
deforms under surface or volume forces (see g.4.1). We consider a material point P at x
0
in a continuum body
with initial volume V
0
. Tis point is moved to x = x (x
0
, t) at time t and deformed by surface or volume forces.
Te vector x
0
is bound to the material point for all congurations parameterized by t and thus when x
0
is constant
we can trace the material point in time where x (x
0
, t) is the kinetic path taken by the material point P.
Partial dierentiation with respect to time t results in the velocity and acceleration elds of the material point
under consideration:
v = x = v (x
0
, t) velocity eld (4.1)
a = v = x = a (x
0
, t) acceleration eld (4.2)
For each vector x, v, a, its three components are scalar elds depending on x
0
and t. Te same would be true for
any other scalar eld, take for example a temperature eld T (x
0
, t) where the temperature is dened on every
point of the body and the temporal evolution monitored.
P
P
V
V(t)
x
x(x,t)
x
y
z
Figure 4.1.: Te material point P is labeled on the reference conguration V
0
and then traced in time.
25
26 4.1 Material and spatial temporal derivatives
Lagrangian and Eulerian frame
We can dene two dierent types of time derivatives:
(1) Material time derivative

T =
T
t
x
0
=const
T is the change in unit time of the temperature at material point P identied by x

0
and moving through space.
When the the temperature eld in the conguration V (t) is mapped onto the reference conguration V
0
the de-
scription is referred to as Lagrangian. Te Lagrangian description is the natural choice for solids where it is always
possible to return to the initial conguration by removing surface or volume forces.
(2) Spatial time derivative
T
t
=
T
t
x=const
In the Eulerian description, x is a xed point in V (t). At each time t, a dierent material particle might be present at
this location. Tus
T
t
is the change in unit time of the temperature at a xed point x in space. Te Eulerian frame
is the natural choice for uids, because upon deformation there is no way to trace back the reference conguration
of a uid.
In order to derive a relation between the material and the spatial time derivative we start with the Eulerian
description of the scalar temperature eld T = T (x, t). Te total derivative can be rewriten with the gradient
operator :
dT =
T
x
dx +
T
y
dy +
T
z
dz +
T
t
dt (4.3)
= (dx ) T +
T
t
x=const
dt (4.4)
= (vdt ) T +
T
t
dt

Tdt (4.5)
where dx = vdt implies movement with the ow and accordingly a change in temperature dT =

Tdt. Dividing
equation 4.5 by dt gives
T
..
material derivative
= (v ) T +
T
t
..
spatial derivative
(4.6)
Special care has to be taken with the two dierent types of derivatives involved. If v = 0, i.e. there is no move-
ment/deformation, the derivatives are equal. Te dierence between material and spatial derivatives is called the
convective contribution given by (v ).
Te temperature eld was an introductory example, but the same can now also be applied to vectors:
x = (v ) x +
x
t
..
xed position in space
= v 1 + 0 = v
or with the convective part writen in components
(v ) x = (v
j
j
) x
i
= v
j
ij
= v
i
(4.7)
Te velocity is of course the correct result when following the motion in the Eulerian frame.
Applying the same operation to the velocity v and taking the i
th
component of the convective term results in
v = (v ) v
. .
(v
j
j
)v
i
+
v
t
= v (v)
. .
v
j
j
v
i
+
v
t
= (v)
T
v
. .
j
v
i
v
j
+
v
t
= L
..
velocity gradient tensor
v +
v
t
x
y
z
x
P

dx vdt
dx+dxdt
(v+Ldx)dt
P'
Figure 4.2.: Change of a material line element dx expressed by the the velocity gradient tensor L.
Tis is the rst time we encounter tensors and the dyadic product . We also have used b
i
= M
ij
a
j
= a
j
(M
T
)
ji
for b = M a = a M
T
. Te dierence between the material and the spatial derivative is in this case simply a linear
operation on the velocity.
Tensors
Elasticity theory in 3d makes use of 2nd rank tensors, which are linear maps of one vector onto another. Te
simplest way to generate such a tensor is the dyadic or tensor product:
(a b)
ij
a
i
b
j
We take the dyadic product of gradient and velocity
v =
_
_
x
v
x

x
v
y

x
v
z
y
v
x

y
v
y

y
v
z
z
v
x

z
v
y

z
v
z
_
_
(4.8)
therefore the velocity gradient tensor reads
L
T
= (v)
T
=
_
_
x
v
x

y
v
x

z
v
x
x
v
y

y
v
y

z
v
y
x
v
z

y
v
z

z
v
z
_
_
(4.9)
Te tensor L measures the change in unit time of a material line element dx (see g.4.2):
Te tail of dx (P) moves to x + vdt
Te head of dx (P') moves to x +dx + (v +dx(v)
. .
Ldx
)dt
Afer an innitesimal increase in time dt the line element in the volume V will change into dx + d x and we can
read of g.4.2
dx +d xdt = x +dx + (v + Ldx)dt (x + vdt)
d x = Ldx
Te line element thus changes by the linear element L which measures changes both in length and in orientation,
gives complete information as the 3d metric of the system.
Since in the following, extensive use of 2nd rank tensors will me made, this section summarizes some general
statements:
Typically one uses Cartesian coordinates with a set of basis vectors e
i
. Ten an arbitrary 2nd rank tensor can be
writen as
M =
ij
M
ij
(e
i
e
j
) (4.10)
and we note that the dyadic product is compatible with the inner product (by denition and the linearity of vectors)
(a b) f = a(b f) (e
i
e
j
) e
k
= a
i

jk
(4.11)
and thus
b = M a =
_
_
ij
M
ij
(e
i
e
j
)
_
_
k
a
k
e
k
_
=
ij
M
ij
a
j
e
i
b
i
= M
ij
a
j
Te trace of M is dened as
tr(M) =
i
M
ii
(4.12)
i.e. as the sum of diagonal elements.
Te deviatoric part of M is dened as
M
d
= M
1
3
tr(M)1 tr(M
d
) = 0 (4.13)
Most of the tensors we encounter will be symmetric and can therefore always be diagonalized. Te eigenvalues of
M are dened by the caracteristic equation
0 = det (M1) =
3
I
1
2
+I
2
I
3
(4.14)
in 3d with the three invariants of M:
I
1
= tr(M) = M
xx
+M
yy
+M
zz
(4.15)
I
2
=
1
2
_
(tr(M))
2
tr(M M)
_
= M
xx
M
yy
+M
xx
M
zz
+M
yy
M
zz
M
2
xy
M
2
yz
M
2
xz
(4.16)
I
3
= det(M) = M
xx
(M
yy
M
zz
M
zy
M
yz
) M
xy
(M
yx
M
zz
M
zx
M
yz
) +M
xz
(M
yx
M
zy
M
zx
M
yy
) (4.17)
29 4.2 The displacement vector field
P
P
V
V(t)
x
x(x,t)
x
y
z
u
Figure 4.3.: Te displacement vector for a material point P.
4.2. The displacement vector field
We now introduce the most central quantity of elasticity theory, the displacement vector eld. Again considering
the displacement of a material point P from a reference conguration V
0
to the current conguration V (t) (see
g.4.3), the displacement vector u satises
u = x x
0
=
_
x(x
0
, t) x
0
= u(x
0
, t) Lagrangian frame
x x
0
(x, t) = u(x, t) Eulerian frame
(4.18)
Many problems in solid mechanics amount to determining the displacement eld u corresponding to a given set of
applied forces.
For a scalar eld like temperature T, we can dene two gradients depending on the frame that should become
identical for u = 0:
spatial gradient T = e
x
T
x
+ e
y
T
y
+ e
z
T
z
material gradient
0
T = e
x
T
x
0
+ e
y
T
y
0
+ e
z
T
z
0
Tese two can be related by the chain rule as exemplied for the x-component at a xed time t
T
x
0
=
T
x
0
x
x
0
+
T
y
y
x
0
+
T
z
z
x
0
(4.19)
and likewise for the other two components. Using the procedure from section 4.1 we can summarize this as
0
T = F
T
T (4.20)
with the transpose of the deformation gradient tensor F from the current conguration with respect to the reference
conguration, or mathematically with the Jacobian matrix of the coordinate transformation.
F
T
=
_
_
_
x
x
0
y
x
0
z
x
0
x
y
0
y
y
0
z
y
0
x
z
0
y
z
0
z
z
0
_
_
_ F =
_
_
_
x
x
0
x
y
0
x
z
0
y
x
0
y
y
0
y
z
0
z
x
0
z
y
0
z
z
0
_
_
_ (4.21)
We can also write, including the deformation vector u = x x
0
F
T
=
0
x =
0
(x
0
+ u) = 1 +
0
u (4.22)
F = (
0
x)
T
= 1 + (
0
u)
T
(4.23)
If there is no deformation (u = 0):
F
T
= F =
0
(4.24)
Again, if the reference and current state are equal, then the deformation is zero and F = 1 and the gradient
operators are also identical as demanded.
30 4.3 The strain tensor
P
P
V
V(t)
x
x(x,t)
x
y
z
P'
dx
P'
dx
Figure 4.4.: Te change in distance and direction between the two material points can be described with the strain
tensor.
4.3. The strain tensor
We now consider a material line segment dx
0
that changes both length and orientation during deformation:
dl
0
dl length
e
0
e orientation
Te orientation and length of the respective line segments are dened by:
dx
0
= e
0
dl
0
dl
0
=
_
dx
0
dx
0
(4.25)
dx = edl dl =
dx dx (4.26)
For the relation between dx
0
and dx we look at the total dierential again (e.g. for the x-component)
dx =
x
x
0
dx
0
+
x
y
0
dy
0
+
x
z
0
dz
0
(4.27)
and the same for the other components such that in a more compact form the deformation gradient tensor can be
used
dx = F dx
0
(4.28)
Tis can be rewriten with the help of equations 4.25 and 4.26
edl = F e
0
dl
0
[()
2
(4.29)
e edl
2
= e
0
F
T
F e
0
dl
2
0
(4.30)
stretch ratio =
dl
dl
0
=
e
0
F
T
F
. .
C
e
0
(4.31)
Te stretch ratio is dened as the ratio between the lengths of the line segments in the corresponding congurations.
Te stretch ratio for the material line segment is thus determined by the right Caucy-Green deformation tensor
C (Lagrangian description).
Te new orientation of the material line segment can be calculated as
e = F e
0
dl
0
dl
= F e
0
1
=
F e
0
e
0
C e
0
(4.32)
Te procedure can also be inverted, starting from (dl, e) we can get to (dl
0
, e
0
):
F
1
dx = dx
0
F
1
edl = e
0
dl
0
(4.33)
e F
T
F
1
edl
2
= e
0
e
0
dl
2
0
(4.34)
=
dl
dl
0
=
1
e F
T
F
1
. .
B
1
e
(4.35)
Te tensor B F F
T
is called the lef Caucy-Green deformation tensor in the Eulerian frame. Te old direction of
the material line segment is then given by
e
0
= F
1
e
dl
dl
0
=
F
1
e
e B
1
e
(4.36)
As explained earlier, ofen it is more convenient to introduce a variable which vanishes for vanishing deforma-
tion (strain). We now dene various strain tensors. One can show that each of them is invariant under a rigid body
transformation (translation and rotation).
In the Lagrangian frame we found
2
= e
0
C e
0
(4.37)
coupled to this one can now introduce the Green-Lagrange strain
GL
First denition:
GL
=

