MMC Problems PDF

[
y
y +
x
y
z

yz + b = 0
z =
b = 0
xz
z
E y
+ z
+
+
3,33
(
y +
= 1
x
z
y
3
z
r
r r
x +
z =
10 5
t (n ) = + b = 0
n
3,33
z )=
E z
( +
2 xa 4ax 2
2
31
2 x1 a 4ax 2
rn
x
=
,31
10
x
t () x 0
x +
8
4ax 2 5 2ax1 = +n xy + xz +y = 1 = 0

z )=
b =
10 5
4ax 2 2ax1x 0
2
y =
x
y 2 2 z
G xy =
,31
4
) 16 x a = 0
+
2, 2 2
8 (2 x1 a )(2ax1
2 xy 1
16 0
10 5
xy y (2 x1 ayz= )(2ax1 )a5 x 2 a 5 =4ax 2 ,348
x xy
y =
b = 0 2 x1 10
10 6
=0
2
2
+
+
+
+
+G2 y
4,1
(2 x1 a ) = (4ax 2 ) (
24ax 2 2ax1
y
( ax
x = 34
2 a
y x1x z ) =xy 4= 02 )x1
x
8
x
5
dx1 t
0
1
xx a
1 = 2y 1= 14ax 2
2 2
xy y
E x 0 6
a
2
x1 a = xz + b = (2 x11 a= xd)(2 41ax 2 ) 16 x 2 a = 0
4ax 2 + yz 2 x a z + = 4ax0 x = 2 1t ax
+
+
G y 2
1
x
2 = 21x1 a + 4axz= 0 2 X1 = 0 2 + x =l2 1 x

t
y = 1
y
x
y
y + 4ax x
z
2 x1 a
2 x ) 4ax 2
x
(2 x2 a 1 a = (4ax 2 )n 1
2
x2 1
=0
z =
E y
dt
E X 1x = t

dx2
t
3,3
0
ln x14ax 2 2ax1 33
= dt y = 1 =2 xa 4ax 2 x1 = X xz + yz +
1
z = 1
r (n )
1
x =
+t
= 4ax 2
10 52 2 t = n Xx2 x1 a
y +
1 exp t y
2
2
X
2 =x2 x1 a + 4axz = x
z ))(2ax
E z 1(2 x1 a 1 x ) 16 x 2 a = 0
0 2t + 3
=
2
E lyn 2

dt

x3 = X
3,33
1 a X e
2,2 x1= 1 xp 4ax 2
( l
= 1
t
318 )
(2 x a )2 = (4ax 22
3
3
=
x +
=0
z
X 2 x=+n 2t +
2t + 1
r
1
3 ln x1
4ax 20 5 2ax1
3 l n 10 5
t(
2y
G xy =
z )=
E z
==n4
3
r 2
2
2,5 Xx a l = 4ax 1 2 1+ b ax0
,2t4+ 2 ( = )(x a 4 )r
3 8 2 x a 2ax = 16 x 2 a 2 = 0
r r
,31
1
l1 =
31 n
x
10 2 521

+ xb = 1
1 + 4 ax
2
y = 0
x +
8
2 2 x1 a
2
0 6 3 2
10 5
x
2
y =
G xy = x1
G y =
(2 x1 a ) = (4ax 2 )
4
0 dx
x2 = X
2,5
t
2
1
r r ,348
xy = 1
2
1 + 5
= d
= 2 x ta+ 14ax 2
10 6
x
0 b = 0
t
X1 x1
x x1 a = 4ax 2 xy 1 3 1
2
G xy = x1
y =
xz
2 x1 a
0
ln 1
0 x2
x
xy xz d W. V. Chaves 4a
0
+ b + x4ax
+
+ 2 = 2 x1 a 0 2+ Eduardo =t
=t
x1
x = x
+ bx 0
+

y = 1
4ax 2 2ax1
z
y
dx2
X1
t
x
2 x1 a
zX x = dt
x G xy
x1 = X
=
dt
1
1
x (
xy r y 1 exp t
y =
yz
x2
X2
0
0 yz
2 x1 4 )( 2
+ r +
+ b = xy + y + x2 b = xy 2 xz ln 1 =4 a ax 2 2a
0 2t + 3
y
ln x2 + b = y
x1 0 x z
d+2 x yt+ 0 + x1 a b x 0ax 2 0

= 1
X1 = t
x3 = X x
2
+
y
x z x =
a = =
x
4
x =
3
dty
z ax 2 (2 x1 1 )
X 2 = ln 2 +
2ax1 (2 x a
E x
t 3 yz
x x
X2
ln
xz +x 3z + b xxza yz 4axz2 xy 2t y yz x (2 x1 a
+
2 1
+
z
y = 1
= 0 + + x = =00z + 0 3 + (2 x+a )(21 a 1 )4
l 2
+ b =
y + y = 1
1 n 2 xax =
x
y y z
b y =0
r1 rx
+
X
z
y
x z
E y
b = 2 = 2 z
X 2 2 = ln 2t 2+
=3 x xy xz+= = 0x4ax 2X y 2ax1 3x
y
E
3
xz
(
) 24ax 2
= (x a )
2
2 x=a 1 4ax(
t+
,33+ (+ r + b x = ,0 3 (
1
2
1
E 3 x x a 3 2ax1 16 x a 2 = 2 x1 a
2
3 y z n
z
+
x = 0 z = 1x
10 5 x + t ( ) = n 3 2 1 )( r (n1 )xz + 2 yz + 0 z + b = 0
1 5
)
2ax1
4ax 2
0
z
z )=
E z
=
ax
) =1
2
E 2 xy y yzyz = (2 x1 a y + )2 t (4ax)2n y z x1 a 2 x= 4 2 4a
z
z = 2 x
,
1a
=
+ Eb 2=
xy 2 x a )(2ax + ) 16 x 2 a 2 310 + +
1 1
y
=
= (x = 1
x 1 x
1 x 8 10y5 + x ,xy 08 xz
y = 2
(1 +5 b = 0 3,333 1 = 2 x1a 4ax2 4ax 2 2ax
x x= 1 y 31
y
1
z z
+
+ x0 +
r
G E ) =
x
y =
(2 xxya = 2,x 2 (4ax 2 )2 = 4, y =
G 3x
1
t (n ) = n
x 4 y 2za z =x4ax 2 10=5
)=
5
2 x1 a + 4ax 2 (2 x1 a )(2a
xz
48 2,5 yz + z E+ b34=
2
xy =
1
10 5 xy+ = 1 = + 1 ,z z
80
y = 1
y
x
a 4 6
1 2 x1x 10 ax 2y 0 xy5 = z 1 xy + y 10 6yz + b = 2,318
x +
G 2E1 a 1 = 4ax 2 z =
+
yx = x
0
(2 x1 a )2 =

y
10 5
0 y x t x =G1 =y3,=10 +x14ax 2
2 x 23 a
d (
x xy = y
x z y =
x3
G
xy =
1 = 1
1
3 dx t r
4,3
z
=xxy+= 1E
)
x
dt x 10 1x= =d t1(n= n2,5 1
4
5y
X1 x1
2 x1 a = 4
G yE =
2xx a1 axxz + yz + 0 5z + b = 8 10 6
x
t4
z)x
X1
0 y =G1 =y = ln 1
0
0 z 2
1 y 2 x a 0 2 ax
z
G x 4 =2 0 x1
x
lx1 y
n=
=t
2,0 1 x2 + 1
0
dxx +t E 3 8X 4ax 2 z 2ax1 y x 0 dX =zt
xy = 1
t = x t
11 5= 2= 2ax
2
y
1
x
0
r r
(dx2xy=4ax1 t3,33311=XXx x1p=t2r 2dt= x1 = X + b = 0

G xy =X 2 x2 =z = y dt =
e
1 4
+ ( dxy )( 01 ) ( 2 =
( 2 x a = 2ax 51 16 x a 0
x x22 x1 aG )(2ax1 1 ) 1 16 x 2 a n )2 0
e
2,5 0 2t + 3 , X 2 x
x 1 xp t
348 l n
t
xy = 1
0 2 = n ln 1
0
2 z) 21t + 0 1
x3 = X 10 5 E z
x1Xrx a =n223 = (4axx2 )2xln2 x2 =

0 ( l
x
3 r=X2 1 = ),3 2 = 2 d
3
X1 = t
x = 1
= 0 (2 x1 a 2t1+ (4ax2 Xt ln
G y = x1 xy = 1
)3
) d
+ b 23
x +
8 ln 2 = 2 t
0 dx
x1 = X E
2t + 3
t
120 x5 3 = 2 x1 a 4ax 2
ln
x
e
X
2= 1
1
y =
1 xxp t x1
2 1 x a 4ax
=2
=0
=
= G y =
1
x1 2 a = = 2
3
dt
2,5 2a4, 4 4axx ax 2 = 2tx+a +1 4ax= 2 ln2 x x + xy y = xz1 + = y +
x1 3
4 = 0 2 3 x2
X1 x1
G xy =
2
+
b
x
1
E 0
8 3 X 3 2 2 = 2 x1 a + 2 ax22 = ln
4X X +
ln10x1 5
0 x2 xy = 01

X
10 6
t 1 x x 2t 1 2 = 2 y2 tx+ x
E
y
z
=+
3 2
3 ln = 31 1
y
dx2 G xy = x Xxy= t
t
1
1
xy E = 3 ( x +
1
x
x1 a 4ax 2
zy
x =0
x yz E
=
dt 0 dx1 + t + xz + b1x = X
b y0
xx
1
r r + 1 +E + y = x
X 2 y 2=
z
1 ex p t 1
x = y
x =
2
x1 a + 4ax 2
y
t
+ b = 0 x y = y z x z z 1 + (
G 0 2yt + 3 Xx1 ln dx2
1
x =
x
E =
x
z =
x3 = X
0 x2 xy 0 y = yz ln 1
x
E x

y = y1 E x + 3
0
z
+b =
x + X 2 + ln 2t +X y = t
d x t y
z xz + yzG+ x(+ zb = 0 y ,
x = 1
= 1
31 lny =
1
x z
z
y +
y1
2=
x y xy = z =x2,+ x +
dt
3 x1 r= Xr e t z
5 z ) =
x2
E x
X2
E
E y+ b = 0 xp
x

1
1
=
x2
( = X 2 3,33 y = z xy1 = 10 5 2,3
x
G
=
xz 0+2t +yz + + nz =20 1
3
lb z =
y = 1
2
= x r x( x1
y + x3 = X
x +
t + xy 31 10 G5 = xy 1=xn ) + 2,5
3x
y
z
1
y
=
t 0 E x x= t 10 5
= dn
3
X 2 E l n 2t +
z =
y 1
E y
z )=
xy = xy =
G 1
z 3
3,33
(
r
G
ln 3 2,3
4
r= 1
25 xy1X= x x1= ,34
x , y = 1 01
d
3
b = 0
z
xy
1
xy = 1
r (n )
x +
d0x1 t yt8+
10 5
x xy xz = 18 G = y 1= 0 10 5
+

y
t = n
b G= 2 =1x0 5 2 E x2 y
x
x
z )=
= d
y = +
E z
G x = + x + +xy + xz +xb0 x= X 2 x1
y 0 G
=0 t +
2
y
x(
t
dx X1 t1
x
2
,31
x ,5 x y y zxy4=341 x
z = x11 = 0 2 =
0
z, 8
3 d y1 t x
+
xdt
= 1
x +
xy = 1
8
10 5
2
10 5
xy
1
X
2
y =

E = xt
x
+ b = 0 6
G xy = 1 xy y + y yz yzG yxy 0= X1 x1 2z d d0 22t +t 3 z
x
4
Gxy xy = xz,
1
0 + +
2 5
=
0
+
,34
x =
x dx1 yt x z + b y = 0xy x2= 1 x30 = X x x + lnd
+ 1b x = 0
+
1
1 x
X2
8
x
x
2 X
xz
3 2
0 5
dx2 t
10 6 E = y = 1
z
y
= dy
0 yt +
G =xy = x t =
t
G xy = x1
xz x
Ex x G xy = X1 1 yz x

d
1
235 X 3
0
y 0 yz dx t
,
0 x2 +xz + +yz +z + z b zb1 y 0=2 1 2
ln = 0 x X x
y =
y +
1 b =0
5
2t + 3 10
x y z + z == t
y = 1
+
+ = y
0
y + d
y
z X1 x =
ln
z =
x2 x t
x y
E y
dt
x z x1
X1
3G Xxy =
z3 =
E y (
x x1 X
y =
dt
,33x , =
3
3233
0 1 = dx1 expt t
(
0
0 x
lnzx1= 1
3 3
X
+
1
)]
ij =
X 0
0
2X
0
X 0
l2 3
2X
0
0
2 3
X 2

l
3
0
X 2 X 2
l2
X 2
2
l
0
l2
ij =
3
X
2
2X
X l
0
0
l 2 3
X
2X
2X
l 2 3
2 3
0
ij = 0
X 2
3
X 2
l2
0
3
X 0
X
2X
2X
0
l2 3
l 2 3
0
X 2
3
X 2
l2
3
X l
2X
0
ij =
l 2 3
X 0
2X
0
l2 3
X 2
3
X 2
2
l
3
2
X l
2X
l 2 3
)]
[ [
[ [
( (
)] )]
)]
=
0
ij
() ( )
)]
)]
) ( )[
[(
[(
[
(
) ( )[
[
[ (
[
[ ( )]
( )]
[
]
(
[)]
(
[
(
)]
=
0
ij
0
ij =

00
0
ij =
X

2X
0 0
X 20 3
l2 X
0
0
l 3 22 3
X 2
l 2
0
X 2 23
3
X l X 2
l2
3
2lX32
X2
l
2X
l 2 3
ij
X 0
2X
2 3
0
l ij =
0
X 2 0
3 0
X0
2
X 0
ij 2
X3
l =
2X
00 X 0l 2 2 X3
2 3
l
X 2 Xl 2
2 X
0
0
X 2
X3 2
X 2
ll 22 2 33
X 2
2
3
X 2l
0
3
3
X 2l 2
X l
l2
2X
3
2X
l 2 3
X ll 2 3
2X
l 2 3
2
2
=
0
X 2
3
2
X l
2X
l 2 3
l
X 2
3
l2
] ) ( )
)]
)]
=
0
) ( )
([ )
(( )
)] [
)] [
(
] [
] [
(
)]
(
)]
)] [
]
(
(
)]
)]
[
]
(
[
)]
)]
[
]
(
( )
( )
( )
)]
( )
X 0
2X
l2 3
X 2
3
X 2
l2
3
2
X l
2X
l 2 3
CONTINUUM MECHANICS
(
[
)]
ij
ij
BY MEANS OF
=
0
X 0
2X
2 3
l
0
X 2
3
X 2
l2
3
2
X l
2X
l 2 3
X l
2X
)]
SOLVING PROBLEMS
X
2
)]
)]
2
l
X 2
3
l2
)]
Nomenclature
III
Solving Problems by means of

Continuum Mechanics
EDUARDO WALTER VIEIRA CHAVES
IV
MECNICA DEL MEDIO CONTINUO
Presentacin
s
anic
mec
h
dif
fu
sio
Flu
x
Flu
ids
n-
Soil
Co
nv
ec
tio
nsfer
eat tra
H
Presentation
ics
ul
ra
yd
H
Structure
ms
ea
B
Plate
s
lid
So
IBVP and Numerical Treatment
Rigid Body Motion
Constitutive equations
Fundamental equations of C.M.

Stress
Continuum kinematics
Tensors
VI
Abbreviations
Abbreviations
IBVP
BVP
FEM
BEM
FDM
Initial Boundary Value Problem

Boundary Value Problem
Finite Element Method
Boundary Element Method
Finite Difference Method
Latin
i.e.
et al.
e.g.
etc.
Q.E.D.
v., vs.
viz.
id est
et alii
exempli gratia
et cetera
Quod Erat Demonstrandum
versus
vidilicet
that is
and the others
for example
and so on
which had to be demonstrated
versus
namely
Operators and Symbols
Operators and Symbols

=
+
2
Tr ()
() T
() 1
() T
() sym
() skew
() sph
() dev
[[]]
det()
D
&
Dt
cof ()
Adj()
Tr ()
:
2
grad()
div ()
I , II , III
r
1
I
I sym I
Macaulay bracket
Euclidian norm of
trace of ()
transpose of ()
inverse of ()
inverse of the transpose of ()
symmetric part of ()
antisymmetric (skew-symmetric) part of ()
spherical part of ()
deviatoric part of ()
module of
jump of
scalar product
determinant of ()
material time derivative of ()
cofactor of ;
adjugate of ()
trace of ()
double scalar product (or double contraction or double dot product)
Scalar differential operator
tensorial product
gradient of
divergence of
vector product (or cross product)
First, second and third principal invariants of the tensor
Vector
Unity vector
Second-order unit tensor
fourth-order unit tensor
symmetric fourth-order unit tensor
SI-Units
SI-Units
J = Nm - Joules
J
W watt
s
length
m - metro
energy, work, heat
mass
kg - kilogram
power
time
temperature
s - second
K - Kelvin
m
s
m
s2
permeability
m2
dynamic viscosity
Pa s
energy
J = Nm - Joules
energy flux
force
N - Newton
thermal conductivity
pressure, stress
Pa
velocity
acceleration
Prefix
pico
nano
micro
mili
centi
deci
N
- Pascal
m2
Symbol
10 n
10 12
10 6
m
c
d
kg
m2s
J
m2s
W
mK
kg
m3
mass flux
10 9
10 3
10 2
10
mass density
Prefix
kilo
Mega
Giga
Tera
Symbol
10 n
10 3
M
G
10 6
1012
10 9
Nomenclature
Nomenclature
r r
r r
A( X , t ) a ( X , t )
r r
a ( x, t )
B0
B
B
r r
b( x , t )
Acceleration (reference configuration)
m
s2
Transformation matrix
Acceleration (current configuration)
m
s2
Continuum medium in the reference configuration at

t=0
Continuum medium in the current configuration at

time t
Boundary of B
Body force (per unit mass)
Left deformation Cauchy-Green tensor, Finger

deformation tensor
Piola deformation tensor
Entropy created inside
local entropy per unit mass per unit time
Ce
Elasticity tensor
Elasticity matrix (Voigt notation)
Inelasticity tensor
Cauchy deformation tensor
Calor especfico a volumen constante
Calor especfico a presin constante
Cohesion
cc
solute concentration
Right deformation Cauchy-Green tensor
DV
Dilation
D
r
dA
r
da
dV
Rate-of-Deformation tensor
Area element vector in the reference configuration
Area element vector in the current configuration
Volume element
N
m3
[C ]
in
C
c
Cv
Cp
J
sK
J
kg s K
Pa
Pa
Pa
Pa
mol
m3
m
m
m2
m2
m3
Nomenclature
Green-Lagrange strain tensor, or Lagrangian finite

strain tensor, or Green-St_Venant strain tensor
Almansi strain tensor, or Eulerian finite strain tensor
ei
, , k
i j
Youngs modulus, or elastic modulus

Cartesian basis in symbolic notation
Cartesian basis
Deformation gradient (pseudo-tensor)
G
H
shear modulus
Biot strain tensor
Total entropy
r
HO
Angular momentum
Jacobian determinant
r
J ( X , t)
r
material displacement gradient tensor
j ( x, t )
spatial displacement gradient tensor
r
J
diffusion tensor
thermal conductivity tensor
Kinetic energy
r
L
Linear momentum
Spatial velocity gradient
m
M
mass
Mandel stress tensor
outward unit normal to the boundary (current
configuration)
outward unit normal to the boundary (reference
configuration)
r
p
body force (per unit volume)
P
p
First Piola-Kirchhoff stress tensor

thermodynamic pressure
r r
q( x , t )
Cauchy heat flux (non-convective vector)
Q
Q
r
r ( x, t )
Orthogonal tensor
Thermal work
radiant heat constant, or heat source (per unit mass)
XI
m
m
m
m
Pa
m
m
Pa
J
K
kgm 2
= Js
s
m3
m3
m
m
m
m
mol
m2s
W
J
=
mK smK
J
kg m
s
m
sm
kg
Pa
N
m3
Pa
Pa
J
m2s
J
J
kg s
XII
R
S
Orthogonal tensor of polar decomposition

Second Piola-Kirchhoff stress tensor
Entropy flux
T
r r
t (n) ( x , t , n)
r (N)
t0
r
T ( x, t )
t
t0 t = 0
Biot stress tensor

Traction vector (current configuration)
Traction pseudo-vector (reference configuration)
Pa
Temperature
Time
Initial time
&
U
rate of change of the internal energy
Specific internal energy
K
s
s
J
=W
s
J
kg
r r
u( x , t )
r r
u( X , t )
r
U( X , t )
r
V ( x, t )
Pa
J
kg s m 2
Pa
Pa
Displacement vector (Eulerian)

m
Displacement vector (Lagrangian)
m
Right stretch tensor, or Lagrangian stretch tensor, or
material stretch tensor
left stretch tensor, or Eulerian stretch tensor, or
spatial stretch tensor
r r
r r
V ( X , t) v ( X , t)
Velocity (reference configuration)
r r
v ( x, t )
Velocity (current configuration)
Spin tensor, rate-of-rotation tensor, or vorticity

tensor
w int
r
X
r
x
Vector position (material coordinate)

Vector position (spatial coordinate)
coefficient of thermal expansion
ij
1 , 2 , 3
Kronecker delta
Principal strains (infinitesimal strain)
Unit Extension
ijk
Permutation symbol, or Levi-Civita tensor

components
Linear dilatation (volume ratio) (small deformation
regime)
m
s
m
s
m rad
=
ms
s
J
=W
s
m
m
1
K
Stress power
Infinitesimal strain tensor
specific entropy
bulk modulus
m
m
m
m
m
m
J
kg K
Pa
Nomenclature
thermal diffusivity
Stretch
Lam constants
Poissons ratio
Solute mass density
fluid mass density

r
0 (X )
r
( x, t )
1
Mass density (reference configuration)

Mass density (current configuration)
Specific volume
r
N
r
S
m
1 , 2 , 3
r
oct
r
oct
max
Cauchy stress tensor, or true stress tensor

Normal traction vector
Tangential traction vector
Mean stress
Principal stresses
normal octahedral vector
tangential octahedral vector
Maximum shear stress
Kirchhoff stress tensor
angle of internal friction
Helmholtz free energy, specific (per unit mass)
Helmholtz free energy (per unit volume)
( ) = e
Strain energy density
Vorticity vector
m2
s
m
m
Pa
kg
m3
kg
m3
kg
m3
kg
m3
m3
kg
Pa
Pa
dilatancy
XIII
Pa
Pa
Pa
Pa
Pa
Pa
J
kg
J
m3
J
m3
rad
s
XIV
The indicial notation was introduced by Einstein (1916, sec.