2
1
2
= e
0
1
2
(C 1)
. .
E
e
0
(4.38)
with the Green-Lagrange strain tensor E.
Above we showed
F
T
= 1 + (
0
u) F = 1 + (
0
u)
T
(4.39)
Substitution gives an expression for E
E =
1
2
_
F
T
F 1
_
=
1
2
_
_(
0
u) + (
0
u)
T
. .
linear in u
+(
0
u) (
0
u)
T
. .
quadratic
_
_ (4.40)
Te linear strain
lin
is dened as
Second denition:
lin
= 1 =
_
e
0
F
T
F e
0
1 (4.41)
Due to the square root this expression is not easy to use. For small deformations F 1 we can write
lin
=
_
1 + e
0
(F
T
F 1) e
0
1 (4.42)
1
2
e
0
_
F
T
F 1
_
e
0
= e
0
E e
0
=
GL
(4.43)
We further linearise to dene the linear strain tensor expressed as
=
1
2
_
F
T
+ F 21
_
=
1
2
_
(
0
u) + (
0
u)
T
_
(4.44)
Tis is the most ofen used formulation of a strain tensor in linear elasticity theory (LET).
In the Eulerian frame we found
1
2
= e B
1
e (4.45)
Coupled to this we dene the Almansi Euler strain
AE
as
Tird denition:
AE
=
1
1
2
2
= e
1
2

_
1 B
1
_
. .
A
e (4.46)
with the Almansi-Euler strain tensor A.
In summary we have found that for large deformation, geometrical non-linearities appear in the various strain
tensors. In linear elasticity theory (LET), one assumes small deformations. Ten it is sucient to consider the
linear strain tensor
ij
=
1
2
(
i
u
j
+
j
u
i
) =
_
_
_
ux
x
0
1
2
(
ux
y
0
+
uy
x
0
)
1
2
(
ux
z
0
+
uz
x
0
)
1
2
(
uy
x
0
+
ux
y
0
)
uy
y
0
1
2
(
uy
z
0
+
uz
y
0
)
1
2
(
uz
x
0
+
ux
z
0
)
1
2
(
uz
y
0
+
uy
z
0
)
uz
z
0
_
_
_ (4.47)
Tis is a symmetrical 3x3 matrix which can be interpreted as follows:
Te diagonal terms
ii
are the linear strains of material line segments of the reference conguration in the
i-directions.
Te o-diagonal terms represent the shear in the material (compare g. 4.5 )
Volume change
We consider a small parallelepiped spanned by three linearly independent vectors dx
a
0
, dx
b
0
, dx
c
0
. Te Volume is
then given by
dV
0
=
_
dx
a
0
dx
b
0
_
dx
c
0
(4.48)
In the deformed conguration we have
dx
i
= F dx
i
0
(4.49)
One can then calculate for the relative volume
J =
dV
dV
0
= det(F) (4.50)
x
y
dx
dy
(u/y)dy
(uy/x)dx
u
u
y
Figure 4.5.: Graphical interpretation of the linear strain tensor
33 4.4 The stress tensor
Note that this is simply the Jacobian for the transformation x
0
x : dx = F dx
0
.
In linear approximation one nds
dV
dV
0
= 1 + tr()
dV V
0
dV
0
= tr() (4.51)
Tus the trace of the linear strain tensor is simply the relative volume change.
Tis result also motivates to decompose the strain tensor into a pure shear and a pure compression/dilation part:
ij
=
_
ij
1
3
ij
ll
_
. .
pure shear
+
_
1
3
ij
ll
_
. .
compression/dilation
(4.52)
Te trace of the strain tensor
ll
gives the volume change in terms of compression and dilation, whereas the o-
diagonal entries give the pure shear without a change of the volume (deviatoric part).
Relation to deformation in time:
Above we have considered a material line segment dx which evolves into a line segment dx +d xdt from time t to
time t + t. We have shown that
d x = L dx with L = (
0
v)
T
(4.53)
where L is yet another tensor, the velocity gradient tensor, a purely kinematic variable not related to the reference
conguration. On the other hand we now have
dx = F dx
0
(4.54)
with the deformation gradient tensor F = (
0
x)
T
. We can thus relate the two and nd
d x =

F dx
0
=

F F
1
dx = L dx (4.55)
L =

F F
1
(4.56)
4.4. The stress tensor
We have seen before that for a continuum body, stress =
F
/A rather than force F is the correct concept for the
cause of a deformation.
For a 1d bar with constant cross-section A and no volume forces, we have derived as condition of mechanical
equilibrium:
d
dx
= 0. We now generalize these results to 3d.
We rst dene a stress vector s =
3
i=1
s
i
e
i
by decomposing the force F onto an innitesimally small surface
element of area A in a 3d continuum body:
F =
3
i=1
F
i
e
i
s
i
=
F
i
A
(4.57)
Like before, stress is dened in the limit A 0, but now we deal with a 3d vector.
We next investigate the equilibrium conditions in 2d. Terefore we decompose the stress vector into dierent
directions (see g. 4.7(B)):
Following the notation for
ij
in g.4.7, the subscript i denotes the direction in which the stress is acting,
whereas the subscript j gives the direction of the normal of the surface element.
A
F
x
y
z
Figure 4.6.: Force F acting on a small surface element A in a contiuum body.
x
y
s
t
s
r
s
b
s
l
x x +x
y
y+y
0 0
0
0
(A)
x
y
y
y+y
0 0
0
0
x x +x
xx
xx
yy
yy
yx
yx
xy
xy
(B)
Figure 4.7.: (A) Free body diagram of the volume element with the stress vector s decomposed into directions (t)
top, (r) right, (b) botom and (l) lef.
(B) All stress components are a function of the x- and y-position in space, but assumed constant in the
z-direction.
x
y
y
y+y
0 0
0
0
x x +x
F
F
F
F
F
F
F
ty
F
tx
rx
ry
by
bx
ly
lx
Figure 4.8.: Illustration of the forces acting on the surface element.
As for the sign convention,
ij
> 0 if i, j have the same orientation in regard to their respective coordinate
direction.
Note that all
ij
=
ij
(x, y) are functions of position!
For equilibrium the forces in each direction have to add up to zero:
x-direction: F
lx
+ F
bx
= F
rx
+ F
tx
y-direction: F
ly
+ F
by
= F
ty
+ F
ry
We now transform the equations for forces into one equation for stresses and rst take the x-direction with refer-
ence point (x
0
, y
0
):
F
lx
= h
y
0
+y
_
y
0
dy
xx
(x
0
, y) (4.58)
= h
y
0
+y
_
y
0
dy
_
xx
(x
0
, y
0
) +

xx
y
x=x
0
,y=y
0
(y y
0
) +...
_
(4.59)
=
xx
hy +

xx
y
h
y
2
2
(4.60)
Here h is a constant thickness in z-direction and we suppress dependances on (x
0
, y
0
). In a similar way we nd
F
rx
= h
y
0
+y
_
y
0
dy
_
xx
+

xx
x
x +

xx
y
(y y
0
) +...
_
(4.61)
=
xx
hy +

xx
x
hxy +

xx
y
h
y
2
2
(4.62)
F
tx
= h
x
0
+y
_
x
0
dx
_
xy
+

xy
x
(x x
0
) +

xy
y
y +...
_
(4.63)
=
xy
hx +

xy
x
h
x
2
2
+

xy
y
hxy (4.64)
F
bx
= h
x
0
+x
_
x
0
dx
_
xy
+

xy
x
(x x
0
) +...
_
(4.65)
=
xy
hx +

xy
x
h
x
2
2
(4.66)
Te equilibrium condition in the x-direction now yields
xx
y +

xx
y
y
2
2
+
xy
x +

xy
x
x
2
2
(4.67)
=
xx
y +

xy
y
xy +

xx
y
y
2
2
+
xy
x +

xy
x
x
2
2
+

xy
y
xy (4.68)

xx
x
+

xy
y
= 0 (4.69)
For the y-direction one nds in a similar manner
yx
x
+

yy
y
= 0 (4.70)
Tus the equilibrium conditions amount to PDEs for
ij
. Tis is the generalization of

x
= 0 for 1d.
x
y
F
ry
F
bx
F
ly
F
tx
x/2
y/2
Figure 4.9.: Shear forces contribute to rotation of the prism.
We next show that
xy
=
yx
by balancing the moments around the midpoint (see g.4.9). We note that only
the shear forces create a moment:
x
2
F
ly
x
2
F
ry
+
y
2
F
tx
+
y
2
F
bx
= 0 (4.71)
Inserting the expressions from above gives the desired result
xy
=
yx
(4.72)
Te stress tensor
ij
has to be symmetric in order to avoid rotation.
In 2d, we have three independent stresses
xx
,
yy
,
xy
. How do we calculate the stress vector s acting on an
arbitrary area element from these stresses? We consider the following triangular prism (see g. 4.10): Te normal
to the inclined plane has components n
x
= sin and n
y
= cos and the stress vector can be decomposed as
s = s
x
e
x
+s
y
e
y
. Te forces over the whole prism have to sum up to zero again:
x-direction: s
x
l h =
xx
sin l h +
xy
cos l h s
x
=
xx
n
x
+
xy
n
y
y-direction: s
y
l h =
yx
sin l h +
yy
cos l h s
y
=
yx
n
x
+
yy
n
y
s = n (4.73)
Te stress vector is simply the product of the stress tensor with the normal vector n.
yy
xy
xx
yx
n
s
l
Figure 4.10.: A prism with two faces along cartesian coordinates and one face that is inclined by an angle .
Generalization to 3d
In 3d there are 6 independent stress components building up the symmetric stress tensor:
=
3
i,j=1
ij
e
i
e
j
,
ij
=
ji
(4.74)
Te stress vector on an arbitrary surface element with normal n is simply
s = n (4.75)
Te condition of mechanical equilibrium is
ij
= 0 (4.76)
the divergence of the stress tensor has to vanish.
Tis central result follows in a more elegant way from the divergence theorem (DT). Since all elastic forces act over
surfaces, then there must exist a tensor
ij
, called Caucy's stress tensor, such that
0 =
_
V
f
i
dV =
_
V
ij
dA
j
DT
=
_
V
ij
x
j
dV (4.77)
Tis holds for any volume V and thus

j
ij
= 0 (4.78)
Te detailed derivation above gave the same result and showed how to calculate and interpret in detail. A more
general derivation starts frommomentumconservation and considers Newton's second lawfor a small and arbitrary
material volume:
d
dt
_
V
u
i
t
dV
..
JdV
0
=const
=
_
V
g
i
dV +
_
V
ij
dA
j
(4.79)

2
u
i
t
2
= g
i
+

ij
x
j
Cauchy's momentum equation
(valid both for uids and solids)
(4.80)
Te condition = 0 thus arises as a steady state solution for Cauchy's equation without volume forces. With
the constitutive equation for LET between and , Cauchy's equation becomes the Navier equation.
We now formulate an energy equation from equation 4.80 by multiplying with
u
t
and integrating over a large
volume (without volume forces, g
i
= 0):
_
V
2
u
i
t
2
u
i
t
dV =
d
dt
_
V
1
2
u
t
2
dV
. .
kinetic energy T
=
_
V
ij
x
j
u
i
t
dV (4.81)
DT, PI
=
_
V
u
i
t

ij
dA
j
. .
surface traction term, can usually be neglected
_
V
ij
ij
t
dV
. .
bulk term describing the rate at which en-
ergy is stored in the material as it deforms
(4.82)
where we again used the divergence theorem (DT), partial integration (PI), the symmetry of the stress tensor
ij
=
ji
and the linear strain tensor
ij
=
1
2
(
u
i
x
j
+
u
j
x
i
) to get to the last line.
We introduce a scalar function w such that
w
ij
=
ij

T +
_
V
w
ij
ij
t
dV
. .
U
= 0 energy conservation: T +U = const (4.83)
dw =
ij
d
ij
strain energy density (4.84)
analogous to the energy stored in a stretched spring. U is the potential energy.
Principal stresses
For each point in the continuum body, the stress tensor describes its local stress state. Because is symmetric, it
can be diagonalized, giving three principal stresses
i
and the corresponding principal stress directions n
i
. Ten
s
i
= n
i
=
i
n
i
1 i 3 (4.85)
Tus for the directions, only normal and no shear forces are acting. We arrange the principal stresses such that
1

2

3
(ordered in rising magnitude).
We consider an arbitrary surface element with normal n. Ten the stress vector s = n has normal and tangential
components
s
n
= (s n)n s
n
= s n (4.86)
s
t
= s s
n
s
t
= [s
t
[ (4.87)
One can prove that all possible combinations of (s
n
, s
t
) are located in the marked area between the three Mohr's
circles in g. 4.11:
(s
n
)
max
=
3
(s
n
)
min
=
1
(s
t
)
max
=

3
1
2
(4.88)
Te eigenvalues of
ij
give upper bounds for maximal stresses which are the starting point for failure mecanics.
Next we observe that if all three shear components are zero (
xy
=
xz
=
yz
= 0) and all normal stresses are
equal (
xx
=
yy
=
zz
= p), then
= p1 (4.89)
In this case, p can be identied with the pressure. Tis motivates to identify
p =
1
3
tr() =
1
3
(
1
+
2
+
3
) (4.90)
h
= p1 hydrostatic stress tensor (4.91)
=
h
+
d
(4.92)
with
d
the deviatoric stress tensor.
Depending on the nature of the material under consideration, it might fail (break) if dierent stresses are exceeded.
For example for metals, the maximum shear is relevant, whereas ceramics have a threshold in extension.
In this context, ofen one considers the von Mises stress:
M
=
_
3
2
tr(
d

d
) =
_
1
2
(
1
2
)
2
+ (
2
3
)
2
+ (
3
1
)
2
(4.93)
Te von Mises stress is ofen used to color-circle the stress eld in a loaded piece of material.
s
n
s
t
2

3
1
Figure 4.11.: Te shaded area between the three Mohr's circles gives all possible combinations for (s
n
, s
t
).
39 4.5 Linear elasticity theory
4.5. Linear elasticity theory
For a purely elastic system, the deformation history is not relevant and therefore there must exist a constitutive
relation between stress and deformation gradient tensors
= (F) (4.94)
We also assume that the reference state is stress-free,
(F = 1) = 0 (4.95)
thus excluding pre-stressed material (typical for biomaterials, for example wood, carots or skin, which spring open
when being cut).
We rst consider linear elasticity theory (LET), where one assumes that
i
u
j
is small. Ten
ij
depends only on
the linear strain tensor
ij
=
1
2
(
i
u
j
+
j
u
i
) as
ij
= C
ijkl

kl
(4.96)
where C
ijkl
is the tensor of elastic moduli (of rank 4).
Te symmetry of
ij
and
kl
allows one to reduce the number of unknowns from 81 to 36. Te minimal number of
elastic moduli depends on the symmetry group of the material:
triclinic 21
hexagonal 5
cubic 3
isotropic 2
Isotropic LET
Te isotropic case can be introduced as follows. We decompose both stress and strain tensors into isotropic and
deviatoric parts:
= p1 +
d
=
1
3
tr()1 +
d
(4.97)
where p =
1
3
tr() is the hydrostatic pressure and tr() =
dV dV
0
dV
0
is the relative volume change. Linear elastic
isotropic behaviour then assumes linear relations between the corresponding parts:
p = K tr(),
d
= 2G
d
(4.98)
with K the compression or bulk modulus and G the shear modulus. Both are positive for thermodynamic stability.
Rewriting equation 4.97 with the help of equation 4.98 gives the 3d version of Hooke's law:
= K tr()1 + 2G (
1
3
tr()1)
. .
d
. .
d
(4.99)
= (K
2
3
G) tr()1 + 2G (4.100)
= tr()1 + 2
3d Hooke's law, gener-
alization of = E
(4.101)
with the Lam constants = K
2
3
G and = G being an alternative choice to (K, G). Tis choice corresponds
to
C
ijkl
=
ij
kl
+ 2
ik
jl
(4.102)
40 4.5 Linear elasticity theory
Te two terms represent the two possibilities to construct an isotropic tensor of rank 4 and explain why one has at
least two elastic constants.
Te relation between stress and strain tensors can be easily inverted:
ll
= 3
ll
+ 2
ll
= (3 + 2)
ll
(4.103)