5), who later jested to a friend, "I have made a great discovery
in mathematics; I have suppressed the summation sign every
time that the summation must be made over an index which
occurs twice..." (Kollros 1956; Pais 1982, p. 216).
Ref. (Wolfram MathWorld (Einstein Summation))
1 Tensors
1.1 Solved Problems

1.1.1
Vectors, Indicial Notation
Problem 1.1
r
Let a and b be arbitrary vectors. Prove that the following relationship is true:
r r r r
r r r r
r r
(a b) (a b) = (a a)(b b) (a b)
Solution:
r r r r
(a b) (a b)
2
r r 2
r r
= a b = a b sin
r 2 r 2
r 2 r 2
r 2 r
= a b sin 2 = a b 1 cos 2 = a b
2
r 2 r 2
r r
r 2 r 2 r r 2
= a b a b cos = a b a b
r r r r
r r 2
= (a a) b b a b
( ) ( )
r r
( )
r r
r
a
r 2
b cos 2
where we have taking into account that a a = a and b b = b .
Problem 1.2
r
Show that: if c = a + b , the module of c can be expressed by means of the following

relationship:
r
c =
r
a
r
r r
+ 2 a b cos + b
r
where is the angle formed by the vectors a and b .

Solution:
Starting from the module definition of a vector it fulfills that:
SOLVING PROBLEMS BY MEANS OF CONTINUUM MECHANICS
r r
a+b
)(
r r r r
= a+b a+b
r r r r r r r r
= aa + ab + b a + b b
r r r r
r r r 2 r r r 2
Taking into account that a a = a , b b = b and a b = b a (commutative), we can
2
conclude that:
r r
a+b
r r r r r r r r
= aa + ab + b a + b b
r
= a
r r r
+ 2a b + b
r
= a
r
r r
+ 2 a b cos + b
r
b
with which we demonstrate a + b =

r r
ab =
r
a
r
r r
2 a b cos + b
r
a
r
r r
+ 2 a b cos + b
r r
a+b
r r
we can conclude that the
is maximum when = 0 holds, then
r
= a
r r r
+ 2a b + b
r
= a
. Then, it is easy to show that
NOTE: Starting from the equation a + b = a + 2a b + b

value a + b
r
a
r
r r
+2a b + b
r
r
= a + b
= 0
2
r
b
r
a
r
r r
r
a+b = a + b
Then, for any value of 0 < 180 the outcome a + b will be less than a + b . Then,
r
r r
r
a + b a + b holds:
r
b
r
r
r r
r
c = a+b a + b
r r r
c = a+b
r
b
r
a
r
In a similar fashion we can show that a c + b and b a + c which is known as the

triangle inequality, where it holds that:
a<c+b
b<a+c
c<a+b
University of Castilla-La Mancha

Ciudad Real - Spain
Draft
By: Eduardo W. V. Chaves (2013)
1 TENSORS
Problem 1.3
1
2
Given the following functions () = E and () = E 2 , demonstrate whether these

functions show a linear transformation or not.
Solution:
( 1 + 2 ) = E [1 + 2 ] = E 1 + E 2 = (1 ) + ( 2 ) (linear transformation)
()
( 1 + 2 ) = ( 1 ) + ( 2 )
( 2 )
( 1 )
1 + 2
1
2
(1 + 2 ) = (1 ) + ( 2 ) has not been satisfied:
The function () = E 2 does not show a linear transformation because the condition
2
2
(1 + 2 ) = E[1 + 2 ]2 = E [1 + 21 2 + 2 ] = E1 + E 2 + E 21 2
2
2
1
2
1
1
2
2
= (1 ) + ( 2 ) + E 1 2 ( 1 ) + ( 2 )
1
2
1
2
()
(1 + 2 )
(1 ) + ( 2 )
( 2 )
(1 )
1
1 + 2
Problem 1.4
Consider the points: A(1,3,1) , B (2,1,1) , C (0,1,3) and D(1,2,4 ) , defined in the Cartesian
coordinate system.

Ciudad Real - Spain
Draft
1) Find the parallelogram area defined by AB and AC ; 2) Find the volume of the
parallelepiped defined by AB , AC and AD ; 3) Find the projection vector of AB onto BC .

Solution:
1) Firstly we calculate the vectors AB and AC :
) (
)
r
b = AC = OC OA = (0 + 1 + 3k ) ( + 3 + 1k ) = 1 2 + 2k
i j 1i j
i
j
a = AB = OB OA = 2 1 + 1k 1 + 3 + 1k = 1 4 + 0k
i j
i
j
i
j
Next, we evaluate the vector product as follows:
i
r r
ab= 1
4 0 = ( 8) 2 + ( 6 ) k
i
j
1 2
Then, the parallelogram area can be obtained by using the following definition:
r r
A = a b = (8) 2 + (2) 2 + ( 6) 2 = 104
2) Next, we can evaluate the vector AD as:
) (
c = AD = OD OA = 1 + 2 + 4k 1 + 3 + 1k = 0 1 + 3k
i
j
i
j
i j
we can obtain the volume of the parallelepiped as follows:
r r r
r r r
V (a, b, c ) = c a b
i j
= 0 1 + 3k
) ( 8i 2j 6k )
= 0 + 2 18 = 16
3) The BC vector can be calculated as:
) (
BC = OC OB = 0 + 1 + 3k 2 1 + 1k = 2 + 2 + 2k
i j
i j
i
j
Hence, it is possible to evaluate the projection vector of AB onto BC as follows:
proj BC AB =
BC AB
BC 4
123
4 BC
BC
BC
( 2i + 2j + 2k ) (1i 4j + 0k ) ( 2i + 2j + 2k )
+ 2 + 2k ) ( 2 + 2 + 2k )
( 2i j
i
j
( 2 8 + 0 ) ( 2 + 2 + 2k ) = 5 5 5 k
i
j
i
j
(4 + 4 + 4 )
3
3
3
Problem 1.5
Rewrite the following equations using indicial notation:
1) a1 x1 x 3 + a 2 x 2 x 3 + a 3 x3 x 3
Solution:
a i xi x 3
(i = 1,2,3)
2) x1 x1 + x2 x2

Ciudad Real - Spain
Draft
1 TENSORS
Solution:
xi x i
(i = 1,2)
a11 x + a12 y + a13 z = b x
3) a 21 x + a 22 y + a 23 z = b y
a 31 x + a 32 y + a 33 z = b z
Solution:
a11 x1 + a12 x 2 + a13 x 3 = b1
a 21 x1 + a 22 x 2 + a 23 x 3 = b2
a x + a x + a x = b
32 2
33 3
3
31 1
a1 j x j = b1
a 2 j x j = b2
a 3 j x j = b3
dummy
index j
free
index i
a ij x j = bi
As we can appreciate in this problem, the use of the indicial notation means that the equation
becomes very concise. In many cases, if algebraic operation do not use indicial or tensorial
notation they become almost impossible to deal with due to the large number of terms
involved.
Problem 1.6
Show that:
a) 3 p v p = v3 ;
b) 3i A ji = A j 3 ;
c) ij ijk ;
d) i 2 j 3 Aij .
Solution:
The Kronecker delta components are:
11
ij = 21
31
12 13 1 0 0
22 23 = 0 1 0

32 33 0 0 1

(1.1)
a) The expression ( 3 p v p ) has no free index, then the result is a scalar:
3 p v p = 31v1 + 32 v 2 + 33 v 3 = v3
(1.2)
b) The expression 3i A ji has one free index ( j ), then the result is a vector:
3i A ji = 31 A j1 + 32 A j 2 + 33 A j 3 = A j 3
(1.3)
c) The expression ij ijk has one free index ( k ), then the result is a vector:
ij ijk = 1 j 1 jk
123
1111k
+ 2 j 2 jk + 3 j 3 jk
123
4 4
123
+ 21 21k
+
+ 31 31k
+
12 12 k + 22 22 k
+ 32 32 k
+
+
13 13k

Ciudad Real - Spain
+
+ 23 23k
Draft
(1.4)
+
+ 33 33k
thus ij ijk = 0 k is the null vector. Note that ij ijk = iik = 11k + 22 k + 33k = 0 k .
d)
i 2 j 3 Aij = A23
(1.5)
Problem 1.7
Expand the equation: Aij x i x j
(i, j = 1,2,3)
Solution: The indices i, j are dummy indices, and indicate index summation and there is no
free index in the expression Aij x i x j , therefore the result is a scalar. So, we expand first the
dummy index i and later the index j to obtain:
expanding j
expanding
Aij x i x j i A1 j x1 x j + A2 j x 2 x j + A3 j x 3 x j
1 2 3 1 24 1 24
4 4
4 3 4 3
A11 x1 x1 A21 x 2 x1 A31 x 3 x1
+
+
+
A12 x1 x 2
A22 x 2 x 2
A32 x 3 x 2
A13 x1 x 3
A23 x 2 x 3
A33 x 3 x 3
Rearranging the terms we obtain:

Aij x i x j = A11 x1 x1 + A12 x1 x 2 + A13 x1 x3 + A21 x 2 x1 + A22 x 2 x 2 +
A23 x 2 x3 + A31 x3 x1 + A32 x 3 x 2 + A33 x 3 x3
Problem 1.8
Obtain the numerical value of:
1) ii jj
Solution:
ii jj = ( 11 + 22 + 33 )( 11 + 22 + 33 ) = 3 3 = 9
2) 1 1
Solution:
1 1 = 1 1 = 11 = 1
NOTE: Note that the following algebraic operation is incorrect 1 1 = 3 11 = 1 ,

since what must be replaced is the repeated index, not the number
Problem 1.9
a) Prove the following is true ijk pjk = 2 ip , ijk ijk = 6 and ijk a j a k = 0 i . b) Obtain the
numerical value of ijk 2 j 3k 1i .
Solution: a) Using the equation ijk pqk = ip jq iq jp , and by substituting q for j , we
obtain:
ijk pjk = ip jj ij jp = ip 3 ip = 2 ip
Based on the above result, it is straight forward to check that:

Ciudad Real - Spain
Draft
1 TENSORS
ijk ijk = 2 ii = 6
c) Note that ijk = ikj , i.e. it is antisymmetric in jk and also note that a j a k is a symmetric
second-order tensor. So, as we know, the double scalar product between a symmetric and an
antisymmetric second-order tensors is zero, thus:
r
ijk a j a k = ijk (a a ) jk = 0 i
r r
= (a a ) i = 0 i
b) ijk 2 j 3k 1i = 123 = 1
Problem 1.10
Get the value of the following expressions:
a) ijk i1 j 2 3k
b) ijk pqk = ip jq iq jp for the following cases:
b.1) i = 1, j = q = 2, p = 3
b.2) i = q = 1, j = p = 2
c) ( ijk A jp c p A kq c q + i1 )( ist A sa c a A tb c b + i1 )
where ijk is the permutation symbol and ij is the Kronecker delta.
Solution:
a) ijk i1 j 2 3k = 123 = 1 ; b.1)
b.2)
12 k 32 k = 121 321 + 122 322 + 123 323 =

= 0 (1) + 0 0 + 0 0 = 0
12 k 21k = 121 211 + 122 212 + 123 213

= 0 0 + 0 0 + 1 (1) = 1
c) Note that the result of A jp c p = b j is a vector, and also note that the following is true
r
ijk A jp c p A kq c q = [( A c) ( A c )]i = (b b) i = 0 i , with which we can obtain:

r
( ijk A jp c p A kq c q + i1 )( ist A sa c a A tb c b + i1 ) = i1 i1 = 11 = 1
Reminder: Permutation symbol

ijk = 1
ijk = jki = kij
ijk = ikj = kji = jik
ijk = 1

Ciudad Real - Spain
Draft
k =3
j =1
i =1
k =2
i=2
j =1
i =1
k =1
i=2
j =1
i =1
i=2
i=3
j=2
0
0
-1
i=3
j =3
j=2
0
1
0
0
0
i=3
j =3
0
-1
0
-1
j=2
1
0
0
j =3
0
0
0
ij 3
0
0
ij 2
0
ij1
Problem 1.11
r
Write in indicial notation: a) the modulus of the vector a ; b) cos , where is the angle
r
r
between the vectors a and b .
Solution:
r
a
r r
= a a = a i e i a j e j = a i a j ij = a i a i = a j a j
r
thus, it is also true that b = b i b i .
2
r r
r
a = ai ai
r r
By definition a b = a b cos where:

r
a b = a i e i b j e j = a i b j ij = a i b i = a j b j
Taking into account that the index cannot appear more than twice in a term of the expression,
we can express cos as follows:
r r
a jb j
ab
cos = r r =
ai ai b k b k
a b
Problem 1.12
Show the Schwarz inequality:
r r
r r
ab a b
Schwarz inequality
(1.6)
Solution:
Let us consider a scalar , then the following is true:
r
r
a b

Ciudad Real - Spain
r r
r
r
= (a b) (a b) 0
r r
r r
r r
r r
= a a 2 a b b a + b b 0
r 2
r 2
r r
= a 2 2a b + b 0
Draft
1 TENSORS
r r
r 2
r 2 2
r r
ab
f ( ) = a 2a b + b 0 , if we now obtain the value when = r 2 we obtain that:
a
r r
r r 2
r r
r
r
r r
= (a b) = a 2 (a b) 2(a b) (a b) + b 2 0
f
r 2
r 2
r 2
a
a
a
r
r
r r
r r
r r
r 2
r 2 ( a b) 2
r r (a b) r 2 (a b) 2
( a b) 2
= a
r 4 2 ( a b) r 2 + b = r 2 2 r 2 + b 0
a
a
a
a
r r 2
r 2
(a b )
= r 2 + b 0
a
r r
r 2 ( a b) 2
b r 2
a
r 2 r 2
r r
a b ( a b) 2
r r
r r
a b ab
Q.E.D.
Alternative solution
r r
r r
r r
Taking in account that 0 cos 1 we obtain a b = a b cos a b , thus we conclude

r r
r r
that a b a b .
Problem 1.13
(r r ) (r r )
r r
r r
Solution: The vector product (a b ) can be expressed as (a b ) = a e b e =

r r
r r
Likewise, it is possible to express (c d) as (c d) = c d e , thus:

r r r r

(a b) (c d) = a b e ) ( c d e ) = a b c d e e
Rewrite the expression a b c d without using the vector product symbol.

j
nlm
ijk
nlm
nlm
ijk a j b k e i
ijk
= ijk nlm a j b k c l d m in = ijk ilm a j b k c l d m
Taking into account that ijk ilm = jki lmi

jki lmi = jl km jm kl = jki ilm , we obtain:
and
by
applying
the
equation
ijk ilm a j b k c l d m = ( jl km jm kl ) a j b k c l d m = a l b m c l d m a m b l c l d m
r r
(r r )
r r r r
r r r r
r r r r
(a b) (c d) = (a c ) (b d) (a d)(b c )
Since a l c l = (a c ) and b m d m = b d holds true, we can conclude that:

r
Therefore, it is also valid when a = c and b = d , thus:

r r r r
r r
(a b) (a b) = a b
( ) ( )( )
r r r r
r r r r
r
= (a a) b b a b b a = a
r
b
( )
r r
ab
which is the same equation obtained in Problem 1.1.

r
r r
NOTE: We can start from the above equation to show a b = a b sin , i.e.:

Ciudad Real - Spain
Draft
10
r r
ab
r 2 r 2 r r
r 2 r 2
r r
= a b (a b) 2 = a b a b cos
r
)=a
2
r 2
r 2 r 2
b 1 cos 2 = a b sin 2
r r
r r
a b = a b sin
r
r r
r r
Note that 0 sin 1 , with that we can prove that a b = a b sin a b , thus
r r
r r
ab a b
Problem 1.14
Show that:
a) ijk a i a j b k = 0 ;
r r
b) ijk (a k b 3 i1 j 2 + a j b 2 i1 k 3 + a i b1 j 2 k 3 ) = a b ;
c) Aij A ji is an invariant.
Solution:
a) ijk a i a j b k = ij1a i a j b1 + ij 2 a i a j b 2 + ij 3 a i a j b 3 . The term ij1a i a j b1 can be evaluated as
follows:
ij1a i a j b1 = 1 j1 a1 a j b1 + 2 j1a 2 a j b1 + 3 j1a 3 a j b1
= 111 a1 a1b1 + 211 a 2 a1b1 + 311 a 3 a1b1 +
+ 121 a1 a 2 b1 + 221 a 2 a 2 b1 + 321 a 3 a 2 b1 +
+ 131 a1 a 3 b1 + 231 a 2 a 3b1 + 331 a 3 a 3 b1
= 321 a 3 a 2 b1 + 231 a 2 a 3b1 = a 3 a 2 b1 + a 2 a 3b1
=0
In the same way we obtain ij 2 a i a j b 2 = ij 3 a i a j b 3 = 0 .