ij
=
1
2
ij

2
1
(3 + 2)
ll
(4.104)
Our equilibrium condition for stress was
j
ij
+ g
i
= 0. We now can replace
ij
by
ij
and then
ij
by u
i
and
thus obtain an equation for the displacement eld u
i
:
ij
=
j
(
ll
ij
+ 2
ij
) =
i
ll
+ 2
j
ij
(4.105)
=
i
l
u
l
+ (
j
i
u
j
+
j
j
u
i
) (4.106)
= ( +)
i
(
l
u
l
) +
j
j
u
i
= g
i
(4.107)
In vector notation:
u + ( +)( u) +g = 0 (4.108)
where is the Laplace-operator and u couples dierent components of u. Tere are two elastic constants
because there are two ways to write a 2nd order derivative in 3d.
Due to the linearity of this ODE for u, it can be solved by a Green's function:
u
i
(r) =
_
dr
G
ij
(r r
)F
j
(r
) (4.109)
where the Green tensor G follows from solving
u + ( +)( u) = F(r) (no volume force) (4.110)
For the innite isotropic elastic space, the solution was given in 1848 by Lord Kelvin:
G
ij
=
1
8(2 +)
_
_
(3 +)
ij
. .
compression
+( +)
x
i
x
j
r
2
. .
shear
_
_
1
r
(4.111)
Note the
1
r
-scaling factor fromthe solution of the Laplace equation in 3d, implying a long-ranged kernel with strong
dependence on boundary conditions.
We nally give an expression for the strain energy density w:
dw =
ij
d
ij
= (
ll
ij
+ 2
ij
)d
ij
(4.112)
w =
1
2
(
ll
)
2
+
ij
ij
(4.113)
Tere are two elastic constants because there are two ways to contract
ij
to a scalar (in quadratic order).
41 4.6 Non-linear elasticity theory
F
1/r
Figure 4.12.: Te displacement decays with
1
/r and points move according to the change in angle.
4.6. Non-linear elasticity theory
For linear elasticity theory, the dierence between Lagrangian and Eulerian variables could be neglected. For non-
linear elasticity theory, when deformations can be large, we now have to confront the diculty that the balance of
stresses is performed in the deformed state, while the constitutive relation refers to the reference conguration.
In the deformed state, we have for the stress vector
ds = dA (4.114)
We have to relate dA to the surface element dA
0
in the undeformed state. We consider the deformation of a small
cylinder dened by dA
0
and dx
0
:
dA
0
n
dx
0
n
dx
dA
Figure 4.13.: For innitesimal deformations of the cylinder, both normal vector n and axis vector dx
0
change direc-
tions and lengths.
Te volumes are related by
dV = dx dA = J dV
0
= J dx
0
dA
0
= (F dx
0
) dA = dx
0
(F
T
dA) (4.115)
with J = det(F). Tis is valid for all dx
0
and hence
dA = J (F
T
)
1
dA
0
(4.116)
Via equation 4.114 we dene
ds = J (F
T
)
1
. .
P
dA
0
(4.117)
the rst Piola-Kicho stress tensor P. P describes the stress in the reference conguration. In contrast to the Cauchy
stress tensor , it is not symmetric.
Te second Piola-Kircho stress tensor
S F
1
P = J F
1
(F
T
)
1
(4.118)
is symmetric and can be shown to be energy-conjugate to the Green-Lagrange strain tensor E. It is therefore the
standard choice for a constitutive relation.
We now can write
ds
0
= S dA
0
(4.119)
in the undeformed state, where s
0
is the back-transformed stress vector. Note however that this is a mathematical
denition, because in physical reality, the stress tensor can only be measured in the deformed state.
We now revisit Cauchy's momentum equation and transfer it to the reference (Langrangian) frame. We again
start with Newton's second law:
d
dt
_
V
x
i
t
dV =
_
V
g
i
dV +
_
V
ij
dA
j
(4.120)
d
dt
_
V
0
x
i
t
J
..
0
dV
0
=
_
V
0
g
i
J
..
0
dV
0
+
_
V
0
J(F
T
)
1
. .
P
dA
0
(4.121)
DT
_
V
0
2
x
i
t
2

0
dV
0
=
_
V
0
g
i
0
dV
0
+
_
V
0
P
ij
x
0j
dV
0
(4.122)
V

0
2
x
i
t
2
=
0
g
i
+
P
ij
x
0j
Lagrangian form of
Cauchy's equation
(4.123)
In steady state and without volume forces the equilibrium condition = 0 is thus replaced by
0
P =
0
(F S) = 0.
NLET in 1D
We consider a unidirectional displacement u(x, t)
F =
_
_
1 +
u
x
0 0
0 1 0
0 0 1
_
_
E =
1
2
(F
T
F 1) (4.124)
E
xx
=
u
x
+
1
2
_
u
x
_
2
. .
geometrical NL
(4.125)
Te equilibrium condition is
P
xx
x
= 0 P
xx
= const. (4.126)
Te most general constitutive law is
S
xx
= (E
xx
) (4.127)
with some function (possibly non-linear, a material NL).
P
xx
= const = (F S)
xx
=
_
1 +
u
x
_
_
u
x
+
1
2
_
u
x
_
2
_
(4.128)
Tis equation is both geometrically and mechanically non-linear. For mechanical linearity, one has
(E
xx
) = ( + 2)E
xx
(4.129)
P
xx
=
_
1 +
u
x
_
( + 2)
_
1 +
1
2
_
u
x
__
u
x
(4.130)
a geometrically non-linear, cubic equation for
u
x
. In LET this reduces to
P
xx
= ( + 2)
u
x
(4.131)
P
xx
= det J
xx
1
1 +
u
x

xx

xx
= ( + 2)
xx
1d LET (4.132)
NLET in 3D
For LET in 3D, we have assumed
= (F), (F = 1) = 0 (4.133)
For NLET in 3D, one can argue that the corresponding assumption should be
S = S(C), S(C = 1) = 0 (4.134)
Assuming further that the material is locally isotropic, one can show that S has to have the form
S(C) =
0
(I
1
, I
2
, I
3
)1 +
1
(I
1
, I
2
, I
3
)C +
2
(I
1
, I
2
, I
3
)C
2
(4.135)
where
0
,
1
,
2
are functions of the three invariants of C:
I
1
= tr(C) =
2
1
+
2
2
+
2
3
(4.136)
I
2
=
1
2
_
(tr(C))
2
tr(C
2
)
_
=
2
1
2
2
+
2
2
2
3
+
2
3
2
1
(4.137)
I
3
= det(C) =
2
1
2
2
2
3
(4.138)
Tere are no higher order terms of C due to the Cayley-Hamilton theorem from Linear Algebra. Te
2
i
are the
eigenvalues of C. In LET, the
i
become the stretch ratios. Terefore an isotropic constitutive relation amounts to
specifying the three scalar functions
1
,
2
,
3
.
S(1) = 0 leads to the condition
0
(3, 3, 1) +
1
(3, 3, 1) +
2
(3, 3, 1) = 0 (4.139)
Choosing arbitrary
i
usually leads to models with unphysical behaviour (e.g. material which can be used as a
limitless energy source during cyclic deformation). Te best solution to this problem is the use of appropriate
strain energy density functions (hyperelastic materials).
Energy equation for NLET
We again multiply the momentum equation by the velocity and integrate over a large material volume V
0
:
_
V
0
2
x
i
t
2
x
i
t
dV
0
=
_
V
0
0
g
i
x
i
t
dV
0
+
_
V
0
P
ij
x
0j
x
i
t
dV
0
(4.140)
DT, PI
d
dt
_
V
0
0
2
_
x
i
t
_
2
dV
0
. .
rate of change in
kinetic energy =

T
+
_
V
0
P
ij
t
_
x
i
x
0j
_
. .
F
ij
dV
0
. .
rate at which elastic
energy is stored in
the material =

U
(4.141)
=
_
V
0
0
g
i
x
i
t
dV
0
. .
rate of work by
body forces
+
_
V
0
x
i
t
P
ij
dA
j
. .
rate of work on the
surface
(4.142)
We postulate the existence of a strain energy density w(F
ij
) such that
P
ij
=
w
F
ij

U =
d
dt
_
V
0
w dV
0
(4.143)
U is the rate of change in elastic energy stored int the material. We also need to require w to have a minimum for
F = 1. If w is a function only of C, then one can show that
S
ij
= 2
w
C
ij
(4.144)
For isotropic material, w = w(I
1
, I
2
, I
3
), it holds that
S
ij
= 2
w
I
k
I
k
C
ij
(4.145)
Tis leads to an explicit procedure to calculate
0
,
1
,
2
from a given w.
Using minimization of an energy functional is computationally much easier than solving the non-linear PDEs
following from
0
(F S) = 0. Tis procedure naturally leads to the nite element method (FEM).
For hyperelastic material (a typical example is rubber, which can have very large deformations), one usually assumes
incompressibility, that is I
3
= 1.
Te commonly used constitutive relations are:
(a) Neo-Hookean: w =

2
(I
1
3)
Te Neo-Hookean description is good for plastic and rubber up to 20% strain. is the classical shear modulus
known from LET.
(b) Mooney-Rivelin: w = C
1
(I
1
3) +C
2
(I
2
3)
Te Mooney-Rivelin formulation becomes Neo-Hookean with C
2
= 0 and is good for rubber up to 100%
strain.
(c) Ogden: w =
N
p=1
p
p
(
p
1
+
p
2
+
p
3
3)
Te Ogden relation is a generalization of 1 and 2 in the sense that it contains both the Neo-Hookean and
the Mooney-Rivelin description. Usually N = 3 with 6 independent parameters gives a very good t to
experiments. Te relation to LET is such that
N
p=1
p
= 2.
Te results of a typical stress test with uni-axial loading are visualized in g. 4.14.
LET
Neo-Hookean
MR
F
Ogden 1<<2
F
Ogden >2
Ogden <1
Figure 4.14.: In each case the slope of the curve corresponds to the elastic constant of the material probed. Neo-
Hookean behaviour with strain-sofening is typical for synthetic polymer networks, whereas strain-
stiening (Ogden > 2) is typical for biopolymer networks. Material instability or failure behavior
occurs in the Ogden model for < 1.
Example: Blowing up a balloon
Balloons as rubber-like material with very large displacements are ideal examples for NLET.
We consider a thin, spherical, incompressible rubber membrane of initial radius R and thickness H R. Te two
angular stretches have to be the same (biaxial loading in the sheet):
=
2r
2R
=
r
R
(4.146)
Te normal stretch is determined by incompressibility:
I
3
=
2
2
r
= 1
r
=
R
2
r
2
(4.147)
A typical value for the classical shear modulus is = 0.4 MPa. Tis denes the Neo-Hookean model.
For Mooney-Rivelin one can use C
1
= 0.44, C
=
C
2
C
1
=
1
7
.
For the Ogden relation one can use N = 3 and
1
= 1.3
1
= 0.6 MPa (4.148)
2
= 5.0 = 0.01 MPa (4.149)
3
= 2.0 = 0.01 MPa (4.150)
All three material laws give similar results for this example.
We now consider that the balloon is inated by some internal pressure p. Tis gives rise to a Laplace relation
d(E
p
) = d(E
T
) d(
4
3
r
3
p) = d(4r
2
T) (4.151)
pr
2
dr = T2rdr p =
2T
r
(4.152)
For Mooney-Rivelin, one can combine these elements to show:
p =
4C
1
H
R

(1 +C
)(
6
1)
(4.153)
which is only dened for
1.
Equation 4.153 is ploted in g. 4.15.
R
H
P
Figure 4.15.: Te pressure required to inate a balloon with radius R much larger than thickness H initially in-
creases, but then decreases. Tis corresponds to the familiar experience that blowing up a balloon
becomes easier afer an initial barrier.
CHAPTER 5
Applications of LET
Te present chapter deals with solution strategies for simple solid mechanics problems in LET. Exact analytical
solutions to realistic problems are ofen not possible. However, a lot can already be deduced from models where
simplyfying assumptions are made. Before addressing the actual problems, we will quickly recapitulate the most
important concepts of LET.
5.1. Reminder on isotropic LET
In LET we have two elastic constants and we have already encountered two dierent possible choices:
(K, G) the bulk and shear moduli
(, ) the Lam coecients
which are related by = K
2
3
G, = G.
Te three main concepts are the displacement vector eld u(x) and the tensors for strain and stress,
ij
and
ij
.
Teir interconncection is shown in g. 5.1:
displacement
u(x)
strain tensor
= 1/2 ( u + u )
ij i j j
i
stress tensor
= + 2
ij ll ij ij

= -
ij
1
2
ij
2 (3 + 2 )
ll
integration
inversion
inversion:
Figure 5.1.: Overview of the three important concepts of LET. Te integration to get back the displacement eld is
ofen not trivial, due to the partial derivatives. Physical intuition is needed.
Te equilibrium condition for stress and displacement respectively is
0 = +g = u + ( +) ( u) +g (5.1)
and the strain energy density w is given by
w =
ij
ij
=
1
2
(
ll
)
2
+
ij

ij
(5.2)
46
47 5.2 Pure Compression
5.2. Pure Compression
As a rst example, we look at a case where the material is compressed and there are no shear forces, as illustrated
in g. 5.2. Te stress tensor is then given by
ij
= p
ij
(5.3)
where the pressure p is related to the relative volume change by
dw =
ij
d(
ij
) = p d(
ll
) = p
dV
V
0
(5.4)
ll
= 3p = 3K
ll

1
K
=
1
V
0
V
p
(5.5)
Tat the volume change should be negative as pressure increases is a familiar result from thermodynamics. Te
bulk modulus K is the isothermal compressibility and has to be positive, K > 0, as a stability criterium.
Figure 5.2.: We take a piece of material and compress it equally from all sides. Tere are only normal forces acting.
5.3. Pure shear
As a second example we look at a plate under force F and with area A leading to pure, one-dimensional shear in
the x-direction (g. 5.3). Te stress is the force per area s =
F
A
and the stress tensor is now given by
=
_
_
0 s 0
s 0 0
0 0 0
_
_
(5.6)
With vanishing trace,
ll
= 0, the inverted relation is obtained easily and the strain tensor has the same symmetrical
structure:
=
_
_
0
s
2
0
s
2
0 0
0 0 0
_
_
(5.7)
A
F
Figure 5.3.: Shear forces acting on a plate of area A. Te deformed state is characterized by the shear angle .
48 5.4 Uni-axial stretch
Te displacement eld corresponding to these stress and strain tensors is then
u =
_
sy
, 0, 0
_
=
u
1
y
=
s
(5.8)
Te displacement has a linear prole in the x-direction and is dependent only on the shear modulus and not on
the bulk modulus. Te larger the shear modulus, the smaller the displacement and the shear angle .
5.4. Uni-axial stretch
A
F
F
0
z
Figure 5.4.: Illustration of uni-axial stretching. Te material is stretched in the z-direction.
In the case of uni-axial stretching (g. 5.4), we again know the stress p =
F
A
and can immediately write for the
stress tensor
zz
= p,
ij
= 0 for all other components (5.9)
In order to construct the strain tensor, we procede component-by-component:
zz
=
1
2
zz

2(3 + 2)

ll
..
=zz
=
( +)
(3 + 2)
p
p
E
(5.10)
xx
=
yy
=

2(3 + 2)
p
zz
=
E
p (5.11)
Here we have dened a new set of elastic constants:
E =
(3 + 2)
( +)
Young's modulus (5.12)
=

2( +)
Poisson's ratio (5.13)
Te strain tensor and matching displacement are then given by
=
p
E

_
_
0 0
0 0
0 0 1
_
_
, u =
p
E
_
_
x
y
z
_
_
(5.14)
ll
=
p
E
(1 2) (5.15)
49 5.4 Uni-axial stretch
E [Pa] material
TPa graphene
GPa crystals
MPa rubber
kPa cells
Table 5.1.: Typical values for the Young's modulus for dierent types of materials
Te Young's modulus is commonly stated in the literature to characterize the stiness or rigidity of a given material.
Some typical values are given in table 5.1.
Te Poisson ratio has no physical unit and describes the coupling between dierent directions in the material.
It is the ratio between longitudinal expansion and lateral contraction. In the example of uni-axial stretch, the
displacement in x- and y- direction is negative and the material moves in from the sides when being stretched
(Poisson eect).
Since bulk and shear moduli must be positive for thermodynamic reasons, you can also convince yourself that the
values for the dimensionless Poisson's ratio lie in a narrow range:
G, K > 0 1 < <
1
2
(5.16)
Te upper bound =
1
2
corresponds to the limit (K ) and the material becomes incompressible
(
ll
=
p
E
(1 2) = 0). Tis is the case for most biomaterials which are incompressible due to the large amount of
water in the material (volume conservation in biological systems).
For negative Poisson's ratio, < 0, so-called auxetic materials, when expanded in one direction, they also expand
in the other directions (take for example a crumpled piece of paper and expand it uni-axially).
(E, ) is an alternative choice to (, ) or (K, G)
=
E
(1 +)(1 2)
, =
E
2(1 +)
(5.17)
With this choice, the constitutive relation is given by
ij
=
1 +
E

ij
ll
ij
(5.18)
and the strain energy density reads
w =
p
2
2E
compare spring: U =
F
2
2k
(5.19)
50 5.5 Biaxial strain
x
y
z
Figure 5.5.: Te plate is strained in the (x,y)-plane.
5.5. Biaxial strain
Compression and shearing forces act as shown in g. 5.5. We assume linear in-plane deformations and shrinking
by a factor in the z-direction:
u =
_
_
ax +by
cx +dy
z
_
_
=
_
_
a
1
2
(b +c) 0
1
2
(b +c) d 0
0 0
_
_
(5.20)
ij
and
ij
are both constant.
By a linear mapping we get for the stress tensor (assuming free surfaces on the top and botom):
zz
=
ll
+ 2
zz
= (a +d ) + 2() = 0 (5.21)
=
(a +d)
( + 2)
= (