b)
ijk a k b 3 i1 j 2 + ijk a j b 2 i1 k 3 + ijk a i b1 j 2 k 3 =
r r
12 k a k b 3 + 1 j 3 a j b 2 + i 23 a i b1 = a 3b 3 + a 2 b 2 + a1b1 = a i b i = a b
Problem 1.15
(r r ) (r r ) r [r
] r [r
Prove that a b c d = c d (a b) d c (a b)
Solution: Expressing the correct equality term in indicial notation we obtain:
] [
rr r r
rr r r
c d ( a b ) d c ( a b ) = c d a b d c a b
p
i
ijk j k
p
i
ijk j k
[ (
ijk a j b k c p d i ijk a j b k c i d p
)]
)]
ijk a j b k (c p d i c i d p )
Using the Kronecker delta the above equation becomes:

ijk a j b k pm c m d n ni im c m d n np
[ (
ijk a j b k
)c
m dn
pm
ni im np )
and by applying the equation pm ni im np = pil mnl , the above equation can be rewritten
as follows:
Ciudad Real - Spain
Draft
1 TENSORS
( ijk a j b k ) c m d n ( pil mnl )
11
pil ( ijk a j b k ) ( mnl c m d n )
r r
r r
and mnl c m d n represent the components of a b and c d , respectively,
Since ijk a j b k
we can conclude that:
[(r r ) (r r )]
pil [( ijk a j b k ) ( mnl c m d n )] = a b c d
Problem 1.16
r
Let a , b , c be linearly independent vectors, and v be a vector, demonstrate that:

r
r
r r
r
v = a + b + c 0
v i = a i + b i + c i 0 i
components
where the scalars , , are given by:
ijk v i b j c k
pqr a p b q c r
; =
ijk a i v j c k
pqr a p b q c r
; =
ijk a i b j v k
pqr a p b q c r
b) Given three linearly independent vectors, show that: when interchanging two rows or two
r r r
columns the sign of the determinant a (b c ) changes.
r
Solution: a) The scalar product made up of v and ( b c ) becomes:

r r r
r r r
r r r
r r r
v (b c ) = a (b c ) + b (b c ) + c (b c )
1 24
4 3
1 24
4 3
=0
=0
r r r
v (b c )
= r r r
a (b c )
which is the same as:

v1
b1
c1
c2
c3
a1
b1
a2
b2
a3
b3
c2
c3
v1
v2
=
b1
b2
c1
c2
v3
b3
c3
a1
a2
b1
b2
c1
c2
a3
v3
b3
c1
v2
b2
b3
c3
ijk v i b j c k
pqr a p b q c r
One can obtain the parameters and in a similar fashion, i.e.:

r r r
r r r
r r r
r r r
v (a c ) = a (a c ) + b ( a c ) + c ( a c )
1 24
4 3
1 24
4 3
=0
=0
r r r
r r r
jik a j v i c k
v (a c ) ijk v i a j c k
a (v c)
= r r r
= r r r =
=
b (a c ) pqr b p a q c r qpr a q b p c r a (b c )
r r r
r r r
r r r
r r r
v (a b ) = a ( a b ) + b ( a b ) + c (a b )
1 24
4 3
1 24
4 3
=0
=0
r r r
r r r
jki a j b k v i a (b v )
v (a b) ijk v i a j b k
= r r r =
=
= r r r
c (a b) pqr c p a q b r qrp a q b r c p a (b c )
r
NOTE 1: We can restructure the v -components as follows:
v 1 a1

v i = v 2 = a 2
v a
3 3
Ciudad Real - Spain
b1
b2
b3
c 1 a1

c 2 = a 2
c 3 a 3

Draft
b1
b2
b3
c 1 z1

c 2 z 2 = B ij z j
c 3 z 3

12
where we have denoted by z1 = , z 2 = , z 3 = , in which:

v1
v2
ijk v i b j c k
v3
=
a1
pqr a p b q c r
b3
c3
b1
b2
c1
c2
a3
b3
c3
a1
a2
=z 3 =
c1
c2
a2
=z1 =
b1
b2
b1
b2
v1
v2
ijk a i b j v k
a3 b 3
=
a1 b 1
pqr a p b q c r
v3
a2
b2
c1
c2
a3
b3
; =z 2 =
c1
c2
ijk a i v j c k
a3
=
a1
pqr a p b q c r
v3
c3
b1
b2
c1
c2
a3
B (1)
v1
v2
a2
a1
a2
b3
c3
B (2)
B
B (3)
c3
where B (i ) is the determinant of the resulting matrix by replacing the column (i) of the
r
matrix B by the v -components. With that we can state that:

Given v i = B ij z j
zi =
B (i )
B
Cramers rule
NOTE 2: Although we have demonstrated for 3 3 matrix, this procedure is also valid for
matrices of n-dimensions, which is known, in the literature, as Cramers Rule.
NOTE 3: The solution ( z i ) is possible if B 0 .
NOTE 4: If v i = 0 i we have B ij z j = 0 i and B (i ) = 0 i , with that according to Cramers rule
we have:
z i B = B (i ) = 0 i
Note that the non-trivial solution z i 0 i is only possible if and only if B = 0 , (see Problem
1.51).

Ciudad Real - Spain
Draft
1 TENSORS
13
r r r
b) The determinant defined by a (b c ) = [a, b, c ] in indicial notation becomes ijk a i b j c k .

Note that the following holds:
ijk = jki = kij
ijk = ikj = kji = jik

i
Thus:
a1
ijk a i b j c k = b 1
a2
b2
a3
b3
c1
c2
c3
r r r
ijk a i b j c k = [a, b, c ]
a1
a2
a3
= c1
b1
r r r
= ikj a i b j c k = [a, c , b]
r r r
= jki a i b j c k = [b, c , a]
c2
b2
c 3 = ikj a i b j c k
b3
b1
b2
b3
c1
a1
c2
a2
c 3 = jki a i b j c k
a3
Problem 1.17
Show that
( ) (
r r
r
r r r r r r r r r r r
a (b c ) = (a c ) b a b c = b c c b a
r r
r
r r
r r r
a (b a) = [(a a)1 a a] b
r
r r
r r
Solution: Taking into account that (d) i = b c i = ijk b j c k and that a d q = qjk b j c k , we
obtain:
r r r
[a (b c )]
= rsi a s ( ijk b j c k )
= rsi ijk a s b j c k = rsi jki a s b j c k
= rj sk rk sj a s b j c k
= rj sk a s b j c k rk sj a s b j c k
= asb r c s asb s c r
= ak br c k a jb j c r
r r
r r
= b r (a c ) c r a b
r r r rr r
= b(a c ) c a b r
= (b r c s b s c r )a s
r r r r r
= b c c b a r
[(
( )
( )]
So, proving that:
( ) (
) ]
r r r
r r r r r r r r r r r
a b c = (a c ) b a b c = b c c b a
r r
In the particular case when a = c we obtain:
r r r
a b a r = (a k a k )b r (a j b j )a r = (a j a j )b p rp (a j b p jp )a r
[ (
)]
= (a j a j ) rp (a j jp )a r b p = (a j a j ) rp a p a r b p
r r
r r r
= [(a a)1 a a] b r

Ciudad Real - Spain
Draft
14
Problem 1.18
Show the Jacobi identity:
r r r r r r r r r r
a b c + b (c a) + c a b = 0
r
(r r )
r r r
(r r ) r
Solution: By means of Problem 1.17 in which a b c = (a c ) b a b c was proven, we

can obtain that:
r r r r r r
(b ra)c (b c )a
r
r r r r
) (c b)a (c a)b
r r r
b (c a) =
r r r
c ab =
r r
r r
(r r ) (r r )
Then, by considering that the dot product is commutative, i.e. (a c ) = (c a) , a b = b a ,

r r
r r
(b c ) = (c b), we can conclude that:
(a c )b (a b )c
r r r
r r r r r
) ( )c + (b c )a = 0
r r r r
( )a (c a)b
r r
r r r r
r r r r r
a b c + b (c a) + c a b = b a
r r
c b
1.1.2
r r r
Algebraic Operations with Higher Order Tensors
Problem 1.19
Define the order of the tensors represented by their Cartesian components: v i , ijk , Fijj , ij ,
C ijkl , ij . Determine the number of components in tensor C .
Solution: The order of the tensor is given by the number of free indices, so it follows that:
r r
First-order tensor (vector): v , F ; Second-order tensor: , ; Third-order tensor: ;

Fourth-order tensor: C
The number of tensor components is given by the maximum index range value, i.e.
i, j , k , l = 1,2,3 , to the power of the number of free indices which is equal to 4 in the case of
C ijkl . Thus, the number of independent components in C is given by:
3 4 = (i = 3) ( j = 3) (k = 3) (l = 3) = 81
The fourth-order tensor C ijkl has 81 components.

Problem 1.20
(r r ) r (r r ) r
(r r ) (r r ) (r r ) r
Show that a) a b c = b c a ; b) a b c d = b c a d
Solution:
r r r
(a b) c = (a e
i
a)

Ciudad Real - Spain
b je j ) ck ek
= ai e ib j c k
jk
= (b k c k )a i e i
r r r
r r
r
= (b c )a (b c ) a
Draft
1 TENSORS
15
(r r ) (r r )
b) The expression a b c d , which is a second-order tensor, can be expressed in indicial

notation as follows:
r r r r
r r r r
[(a b) (c d)] = (a b) (c d)
ij
ik
kj
= (a i b k ) c k d j = a i b k c k d j
= b k c k ai d j
r r r
= (b k c k )(a i d j ) = (b c )(a d) ij
123
scalar
Problem 1.21
Expand and simplify the expression A ij xi x j when a) A ij = A ji ; b) A ij = A ji .
Solution:
By expanding A ij xi x j we obtain:
A ij xi x j = A 1 j x1 x j + A 2 j x 2 x j + A 3 j x3 x j =
= A 11 x1 x1 + A 21 x 2 x 1 + A 31 x3 x 1 +
(1.7)
A 12 x1 x 2 + A 22 x 2 x 2 + A 32 x 3 x 2 +
A 13 x1 x 3 + A 23 x 2 x 3 + A 33 x3 x 3
a) If A ij = A ji (symmetry) we have
A ij xi x j = A 11 x12 + 2 A 12 x1 x 2 +2 A 13 x1 x 3 +
(1.8)
2
2
A 22 x 2 + 2 A 23 x 2 x 3 + A 33 x3
b) If A ij = A ji (antisymmetry) we have
A ij xi x j = 0
(1.9)
as expected, since:
r
r
r r
A ij xi x j = x A x = A : ( x x )
(1.10)
r r
That is, if A is an antisymmetric and ( x x ) is a symmetric tensor, the double scalar product
between them is always equal to zero.

Problem 1.22
Let and T be second-order tensors, whose Cartesian components are:
5 2 4
ij = 1 2 1
4 3 6
3 1 2
Tij = 4 2 1
1 3 8
(1.11)
Obtain T : .
Solution:
T : = Tij ij

Ciudad Real - Spain
Draft
(1.12)
16
Tij ij = T1 j 1 j + T2 j 2 j + T3 j 3 j
1 3
2
123
12
3
T1111
+
T21 21
+
T31 31
+
T12 12
T22 22
T32 32
+
T23 23
+
T33 33
+
T13 13
(1.13)
thus,
Tij ij = 5 3 + 2 1 + 4 2 + (1) 4 + 2 2 + 1 1 + 4 1 + 3 3 + 6 8 = 87
(1.14)
Problem 1.23
Given the B tensor components:
3 2 4
B ij = 1 5 3
5 7 9
(1.15)
Obtain:
a) C ij = B ik B kj ; b) D ij = B ik B jk ; c) E ij = B ki B kj ; d) C ii , D ii , E ii
Solution:
C = B B
3 2 4 3 2 4 31 44 54
C ij = B ik B kj = 1 5 3 1 5 3 = 23 48 46
5 7 9 5 7 9 67 108 122
(1.16)
D = B BT
3 2 4 3 2 4
29 25 65
1 5 3 1 5 3 = 25 35 67
=
5 7 9 5 7 9
65 67 155
D ij = B ik B jk
3 2 4 3 2 4 35 46 60
E ij = B ki B kj = 1 5 3 1 5 3 = 46 78 86
5 7 9 5 7 9 60 86 106
(1.17)
E = BT B
(1.18)
Then:
C ii = C 11 + C 22 + C 33 = 31 + 48 + 122 = 201
D ii = D11 + D 22 + D 33 = 29 + 35 + 155 = 219
(1.19)
E ii = E11 + E 22 + E 33 = 35 + 78 + 106 = 219
NOTE: We can verify that the following is true: Tr (B B T ) = Tr (B T B) = B : B .

Ciudad Real - Spain
Draft
1 TENSORS
17
Problem 1.24
Given the B second-order tensor components:
1 0 2
B ij = 0 1 2
3 0 3
Obtain: a) B kk
b) B ij B ij
c) B jk B kj
Solution:
a) B kk = B 11 + B 22 + B 33 = 1 + 1 + 3 = 5
b) B ij B ij = B 1 j B 1 j
1 3
2
+ B 2 jB 2 j + B 3 jB 3 j
123
123
B 11B 11
+
B 21B 21
+
+ B 31B 31
+
B 12B 12
+ B 22B 22
+ B 32B 32
+
B 13B 13
+
+ B 23B 23
+
+ B 33B 33
which the result is:

B ij B ij = 1 1 + 0 0 + 2 2 + 0 0 + 1 1 + 2 2 + 3 3 + 0 0 + 3 3 = 28
c) B jk B kj = B 1k B k1 + B 2k B k 2 + B 3k B k 3
12
3
123
B 11B 11
+
B 12B 21
+
B 13B 31
+ B 22B 22
+
+ B 23B 32
B 21B 12
+
123
B 31B 13
+
+ B 32B 23
+
+ B 33B 33
B jk B kj = B 11B 11 + B 22B 22 + B 33B 33 + 2B 21B 12 + 2B 31B 13 + 2B 32B 23

= 1 1 + 1 1 + 3 3 + 2(0 0) + 2(3 2 ) + 2(0 2 ) = 23
Problem 1.25
The D tensor is given by the algebraic operation D = A : B . Obtain the D components for
the following cases:
a)
b)
when
2 3 2
Aij = 4 1 1
1 1 5
when
7 13 14
Aik B kj = 11 18 11
16 27 31
2 3 1
Bij = 1 2 1
1 2 5
Aik B jk
13 9 17
= 15 9 13
18 12 32
Solution:
a) A : B = 2 2 + 3 3 + 2 1 + 4 1 + 1 2 + 1 1 + 1 1 + 1 2 + 5 5 = 50

Ciudad Real - Spain
Draft
18
b) Taking into account that Tr ( A B T ) = Tr ( AT B) = A : B and Aik B jk = A B T , we can

conclude that A : B = Tr ( A B T ) = 13 + 9 + 32 = 54 .
Problem 1.26
Let us consider the following second-order tensor T = Tr ( E )1 + ( F : E ) E which in indicial
notation is Tij = E kk ij + ( Fkp E kp ) E ij . If the components of E and F are given by:
2 1 4
E ij = 1 5 0
2 0 1
4 3 1
Fij = 2 0 3
2 0 0
a) Obtain the T tensor components. b) Are T and E coaxial tensors? Prove it.
Solution:
Next, we obtain the following scalars:
Tr ( E ) = 2 + 5 + 1 = 8
F : E = 2 4 + 1 3 + 4 1 + 1 2 + 5 0 + 0 3 + 2 2 + 0 0 + 1 0 = 21
Then
2 1 4 50 21 84
1 0 0
0 1 0 + 211 5 0 = 21 113 0
Tij = 8
2 0 1 42 0 29
0 0 1
Two tensors are coaxial when they have the same eigenvectors or when the relationship
T E = E T holds:
50
Tik E kj = 21
42
2
E ik Tkj = 1
21 84 2 1 4 289
113 0 1 5 0 = 155

0 29 2 0 1 142

1 4 50 21 84 289
5 0 21 113 0 = 155

0 1 42 0 29 142
155 284
586 84
42 197
155 284
586 84
42 197
with that we conclude that they are coaxial.

Problem 1.27
Obtain the result of the following algebraic operations: I : I , I : I , I : I , I : I , I : I , I : I ,
I : I , I : I , I sym : I sym , I sym : I , I : I sym , where
I = 11 = I ijkl e i e j e k e l
where
I ijkl = ik jl
(1.20)
I = 11 = I ijkl e i e j e k e l
where
I ijkl = il jk
(1.21)
I ijkl = ij kl
(1.22)
I = 1 1 = I ijkl e i e j e k e l
where
Solution:
(I : I ) ijkl = I ijpq I pqkl = ip jq pk ql = ik jl = I ijkl
( I : I ) ijkl = I ijpq I pqkl = iq jp pl qk = ik jl = I ijkl

Ciudad Real - Spain
Draft
1 TENSORS
19
( I : I ) ijkl = I ijpq I pqkl = ij pq pq kl = qq ij kl = 3I ijkl

( I : I ) ijkl = I ijpq I pqkl = iq jp pk ql = il jk = I ijkl
(I : I ) ijkl = I ijpq I pqkl = ip jq pl qk = il jk = I ijkl
(I : I ) ijkl = I ijpq I pqkl = ip jq pq kl = iq jq kl = ij kl = I ijkl

( I : I ) ijkl = I ijpq I pqkl = iq jp pq kl = iq jq kl = ij kl = I ijkl
We summarize the above in tensorial notation as follows:

I : I = (11) : (11) = 11 = I
I : I = (11) : (11) = 11 = I
I : I = (1 1) : (1 1) = 3(1 1) = 3I
I : I = (11) : (11) = 11 = I
I : I = (11) : (11) = 11 = I
I : I = (11) : (1 1) = 1 1 = I
I : I = (11) : (1 1) = 1 1 = I
Taking into account the definition I sym =

I sym : I sym
(
[(
[
(
)(
) (
)
) (
]
1
11 + 11 : 11 + 11
4
1
= 11 : 11 + 11 : 11 + 11 : 11 + (11 : 11)
4
1
= 11 + 11 + 11 + 11
4
1
= 11 + 11
2
= I sym
=
(1 1) : I sym = I : I sym
(
(
) ( )
) ( )
1
1
1
I + I : I = I : I + I : I = I + I = I =1 1
2
2
2
1
1
1
= I : I + I = I :I + I : I = I + I = I =1 1
2
2
2
I sym : (1 1) = I sym : I =
1.1.3
1
1
I + I = 11 + 11 , we conclude that:
2
2
Tensor Transpose
Problem 1.28
Let A , B and C be arbitrary second-order tensors. Show that:
A : (B C ) = B T A : C = A C T : B
Solution: Expressing the term A : (B C ) in indicial notation we obtain:

Ciudad Real - Spain
Draft
20
A : (B C ) = A ij e i e j : B lk e l e k C pq e p e q
= A ij B lk C pq e i e j : kp e l e q
= A ij B lk C pq kp il
jq
= A ij B ik C kj
Note that, when we are dealing with indicial notation the position of the terms does not
matter, i.e.:
A ij B ik C kj = B ik A ij C kj = A ij C kj B ik
We can now observe that the algebraic operation B ik A ij is equivalent to the components of
the second-order tensor (B T A ) kj , thus,
B ik A ij C kj = (B T A ) kj C kj = B T A : C .
Likewise, we can state that A ij C kj B ik = (A C T ): B .

Problem 1.29
r
Let u , v be vectors and A be a second-order tensor. Show that the following relationship
holds:
r
r r
r
u AT v = v A u
Solution:
r
r
u AT v
u i e i A jl e l e j v k e k
u i A jl il v k jk
u l A jl v j
1.1.4
r
r
= v A u
= v k e k A jl e j e l u i e i
= v k kj A jl u i il
= v j A jl u l
Symmetry and Antisymmetry
Problem 1.30
Show that : W = 0 is always true when is a symmetric second-order tensor and W is an
antisymmetric second-order tensor.
Solution:
: W = ij (e i e j ) : Wlk (e l e k ) = ij Wlk il
jk
= ij Wij (scalar)
Thus,
ij Wij = 1 j W1 j + 2 j W2 j + 3 j W3 j
123
123
4 4
123
4 4
31W31
21W21
11W11
+
+
+
32 W32
22 W22
12 W12
+
+
+
33W33
23W23
13W13

Ciudad Real - Spain
Draft
1 TENSORS
21
Taking into account the characteristics of a symmetric and an antisymmetric tensor, i.e.
12 = 21 , 31 = 13 , 32 = 23 , and W11 = W22 = W33 = 0 , W21 = W12 , W31 = W13 ,
W32 = W23 , the equation above becomes:
:W =0
Q.E.D.
Problem 1.31
Show that a) M Q M = M Q sym M ; b) A : B = A sym : B sym + A skew : B skew where M is a

vector, and Q , A , B are arbitrary second-order tensors; c) Show that the relationship
ijk T jk = 0 i holds, where T is symmetric, i.e. Tij = T ji .
Solution:
r
a) M Q M = M (Q sym + Q skew ) M = M Q sym M + M Q skew M

r
(r
Since the relation M Q skew M = Q skew : 1 2M = 0 holds, it follows that:

M
4 4
3
symmetric tensor
r
r r
r
M Q M = M Q sym M
r
r
NOTE: We can make the geometric interpretation of M Q skew M = 0 as follows. Note that
r r r
r
r r r r
the algebraic operation Q skew M = q (M) is a vector, thus M Q skew M = M q (M) = 0 , which
r
r r
implies that M and q (M) are orthogonal vectors. With that we conclude that: the projection of
r
r r
an antisymmetric second-order tensor according to the direction ( M ) is a vector ( q (M) ) which
r
is orthogonal to M , see figure below:
r
Q M
r r r
q (M ) M = 0
r
r r
q (M) = Q skew M
r
M
r
M
b)
A :B
= ( A sym + A skew ) : (B sym + B skew )

= A sym : B sym + 1sym : B skew + 1skew : B sym + A skew : B skew
A 4243 A 4243
=0
= A sym : B sym + A skew : B skew
=0
Then, it is also valid that:

A : B sym = A sym : B sym

Ciudad Real - Spain
Draft
A : B skew = A skew : B skew
Q.E.D.
22
c)
ijk T jk = ij1 T j1 + ij 2 T j 2 + ij 3 T j 3 = 0 i
= i11 T11 + i 21 T21 + i 31 T31 + i12 T12 + i 22 T22 + i 32 T32 + i13 T13 + i 23 T23 + i 33 T33
= i 21 T21 + i 31 T31 + i12 T12 + i 32 T32 + i13 T13 + i 23 T23 = 0 i
Then, the vector components are:

i =1
1 jk T jk = 132 T32 + 123 T23 = T32 + T23 = 0 T32 = T23
i=2
2 jk T jk = 231 T31 + 213 T13 = T31 T13 = 0 T31 = T13
i=3
3 jk T jk = 321 T21 + 312 T12 = T21 + T12 = 0 T21 = T12
with that we demonstrate that if ijk T jk = 0 i holds, this implies that T is symmetric, i.e.
T = TT .
Problem 1.32
Given a second-order tensor A in which the components of the symmetric part is known in
the Cartesian system:
sym
A ij
4 2 0
= 2 1 0
0 0 3
Obtain N A N , where the unit vector components are Ni = [1 0 0].
Solution:
In Problem 1.31 it was shown that N A N = N A sym N with that we obtain:

N A N = N A
sym
sym
N = N i A ij N j
4 2 0 1
= [1 0 0] 2 1 0 0 = 4

0 0 3 0
Problem 1.33
Let W be an antisymmetric tensor. a) Show that W W is a symmetric second-order tensor.
b) Show also that (W T W W) : 1 = 0 .
Solution:
a) If we show that (W W ) skew = 0 holds, then we prove that W W is symmetric.
] [
1
1
(W W) (W W ) T = (W W) W T W T
2
2
1
= [(W W ) W W ]
2
=0
(W W ) skew =
where we have applied the antisymmetric tensor property W = W T .