1
)(a +d) (5.22)
=
_
_
_
E(a+d)
1
2
E(b+c)
2(1)
0
E(b+c)
2(1)
E(a+d)
1
2
0
0 0 0
_
_
_ (5.23)
If no force is applied in the y-direction:
yx
=
xy
=
yy
= 0
d = a, c = b (5.24)
=
_
_
Ea 0 0
0 0 0
0 0 0
_
_
, =
_
_
a 0 0
0 a 0
0 0 a
_
_
(5.25)
We recover the solution for uni-axial stretching for E a = p.
Alternatively, one can also obtain displacement only in the (x,z)-plane, no displacement in the y-direction:
b = c = d = 0 =
_
_
Ea
1
2
0 0
0
Ea
1
2
0
0 0 0
_
_
(5.26)
Tus, a transverse stress
yy
must be applied to prevent the plate from contracting in the y-direction as we stretch
in x-direction. We obtain an eective elastic modulus
E
1
2
> E, so that 2d stretching is more strenuous than
uni-axial stretching.
51 5.6 Elastic cube under its own weight
L
z
0
g
F
Figure 5.6.: Elastic cube with dimenson L subject to gravity. Homogeneous stress at the top surface is holding the
cube.
5.6. Elastic cube under its own weight
Te cube and corresponding stresses are illustrated in g. 5.6. Te cube under its own weight is subject to the body
force f
i
= g
iz
. We ask for stress free boundaries at the sides and botom, n = 0, but hold the cube from
above by a homogeneous surface stress.
Te equilibrium condition is given by the steady Navier equation with constant gravity
ij
+f
i
= 0 (5.27)

zz
z
= g
zz
= gz (5.28)
Gravitation is balanced everywhere in the material and the total force on the upper surface, gL L
2
, exactly
balances the overall gravitation.
Strain tensor and displacement eld are given by
=
_
_
E
gz 0 0
0
E
gz 0
0 0
g
E
z
_
_
u =
_
_
g
E
xz
g
E
yz
g
2E
(z
2
+(x
2
+y
2
))
_
_
(5.29)
Upper and lower surfaces become parabolic and the cube, with the point (0, 0, 0) xed in space, broadens from top
to botom (see g. 5.7).
Figure 5.7.: Image of the deformed cube from one side. Te cube is xed at the point (0, 0, 0). Again the Poisson
eect brings the sides of the cube in, but lower and upper side are parabola shaped. Gravity lets the
system sag down.
52 5.7 Torsion of a bar
D
x
y
z
Figure 5.8.: A moment couple applied to a general elastic rod with domain D. Due to symmetry there will always
be one plane without movement, here at z = 0.
5.7. Torsion of a bar
We twist a bar of arbitrary cross-section D by applying moments to its ends and make the following ansatz for the
displacement vector:
u =
_
_
yz
xz
(x, y)
_
_
(5.30)
Hence we assume a rotation, with representing the twist of the bar, in the (x,y)-plane, and translational invariance,
with the torsion function yet to be determined, in the z-direction.
Te strain and stress tensors are then given by
=
_
_
0 0

2
(
x
y)
0 0

2
(
y
+x)
2
(
x
y)

2
(
y
+x) 0
_
_
= 2 (5.31)
Te trace of the strain tensor vanishes,
ll
= 0, so that this is a pure shear experiment. Te stress tensor has the
same structure as the strain tensor.
Te condition for equilibrium is again given by the Navier equation:
ij
= (
2
x
+
2
y
) = 0 (5.32)

2
= 0
has to satisfy the
Laplace equation on D
(5.33)
We parameterize the boundary D as
_
X(s)
Y (s)
_
. Te normal in the (x, y)-plane is given by n =
_
Y
_
.
Assuming stress-free boundary conditions, we get
zj
n
j
=
_
(
x
Y )Y
(
y
+X)X
= 0 (5.34)

x
Y
y
X
= n (5.35)
=
n
=
1
2
d
ds
(X
2
+Y
2
) on D (5.36)
Te solution (x, y) for this Neumann problem is unique up to an arbitrary constant corresponding to an arbitrary
uniform translation.
53 5.7 Torsion of a bar
Once the solution is known, the moment applied at each end of the bar is given by
M =
_
D
dxdy r F =
_
D
dxdy
_
_
x
y
0
_
_
_
_
xz
yz
0
_
_
(5.37)
=
_
D
dxdy
_
_
0
0
(x
yz
y
xz
)
_
_
(5.38)
M = M
z
=
_
D
dxdy
_
x
y
y
x
+ (x
2
+y
2
)
. .
R torsional rigidity
(5.39)
Te torsional rigidity R is the factor of proportionality between moment M and twist , analogue to the spring
constant.
For simple cross-sectional shapes, R can be calculated analytically. Te simplest case is the circular bar. Ten, D
is a disc of radius a and our Neumann problem is
=
1
r
r
(r
r
) = 0, r < a (5.40)
n
= 0, r = a (5.41)
We then nd
= const (5.42)
R = 2
a
_
0
rdr r
2
=
a
4
2
= R (5.43)
Te torsional rigidity of a circular bar is linear in the shear modulus , as expected in LET, but increases with the
4
th
power of its radius, showing a large geometrical dependence! Tis is formally the same problem as the Hagen-
Poiseuille law for viscous uid ow due to a pressure gradient through a pipe of radius a in hydrodynamics.
Only for the circular bar we get
n
= 0 and = const. For all other cases, one would get more complicated
boundary conditions and a non-trivial component u
z
= (x, y).
We now introduce an alternative and more general way to solve this problem.
Due to the special form of the stress tensor, the steady Navier equation reads
zx
+
y
zy
= 0 (5.44)
Tis can be satised by postulating the existence of a stress function (x, y) such that
zx
=
y
,
zy
=
x
(5.45)
Te factors of are introduced for later convenience. has the role of a scalar potential and is dened up to
addition of an arbitrary constant.
Comparing this with
zx
= (
x
y),
zy
= (
y
+x) (5.46)
we can relate to by
x
=
y
+y,
y
=
x
x (5.47)
= (
x
y
1) + (
y
x
1) = 2 (5.48)
54 5.8 Contact of two elastic spheres (Hertz solution 1881)
Tus satises Poisson's equation
= 2 (5.49)
in the domain D.
Te zero-stress boundary condition now reads
0 =
zx
Y
zy
X
= (
y
Y
+
x
X
) =
on D (5.50)
= const = 0 on Dwithout loss of generality (5.51)
Tus the Neumann problem for has now been converted into a Dirichlet problem for , which is easier to solve.
For the torsional rigidity R we nd:
R =
_
D
dxdy (x
zy
y
zx
) (5.52)
=
_
D
dxdy (x
x
y
y
)
. .
=(r)+2
= 2
_
D
dxdy (5.53)
In application to a circular bar this becomes:
=
1
r
r
(r
r
) = 2, = 0 for r = a (5.54)
=
a
2
r
2
2
R = 2
a
_
0
(a
2
r
2
)rdr =
a
4
2
(5.55)
We recover the same result as in the ansatz with the torsion function above!
5.8. Contact of two elastic spheres (Hertz solution 1881)
We consider two elastic spheres of radii R and R
, which are pressed onto each other by a force F (see g. 5.9).

Teir elastic constants are (E, ) and (E
), respectively. Te undeformed spheres around the point of contact

have the shapes
z = r
2
, z
r
2
(5.56)
with =
1
2R
the mean curvature and the radial distance r
2
= x
2
+y
2
.
Te indentation length is denoted by h. We then have (see Fig. 5.9)
(z +u
z
) + (z
+u
z
) = h (5.57)
( +
)
. .
A
r
2
+u
z
+u
z
= h (5.58)
within the contact area. For symmetry reasons, this has to be a circular disc of radius a.
We assume that only a normal stress p
z
(x, y) acts inside the contact area. Te resulting displacement elds are
obtained from the Green's function for a surface force acting on an elastic halfspace (Boussinesq solution):
u
z
(x, y) =
1
2
E
_
dx
dy
p
z
(x
, y
)
s
(5.59)
u
z
(x, y) =
1
2
E
_
dx
dy
p
z
(x
, y
)
s
(5.60)
s =
_
x
y
_
_
x
=
_
(x x
)
2
+ (y y
)
2
(5.61)
z
x,r
R, (E,)
R', (E',')
h
x,r
z
u'
z
u
z
z
-a a
z'
Figure 5.9.: Te Hertz problem is a mixed boundary contact problem with restrictions on u and . Two elastic
spheres are pressed onto each other by a force F. If they would not feel each other, the spheres would
move into each other with indentation depth h. In reality, the spheres develop a contact area A, a
circular disc of radius a. In general, the contact area does not need to be at as ahown here. For
(R
, E
) , the problem becomes one of pressing an elastic ball onto a at rigid substrate.
Note the
1
/r-relation typical for 3d LET. From (5.58) we now get
1
_
1
2
E
+
1
2
E
_
. .
4
3
D
_
dx
dy
p
z
(x
, y
)
s
= h Ar
2
(5.62)
Tis integral equation determines the stress distribution p
z
(x, y) in the contact area. We solve this by noting that
it corresponds to the potential of a uniformly charged disc known from electrostatics:
p
z
(r) =
3F
2a
2
_
1
_
r
a
_
2
Hertz-stress (5.63)
Te stress is normalized such that
_
dx
dy
p
z
(x
, y
) = F (5.64)
Te maximal stress (at r = 0) is
3F
2a
2
=
3
2
(
F
a
2
)
. .
average stress
(5.65)
Combining the expression for the Hertz-stress with the integral equation (equation 5.63) allows us to express in-
dentation length h and contact area with radius a as a function of the force F (this requires more integrals from
potential theory). One then nds:
a
3
= FD
RR
R +R
, h
3
= F
2
D
2
(
1
R
+
1
R
) (5.66)
Te Hertz law a F
1
/3
and the relation h F
2
/3
for a spherical indenter are the most famous results of contact
mechanics.
spherical indenter conical indenter flat indenter
h~F
2/3
h~F
1/2
h~F
Figure 5.10.: A standard way to measure the rigidity of a material is by means of measuring the indentation length
h as a function of force F.
From F =
U
h
we obtain for the potential energy:
U =
2
5
h
5
/2
1
D
_
RR
R +R
_
1
/2
(5.67)
In the limit R
: a
3
= FDR, h
3
= F
2 D
2
R
, U =
2
5
h
5
/2
R
1
/2
/D.
Tese relations are ofen used to measure the stiness of a material, for example by placing a steel ball of ra-
dius R on a material with modulus E
(E , F = G =
4R
3
3
g) or by indenting with an AFM.
Te scaling laws a F
1
/3
and h F
2
/3
hold true also for non-spherical indenters with nite curvature (but not
for conical or at indenters), compare Fig. 5.10.
Here is a simple way to predict the scaling law for the spherical indenter (see Fig. 5.11):
U V E
2
(5.68)
where V is the deformed volume and is the strain. Te indentation length is usually the quantity monitored.
From Fig. 5.11 we obtain:
R
2
= l
2
+ (R h)
2
l
Rh (5.69)
V l
3
,
h
l
(5.70)
Ten, force and potential energy scale like
U (Rh)
3
/2
E
_
h
(Rh)
1
/2
_
2
= E R
1
/2
h
5
/2
(5.71)
F
U
h
E R
1
/2
h
3
/2
(5.72)
57 5.9 Compatibility conditions
h
R
l
Figure 5.11.: Illustration of a spherical indenter.
5.9. Compatibility conditions
Te steady Navier equation
j
ij
+ g
i
= 0 can be regarded as 3 equations for the 3 displacment components u
i
.
However, the strain tensor
ij
has 6 components, correspondong to 6 equations for u if
ij
and the constitutive
equations are known. Terefore
ij
is actually overdetermined and has to satisfy 3 additional requirements , the 3
compatibility conditions.
Only if these conditions are satisied, then
ij
corresponds to a single-valued, and thus physically acceptable,
displacement eld.
We rst consider a plain strain problem, that is u = (u
x
(x, y), u
y
(x, y), 0). If u
x
and u
y
are twice continuously
dierentiable single-valued functions:
y
u
x
=
y

x
u
x
..
xx
,
x
y
u
y
..
yy
=
y
x
u
y
(5.73)

2
y
xx
+
2
x
yy
=
y
x
(
y
u
x
) +
x
y
(
x
u
y
) =
x
y
(
y
u
x
+
x
u
y
) = 2
x
xy
(5.74)

2
y
xx
+
2
x
yy
2
x
xy
= 0 (5.75)
Te same considerations for the (x, z)- and (y, z)-planes give
2
z
yy
+
2
y
zz
2
y
yz
= 0 (5.76)
2
x
zz
+
2
z
xx
2
z
zx
= 0 (5.77)
One can derive 3 more compatibility conditions, but only 3 out of the 6 are independent. Te rst equation from
above (equation 5.76) in linear elasticity can be writen in terms of stresses as
2
y
xx
+
2
x
yy
2
x
xy
=

1 +
(
2
x
+
2
y
) tr() (5.78)
and similar expressions result from the other 5 compatibility conditions in strain.
5.10. Bending of a plate
We consider a weakly bent plate with the surface normal in z-direction and a thickness t which is much smaller
than the lateral extension (see Fig. 5.12). Te upper side is in compression and the lower side is in tension. No
stresses thus exist in the neutral middle surface, which we describe by the height function h(x,y). To rst order in
h, the displacement vector for the middle surface is
u
0
=
_
_
0
0
h(x, y)
_
_
Monge parametrization (5.79)
At the botom and top surfaces we have the boundary conditions:
xz
=
yz
=
zz
= 0 (5.80)
58 5.10 Bending of a plate
z, h
x
t
tension
compression
Figure 5.12.: Bending of a thin plate in z-direction. Tere must exist a plane with neither compression nor tension
which denes a neutral surface.
Because the plate is very thin, these equations must be valid also in its interior

zx
=
E
1 +
zx
=
E
2(1 +)
_
u
x
z
+
u
z
x
_
= 0 (5.81)
zy
=
E
1 +
zy
=
E
2(1 +)
_
u
y
z
+
u
z
y
_
= 0 (5.82)
zz
=
E
(1 +)(1 2)
[(1 )
zz
+(
xx
+
yy
)] = 0 (5.83)
Since
ij
=
1
E
[(1 +)
ij
ll
ij
] (5.84)
the inverted expression gives the strain tensor
ij
=
E
(1 +)
ij
+
E
(1 +)(1 2)
ij
ll
(5.85)
and with
u
x
z
=
u
z
x
=
x
h u
x
= z
x
h (5.86)
u
y
z
=
u
z
y
=
y
h u
y
= z
y
h (5.87)
zz
=

(1 )
(
xx
+
yy
) (5.88)
we now know the strain tensor:
ij
=
_
_
z
2
x
h z
x
y
h 0
z
x
y
h z
2
y
h 0
0 0
z
(1)
(
2
x
h +
2
y
h)
_
_
(5.89)
Note that for
31
we used, to rst order in h, that
31
=
1
2
(
x
u
z
+
z
u
x
) =
1
2
(
x
h
x
h) = 0 (5.90)
59 5.10 Bending of a plate
Te strain energy density w leads to the bending energy of the plate:
w =
E
2(1 +)
(
2
ij
+

1 2
2
ll
) (5.91)
= z
2
E
(1 +)
_
_
2
(1 )
(
1
2
(
2
x
h +
2
y
h)
. .
H mean curva-
ture
)
2
2
x
h
2
y
h (
x
y
h)
2
_
. .
K Gaussian cur-
vature
_
_
(5.92)
Mean curvature and Gaussian curvature are denitions from dierential geometry
H =
1
2
(
1
+
2
) mean curvature (5.93)
K =
1

2
Gaussian curvature (5.94)
where
1
and
2
are the principal curvatures of the surface. Tese geometric quantities are fundamental in all
theories of plates and shells.
For a thin plate, we intend to integrate out the z-component and obtain for the net strain energy:
U =
t
/2
_
t
/2
dz
_
dxdy w (5.95)
= 2
_
dxdy H
2
. .
bending energy
+
_
dxdy K
. .
= const, due to
Gauss-Bonnet the-
orem
(5.96)
Te bending stiness is then given by
=
E
(1
2
)
t
/2
_
t
/2
dz z
2
=
Et
3
12(1
2
)
= (5.97)
Again, the bending rigidity (in units of energy) shows a strong dependence on geometry, in terms of the thickness,
of the material. = (1 ) is another curvature elastic constant of interest, called the splay modulus.
Minimizing U a a functional of h using the calculus of variation is non-trivial but gives a clear result:
4
h = tg biharmonic equation (5.98)
In addition to derivatives of 4
th
order, the boundary conditions for the plate problem with body forces are non-
trivial and simplest for the clamped case (h =
n
h = 0 at the rim).
60 5.11 Bending of a rod
(E, )
g g
-R R
Figure 5.13.: Te plate is clamped at both sides and gravity pulling everywhere lets it sag down.
Example: Circular plate with gravity
For example, in order to describe the sagging of an elastic disc of radius R under gravity with clamped boundary
conditions, (5.98) reads
4
h =
_
1
r
d
dr
(r
d
dr
)
_
2
h = 64 with =
3g(1
2
)
16t
2
E
(5.99)
h = r
4
+ar
2
+b +cr
2
ln(
r
R
) +d ln(
r
R
) (5.100)
We make a polynomial ansatz for the height function h(r) in cylindrical polar coordinates and demand d = c = 0
to avoid singularities. Te constants a and b are calculated from the boundary conditions h =
r
h = 0 at r = R
and we nd that
h = (R
2
r
2
)
2
(5.101)
Te height of the clamped plate varies parabolically in r (compare Fig. 5.13).
A tecnical application of the mechanics of bending a plate is, for example, the manufacture of curved wind
screens. A glass plate is heated with an inhomogeneous temperature eld to achieve the required sag.
5.11. Bending of a rod
Like for the plate, we have tension on one side, compression on the other, and a neutral surface inbetween. For
weak bending, torsion is a higher order eect and can be neglected. We choose the z-axis for the long axis of the
rod. In this case
n
z
= 0
ix
n
x
+
iy
n
y
= 0 (5.102)
for example, for i = x. At point P we have
n
y
= 0
xx
= 0 (5.103)
z
x
R
tension
compression
Figure 5.14.: Bending of a rod. Basically this is a stretch experiment with tension for x > 0 and compression for
x < 0.
y
x
(a)
y
x
u
y
u
x
(b)
Figure 5.15.: Te sides of the rectangular cross-section are tilted with linear dependence in x-direction, but remain
planar, whereas the upper and lower planes deform parabolically.
everywhere since the rod is thin.
ij
vanishes except for
zz
and we only have tension or compression along the
z-axis. Basically this is a stretch experiment.
For the stretch in x-direction we get
dz
dz
=
2(R +x)
2R
= (1 +
x
R
)
dz
dz
dz
=
zz
=
x
R
(5.104)
We then nd for the displacement eld and stress tensor:
=
_
_
x
R
(y+y)
R
= 0
(zz)
R
= 0
0
x
R
0
0 0
x
R
_
_
u =
1
R
_
_
1
2
(z
2
+(x
2
y
2
))
xy
xz
_
_
(5.105)
where we used the same intermediate steps as in section 5.10 for the bending of a plate, respectively
zz
= E
x
R
(5.106)
xx
=
yy
=
zz
=
x
R