Alternative solutions a) Taking into account the definition of antisymmetric tensor W = W T ,
we obtain:

Ciudad Real - Spain
Draft
1 TENSORS
23
W W = W T W = W T W T = (W W ) T
We can also check the symmetry by means of the tensor components:

0
( W W ) ij = W12
W13
W12
0
W23
= W13 W23
W12 W23
2
W12
W13 0
W23 W12
0 W13
2
W13
W13 W23
2
W12
2
W23
W12 W13
W12
0
W23
W13
W23
W12 W13
2
2
W13 W23
W12 W23
b) (W T W W ) : 1 = (W pi W pk Wkj ) ij = W pi (W pk Wki ) = W : (W W) = 0 , since the double

scalar product between a symmetric tensor (W W ) and an antisymmetric tensor ( W ) is zero.
Problem 1.34
1
2
Let B be a second-order tensor such that B pq = pqs a s with a i = ijk B jk . Prove that B is an
antisymmetric tensor.
Solution:
1
1
1
B pq = pqs a s = pqs sjk B jk = pqs sjk B jk = pqs jks B jk
2
2
2
Taking into account the relationship pqs jks = pj qk pk qj we obtain:

1
1
pqs jks B jk = ( pj qk pk qj )B jk
2
2
1
= ( pj qk B jk pk qj B jk )
2
1
= (B pq B qp )
2
= B skew
pq
B pq =
Alternative solution:
Taking into account that B qp = qps a s and pqs = qps , we can conclude that:
B pq = pqs a s = qps a s = B qp
B = B T (antisymmetric)
Problem 1.35
Show that the tensor A skew A sym + A sym A skew is an antisymmetric tensor.
Solution: Denoting by B = A skew A sym + A sym A skew , and by taking into account that
A skew = (A skew ) T , A sym = (A sym ) T , we can conclude that:
B = A skew A sym + A sym A skew
= A skew A sym A sym ( A skew ) T
= A skew A sym ( A skew A sym ) T
= 2( A skew A sym ) skew

Ciudad Real - Spain
Draft
24
Problem 1.36
Let T be ran arbitrary second-order tensor, and n be a vector. Check if the relationship
r
n T = T n is valid.
Solution:
n T = n i e i Tkl (e k e l )
= n i Tkl ik e l
T n = Tlk (e l e k ) n i e i
= n i Tlk ki e l
and
= n k Tkl e l
= n k Tlk e l
= (n1 T1l + n 2 T2 l + n 3 T3l )e l
= (n1 Tl1 + n 2 Tl 2 + n 3 Tl 3 )e l
With the above we can prove that n k Tkl n k Tlk , then:

r
r
n T T n
If T is a symmetric tensor, it follows that the relationship n T sym = T sym n holds.

Problem 1.37
r
Obtain the axial vector w associated with the antisymmetric tensor ( x a ) skew .
r
Solution: Let z be an arbitrary vector, it then holds that:

r r
r r r
( x a ) skew z = w z
r
r r
where w is the axial vector associated with ( x a ) skew . Using the definition of an
antisymmetric tensor:
r r
r r
1 r r
1 r r r r
( x a ) skew = ( x a ) ( x a ) T = [ x a a x ]
2
2
r r skew r r r
and by replacing it with ( x a ) z = w z , we obtain:
r r r r r
r r
1 r r r r r r r
[ x a a x ] z = w z [ x a a x ] z = 2w z
2
r r r r r r
r r
By using the equation [x a a x ] z = z ( x a ) , (see Problem 1.17), the above
equation becomes:
r r r r r r
r r
r r
r r
r
[ x a a x ] z = z ( x a ) = (a x ) z = 2 w z
with the above we can conclude that:

r 1 r r
r r
w = (a x ) is the axial vector associated with ( x a ) skew
2
Problem 1.38
Let us consider two symmetric tensors W (1) and W ( 2) , and their axial vectors represented
r
r
respectively by w (1) and w ( 2) . Show that:
r
r
r
r
W (1) W ( 2 ) = ( w ( 2 ) w (1) ) ( w (1) w ( 2) )1
r
r
Tr W (1) W ( 2 ) = 2( w (1) w ( 2 ) )
Solution: Given antisymmetric tensor properties, we can obtain that:

Ciudad Real - Spain
Draft
1 TENSORS
25
r r
r
W (1) a = w (1) a
r
r r
T
r r
r
a W (1) = a w (1)
and
W ( 2) a = w ( 2 ) a
r
r r (1)
a W (1) = a w
r
r r
a W (1) = a w (1)
r
r
Then, by applying the dot product a W (1) W ( 2) a we obtain:
r
r
r r
r
r
a W (1) W ( 2) a = (a w (1) ) ( w ( 2 ) a)
(
)(
)(
)
We will continue the development in indicial notation:

(1
(
(
(a i Wij1) )(W jk) a k ) = ( ijk a j wk1) )( ipq w (p2 ) a q )
(
(
(
a i (Wij1) W (1) )a k = a j ( ijk ipq wk1) w (p2 ) )a q = a j ( jp kq jq kp ) wk1) w (p2) a q
jk
[
[w
(
(
= a j jp kq wk1) w (p2 ) jq kp wk1) w (p2) a q
=aj
(
(
jq wk1) wk2) a q
(1) ( 2 )
q wj
In tensorial notation the above equation becomes:
r
] a
r
r r r
r
r
r
a W (1) W ( 2 ) a = a ( w ( 2 ) w (1) ) ( w (1) w ( 2 ) )1
With that we demonstrate that W (1) W ( 2) = ( w ( 2) w (1) ) ( w (1) w ( 2) )1 .

b)
Tr W (1) W ( 2)
[
[
r
r
r
r
= Tr ( w ( 2) w (1) ) ( w (1) w ( 2 ) )1
r
r
r
r
= Tr ( w ( 2) w (1) ) Tr ( w (1) w ( 2 ) )1
r
r
r
r
= ( w ( 2 ) w (1) ) ( w (1) w ( 2) ) 1[1]
Tr3
2
=3
r
r
= 2( w (1) w ( 2) )
Alternative solution
In this alternative solution we use the tensor components in which it fulfills:
(
Wij1)
(1
= W12 )
W (1)
12
(
Wij 2 )
(
= W122 )
W ( 2 )
12
(1
W12)
0
(1
W12)
(
W122 )
0
(2
W12 )
(1
W13) 0
(
(1
W23) = w31)
(
0 w21)

(
w31)
0
w1(1)
(2
W13 ) 0
(
(2
W23 ) = w32 )
(
0 w22)

(
w21)
w1(1)
0
(
w32 )
0
w1( 2 )
(
w22 )
w1( 2 )
0
With that we obtain:
[W
(1)
W ( 2) ]ij
(
(
Wik1) Wkj2 )
(
(
= Wik1) Wkj2 )
0
(
= w31)
w (1)
2
(
(
(
(
w31) w32 ) w21) w22 )
(
(
w11) w22)
=
(
w32 ) w1(1)

Ciudad Real - Spain
(
w31)
0
w1(1)
(
w21) 0
(
w1(1) w32 )
(
0 w22)
(
w32 )
0
w1( 2 )
(
w21) w1( 2 )
(
(
w32 ) w31) w1(1) w1( 2 )
(
(
w32 ) w21)
Draft
(
w22 )
w1( 2 )
0
(
(
w31) w1 2 )
(
(
w22) w31)
(1) ( 2 )
( 2 ) (1)
w2 w2 w1 w1
26
In the term (11) we sum and subtract the term w1( 2) w1(1) , in the term (22) we sum and subtract
(
(
(
(
the term w22) w21) and in the term (33) we add and subtract the term w32) w31) , so,
(
(
Wik1) Wkj2 )
w1( 2) w1(1)
(
= w22) w1(1)
w ( 2) w (1)
3 1
(
w1( 2 ) w31)
(
(
w22 ) w31) +
(
(
w32 ) w31)
(
w1( 2) w21)
(
(
w22) w21)
(
(
w32) w21)
(
(
(
(
w1(1) w1( 2 ) w21) w22) w31) w32)
0
+
(
(
(
(
w1(1) w1( 2 ) w21) w22 ) w31) w32 )
0
(
w1(1) w1 2 )
(
(
w21) w22 )
(
(
w31) w32 )

(
(
(
(
(
(
(
(
(
Wik1) Wkj2 ) = wi( 2) w (j1) ( w11) w1( 2) + w21) w22) + w31) w32) ) ij = wi( 2) w (j1) ( wk1) wk2) ) ij
We leave to the reader to obtain the trace.

NOTE: The alternative solution by means of components was made only as a check. The
reader must give priority to the solution via indicial or tensorial notation, since the solution via
components is not always so simple to obtain.
1.1.5
Cofactor. Adjugate. Inverse. Particular Tensors.

Determinant
Problem 1.39
r r
Show that Tr (a b) = a b .
Solution:
r r
Tr (a b) = Tr (a i e i ) (b j e j )
= a i b j Tr e i e j

= a i b j (e i e j ) = a i b j ij
= aibi
r r
= ab
Problem 1.40
1
2
Given that Tij = E kk ij + 2 Eij , W = Tij E ij , and P = Tij Tij , show that:
W = E : E +
[Tr( E )]2
and
P = 4 2 E : E + (3 + 4 )[Tr ( E )]
Solution 1: (Indicial notation)
1
1
1
1
W = Tij E ij = E kk ij + 2 E ij E ij = E kk ij E ij + 2 E ij E ij = E kk E ii + 2 E ij E ij
2
2
2
2
since E kk = E ii = Tr (E ) and Eij Eij = E : E , we can conclude that W = E : E +

Ciudad Real - Spain
Draft
[Tr( E )]2 .
1 TENSORS
27
)(
P = Tij Tij = E kk ij + 2 E ij E qq ij + 2 E ij
= E kk ij E qq ij + E kk ij 2 E ij + 2 E ij E qq ij + 2 E ij 2 E ij
= 2 E kk ii E qq + 2 E kk E ii + 2 E ii E qq + 4 2 E ij E ij
= 3 2 E kk E qq + 4 E kk E ii + 4 2 E ij E ij
= (3 + 4 )E kk E qq + 4 2 E ij E ij
With that we demonstrate that P = 4 2 E : E + (3 + 4 )[Tr ( E )]2 .

Solution 2: (Tensorial notation)
In tensorial notation we obtain:
1
T = Tr ( E )1 + 2 E , W = T : E , and P = T : T
2
Then
1
1
1
W = T : E = (Tr ( E )1 + 2 E ) : E = (Tr ( E )1 : E + 2 E : E )
2
2
2
1
= (Tr ( E ) Tr ( E ) + 2 E : E )
2
=
[Tr( E )]2 + E : E
P = T : T = (Tr ( E )1 + 2 E ) : (Tr ( E )1 + 2 E )
= [Tr ( E )] 1 : 1 + 2 Tr ( E ) 1 : E + 2 Tr ( E ) { + (2 ) 2 E : E
E :1
{
{
2
=3
= Tr ( E )
= Tr ( E )
= 3 2 [Tr ( E )] + 4 [Tr ( E )] + 4 2 E : E
2
= (3 + 4 )[Tr ( E )] + 4 2 E : E
2
Problem 1.41
Let ij be the second-order tensor components which are a function of ij , ij = ij ( ij ) , and
is given by:
ij = kk ij + 2 ij
Tensorial
= Tr ( )1 + 2
where and are scalars. Starting by the above equation, obtain an expression for ij in
function of ij , i.e. ij = ij ( ij ) . Express the result in indicial and tensorial notation.
Solution:
Indicial notation
Tensorial notation
ij = kk ij + 2 ij
= Tr ( )1 + 2
2 ij = ij kk ij
2 = Tr ( )1
ij =
1
ij
kk ij
2
2

Ciudad Real - Spain
Draft
Tr ( )1
2
2
28
Next, we need to obtain the following trace kk , to do this we obtain the trace of ij :
Indicial notation
ij = kk ij + 2 ij
Tensorial notation
(i = j )
: 1 = Tr ( )1 : 1 + 2 : 1
ii = kk ii + 2 ii = kk 3 + 2 kk
kk = (3 + 2 ) kk
kk =
Tr ( ) = Tr ( )3 + 2 Tr ( )
1
kk
(3 + 2 )
Tr ( ) =
1
Tr ( )
(3 + 2 )
Then
Indicial notation
ij =
=
Tensorial notation
ij
kk ij
2
2
Tr ( )1
2
2
1
1
ij
kk ij
2
2 (3 + 2 )
Tr ( )1
2
2 (3 + 2 )
Problem 1.42
Let T be a second-order tensor. Show that:
(T ) = (T )
m T
T m
and
( )
Tr T T
( )
= Tr T m .
Solution:
(T )
( )
For the second demonstration we can use the trace property Tr (T ) = Tr (T ) , thus:
Tr (T ) = Tr (T ) = Tr (T )
m T
= (T T L T ) = T T T T L T T = T T
T
T m
m T
Problem 1.43
Show that T : 1 = Tr (T ) , where T is an arbitrary second-order tensor.
Solution:
T : 1 = Tij e i e j : kl e k e l
= Tij kl ik jl
= Tij ij = Tii = T jj
= Tr ( T )
Problem 1.44
Show that if and D are second-order tensors, the following relationship is valid:
D = Tr ( D )
Solution: We start with the following definition:

Ciudad Real - Spain
Draft
1 TENSORS
29
D = ij D ji
= kj D jl ik il = kj D jl lk
= kj D jl lk
1 3
2
( D )
kl
= ( D) kl lk = ( D) kk = ( D) ll
= Tr ( D)
An alternative demonstration would be:

D = ij D ji = ij D jk ik
= ( D ) : 1
= Tr ( D )
Problem 1.45
Show that:
det (S ) S =
1
rjk tpq S rt S jp S kq
6
(1.23)
Solution:
det (S ) = ijk S 1i S 2 j S 3k
(1.24)
pqr det(S ) = ijk S pi S qj S rk
(1.25)
pqr pqr det (S ) = pqr ijk S pi S qj S rk
(1.26)
1 24
4 3
6
det (S ) =
1
pqr ijk S pi S qj S rk
6
(1.27)
Problem 1.46
Show that A tpq = rjk A rt A jp A kq .
Solution:
We start with the following definition:
A = rjk A r1 A j 2 A k 3
A tpq = rjk tpq A r1A j 2 A k 3
(1.28)
and also taking into account that the term rjk tpq can be replaced by:
rjk tpq
rt rp rq
= jt jp jq
kt kp kq
(1.29)
= rt jp kq + rp jq kt + rq jt kp rq jp kt jq kp rt kq jt rp
Then, by substituting (1.29) into (1.28) we can obtain:

A tpq
= A t1 A p 2 A q 3 + A p1 A q 2 A t 3 + A q1 A t 2 A p 3 A q1 A p 2 A t 3 A t1 A q 2 A p 3 A p1 A t 2 A q 3
= A t1 1 jk A pj A qk + A t 2 2 jk A pj A qk + A t 3 3 jk A pj A qk
= rjk A rt A jp A kq = rjk A tr A pj A qk

Ciudad Real - Spain
Draft
30
NOTE: Let us consider that C = A B ( C ij = A ik B kj ), then we can obtain

C = A B = rjk C r1C j 2 C k 3 = rjk [A B ]r1 [A B ] j 2 [A B ]k 3 = rjk ( A rt B t1 )( A jpB p 2 )( A kqB q 3 )
= rjk A rt A jp A kqB t1B p 2B q 3 = A tpq B t1B p 2B q 3 = A B
So, we have shown that A B = A B .

Considering that tpq
1t 1 p 1q 1t 2t 3t
= 2t 2 p 2 q = 1 p 2 p 3 p
3 t 3 p 3 q 1q 2 q 3 q
and that A B = A B we can
obtain:
A tpq
1t 2t 3t A11 A12 A13

= rjk tpq A r1A j 2 A k 3 = 1 p 2 p 3 p A 21 A 22 A 23
1q 2 q 3q A 31 A 32 A 33
1t 2t 3t A11
= 1 p 2 p 3 p A 21
1q 2 q 3q A 31
A12
A 22
A 32
A13
A 23
A 33
Note that 1t A11 + 2t A 21 + 3t A 31 = st A s1 = A t1 , with that we can obtain:

A tpq = rjk tpq A r1A j 2 A k 3
A t1
= A p1
A q1
At2
A p2
A q2
At3
A p 3 = rjk A tr A pj A qk
A q3
Problem 1.47
1
6
Show that A = rjk tpq A rt A jp A kq .

Solution:
Starting with the definition A tpq = rjk A rt A jp A kq , and by multiplying both sides of the
equation by tpq , we obtain:
A tpq tpq = rjk tpq A rt A jp A kq
(1.30)
Note that tpq tpq = tt pp tp tp = tt pp tt = 6 . Then, the relationship (1.30) becomes:

A =
1
rjk tpq A rt A jp A kq
6
Problem 1.48
Show the following property:
r r r
r r r
r r r
r r r
(B a) (b c) (B b) (a c) + (B c ) (a b) = Tr (B) a (b c)
(1.31)
Solution:
Expressing in Voigt notation the left side of the above equation we obtain:
Ciudad Real - Spain
Draft
1 TENSORS
31
ijk (B a) i b j c k ijk (B b) i a j c k + ijk (B c ) i a j b k =

= ijk [(B i1a1 + B i 2 a 2 + B i 3 a 3 )b j c k (B i1b1 + B i 2 b 2 + B i 3b 3 )a j c k +
+ (B i1 c 1 + B i 2 c 2 + B i 3 c 3 )a j b k ]
= ijk [(B i1a1b j c k + B i 2 a 2 b j c k + B i 3 a 3b j c k ) (B i1b1 a j c k + B i 2 b 2 a j c k + B i 3b 3 a j c k ) +
+ (B i1 c 1 a j b k + B i 2 c 2 a j b k + B i 3 c 3 a j b k )]
= ijk [B i1 (a1b j c k b1a j c k + c 1a j b k ) + B i 2 (a 2 b j c k b 2 a j c k + c 2 a j b k ) +
+ B i 3 (a 3b j c k b 3 a j c k + c 3 a j b k )]
= (1 jk B 11 + 2 jk B 21 + 3 jk B 31 )(a1b j c k b1a j c k + c 1 a j b k ) +
+ (1 jk B 12 + 2 jk B 22 + 3 jk B 32 )(a 2 b j c k b 2 a j c k + c 2 a j b k ) +
(1.32)
+ (1 jk B 13 + 2 jk B 23 + 3 jk B 33 )(a 3b j c k b 3 a j c k + c 3 a j b k )
Note that:
a1 a 2
1 jk (a1b j c k b1a j c k + c 1a j b k ) = b1 b 2
c1
c2
a3
b 3 = ijk a i b j c k
c3
2 jk (a1b j c k b1a j c k + c 1a j b k ) = 3 jk (a1b j c k b1a j c k + c 1a j b k ) = 0

whereby the equation in (1.32) becomes:
r r r
B 11 ijk a i b j c k + B 22 ijk a i b j c k + B 33 ijk a i b j c k = (B 11 + B 22 + B 33 ) ijk a i b j c k = Tr (B) a (b c )
Q.E.D.
Note also that:
r r r
r r r
r r r
r r r
(B T a) (b c ) (B T b) (a c ) + (B T c ) (a b) = Tr (B) a (b c )
since Tr (B) = Tr (B T ) . It is also valid the following:
r
r r r r
r r r
r
r r r
(B a) (b c ) + a ((B b) c) + a (b (B c )) = Tr (B) a (b c )
r r
r r r
r
r r
r
r r r
[(B a), b, c ] + [a, (B b), c ] + [a, b, (B c )] = I B [a, b, c ]
(1.33)
Problem 1.49
Show the following property:
r
r
r
r r r
( A a) ( A b) ( A c ) = det ( A ) a (b c )
r
(1.34)
r
where A is a non-singular second order tensor, and a , b and c are linearly independent
vectors.
Solution:
A non-singular tensor det( A ) A 0
r r r
r r r
a , b , c linearly independent vectors a b c 0 .

Ciudad Real - Spain
Draft
32
(r r )
We express the scalar triple product in indicial notation, i.e. a b c = ijk a i b j c k , and by
multiply both sides of this equation by the determinant of A we obtain:
r r r
a b c A = ijk a i b j c k A
It was proven in Problem 1.46 that A ijk = pqr A pi A qj A rk , thus:
r r r
a b c A = ijk a i b j c k A
= pqr A pi A qj A rk a i b j c k
= pqr ( A pi a i )( A qj b j )( A rk c k )
r
r
r
= ( A a) ( A b) ( A c )
Problem 1.50
r
Let a , b be arbitrary vectors and , be scalars. Show that:
r r
r r
det 1 + a b = 3 + 2 a b
(1.35)
Solution: The determinant of A is given by A = ijk A i1 A j 2 A k 3 . If we denote by

A ij ij + a i b j , thus, A i1 = i1 + a i b1 , A j 2 = j 2 + a j b 2 , A k 3 = k 3 + a k b 3 , then the
equation in (1.35) can be rewritten as:
r r
det 1 + a b = ijk ( i1 + a i b1 )
j2
+ a j b 2 ( k 3 + a k b 3 )
(1.36)
By developing the equation (1.36), we obtain:
r r
det 1 + a b = ijk 3 i1 j 2 k 3 + 2 a k b 3 i1 j 2 + 2 a j b 2 i1 k 3 + 2 a i b 1 j 2 k 3 +
+ 2 a j b 2 a k b 3 i1 + 2 a i a k b 1b 3 j 2 + 2 a i a j b 1b 2 k 3 + 3 a i a j a k b 1b 2 b 3
Note that: 3 ijk i1 j 2 k 3 = 3 123 = 3 ,
2 ( ijk a k b 3 i1 j 2 + ijk a j b 2 i1 k 3 + ijk a i b1 j 2 k 3 ) =
r r
2 (12 k a k b 3 + 1 j 3 a j b 2 + i 23 a i b1 ) = 2 (a 3b 3 + a 2 b 2 + a1b1 ) = 2 (a k b k ) = 2 (a b)
ijk a i a k b1b 3 j 2 = i 2 k a i a k b1b 3 = a1a 3b1b 3 a 3 a1b1b 3 = 0
ijk a i a j b1b 2 k 3 = ij 3 a i a j b1b 2 = 123 a1a 2 b1b 2 213 a 2 a1b1b 2 = 0
ijk a i a j a k b1b 2 b 3 = 0
Notice that, there was no need to expand the terms ijk a i a k b1b 3 j 2 , ijk a i a j b1b 2 k 3 , and
ijk a i a j a k b1b 2 b 3
to
realize
that
these
terms
equal
zero,
since
r
ijk a i a k b1b 3 j 2 = (a a) j b1b 3 j 2 = 0 , similarly for other terms.
Taking into account the above considerations we can prove that:
r r
r r
det 1 + a b = 3 + 2 a b
For the particular case when = 1 the above equation becomes:

Ciudad Real - Spain
Draft
1 TENSORS
33
r r
r r
det 1 + a b = 1 + a b
r r
Then, it is simple to prove that det a b = 0 , since
r r
r r r
det a b = 3 ijk a i a j a k b1b 2 b 3 = 3b1b 2 b 3 [a (a a)] = 0
NOTE: We can extrapolate the equation in (1.35) in such a way that:
det I sym + A B = 3 + 2 A : B
(1.37)
where I sym is the symmetric fourth-order unit tensor, A and B are second-order tensors.
Note that det ( I sym ) = (1) 3 + (1) 2 (0)(0 : 0 ) = 1 and det (1 1) = (0) 3 + (0) 2 (1)(1 : 1) = 0 .
Problem 1.51
r
Let A be an arbitrary second-order tensor. Show that there is a nonzero vector n 0 so that
r r
A n = 0 if and only if det ( A ) = 0 , Chadwick (1976).
r
Solution: Firstly, we show that, if det ( A ) A = 0 n 0 . Secondly, we show that, if

r r
n 0 det ( A ) A = 0 .
r r r
We assume that det ( A ) A = 0 , and we choose an arbitrary basis {f , g, h} (linearly

independent), then:
r r r
r
r
r
f g h A = ( A f ) ( A g) ( A h)
(see Problem 1.49)
Due to the fact that det ( A ) A = 0 , the implication is that:
r
r
r
( A f ) ( A g) ( A h) = 0
r
r
r
Thus, we can conclude that the vectors ( A f ) , ( A g) , ( A h) , are linearly dependent. This
implies that there are nonzero scalars , , so that:

r
(r
( A f ) + ( A g) + ( A h) = 0 A f + g + h = 0 A n = 0
r
r r r
where n = f + g + h 0 since {f , g, h} is linearly independent.

r
Now we choose two vectors k , m , which are linearly independent to n . Once more, we
apply definition:
r r r
r
r
r
k (m n) A = ( A k ) [( A m) ( A n)]
r r r
r r r
r r
Considering that A n = 0 , and k (m n) 0 owing to the fact that k , m , n are linearly
independent, we can conclude that:
r r r
k (m n) A = 0
14 4
2 3
A =0

Ciudad Real - Spain
Draft
34
Problem 1.52
Let F be an arbitrary second-order tensor. Show that the resulting tensors C = F T F and
b = F F T are symmetric tensors and semi-positive definite tensors. Also check in what condition are
C and b positive definite tensors.
Solution: Symmetry:
C T = (F T F )T = F T (F T )T = F T F = C
b T = (F F T ) T = (F T )T F T = F F T = b
Thus, we have shown that C = F T F and b = F F T are symmetric tensors.

To prove that the tensors C = F T F and b = F F T are semi-positive definite tensors, we
start with the definition of a semi-positive definite tensor, i.e., a tensor A is semi-positive
r
definite if x A x 0 holds, for all x 0 . Thus:
x (F T F ) x = F x F x
= (F x ) (F x )
= F x
x (F F T ) x = x F F T x
T
= (F x) (F T x)
2
= FT x 0
Or in indicial notation:
x i C ij x j
= x i ( Fki Fkj ) x j
= ( Fki x i )( Fkj x j )
= Fki x i
= x i ( Fik F jk ) x j
= ( Fik x i )( F jk x j )
x i bij x j
= Fik x i
Thus, we proved that C = F T F and b = F F T are semi-positive definite tensors. Note that
r
r
2
x C x = F x
equals zero, when x 0 , if F x = 0 . Furthermore, by definition
r
r
F x = 0 with x 0 if and only if det ( F ) = 0 , (see Problem 1.51). Then, the tensors
C = F T F and b = F F T are positive definite if and only if det ( F ) 0 .
Problem 1.53
r
Let dX (1) , dX ( 2) , dX (3) , dx (1) , dx ( 2) , dx (3) be vectors, and they are related to each other as
r
r
r
r
r
r
follows dx (1) = F dX (1) , dx ( 2) = F dX ( 2) , dx (3) = F dX (3) , where F is a non-singular
r
r
r
second-order tensor and F 1 . a.1) Considering dV = dx (1) (dx ( 2) dx (3) ) 0 and
r
r
r
dV0 = dX (1) (dX ( 2 ) dX (3) ) 0 , obtain a relationship between the scalars dV and dV0 in
r
r
r
r
terms of F . a.2) Obtain the relationship between c = dX ( 2) dX (3) 0 and
r
r
r
r
c * = dx ( 2 ) dx ( 3 ) 0 .
Solution
a.1) Taking into account the problem statement it fulfills that:
r
r
r
r
r
r
dV = dx (1) (dx ( 2 ) dx (3) ) = ( F dX (1) ) ( F dX ( 2 ) ) ( F dX (3) )
r
(r r )
In Problem 1.49 it was proven that a b c A = ( A a) ( A b) ( A c) , so
r
r r r
r
r
a b c A = ( A a) ( A b) ( A c )
r
r
r
r
r
r
dX (1) dX ( 2 ) dX (3) F = ( F dX (1) ) ( F dX ( 2 ) ) ( F dX (3) )

Ciudad Real - Spain
Draft
1 TENSORS
35
With that we conclude that:
r
r
r
r
r
r
dV = dx (1) (dx ( 2 ) dx (3) ) = ( F dX (1) ) ( F dX ( 2 ) ) ( F dX (3) )
r
r
r
= F dX (1) (dX ( 2) dX (3) )
thus
dV = F dV0
a.2) Taking into account the previous equations we get:

dV = F dV0
r
r
r
r
r
r
dx (1) (dx ( 2 ) dx (3) ) = F dX (1) dX ( 2 ) dX (3)
r
r
r
r
r
r
dx (1) (dx ( 2 ) dx (3) ) = F ( F 1 dx (1) ) dX ( 2) dX (3)
r
r
r
r
r
r
dx (1) (dx ( 2 ) dx (3) ) = dx (1) F F T dX ( 2 ) dX (3)
r
r
r
r
(dx ( 2 ) dx (3) ) = F F T dX ( 2 ) dX (3)
r
r
c * = F F T c
[
[
]
]
])
NOTE 1: Note that c * F c . We can rewrite the above equation as follows
]
[
r
r
r
r
dx ( 2 ) dx (3) = F F T dX ( 2) dX (3)
r
r
r
r
( F dX ( 2 ) ) ( F dX (3) ) = F F T dX ( 2 ) dX (3)
The tensor F F T is known as the cofactor of F , i.e. cof ( F ) = F F T with this we define
the inverse of a tensor:
cof ( F ) = F F T
F 1 =
T T
= [cof ( F )]
F F 1 = [cof ( F )]
1
[cof (F )]T = 1 [adj( F )]
F
F
r
dX (1)
r
r
r
c = dX ( 2) dX (3)
[F F ]
r
r
dx (1) = F dX (1)
F
r
dX ( 3 )
r
dX ( 2 )
r
r
c* F c
r
r
c * = [cof ( F )] c
r
r
r
c * = dx ( 2 ) dx ( 3 )
r
r
dx (3) = F dX (3)
dV = F dV0
r
r
r
dV0 = dX (1) (dX ( 2 ) dX (3) ) 0
r
r
dx ( 2 ) = F dX ( 2 )
r
dV = dx (1)
(dx ( 2) dx (3) ) 0
r
NOTE 2: Let us suppose now that F = A B , and let us consider three vectors a (b c ) 0 ,
r r
r
r r
r
and a * = B a , b * = B b , c * = B c , thus by apply the previous definitions we can state:

Ciudad Real - Spain
Draft
36
r
r r r
r
r
F a (b c ) = ( F a ) ( F b ) ( F c )
r
r
r
= ( A B a) ( A B b) ( A B c)
r
r
r
= ( A a*) ( A b*) ( A c*)
r
r
r
= A a * (b * c * )
r
r
r
= A (B a ) (B b ) (B c )
r r r
= A B a (b c )
With that we can conclude that: if F = A B then F = A B = A B .

Problem 1.54
Let A and B be orthogonal tensors, show that the tensor C = A B is also an orthogonal
tensor.
Solution: By definition, a tensor is orthogonal if C 1 = C T holds:
C 1 = ( A B) 1 = B 1 A 1 = B T A T = ( A B) T = C T
Q.E.D.
Problem 1.55
Show that adj( A B) = adj(B) adj( A ) and cof( A B) = [cof( A )] [cof(B)] .
Solution:
Based on the definition of the inverse of a tensor we can say that:
B 1 A 1 =
[adj(B)] [adj(A )]
B
A B B 1 A 1 = [adj(B)] [adj( A )] = [cof(B)]
A B (A B ) = [adj(B)] [adj( A )] =
1
AB
[cof( A)]T
( [cof(A)] [cof(B)] )
(1.38)
[adj(A B)] = [adj(B)] [adj(A)] = ([cof(A)] [cof(B)])T

A B
adj( A B) = [adj(B)] [adj( A )] = ([cof( A )] [cof(B)])
where we have used the property A B = A B . Also taking into account the definition of
adjugate and cofactor we can conclude that:
adj( A B) = ([cof( A B)]) = ([cof( A )] [cof(B)])
T
[cof( A B)] = [cof( A )] [cof(B)]
(1.39)
Problem 1.56
Show that:
r
r
r r
( A a) ( A b) = [cof( A )] (a b)
(1.40)
Solution:
Starting from the equation A tpq = rjk A rt A jp A kq (see Problem 1.46), and by multiply both
sides by a t b p , we obtain:

Ciudad Real - Spain
Draft
1 TENSORS
37
A tpq a t b p = rjk A rt A jp A kq a t b p = rjk ( A rt a t )( A jp b p ) A kq
Multiplying both sides by A 1 we obtain:

qs
A tpq a t b p A 1 = rjk ( A rt a t )( A jp b p ) A kq A 1 = rjk ( A rt a t )( A jp b p ) ks = rjs ( A rt a t )( A jp b p )
qs
qs
1
Note that A qs =
[cof ( A )] sq
holds, whereby the above equation becomes:
1
A tpq a t b p A qs = A tpq a t b p
[cof ( A )] sq
A
= [cof ( A )] sq tpq a t b p = rjs ( A rt a t )( A jp b p )
r
r r
r
[cof( A )] (a b) = ( A a) ( A b)
Problem 1.57
Show that:
r
r
r
r
r r
r
r
r
r r r
a ( A b) ( A c ) + ( A a) b ( A c ) + ( A a) ( A b) c = Tr ([cof ( A )]) a (b c )
(1.41)
Solution:
In Problem 1.56 it was demonstrated that [cof( A )] (a b) = ( A a) ( A b) , thus the

following relationships hold:
r r
r
r
r
r
a [cof( A )] (b c ) = a ( A b) ( A c )
r
r
r r
r
r
r r
r
b [cof( A )] (a c ) = b [( A a) ( A c ) ] = ( A a) b ( A c )
r
r
r
r r
r
r
r
r
c [cof( A )] (a b) = c ( A a) ( A b) = ( A a) ( A b) c
Summing the above equations we obtain:
r r
r
r
r r
r
r r
a [cof( A )] (b c ) b [cof( A )] (a c ) + c [cof( A )] (a b) =
r
r
r
r
r r
r
r
r
= a ( A b) ( A c ) + ( A a) b ( A c ) + ( A a) ( A b) c
According to Problem 1.48 the following is true:
r
r
r r
r r
r r
([cof( A)] a) (b c) ([cof( A)] b) (a c) + ([cof( A)] c ) (a b) = Tr ([cof( A)])[c (a b)]
r
r
where II A = Tr [cof( A )] is the second principal invariant of A , thus:
r r
= II A [c (a b)]
r
r
r
r
r r
r
r
r
r r r
a ( A b) ( A c ) + ( A a) b ( A c ) + ( A a) ( A b) c = II A a (b c )
NOTE 1: We can summarize that:
] [
] [
] (see Problem 1.48)
r
r
r
r
r r
r
r
r
r r r
a [( A b) ( A c )] + ( A a) [b ( A c )] + ( A a) [( A b) c ] = II [a (b c )]
r
r
r
r r r
( A a) [( A b) ( A c )] = III [a (b c )] (see Problem 1.49)
r
r r r r
r
r r
r
r r r
( A a) (b c) + a ( A b) c ) + a b ( A c ) = I A a (b c )

Ciudad Real - Spain
Draft
(1.42)
(1.43)
(1.44)
38
r r r
where I A = Tr (A ) , II A = Tr ([cof( A )]) , III A = det (A ) . Using the notation a (b c ) [a, b, c ] ,

the above equations can also be written as follows:
r r
r r r
r
r r
r
r r r
[( A a), b, c ] + [a, ( A b), c] + [a, b, ( A c )] = I A [a, b, c ]
r
r r
r
r
r r
r
r
r r r
[a, ( A b), ( A c )] + [( A a), b, ( A c )] + [( A a), ( A b), c ] = II A [a, b, c]
r
r
r
r r r
[( A a), ( A b), ( A c )] = III A [a, b, c ]
r
r r r
NOTE 2: If we consider three linearly independent vectors [a (b c )] [a, b, c ] 0 , and

three vectors such as:
r
r
r
r
f = 1a + 2 b + 3 c
r
r
r
r
g = 1 a + 2 b + 3 c
r
r
r
r
h = 1a + 2 b + 3 c
r
f
1

r
g = 1
r
h 1

2 3 a
r
2 3 b
r
2 3 c

(1.45)
And according to Cramers rule, (see Problem 1.16), the following relationships are true:
r r r
r r r
r r r
r r r
f (b c ) [ f , b, c ]
[a, f , c ]
[a, b, f ]
;
;
1 = r r r r r r
2 = r r r
3 = r r r
a (b c ) [a, b, c ]
[a, b, c ]
[a, b, c ]
r r
r r
r
r r r
r
[g, b, c ]
[a, g, c ]
[a, b, g]
;
;
1 = r r r
2 = r r r
3 = r r r
[a, b, c ]
[a, b, c ]
[a, b, c ]
r r r
r r
r
r r r
[h, b, c ]
[a, h, c ]
[a, b, h]
;
;
1 = r r r
2 = r r r
3 = r r r
[a, b, c ]
[a, b, c ]
[a, b, c ]
r r r
r r r
By performing the triple scalar product [ f (g h)] [f , g, h] , we can obtain:
r r r
[ f (g h)] =
1
1
2 3
r r r
2 3 [a, b, c ]
2 3
r r r
r r r
r r r
[ f , b, c ] [a, f , c ] [a, b, f ]
r r r
r r r r r r
r r r
r r r
1
r r r [g, b, c ] [a, g, c ] [a, b, g] [a, b, c ] = P [a, b, c ]
r r r
r r r
[a, b, c ] r r r
[h, b, c ] [a, h, c ] [a, b, h]
where
r r r
r r r
r r r
[ f , b, c ] [a, f , c ] [a, b, f ]
1 2 3
r r r
r r r
r r r
1
P = 1 2 3 = r r r [g, b, c ] [a, g, c ] [a, b, g]
r r r
r r r
r r r
1 2 3 [a, b, c ] [h, b, c ] [a, h, c ] [a, b, h]
r
r r
r r
r
For the case when f = A a , g = A b , h = A c , the principal invariants of P are:
(1.46)
r r
r
r r r
r
r r
1
I P = Tr ( P ) = r r r [ A a, b, c] + [a, A b, c] + [a, b, A c ] = I A
[a, b, c]

Ciudad Real - Spain
Draft
1 TENSORS
1
II P = r r r
([a, b, c ]) 2
39
r r
r
r
r r
r r r
r r
r
r r r
r
r r
[a, A b, c ] [a, b, A b] [ A a, b, c ] [a, b, A a] [ A a, b, c ] [a, A a, c]
r
r r r
r r
r r
r r
r +
r r r
r r
r +
r
[a, A c, c ] [a, b, A c ] [ A c, b, c] [a, b, A c ] [ A b, b, c ] [a, A b, c ]
= II A
III P = III A = det (A )
NOTE 3: Let us consider the Cartesian system where

r
a = a1e1 + a 2 e 2 + a 3 e 3
b = b1e1 + b 2 e 2 + b 3 e 3
r
c = c 1e1 + c 2 e 2 + c 3 e 3
r
r
r
Also let us consider that f = e 1 , g = e 2 , h = e 3 ,
r
a a
1

r
b = b1
r
c c
1

a 3 e1

b 3 e 2
c 3 e 3

a2
b2
c2
so, taking into account the above equation
and the equation in (1.45) we can conclude that:

r
f
r 1

g = 1
r
h 1

2 3 a
r
2 3 b
r
2 3 c

1 0 0 e 1 1

0 1 0 e =
2 1
e
0 0 1 3
2 3 a 1 a 2 a 3 e 1

2 3 b1 b 2 b 3 e 2
e
2 3 c1 c 2 c 3 3
thus
1
1
1
2 3 a 1 a 2 a 3 1 0 0
2 3 b1 b 2 b 3 = 0 1 0
2 3 c 1 c 2 c 3 0 0 1
1
1
2 3 a1 a 2 a 3
2 3 = b1 b 2 b 3