x
u
x
=
y
u
y
=
x
R
(5.107)
A cross-section at constant z = z
0
has u
z
= z
0
x
R
. It stays planar but is rotated (except at the origin), as shown in
Fig. 5.15a.
However, the shape of the cross-section is changed as shown in Fig. 5.15b. For example for a rectangular cross-
section at y = y
0
the two sides stay planar, but are rotated:
u
y
=
y
0
R
x (5.108)
At x = x
0
, the top and botom sides become parabolic:
u
x
=
1
2R
(z
2
0
+(x
2
0
y
2
)) (5.109)
From the strain energy density w we now obtain a strain energy per length
U
L
:
w =

ik
ik
2
=

zz
zz
2
=
Ex
2
2R
2
(5.110)
U
L
=
E
2R
2
_
x
2
dA
. .
Iy
(5.111)
where I
y
is the moment of inertia with respect to the y-axis. For a rectangular cross-section with dimensions a
and b in x- and y-directions, respectively, we have I
y
=
a
3
b
12
. For a circular cross-section we have I
y
=
R
4
4
.
For the whole rod we now have
U =
EI
2
_
ds
1
R
2
=
EI
2
_
ds
_
d
2
r
dz
2
_
(5.112)
where I is the moment of inertia with regard to the axis around which we bend. Tis is the basis of the worm-like
cain model for polymers. = E I is called the bending stiness which is related to the persistence length l
p
, i.e.
the length on which the rod stays bend, by = l
p
k
B
T.
For bending in a plane, r = (x, y, 0) and the corresponding Euler-Lagrange equation gives
EI
2
_
d
2
ds
2
/
X
_
= EIX
= K
x
(5.113)
EI
2
_
d
2
ds
2
/
Y
_
= EIY
= K
y
(5.114)
where K
x
, K
y
are external forces per length. Tus for weakly bent rods, we have to solve a dierential equation
of 4
th
order, like for weakly bent plates.
Example 1:
An initially horizontal rod is clamped at s = 0 and free at s = L. How does it deform under its own weight?
Y
=
g
EI
Y =
g
24EI
s
2
(s
2
4Ls + 6L
2
) Y (L) =
gL
4
8EI
(5.115)
Again we encounter a strong dependence on geometry. Cross-checking the solution depicted in Fig. 5.16 gives the
correct results:
Y
=
g
EI
Y (0) = Y
(0) = 0 (5.116)
moment M
x
(L) = EIY
(L) = 0 (5.117)
Y(L)
g
g
L
x
z
r(s)
Figure 5.16.: Clamped rod subject to gravity.
Example 2:
Now the rod is deformed by a point force F at its free end:
Y =
F
6EI
s
2
(3L s) Y (L) =
FL
3
3EI
=
4FL
3
3ER
4
(5.118)
where the last formula is valid for a spherical cross-section with radius R. Te proportionality factor between force
and deformation is the spring constant of the rod, given by
k =
3ER
4
4L
3
(5.119)
where the rod radius R enters to 4
th
power. Cross-checking the solution gives
Y
= 0, Y (0) = Y
(0) = 0 (5.120)
M
x
(L) = 0, F
y
= EIY
= F (5.121)
F
L
Figure 5.17.: Clamped rod subject to a point force at its end.
CHAPTER 6
The Finite Element Method (FEM)
Te FEM is the standard choice to solve partial dierential equations (PDEs), although several other methods exist
(for example nite dierences, boundary element method, nite volumes, spectral method, etc.). In contrast to
ordinary dierential equations (ODEs), there is no general mathematical theory for the solvability of PDEs. Rather
dierent numerical schemes have been developed for dierent classes of PDEs. In each case, one has to check for
which assumption a solution exists, if it is unique, and how it depends on the parameters (existence, uniqueness,
robustness).
6.1. Classification of PDEs
A linear PDE of second order has the form (a
ij
= a
ji
since
j
i
u =
i
j
u):
Lu = a
ij
i
u +b
i
i
u +cu = f (6.1)
It is called
(a) elliptic if all eigenvalues of a are non-zero and have the same sign. For example: Laplace equation
u = 0 a =
_
_
1 0 0
0 1 0
0 0 1
_
_
(6.2)
(b) hyperbolic if all eigenvalues are non-zero, n 1 have the same sign and the remaining one the opposite sign.
For example: Wave equation
2
t
u = u a =
_
_
_
_
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
_
_
_
_
(6.3)
(c) parabolic if one eigenvalue is zero and the remaining ones have the same sign. For example: Heat equation
t
u = u a =
_
_
_
_
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 0
_
_
_
_
(6.4)
64
65 6.2 The weak form
Tese names are taken from the case n = 2 when the quadratic form a
11
x
2
+ 2a
12
xy + a
22
y
2
= 0 describes
an ellipse, a hyperbola or a parabola. For each of these types and depending on boundary conditions, theorems
on existence and uniqueness and robustness can be derived. However, for practical purposes it is important to
note that also functions that do not have the derivatives required for a rigorous solution of a given PDE qualify as
reasonable solutions. It is these weak solutions which are obtained with the FEM.
Below we will choose the elliptical 1d equation (actually an ODE)
d
dx
(c
du
dx
) +f = 0 (6.5)
to introduce the FEM. Here c could have a space-dependance, c = c(x). Note that this equation occurs in many
dierent physical situations. Later we generalise to 3d and PDEs.
diusion equation u concentration
c diusion constant
f particle production
heat conduction (steady state) T temperature
c thermal conductivity
f heat source term
mechanics of a bar u displacement
c E A
f body force
6.2. The weak form
Rather than solving the PDE directly (strong form), we transform it into an integral equation by multiplying with
a weighting function w(x) and integrating over the domain [a, b]:
I =
b
_
a
w
_
d
dx
(c
du
dx
) +f
_
. .
residual R(x)
dx = 0 (6.6)
Tis has to hold for all weighting functions w(x). Ten it is also true for w(x) = R(x) and thus the requirement
I = 0 w is equivalent to R = 0.
Integration by parts gives:
I = w c
du
dx
b
a
. .
= J
b
J
a
B,
with ux J = c
du
dx
b
_
a
dw
dx
c
du
dx
dx +
b
_
a
wf dx (6.7)
b
_
a
dw
dx
c
du
dx
dx =
b
_
a
wf dx +B (6.8)
Te weak form only involves the rst order derivative. Tis formulation forces the residual to vanish in a spatially
averaged sense. Mathematically the weak formulation corresponds to the introduction of a scalar product in a
Sobolev space. Te main idea of FEMis to solve the weak problemin a nite dimensional subspace. Te Lax-Milgram
theorem then ensures solvability in the subspace. For a grid size h going to zero, this solution converges to the full
solution. Te discretized version can be solved algebraically (by matrix inversion).
6.3. Shape functions
67
APPENDIX A
Overview of tensors in elasticity theory
F
ij
=
x
i
x
0j
=
ij
+
u
i
x
0j
deformation gradient tensor
C = F
T
F right Cauchy-Green deformation tensor
B = F F
T
lef Cauchy-Green deformation tensor (aka Finger ten-
sor)
E =
1
2
(C 1) Green-Lagrange strain tensor
A =
1
2
(1 B
1
) Almansi-Euler strain tensor
=
1
2
(F
T
+ F 21) linear strain tensor
Cauchy stress tensor
C
ijkl
tensor of elastic moduli
P = det(F)(F
T
)
1
1st Piola-Kirchho stress tensor
S = F
1
P 2nd Piola-Kirchho stress tensor

Continuum Mechanics

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Continuum Mechanics

Загружено:

Авторское право:

Доступные форматы

F

is the damping coecient and

. Ten the strain rate is zero for t > t

F +BF = f(t) (2.40)

11 2.6 Kelvin-Voigt model

linear ODE for (2.48)

a linear viscous response (2.51)

Figure 2.15.: Linear superposition of responses.

) Boltzmann 1876 (2.66)

) (t) Boltzmann linear viscoelasticity (2.75)

and then propagate it in a linear manner

() as the Fourier transform of G(t).

() is the storage and G

() the loss modulus.

) and viscous at low frequencies (G

(Rubinstein and Colby, Polymer Physics, 2003, p. 293).

(L) = F = q L +a a = F +qL (3.28)

(x) Ku(x) +qa (3.36)

(x)) u(x) +q = 0 (3.37)

+P, P = const (3.45)

T is the change in unit time of the temperature at material point P identied by x

, which are pressed onto each other by a force F (see g. 5.9).

), respectively. Te undeformed spheres around the point of contact

Вам также может понравиться