2 3 c 1 c 2 c 3

With that we can obtain the inverse of a tensor. Let us consider the tensor A where the
components are:
A11
A ij = A 21
A 31
A12
A 22
A 32
A13 a1 a 2
A 23 = b1 b 2

A 33 c1 c 2

a3
b3
c3
r r r
A = [a, b, c ]
Then, the inverse P = A 1 , (see equation (1.46)) becomes:

A 1
r r r
r r r
r r r
[ f , b, c ] [a, f , c ] [a, b, f ]
r r r
r r r
r r r
1
= r r r [g, b, c ] [a, g, c ] [a, b, g]
r r r
r r r
[a, b, c ] r r r
[h, b, c ] [a, h, c ] [a, b, h]
b1
c1
0
1
b1
=
A
c1
0
b1
c1
a1
a2
a3
b2
b3
c2
c3
c1
c2
c3
a1
a2
a3
b2
c2
b3
c3
0
c1
1
c2
0
c3
0
b2
1
b3
a1
0
a2
0
a3
1
c2
c3
c1
c2
c3

Ciudad Real - Spain
a3
b1 b 2 b 3
b2
1 0
0
c2
a1 a 2 a 3
= 1 b1
b1 b 2 b 3
A c1
0 1
0
b1
a1 a 2 a 3
c1
b1 b 2 b 3
0 0
1
a1
a2
Draft
b3
c3
a2
a3
c2
c3
b3
a1
a3
c3
c1
c3
b2
c2
a1
c1
a2
c2
a3
b2 b3
a1 a 3
b1 b 3
a1 a 2
b1 b 2
a2
40
1
[cof( A )]T = 1 [adj( A )] , we can conclude that:
A
A
Taking into account that A 1 =
[cof( A )]ij
b
2
c2
b
= 1
c
1
b1
c1
b3
a2
c2
c3
a3
c3
b3
a1
a3
c3
c1
c3
b2
c2
a1
a2
c1
c2
b
a3
2
b2 b3
c2
a1 a3
a2
= c
b1 b 3
2
a2
a1 a 2
b1 b 2
b2
a2
b3
c3
b1 b 3
c1
c3
a3
a1
a3
c3
c1
c3
a3
b3
a1 a3
b1 b 3
b1 b 2
c1 c 2
a1 a 2
c1 c 2
a1 a 2
b1 b 2
Note that the coefficient of the above matrix, [cof(A )]ij , can be obtained by solving the
determinant of the resulting matrix by removing the i th row and the j th column, which result
we multiply by (1) i + j , for example:
[cof(A)]12 = (1)
1+ 2
a1 a 2
b1 b 2
c1
a3
b b3
b3 = 1
c1 c 3
c3
c2
Problem 1.58
Given the scalars I C , II C , III C in terms of the scalars I E , II E , III E :
I C = 2I E + 3
II C = 4 I E + 4 II E + 3
III C = 2 I E + 4 II E + 8 III E + 1
(1.47)
Obtain the reverse form of the above equations, i.e. obtain I E , II E , III E in terms of I C , II C ,
III C .
Solution:
The equations in (1.47) can be restructured as follows:
I C 2 0 0 I E 3

II C = 4 4 0 II E + 3
III 2 4 8 III 1
E
C
2 0 0 I E I C 3
4 4 0 II = II 3
E C
2 4 8 III E III C 1
2 0 0 2 0 0 I E 2 0 0 I C 3
4 4 0 4 4 0 II E = 4 4 0 II C 3
2 4 8 2 4 8 III E 2 4 8 III C 1

I E 2 0 0 I C 3

II E = 4 4 0 II C 3
III 2 4 8 III 1
C
where

Ciudad Real - Spain
Draft
1 TENSORS
2 0 0
= 4 4 0
2 4 8
[cof( A )]T
1 0
=
64 4
41
8
0
8
0
0
4 0
2 8
2 0
2 8
2 0
4 0
4 4
1
2 4
2
1
2 0
=
2 4
2
1
2 0
8
4 4
1
8
0
1
4
1
8
with that the scalars I E , II E , III E can be obtained as follows:

I E 2 0 0

II E = 4 4 0
III 2 4 8
1.1.6
1
IC 3 2
II C 3 =
III 1 2
1
C
8
2 ( I C 3)
I C 3
1
0 II C 3 = ( 2 I C + II C + 3)
4
1 III C 1 1
8 ( I C II C + III C 1)
8
0
1
4
1
8
Additive Decomposition of Tensors
Problem 1.59
Find a fourth-order tensor P so that P : A = A dev , where A is a second-order tensor.
Solution: Taking into account the additive decomposition into spherical and deviatoric parts,
we obtain:
A = A sph + A dev =
Tr ( A )
1 + A dev
3
A dev = A
Tr ( A )
1
3
By definition the fourth-order tensors are:
I = 11 = ik jl e i e j e k e l = I ijkl e i e j e k e l
(1.48)
I = 11 = il jk e i e j e k e l = I ijkl e i e j e k e l
(1.49)
I = 1 1 = ij kl e i e j e k e l = I ijkl e i e j e k e l
(1.50)
where it holds that:
)(
I : A = ik jl e i e j e k e l : A pq e p e q
= ik jl A pq kp lq e i e j
= ik jl A kl e i e j
= A ij e i e j
(1.51)
=A
)(
I : A = ij kl e i e j e k e l : A pq e p e q
= ij kl A pq kp lq e i e j
= ij kl A kl e i e j
= A kk ij e i e j
= Tr ( A )1

Ciudad Real - Spain
Draft
)
(1.52)
42
Referring to the definition of fourth-order unit tensors seen in (1.51), and (1.52), where the
relations I : A = Tr ( A )1 and I : A = A hold, we can now state:
A dev = A
Tr ( A )
1
1
1
1 = I : A I : A = I I : A = I 1 1 : A
3
3
3
3
Therefore, we can conclude that:

1
P = I 1 1
3
The tensor P is known as a fourth-order projection tensor, Holzapfel(2000).
1.1.7
Transformation Law of Tensor Components. Invariants.
Problem 1.60
Under the base transformation e i = a ij e j and by considering that the second-order tensor
components in this new base are given by:
Tij = a ik a jl Tkl
Show that:
a) Tii = Tkk = Tr (T ) ; b) Tij T ji = Tkl Tlk ; c) det ( T ) = det ( T )
Solution:
=j
a) Tij = a ik a jl Tkl i Tii = a ik a il Tkl = kl Tkl = Tkk = Tll
b) Tij T ji = (a ik a jl Tkl )(a jp a iq T pq ) = a ik a iq a jl a jp Tkl T pq = kq lp Tkl T pq = Tqp T pq = Tkl Tlk

1 31 3
2 2
= kq
= lp
with that we show that Tr ( T 2 ) = Tr ( T T ) = Tij T ji
c) det ( Tij ) = det(a ik a jl Tkl ) = det (a ik )det (a jl )det ( Tkl ) = det( Tkl )
1 24 1 24
4 3 4 3
=1
=1
we have just shown that Tkk = Tr ( T ) , Tkl Tlk = Tr ( T T ) and det ( T ) are invariants.
Problem 1.61
Let T be a symmetric second-order tensor and I T , II T , III T be scalars, where:
I T = Tr ( T ) = Tii
II T =
1
2
I T Tr ( T 2 )
2
III T = det ( T )
Show that I T , II T , III T are invariant with a change of basis.

Solution:
a) Taking into account the transformation law for the second-order tensor components
Tij = a ik a jl Tkl or in matrix form T = A T A T . Then, Tii is:
Tii = a ik a il Tkl = kl Tkl = Tkk = I T

Ciudad Real - Spain
Draft
1 TENSORS
43
Hence we have proved that I T is independent of the adopted system.

b) To prove that II T is an invariant, one only needs to show that Tr ( T 2 ) is one also, since
2
I T is already an invariant.

Tr ( T 2 ) = Tr ( T T ) = T : T = Tij Tij
= ( a ik a jl Tkl )( a ip a jq T pq )
= a ik a ip a jl a jq Tkl T pq
1 31 3
2 2
kp
lq
= T pl T pl
= T : T = Tr ( T T ) = Tr ( T 2 )
c) Matrix form: det ( T ) = det (T ) = det (A T A T ) = det (4)det (T )det24 ) = det (T )

A
(A T
123
4
1
4 3
=1
=1
Problem 1.62
Show that the following relations are invariants:
2
2
C12 + C 2 + C 3
3
3
C13 + C 2 + C 3
4
4
C14 + C 2 + C 3
where C1 , C 2 , C 3 are the eigenvalues of the second-order tensor C .

Solution: Any combination of invariants is also an invariant, so, on this basis, we can try to
express the above expressions in terms of their principal invariants.
2
2
2
2
2
2
I C = (C1 + C 2 + C3 ) = C12 + C 2 + C 3 + 2 C1 C 2 + C1 C3 + C 2 C3 C12 + C 2 + C 3 = I C 2 II C
1444 24444
4
3
2
II C
2
So, we have proved that C12 + C 2 + C 32 is an invariant. Similarly, we can obtain the other
relationships, so, we summarize:
C1 + C 2 + C3 = I C
2
2
2
C12 + C 2 + C3 = I C 2 II C
3
3
3
C13 + C 2 + C3 = I C 3 II C I C + 3 III C
4
4
4
2
2
C14 + C 2 + C3 = I C 4 II C I C + 4 III C I C + 2 II C
5
5
5
2
3
2
C15 + C 2 + C3 = I C 5 II C I C + 5 III C I C + 5 II C I C 5 III C II C
Note also that
n
n
n
n
n
n
n
n
C1n +1 + C 2 +1 + C3 +1 = C1n + C 2 + C3 I C C1 C 2 1 + C3 1 C 2 C1n1 + C3 1 C3 C1n 1 + C 2 1
Problem 1.63
Obtain the components of T , given by the transformation:
T = A T AT
where the components of T and A are shown, respectively, as Tij and a ij . Afterwards, given
that a ij are the components of the transformation matrix, represent graphically the
components of the tensors T and T on both systems.
Solution: The expression T = A T A T in symbolic notation is given by:

Ciudad Real - Spain
Draft
44

Tab (e a e b ) = a rs (e r e s ) T pq (e p e q ) a kl (e l e k )
= a rs T pq a kl sp ql (e r e k )
= a rp T pq a kq (e r e k )
To obtain the components of T one only need make the double scalar product with the basis
(e i e j ) , the result of which is:

Tab (e a e b ) : (e i e j ) = a rp T pq a kq (e r e k ) : (e i e j )
Tab ai bj = a rp T pq a kq ri kj
Tij = a ip T pq a jq
The above equation is shown in matrix notation as:

T = A T A T inverse T = A 1 T A T
Since A is an orthogonal matrix, it holds that A T = A 1 . Thus, T = A T T A . The

graphical representation of the tensor components in both systems can be seen in Figure 1.1.
T = A T AT
x3
T33
T23
x3
T13
T33
T31
T13
T23
T32
T22
T12
x2
T21
T22
T31
T11
T11
T32
T21
T12
x1
x2
x1
T = AT T A
Figure 1.1: Transformation law of the second-order tensor components.

Problem 1.64
Let T be a second-order tensor whose components in the Cartesian system (x1 , x 2 , x3 ) are
given by:
(T )ij
3 1 0
= Tij = T = 1 3 0
0
0 1

Given that the transformation matrix between two systems, (x1 , x 2 , x3 ) - (x1 , x 2 , x 3 ) , is:

Ciudad Real - Spain
Draft
1 TENSORS
0
2
A=
2
45
0
2
2
2
2

Obtain the tensor components Tij in the new coordinate system (x1 , x 2 , x 3 ) .
Solution: The transformation law for second-order tensor components is Tij = aik a jl Tkl .
To enable the previous calculation to be carried out in matrix form we use:
[ ]
Tij = [a i k ] [Tk l ] a l j
Thus
T = A T AT
0
2
T =
2
1
0
3 1 0
0 1 3 0 0
0 1
0
1
0
0
2
2
2
2
2
2
2
2
0
2
2
2
On carrying out the operation of the previous matrices we now have:

1 0 0
T = 0 2 0
0 0 4
Problem 1.65
Find the transformation matrix between the systems: x, y , z and x, y, z . These systems are
represented in Figure 1.2.
z = z
z = z
y
y = y
x
x
Figure 1.2: Rotation.

Ciudad Real - Spain
Draft
46
Solution: Note that: if we have an initial space and successive transformations until the final
space, the transformation law from the initial space to the final space is formed by the product
of the transformations in the opposite direction. That is, we place in the final space and we
follow opposite direction of the arrows until the initial space, (see figure below).
B
r
a
r
a
B 1
r
a
A 1
initial
space
C 1
CBA
C
r
a
final space
A 1B 1 C 1 = (CBA ) 1 if orthogonal s A T B T C T = (CBA ) T
The coordinate system x, y, z can be obtained by different combinations of rotations as

follows:
Rotation along the z -axis

z = z
from x, y , z to x, y, z
cos sin 0
A = sin cos 0
0
0
1

Ciudad Real - Spain
with 0 360
Draft
1 TENSORS
Rotation along the y -axis
from x, y, z to x, y, z
z = z
cos
B= 0
sin
y = y
47
0 sin
1
0
0 cos
with 0 180
y
z = z
Rotation along the z -axis

z = z
z = z
from x, y, z to x, y, z
y
y = y
cos
C = sin
sin
cos
0
0
0
with 0 360
The transformation matrix from ( x, y , x ) to ( x, y, z ) is given by

z
After multiplying the matrices, we obtain:
D = CBA
(sin cos cos + cos sin ) sin cos

(cos cos cos sin sin )
( cos cos sin sin cos ) ( sin cos sin + cos cos ) sin sin
D=
cos sin
sin sin
cos
The angles , , are known as Euler angles and were introduced by Leonhard Euler to
describe the orientation of a rigid body motion.

Ciudad Real - Spain
Draft
48
Problem 1.66
If a ij represent the components of the base transformation matrix, show that the following
equations are fulfilled:
2
2
2
a11 + a12 + a13 = 1
2
2
2
a 21 + a 22 + a 23 = 1
2
2
2
a 31 + a 32 + a 33 = 1
a11 a 21 + a12 a 22 + a13 a 23 = 0
a 21 a31 + a 22 a 32 + a 23 a 33 = 0
a a + a a + a a = 0
12 32
13 33
11 31
or
2
2
2
a11 + a 21 + a 31 = 1
2
2
2
a12 + a 22 + a 32 = 1
2
2
2
a13 + a 23 + a 33 = 1
a11 a12 + a 21 a 22 + a 31 a 32 = 0
a12 a13 + a 22 a 23 + a 32 a 33 = 0
a a + a a + a a = 0
21 23
31 33
11 13
Solution:
We start from the principle that the basis transformation matrix is an orthogonal matrix, i.e.
a ik a jk = a ki a kj = ij . Then:
a ik a jk = a i1 a j1 + a i 2 a j 2 + a i 3 a j 3 = ij
(i = 1, j = 1)
(i = 2, j = 2)
(i = 3, j = 3)
(i = 1, j = 2)
(i = 2, j = 3)
(i = 1, j = 3)
2
2
2
a11 + a12 + a13 = 1
2
2
2
a 21 + a 22 + a 23 = 1
2
2
2
a31 + a 32 + a 33 = 1
a11 a 21 + a12 a 22 + a13 a 23 = 0

a 21 a 31 + a 22 a 32 + a 23 a 33 = 0
a11 a 31 + a12 a 32 + a13 a 33 = 0
T
AA = 1
a11
a
21
a 31
a12
a 22
a 32
a13 a11
a 23 a12
a 33 a13
a 21
a 22
a 23
a 31 1 0 0
a 32 = 0 1 0

a 33 0 0 1

Performing the matrix multiplication we obtain:

2
2
2
a11 + a12 + a13
a11 a 21 + a12 a 22 + a13 a 23

a a + a a + a a
12 32
13 33
11 31
1.1.8
a11 a 21 + a12 a 22 + a13 a 23

2
2
2
a 21 + a 22 + a 23
a 21 a 31 + a 22 a 32 + a 23 a 33
a11 a 31 + a12 a 32 + a13 a 33 1 0 0
a 21 a 31 + a 22 a 32 + a 23 a 33 = 0 1 0
2
2
2
0 0 1
a 31 + a 32 + a 33
Eigenvalues, Eigenvectors, Orthogonal Transformation
Problem 1.67
Let Q be a proper orthogonal tensor, and E be an arbitrary second-order tensor. Show that
the eigenvalues of E do not change with the following orthogonal transformation:
E* = Q E QT
Solution:
We can prove this as follows:

Ciudad Real - Spain
Draft
1 TENSORS
(
)
= det (Q E Q 1)
= det (Q E Q Q 1 Q )
= det [Q (E 1 ) Q ]
= det (Q ) det (E 1 ) det (Q )
123
1 24
4 3
0 = det E * 1
T
T
= det (E 1 )
49
(
= det (Q
= det (Q
0 = det E * ij
ij
ik E kp Q jp
ij
ik E kp Q jp
Q ik Q jp kp
[ (
) ]
= det (Q )det (E ) det (Q )
= det (E )
= det Q ik E kp kp Q jp
ik
kp
kp
kp
jp
kp
Thus, we have proved that E and E * have the same eigenvalues.

Problem 1.68
Let A be a second-order tensor and Q be an orthogonal tensor. If the orthogonal
transformation law to A is given by A * = Q A Q T , show that A 2 = Q A 2 Q T .
*
Solution:
A 2 = A* A*
( A 2 ) ij = ( A * A * ) ij = A * A *
ik kj
= ( Q A Q T ) (Q A Q T )
= QA Q QA Q
123
T
= (Q ip A pr Q kr )(Q ks A st Q jt )
= Q ip A pr Q kr Q ks A st Q jt
123
=1
= rs
= Q A A QT
= Q ip A pr rs A st Q jt = Q ip A ps A st Q jt
= Q A 2 QT
= Q ip ( A A ) pt Q jt
= (Q A 2 Q T ) ij
Problem 1.69
Given the tensor components:
5 3 3
Tij = 2 6 3
2 2 4
a) Obtain the principal invariants of T , i.e. obtain I T , II T and III T ;

b) Obtain the characteristic polynomial associated with T ;
c) If 1 , 2 and 3 are the eigenvalues of T and 1 = 10 . Obtain 2 and 3 > 2 .
Solution:
a) The principal invariants of T are:
I T = Tr ( T ) = 5 + 6 + 4 = 15
II T =
6 3
2 4
5 3
2 4
5 3
2 6
= 56
III T = det ( T ) = 60
b) The characteristic polynomial can be obtained by solving the determinant:

Ciudad Real - Spain
Draft
50
5
3
2
6
2
3
3
=0
3 2 I T + II T III T = 0
thus:
3 152 + 56 60 = 0
c) In the principal space the following is true:

1 = 10 0
2
Tij = 0
0
0
3 > 2
0
0
where the principal invariants are

I T = Tr ( T ) = 1 + 2 + 3 = 15
III T = det ( T ) = 1 2 3 = 60
2 + 3 = 5
23 = 6
By combining these two equations we obtain:

(1)
3 = 3
2
(5 3 ) 3 = 6 3 5 3 + 6 = 0 ( 2 )
2 + 3 = 5
3 = 2
23 = 6
We discard the solution (32) = 2 , thus 3 = 3 . Then:

10 0 0
Tij = 0 2 0
0 0 3
I T = Tr ( T ) = 10 + 2 + 3 = 15
In this space we can check that II T = 2 3 + 10 3 + 10 2 = 56

III = det ( T ) = 10 2 3 = 60
T
Problem 1.70
Find the principal values and directions of the second-order tensor T , where the Cartesian
components of T are:
(T )ij
3 1 0
= Tij = T = 1 3 0
0
0 1
Solution: We need to find nontrivial solutions for (Tij ij ) n j = 0 i , which are constrained by

n j n j = 1 (unit vector). As we have seen, the nontrivial solution requires that:
Tij ij = 0
Explicitly, the above equation is:

Ciudad Real - Spain
Draft
1 TENSORS
T11
T21
T12
T22
T13
T23
T31
T32
T33
51
3 1
= 1 3
0
0
0
=0
Developing the above determinant, we can obtain the cubic equation:
(1 ) (3 ) 2 1 = 0
3 7 2 + 14 8 = 0
We could have obtained the characteristic equation directly in terms of invariants:

I T = Tr ( Tij ) = Tii = T11 + T22 + T33 = 7
II T =
T
1
Tii T jj Tij Tij = 22
T32
2
T23
T33
T11
T13
T31
T33
T11
T12
T21
T22
= 14
III T = Tij = ijk Ti1 T j 2 Tk 3 = 8
Then, the characteristic equation becomes:

3 2 I T + II T III T = 0
3 72 + 14 8 = 0
On solving the cubic equation we obtain three real roots, namely:

1 = 1;
2 = 2;
3 = 4
We can also verify that:

I T = 1 + 2 + 3 = 1 + 2 + 4 = 7
II T = 1 2 + 2 3 + 3 1 = 1 2 + 2 4 + 4 1 = 14
III T = 1 2 3 = 8
Thus, we can see that the invariants are the same as those evaluated previously.
Principal directions:
Each eigenvalue, i , is associated with a corresponding eigenvector, n (i ) . We can use the
equation ( Tij ij ) n j = 0 i to obtain the principal directions.

1 = 1
3 1
1
1
3 1
0 n1 0
0 n 1 3 1 1
n = 1 3 1 0 n = 0
0 2
2
n 3 0
0 1 1 n 3 0
1 1

These become the following system of equations:

2n1 n 2 = 0
n1 = n 2 = 0
n1 + 2n 2 = 0
0n = 0
3
2
2
n i n i = n1 + n 2 + n 3 = 1
2
Then we can conclude that: 1 = 1
n i(1) = [0 0 1] .
NOTE: This solution could have been directly determined by the specific features of the T
matrix. As the terms T13 = T23 = T31 = T32 = 0 imply that T33 = 1 is already a principal value,
then, consequently, the original direction is a principal direction.

Ciudad Real - Spain
Draft
52
2 = 2
3 2
1
1
3 2
0
0 n1 0
n 1 3 2 1
n = 1 3 2
0 n 2 = 0
2

0
1 2 n 3 0
1 2 n 3 0

0
0
n1 n 2 = 0 n1 = n 2
n1 + n 2 = 0
n = 0
3
The first two equations are linearly dependent, after which we need an additional equation:
2
2
2
n i n i = n1 + n 2 + n 3 = 1 2n1 = 1 n1 =
2
1
2
Thus:
2 = 2
n i( 2 ) =
2
1
2
3 = 4
3 3
1
1
3 3
0
0 n 1 3 4 1
0 n1 0
n = 1 3 4
0 2
0 n 2 = 0

1 3 n 3 0
0
1 4 n 3 0

n1 n 2 = 0
n1 = n 2
n1 n 2 = 0
3n = 0
3
2
2
n i n i = n1 + n 2 + n 3 = 1 2n 2 = 1 n 2 =
2
2
1
2
Then:
3 = 4
n i(3) = m
1
2
1
2
Afterwards, we summarize the eigenvalues and eigenvectors of T :

1 = 1
n i(1)
= [0 0 1]
2 = 2
n i( 2)
1
2
1
2
3 = 4
n i(3)
= m
1
2
1
2
NOTE: The tensor components of this problem are the same as those used in Problem 1.64.
Additionally, we can verify that the eigenvectors make up the transformation matrix, A ,

between the original system, (x1 , x 2 , x3 ) , and the principal space, (x1 , x 2 , x 3 ) , (see Problem
1.64).

Ciudad Real - Spain
Draft
1 TENSORS
53
Problem 1.71
Let Q be a proper orthogonal tensor a) show that Q has one real eigenvalue and equals to 1 .
b) Also show that Q can be represented by means of the angle as follows:

Q = p p + cos (q q + r r ) sin (q r r q)

where p , q , r , are unit vectors which form an orthonormal basis, where p is the direction
associated with the eigenvalue = 1 , i.e. p is an eigenvector of Q . c) Obtain the principal

r
invariants of Q in function of the angle . d) Given a vector position x , obtain the new
r

vector originated by the orthogonal transformation Q x in the space formed by p , q .
Solution:
a) Taking into account the definition of the orthogonal tensor we can conclude that:
QT Q = 1
QT Q QT = 1 QT
Q T (Q 1) = (Q T 1)
Q T (Q 1) = (Q 1) T
Then we obtain the determinant of the two previous tensors:
det Q T (Q 1) = det (Q 1) T
[ ]
det Q det[(Q 1)] = det (Q 1)

123
4 4
T
= (1) 3 det (Q 1) T
T
] = det[(Q 1)]
= detQ =1
det[(Q 1)] = det[(Q 1)]
where
[ ]
det A T
we
have used
the
following
determinant
properties:
det[A ] = 3 det[A ] ,
= det[A ] , det[A B ] = det[A ]det[B ] . The unique scalar which satisfies the expression
above is zero, then:

det[(Q 1)] = 0
Taking into account the definition of eigenvalue, det[(Q 1)] = 0 , we conclude that when
= 1 it fulfills that det[(Q 1)] = 0 , then = 1 is eigenvalue of Q . Hence, there is a direction
*
* *
(eigenvector) satisfying that Q e1 = e1 = e1 .
* 2 3
b) We consider that the vectors p e1 , q e * , r e * form an orthonormal basis.
e3
2
q e*
*
e1 p
* *
Q e1 = e1
e2
e1
3
r e*
* 2 3
The symbolic representation of the tensor in the basis e1 , e * , e * is given by:

Ciudad Real - Spain
Draft
54
Q = Q * e * e *j
ij i
* *
* *
* *
*
2
3
= Q11e1 e1 + Q12 e1 e * + Q13 e1 e * +
(1.53)
3
+ Q * e * e1 + Q * e * e * + Q * e * e * +
21 2
22 2
23 2
3
+ Q * e * e1 + Q * e * e * + Q * e * e *
31 3
32 3
33 3
*
Taking into account that e1 is eigenvector of Q associated with the eigenvalue = 1 , it holds
*
* *
that Q e1 = e1 = e1 . In addition, making the projection of Q , given by (1.53), according to
*
direction e1 , we obtain:
* *
Q e 1 = e1
* *
* *
* *
*
*
2
3
Q e1 = [ Q11e1 e1 + Q12 e1 e * + Q13 e1 e * +
3
+ Q * e * e1 + Q * e * e * + Q * e * e * +
21 2
22 2
23 2
3
+ Q * e * e1 + Q * e * e * + Q * e * e *
31 3
32 3
33 3
*
] e1
* *
= Q11e1 + Q * e * + Q * e *
21 2
31 3
*
with that we conclude that Q11 = 1 , Q * = 0 , Q * = 0 .
21
31
Remember that two coaxial tensors have the same principal directions (eigenvectors). A tensor
and its inverse are coaxial tensors, then if Q 1 = Q T , this implies that Q T and Q are coaxial
*
tensors, and e1 is also principal direction of Q T , then it fulfills that:
* *
Q T e1 = e1
* *
*
*
* 2
* 3
Q T e1 = [ Q11e1 e1 + Q * e1 e * + Q * e1 e * +
21
31
*
*
3
+ Q12 e * e1 + Q * e * e * + Q * e * e * +
2
22 2
32 2
*
*
3
+ Q13 e * e1 + Q * e * e * + Q * e * e *
3
23 3
33 3
*
] e1
* *
*
*
= Q11e1 + Q12 e * + Q13 e *
2
3
*
*
*
with that we conclude that Q11 = 1 , Q12 = 0 , Q13 = 0 . Then, the equation (1.53) becomes:
* *
3
Q = e1 e1 + Q * e * e * + Q * e * e * + Q * e * e * + Q * e * e *
22 2
23 2
32 3
33 3
(1.54)
i
In matrix form, the components of Q , in the basis e * , are given by:
*
x2
Q*
ij
1 0
= 0 Q *
22
0 Q *
32
Q*
22
*
Q11 = 1
0
Q*
23
Q*
33
Q*
32
Q*
23
*
x1
Q*
33
*
x3

Ciudad Real - Spain
Draft
1 TENSORS
55
Once more we use the orthogonality condition Q T Q = Q Q T = 1 , or in terms of the

i
components in the space e * :
Q* Q*
ki kj
1
0
= ij
[(Q )
[Q Q
* 2
22
*
*
22 23
1 0
0 Q *
22
0 Q *
23
0 1 0
Q * 0 Q *
32
22
*
0 Q *
Q 33
32
0
] [Q Q
] [(Q )
+ (Q * ) 2
32
*
+ Q 32 Q *
33
*
*
22 23
* 2
33
0 1 0 0
Q * = 0 1 0
23
*
Q 33 0 0 1

1 0 0
= 0 1 0

0 0 1

(1.55)
]
]
+ Q* Q*
32 33
+ (Q * ) 2
23
The determinant of a proper orthogonal tensor is det (Q) = +1 , thus

1 0
0 Q *
22
0 Q *
32
0
Q* = 1
23
Q*
33
Q* Q* Q* Q* = 1
22 33
23 32
(1.56)
Taking into account (1.55) and (1.56) we obtain the following set of equations:
(Q * ) 2 + (Q * ) 2 = 1
22
32
* *
*
*
Q 22 Q 23 + Q 32 Q 33 = 0
* 2
* 2
(Q 33 ) + (Q 23 ) = 1
* *
*
*
Q 22 Q 33 Q 23 Q 32 = 1
cos 2 + sin 2 = 1
cos ( sin ) + sin cos = 0

2
2
cos + sin = 1
cos cos ( sin )(sin ) = 1
whereupon we have demonstrated the existence of an angle that meets the above conditions:
Q*
ij
1 0
= 0 Q *
22
*
0 Q 23
0 1
0
0
0 cos sin
*
Q 32 =
Q * 0 sin cos
33
(1.57)
Returning to the equation (1.54), and taking into account (1.57), we conclude that:
* *
2 2
2 3
3 2
3 3
Q = e1 e1 + (cos ) e * e * + ( sin )e * e * + (sin )e * e * + (cos ) e * e *
* *
2 2 3 3
2 3 3 2
= e1 e1 + cos e * e * + e * e * sin e * e * e * e *
* 2 3
Considering that p e1 , q e * , r e * , we show that:

Q = p p + cos (q q + r r ) sin (q r r q)
It is interesting to note that the additive decomposition of Q in an antisymmetric and a

i
symmetric part, in the space e * , is:
0
0
1
0 cos
0
=
0
0
cos
14444244443
sym
Q*
ij

[pp+cos (qq+rr )]ij
0
0
0
0
0
=
sin
0 sin
0
4
14444 244444
3
skew
Q*
ij
[sin (qr r q) ]ij
skew
Note that the format of Q *

has the same format as the antisymmetric tensor ( W ) in the
ij
space defined by the axial vector:

Ciudad Real - Spain
Draft
56
*
Wij
0 0
= 0 0
where is the magnitude of the axial vector.

c) By means of (1.57) it is easy to show that I Q = II Q = 1 + 2 cos , III Q = 1 .
r

d) We represent the vector x by means its components and the basis p , q , r , as follows:
r
x = pp + qq + rr .
r

Then, it fulfills that: x p = ( pp + qq + rr ) p = p
xq= q
x r = r
Thus, (see Figure 1.3), it holds that:
r
r
~

x = Q x = p p + cos (q q + r r ) sin (q r r q)
= pp + (q cos r sin )q + (r cos + q sin )r
] [ pp + qq + r r ]
*
e1 p
2
q e*
r
x
r
r
~ =Q x
x
3
r e*
Figure 1.3

Ciudad Real - Spain
Draft
1 TENSORS
Problem 1.72
57
Let us consider the tensorial transformations p = U p and p = R p , where R is an

orthogonal tensor and U is a second-order tensor with U U 1 = 1 , i.e. U 1 . Obtain the
r
r
transformation law between p and p .
Solution:
The proposed problem can be represented by the following figure:
U
r
p
r
p
r
p
?
Taking into account that R 1 = R T (orthogonal tensor), we can guarantee that the inverse of
r
r
R exists, and considering that p = R p we obtain:
r
r
p = R p
r
r
R 1 p = R 1 R p
r
r r
R 1 p = 1 p = p
r
r
r
r
Substituting p = R 1 p into p = U p , we obtain:
r
r
r
r
p = U p
p = U p
r
r
r
r
R 1 p = U p
R 1 p = U p
r
r
r
r
R R 1 p = R U p
U 1 R 1 p = U 1 U p
r
r
r
r r
1 p = R U p
(R U) 1 p = 1 p = p
r
r
r
r
p = (R U) p
p = (R U) 1 p
(1.58)
Or in indicial notation:
p = U ij p j
i
p = U ij p j
i
R ij1p = U ij p j
j
R ij1p = U ij p j
j
R ki R ij1p = R ki U ij p j
j
U ki1R ij1p = U ki1U ij p j

j
kj p = R ki U ij p j
j
(R ki U ij ) p j = kj p j = p k
p k = (R ki U ij )p j
p k = (R ki U ij ) 1 p
j

Ciudad Real - Spain
(1.59)
Draft
58
r
p
U 1
R 1 = R T
r
p
r
p
(R U)
(R U) 1 = U 1 R T
1.1.9
Spectral Representation of Tensors
Problem 1.73
Let w be an antisymmetric second-order tensor and V be a positive definite symmetric
tensor whose spectral representation is given by:
V=
n(a) n(a)
a =1
Show that the antisymmetric tensor
w can be represented by:

3
w = w ab n (a ) n (b)
a ,b =1
a b
Demonstrate also that:

3
w V V w = w ab ( b a ) n ( a ) n (b)
a ,b =1
a b
Solution:
It is true that
a =1
a =1
w 1 = w n ( a ) n ( a) = w n (a ) n (a ) = (w n ( a) ) n ( a)
a =1
w (n
3
(b )
n(a) n(a)
a ,b =1
where we have applied the antisymmetric tensor property w n = w n , where w is the axial
vector associated with w . Expanding the above equation, we obtain:
w = wb (n (b) n (1) ) n (1) + wb (n (b) n ( 2) ) n ( 2) + wb (n (b) n (3) ) n (3) =
+ w (n
+ w (n
= w1 n (1) n (1) n (1) + w2 n ( 2) n (1) n (1) + w3 n (3) n (1) n (1) +

1
(n
) n
(n
) n
n ( 2 ) n ( 2 ) + w2 n ( 2 ) n ( 2) n ( 2) + w3 n (3) n ( 2) n ( 2 ) +
(1)
( 3)

Ciudad Real - Spain
) n
(1)
( 3)
+ w2
( 2)
Draft
( 3)
( 3)
+ w3
( 3)
( 3)
( 3)
1 TENSORS
59
On simplifying the above expression we obtain:
w = w2 (n (3) ) n (1) + w3 (n ( 2) ) n (1) +
( )
(n ) n
( )
(n ) n
+ w1 n (3) n ( 2 ) w3 n (1) n ( 2 ) +
( 2)
w1
( 3)
+ w2
(1)
( 3)
Taking into account that w1 = w 23 = w 32 , w2 = w13 = w 31 , w3 = w12 = w 21 , the above

equation becomes:
w = w 31 n (3) n (1) + w 21 n ( 2) n (1) +
+ w 32 n (3) n ( 2 ) + w12 n (1) n ( 2 ) +
+ w 23 n ( 2 ) n (3) + w13 n (1) n (3)

3
w = w ab n (a ) n (b)
a ,b =1
a b
The terms
wV
and V w can be expressed as follows:
3
3
(a)
(b )
w V = w ab n n b n (b) n (b)
b =1
a ,b =1
ab
b w ab n ( a ) n (b ) n (b ) n (b ) =
a ,b =1
a b
ab
n ( a ) n (b )
a ,b =1
ab
and
3
3
3
(a )
(a)
(a )
(b )
V w = a n n
w ab n n = a w ab n ( a) n (b)
a ,b =1
a =1
a ,b =1
ab
ab
Then,
3
3

(a)
(b )
(a)
(b )
w V V w = b w ab n n a w ab n n
a ,b =1
a ,b =1
a b
ab
ab ( b
a ) n ( a ) n (b )
a ,b =1
a b
Similarly, it is possible to show that:

3
w V 2 V 2 w = w ab (2b 2a ) n ( a ) n (b)
a ,b =1
a b

Ciudad Real - Spain
Draft
60
Problem 1.74
Let C be a positive definite tensor, whose Cartesian components are given by:
2 0 1
C ij = 0 4 0
1 0 2
Obtain the following tensors: a) C 2 ; b) U = C . c) Check if the tensors C and U are coaxial.
Solution:
Note that the tensors C 2 and U = C are coaxial with the tensor C . By means of the
spectral representation of C :
C=
N
a
(a)
N(a )
a =1
where a are the eigenvalues of C , and N( a ) are the eigenvectors of C , we can obtain:
C2 =
2 (a)
aN
N( a)
U= C =
a =1
a N( a) N(a )
a =1
Calculation of the eigenvalues and eigenvectors of the tensor C .

Due to the structure of the C tensor components we already know one eigenvalue 2 = 4
which is associated with the principal direction Ni( 2) = [0 1 0] . To calculate the remaining
eigenvalues is sufficient to solve the following characteristic determinant:
2
1
=0
1
2
( 2 ) 2 = 12
1 = 2 1 = 1
( 2 ) = 1
3 = 2 + 1 = 3
Associated with the eigenvalue 1 = 1 we have the following eigenvector:

2 1
1
(
1 N11) 0
(1) =
2 1 N3 0
(
1 1 N11) 0
(1) =
1 1
N 3 0
(
(
N11) = N 31)

with the restriction Ni(1) Ni(1) = 1 , thus
( (

(
N11) N11) + N (1) N (1) + N 31)N (1) = 1
2
2
3
(1) (1)
(1) (1)
N N + N N =1
1
1
(
N11) =
2
1
(
(
N 31) = N11) = m
2
Associated with the eigenvalue 3 = 3 we have the following eigenvector:

2 3
1
(
N13) 0
( 3) =
2 3 N 3 0
(
1 1 N13) 0
(3) =
1 1
N 3 0
(
(
N13) = N 33)

with the restriction Ni(3) Ni(3) = 1 , thus

Ciudad Real - Spain
Draft
1 TENSORS
61
( (

( (
N13)N13) + N (23) N (3) + N 33) N 33) = 1
2

N ( 3) N ( 3) + N ( 3) N ( 3) = 1
1
1
(
N13) =
2
1
(
(
N 33) = N13) =
2
Summarizing we have:
1
2
2
1 0]
1
1
0
2
2
N i(1) =
N i( 2 ) = [0
N i(3) =
1 = 1
2 = 4
3 = 3
0 m
Transformation
Matrix
A=
2
0
1
1
0
2
0
1
2
Then it holds that:

C = A C AT C = AT C A
In the principal space we have:
1 0 0
C ij = 0 4 0
0 0 3
1
2
= 0
C ij
U = C
ij
ij
0
16 0
0 9
1 0
= 0
4
0
0
0 1 0
0 = 0 2
3 0 0

0
0
Note that the tensor C is a positive definite tensor, so, its eigenvalues are positive. In the
original space we have the following components:
2
C ij =
1
2
0
1
2
1
0
2
1
0
1
0
2
1
1 0 0 2
0 16 0 0
0 0 9 1
0
1
0
2 5 0 4
0 = 0 16 0
1
4 0 5
(1.60)
Note that this result could have been obtained easily by means of the operation C 2 = C C ,
which in components becomes:
2
C ij
2 0 1 2 0 1 5 0 4
= C ik C kj = 0 4 0 0 4 0 = 0 16 0
1 0 2 1 0 2 4 0 5
Similarly to (1.60), we obtain the components of U in the original Cartesian system:
U ij =
1
2
0
1
2
0
1
0
2
0
1
2

Ciudad Real - Spain
1 0
0 2
0 0
0
0
Draft
1
2
0
1
2
0
1
0
1 3 +1
0

2 2
2
0 = 0
1 3 1
0
2 2

3 1
2
0
3 + 1
2
62
c) The tensors C and U are coaxial, because the eigenvalues of U were obtained in the
principal space of C . We can also verify that C and U are coaxial by means of C U = U C ,
i.e.:
3 +1
0
2 0 1
2
C ik U kj = 0 4 0 0
2
3 1 0
1 0 2
3 1
3.098 0 2.098
2
8
0
0 = 0
3 + 1 2.098 0 3.098
3 1
2 0 1 3.098 0 2.098
2
0 0 4 0 = 0
8
0

3 + 1 1 0 2 2.098 0 3.098
3 +1
0
2
U ik C kj = 0
2
3 1 0
2
Problem 1.75
Let C be a symmetric second-order tensor and R a proper orthogonal tensor. The
components of these tensors, in the Cartesian system, are given by:
2 0 1
C ij = 0 4 0
1 0 2
R ij =
0
2
2
2
2
0
2
2
2
2
a) Obtain the following tensors: a.1) C 8 ; a2) U = C .

b) Obtain the principal invariants of C .
c) Taking into account that the tensors b and C are related to each other by the following
proper orthogonal transformation C = R T b R , obtain the third principal invariant of b .
Solution:
0
3280
3281
0
65536
0
a) Similarly to Problem 1.74. Answer: C =
3280
0
3281
b)
I C = Tr (C ij ) = C ii = C11 + C 22 + C 33 = 8
II C =
4 0 2 1 2 0
1
C ii C jj C ij C ij =
+
+
= 19 ; III C = C = ijk C i1C j 2 C k 3 = 12
0 2 1 2 0 4
2
c) Taking into account the determinant property, the third principal invariant of b can be
expressed as follows:
C det (C ) = det (R T b R ) = det (R T )det (b)det (R ) = det (b) = III b = 12
1 24
4 3
123
= +1

Ciudad Real - Spain
Draft
= +1
1 TENSORS
63
Problem 1.76
Let S be a symmetric second-order tensor with det (S ) 0 . Considering that S has two
equal eigenvalues, i.e. S 2 = S 3 and S1 S 2 , show that S can be represented by:
S = S 1n (1) n (1) + S 2 (1 n (1) n (1) )
where n (1) is the eigenvector of S associated with the eigenvalue S1 , 1 is the second-order
unit tensor.
Solution: We start from the spectral representation of S :

S=
S n
a
n ( a ) = S 1n (1) n (1) + S 2 n ( 2) n ( 2) + S 3n (3) n (3)
(a)
a =1
= S 1n
(1)
(1)
+ S 2 (n
( 2)
( 2)
+n
( 3)
(1.61)
n )
( 3)
Remember that 1 is a spherical tensor, whereby any direction is a principal direction. Based
on this principle, we adopt the principal space of S to make the spectral representation of 1 ,
i.e.:
1=
(a)
n ( a ) = n (1) n (1) + n ( 2 ) n ( 2 ) + n (3) n (3)
a =1
(1.62)
n ( 2 ) n ( 2 ) + n (3) n (3) = 1 n (1) n (1)
By substituting the above equation into (1.61) we obtain:
S = S 1n (1) n (1) + S 2 (n ( 2) n ( 2) + n (3) n (3) )
= S 1n (1) n (1) + S 2 (1 n (1) n (1) )
1.1.10 Cayley-Hamilton Theorem

Problem 1.77
Let T be an arbitrary second-order tensor, show the Cayley-Hamilton theorem, which states
that any tensor satisfies its own characteristic equation.
Solution:
We start from the characteristic equation of the tensor: 3 2 I T + II T III T = 0 , which
fulfils for each eigenvalue 1 , 2 , 3 , then:
3 2 I T + 1 II T III T = 0
1
1
32 22 I T + 2 II T III T = 0
33 2 I T + 3 II T III T = 0
3
Restructuring the above equations in matrix form we obtain:

3
1
0
32
0
0 2
1

00
33 0

0
22
0
1
0
0 I T + 0
0
2
3
0
2
0
0
1 0 0
0 0 0
0 1 0 III = 0 0 0
0 II T
T
0 0 1
0 0 0
3
(1.63)
Tij3 Tij 2 I T + Tij II T III T ij = 0 ij

Ciudad Real - Spain
Draft
64
Note that in the principal space of T the following relationships are true:
1
Tij = 0
0
0
2
0
0
3

Tij = Tik Tkj = 0
0
0 1
0 0
3 0
0
2
0
pj
Tij = Tik Tkp T = 0
0
0
2
0
0
2
0
0 1
0 0
3 0
0 2
1

0 =0
3 0

0
2
0
0
2
3
0
22
0
0 1
0 0
3 0
3
1

=0
3 0

0
2
0
0
0
32
0
0
33
The component transformation law between spaces for a second-order tensor is
Tij = Fik Tkp F pj1 , where Fij is the transformation matrix from the original space ( Tij ) to the
2
principal space ( Tij ). Note also that the relationships Tij 2 = Fik Tkp F pj1 and Tij 3 = Fik Tkp F pj1
hold, (see Problem 1.68). With that we can conclude that the equation in (1.63) can be
rewritten as follows:
Tij 3 Tij 2 I T + Tij II T III T ij = 0 ij

3
2
Fik Tkp F pj1 Fik Tkp F pj1 I T + Fik Tkp F pj1 II T III T Fik kp F pj1 = 0 ij
3
2
Fik Tkp Tkp I T + Tkp II T III T kp F pj1 = 0 ij
3
2
Fsi 1 Fik Tkp Tkp I T + Tkp II T III T kp F pj1 F jt = Fsi 1 0 ij F jt = 0 st
3
2
sk Tkp Tkp I T + Tkp II T III T kp pt = Fsi 1 0 ij F jt = 0 st

3
2
Tst Tst I T + Tst II T III T st = 0 st
T 3 T 2 I T + T II T III T 1 = 0
In Problem 1.57 (NOTE 1) we have summarized that:
r r
r r r
r
r r
r
r r r
[( A a), b, c] + [a, ( A b), c ] + [a, b, ( A c )] = I A [a, b, c]
r
r r
r
r
r r
r
r
r r r
[a, ( A b), ( A c )] + [( A a), b, ( A c )] + [( A a), ( A b), c ] = II A [a, b, c]
r
r
r
r r r
[( A a), ( A b), ( A c )] = III A [a, b, c ]
r r r
where [a, b, c ] a (b c ) 0 holds with a 0 , b 0 , c 0 . Now if we consider that the

r
vector a is given by a = A f we can get:

r r
r r r
r
r r
r
r r r
[( A a), b, c] + [a, ( A b), c ] + [a, b, ( A c)] = I A [a, b, c ]
r r r
r
r r
r r
r r r
r
[( A A f ), b, c ] + [( A f ), ( A b), c ] + [( A f ), b, ( A c )] = I A [( A f ), b, c ]
r r r
r
r r
r r
r r r
r
[( A 2 f ), b, c ] + [( A f ), ( A b), c] + [( A f ), b, ( A c)] = I A [( A f ), b, c ]
r r r
r r r
r
r r
r r
r
[( A 2 f ), b, c ] I A [( A f ), b, c ] = [( A f ), ( A b), c ] [( A f ), b, ( A c)]
(1.64)
According to the definition of II A it is also true that:

Ciudad Real - Spain
Draft
1 TENSORS
65
r
r
r r
r
r r
r r r
r
r
[f , ( A b), ( A c )] + [( A f ), b, ( A c )] + [( A f ), ( A b), c ] = II A [f , b, c ]
r
r
r r r
r r
r
r r
r
r
[f , ( A b), ( A c )] II A [ f , b, c ] = [( A f ), b, ( A c )] [( A f ), ( A b), c]
Taking into account the above equation into the equation (1.64) we can obtain:
r r r
r r r
r
r r
r r
r
[( A 2 f ), b, c ] I A [( A f ), b, c] = [( A f ), ( A b), c] [( A f ), b, ( A c )]
r r r
r r r
r
r
r r r
r
[( A 2 f ), b, c ] I A [( A f ), b, c] = [f , ( A b), ( A c )] II A [ f , b, c ]
r r r
r r r
r r r
r
r
r
[( A 2 f ), b, c ] I A [( A f ), b, c] + II A [f , b, c ] [f , ( A b), ( A c )] = 0
r r r
r r r
r r r r
r
r
( A 2 f ) (b c ) I A ( A f ) (b c) + II A f (b c ) f ( A b) ( A c ) = 0
r
r r
r
In Problem 1.56 we have shown that ( A b) ( A c) = [cof( A )] (b c ) holds, then the
above equation becomes
r r r
r r r
r r r r
r
r
( A 2 f ) (b c ) I A ( A f ) (b c) + II A f (b c ) f ( A b) ( A c ) = 0
r r r
r r r
r r r r
r r
( A 2 f ) (b c ) I A ( A f ) (b c) + II A f (b c) f [cof( A )] (b c) = 0
r
r
r r
r r
( A 2 f ) I A ( A f ) + II A f f [cof( A )] (b c ) = 0
r
r r r r
r r
r
Note that the vectors ( A 2 f ) , ( A f ) , f 0 , (f [cof(A )]) are not orthogonal to (b c ) 0 ,
so, we can conclude that
r
r
r r
r
( A 2 f ) I A ( A f ) + II A f f [cof( A )] = 0
r
r
r
r r
T
A 2 f I A A f + II A 1 f [cof( A )] f = 0
r r
T
A 2 I A A + II A 1 [cof( A )] f = 0
r
T
A 2 I A A + II A 1 [cof( A )] = 0
Using the definition A A 1 = [cof( A )]T , the above equation becomes

r
T
A 2 I A A + II A 1 [cof( A )] = 0
r
A 2 I A A + II A 1 A A 1 = 0
r
A 2 A I A A A + II A 1 A A A 1 A = 0 A
r
A 3 I A A 2 + II A A A 1 = 0
Problem 1.78
Based on the Cayley-Hamilton theorem, find the inverse of a tensor T in terms of tensor
power.
Solution:
The Cayley-Hamilton theorem states that:

T 3 T 2 I T + T II T III T 1 = 0
Carrying out the dot product between the previous equation and the tensor T 1 , we obtain:
T 3 T 1 T 2 T 1 I T + T T 1 II T III T 1 T 1 = 0 T 1
T 2 TI T + 1 II T III T T 1 = 0
T 1 =

Ciudad Real - Spain
1
T 2 I T T + II T 1
III T
Draft
66
Problem 1.79
Check the Cayley-Hamilton theorem by using a second-order tensor whose Cartesian
components are given by:
5 0 0
T = 0 2 0
0 0 1
Solution:
The Cayley-Hamilton theorem states that:
T 3 T 2 I T + T II T III T 1 = 0
where I T = 5 + 2 + 1 = 8 , II T = 10 + 2 + 5 = 17 , III T = 10 , and

3
5 3
=0
0
0 125 0 0
0 = 0 8 0
1 0 0 1
0
23
0
; T
5 2
=0
0
0
22
0
0 25 0 0
0 = 0 4 0
1 0 0 1
By applying the Cayley-Hamilton theorem, we can verify that the following is true:
125 0 0
25 0 0
5 0 0
1 0 0 0 0 0
0 8 0 8 0 4 0 + 17 0 2 0 10 0 1 0 = 0 0 0
0 0 1
0 0 1
0 0 1
0 0 1 0 0 0
Problem 1.80
Given the matrix P which is represented by its components Pij (i, j = 1,2,3,4) . a) Obtain the
inverse of P , b) the invariants, y c) the characteristic equation. Consider that:
1
2
P=
4
2 3 1
2 1 2
1 5 3
1 2 4
and
1
0
1=
0
0 0 0
1 0 0
0 1 0
0 0 1
Solution:
By applying the Cayley-Hamilton theorem we obtain:
P 4 + P 3 I1 + P 2 I 2 + P I 3 + I 41 = 0
(
P (P (P
P P 3 + P 2 I1 + P I 2 + 1 I 3 + I 41 = 0
2
+ P I1 + 1 I 2 + 1 I 3 + I 41 = 0
P P P P + 1 I 1 + 1 I 2 + 1 I 3 + I 41 = 0
144 44
2
3

C1
P P C1 + 1 I 2 + 1 I 3 + I 4 1 = 0
14 244
4
3
C2
P C2 + 1 I 3 + I 4 1 = 0
C3 + I 4 1 = 0
Ciudad Real - Spain
Draft
1 TENSORS
67
where we denoted by:
C0 = P
(
= P (C
= P (C
)
+1 I )
+1 I )
C1 = P C0 + 1 I 1
C2
C3
We can obtain the trace of C3 + I 41 = 0 as follows:

Tr (C3 + I 4 1) = Tr (0 )
Tr (C3 ) + Tr ( I 4 1) = Tr (C3 ) + I 4 Tr (1) = Tr (C3 ) + 4 I 4 = 0

I4 =
Tr (C3 )
4
Similarly, we can define that:

I3 =
Tr (C2 )
3
I2 =
Tr (C1 )
2
I1 =
Tr (C0 )
1

I1 =
Tr (C0 )
= (1 + 2 + 5 + 4) = 12
1
thus we evaluate the matrix C1 = P C0 + 1 I 1 :

1
2
C1 =
4
2 3 1 1
2 1 2 2

1 5 3 4

1 2 4 3
I2 =
2 3 1
1
0
2 1 2
12
0
1 5 3
1 2 4
0
Tr (C1 ) (8 13 16 21) (42)

=
=
= 21
2
2
2
In turn we get C2 = P C1 + 1 I 2
1
2
C2 =
4
I3 =
2 3 1 8
6
14 14
1
0
8 13
2 1 2
5
7
+ 21
0
1 5 3 13
6
16 3

11
1 2 4
2
4
21
0
0 0 0 37 22
15 17
7
2 5 5
1 0 0
=
0 1 0 10 12 14
2

14 11
0 0 1 9
5
Tr (C2 ) ( 37 2 14 + 5) ( 48)
=
=
= 16
3
3
3
In turn we get C3 = P C2 + 1 I 3
1
2
C3 =
4
I4 =
14 14
0 0 0 8
6
8 13
7
1 0 0
5
=
16 3
0 1 0 13
6

21
0 0 1 11
2
4
2 3 1 37 22
15 17
1
0
7
2 1 2
2 5 5
+ 16
0
1 5 3 10 12 14
2

1 2 4 9
5
14 11
0
0 0 0 32 0 0 0
1 0 0 0 32 0 0
=
0 1 0 0 0 32 0

0 0 1 0 0 0 32
Tr (C3 ) 4(32)
=
= 32 = det (P )
4
4
Then, the characteristic equation becomes:

Ciudad Real - Spain
Draft
68
P 4 + P 3 I1 + P 2 I 2 + P I 3 + I 41 = 0
P 4 12P 3 + 21P 2 + 16P 321 = 0
The characteristic equation coefficients could have been obtained by evaluates the
determinant:
det ( P 1) P 1
c) The inverse can be obtained by starting from:

P (C2 + 1 I 3 ) + I 4 1 = 0
P 1 P C2 + 1 I 3 + I 4 P 11 = 0
P 1 =
1
C2 + 1 I 3
I4
P 1 =
thus:
P 1
C2 + 1 I 3 + I 4 P 1 = 0
1
adj[P ]
det (P )
37 22
15 17
1
2 5 5
1 7
+ 16
=
0
2
(32) 10 12 14
9
14 11 5
0
adj[P ] = C2 + 1 I 3
0 0 0
15 17
21 22
7
1 0 0 1
14
5 5
=
0 1 0 32 10 12 2
2
0 0 1
14 11 21
9
NOTE 1: This procedure just performed, in the literature, is called Faddeev-Leverrier method.
Note that the inverse can also be obtained by using the same procedure as the one used in the
equation (1.46), i.e.:
32

2
4
3
2
4
3
2
4
3
2
4
3
0 0 0
2 1 2
1 5 3
1 2 4
1 0 0
2 1 2
1
1
0
2
1
5 3
2 4
1 0
1 2
5 3
1 2 4
0 0 1
2 1 2
1 5 3
1 2 4
1
4
3
0
4
3
0
4
3
0
4
3
2 3 1
0 0 0
1 5 3
1 2 4
2 3 1
1 0 0
1
1
2
0
1
1 2 4
2 3 1
0 0 1
1 5 3
1 2 4

Ciudad Real - Spain
5 3
2 4
3 1
1 0
5 3
2
1
3
2
0
3
2
0
3
2
0
3
2 3 1
2 1 2
0 0 0
1 2 4
2 3 1
2 1 2
1
1
2
2
0
0 0
2 4
3 1
1 2
1 0
1 2 4
2 3 1
2 1 2
0 0 1
1 2 4
Draft
2
4
1
2
4
0
2
4
0
2
4
0
2 3 1
2 1 2
1 5 3
0 0 0
2 3 1
2 1 2
1 5 3
1 0 0
2 3 1
2 1 2
1 5 3
0 1 0
2 3 1
2 1 2
1 5 3
0 0 1
21
= 1 7
32 10
22 15 17
14
12
14
11 21
1 TENSORS
69
NOTE 2: We can also obtain the characteristic coefficients by means of the following
procedure. Considering P 4 P 3 I 1 + P 2 I 2 P I 3 + I 4 1 = 0
The last coefficient is I 4 = det(P ) = 32 .
The coefficient I 3 is obtained by the sum of the determinants of the resulting matrices by
eliminating 1 row and 1 column associated with the main diagonal, i.e.
1
2
I3 =
4
2 3 1 1 2
2 1 2 2 2
+
1 5 3 4 1

1 2 4 3 1
3 1 1 2
1 2 2 2
+
5 3 4 1

2 4 3 1
3 1 1 2
1 2 2 2
+
5 3 4 1

2 4 3 1
3 1
1 2
5 3
2 4
2 1 2 1 3 1 1 2 1 1 2 3
= 1 5 3 + 4 5 3 + 2 2 2 + 2 2 1 = 16
1 2 4 3 2 4 3 1 4 4 2 5
eliminating 2 rows and 2 columns associated with the main diagonal, i.e.
1
2
I2 =
4
2 3 1
2 1 2
+
1 5 3
1 2 4
1
2
2 3 1
2 1 2
+
1 5 3
1 2 4
1
2
+
4
2 3 1 1
2 1 2 2
+
1 5 3 4

1 2 4 3
2 3 1
2 1 2
+
1 5 3
1 2 4
1
2
+
4
2 3 1
2 1 2
1 5 3
1 2 4
1
2
2 3 1
2 1 2
+
1 5 3
1 2 4
5 3 2 2 2 1 1 1 1 3 1 2
=
= 21
+
+
+
+
+
2 4 1 4 1 5 3 4 4 5 2 2
eliminating 3 rows and 3 columns associated with the main diagonal, i.e.
1
2
I1 =
4
2 3 1 1 2
2 1 2 2 2
+
1 5 3 4 1

1 2 4 3 1
3 1 1
1 2 2
+
5 3 4

2 4 3
= [4] + [1] + [2] + 5 = 12 = Tr (P )

Ciudad Real - Spain
2 3 1 1 2
2 1 2 2 2
+
1 5 3 4 1

1 2 4 3 1
Draft
3 1
1 2
5 3
2 4
70
Problem 1.81
Let A be a second-order tensor, show that:
a) II A =
1
( I A ) 2 Tr ( A 2 )
2
b) det ( A ) =
1
[Tr (A )]3 + 2 Tr( A 3 ) 3Tr(A ) Tr(A 2 )
6
Solution:
a) It was shown in Problem 1.78 that III A A 1 = (A 2 AI A + 1 II A ) , then, by applying the

double scalar product with the second-order unit tensor we obtain:
III A A 1 : 1 = A 2 AI A + 1 II A : 1 = A 2 : 1 A : 1 I A + 1 : 1 II A
1
III A Tr ( A ) = Tr ( A ) Tr ( A ) I A + Tr (1) II A = Tr ( A 2 ) ( I A ) 2 + 3 II A
Taking into account the inverse of a tensor A 1 =
[cof ( A )]T
III A Tr ( A 1 ) = Tr ( III A A 1 ) = Tr III A
III A
[cof ( A )]T , we can conclude that:

III A
= Tr [cof ( A )]T = Tr ([cof ( A ) ]) = II A
With that, we can obtain:

III A Tr ( A 1 ) = II A = Tr ( A 2 ) ( I A ) 2 + 3 II A
II A 3 II A = Tr ( A 2 ) ( I A ) 2
II A =
1
( I A ) 2 Tr ( A 2 )
2
b) We start from the Cayley-Hamilton theorem, which states that any tensor satisfies its own
characteristic equation, i.e.:
A 3 A 2 I A + AII A III A 1 = 0
where I A = [Tr (A )] , II A =
1
[Tr( A )]2 Tr( A 2 ) , III
2
(1.65)
= det (A ) are the principal invariants of
A . Applying the double scalar product between the second-order unit tensor ( 1 ) and the
equation in (1.65) we obtain:

A 3 : 1 A 2 : 1 I A + A : 1II A III A 1 : 1 = 0 : 1
Tr ( A 3 ) Tr ( A 2 ) I A + Tr ( A ) II A III
[Tr(1)] = [Tr(0)]
1
[Tr(A )]2 Tr(A 2 ) III A 3 = 0
2
1
1
3
Tr ( A 3 ) Tr ( A 2 ) Tr ( A ) + [Tr ( A )] Tr ( A ) Tr ( A 2 ) III A 3 = 0
2
2
1
3
2 Tr ( A 3 ) 3 Tr ( A 2 ) Tr ( A ) + [Tr ( A )] III A 3 = 0
2
Tr ( A 3 ) Tr ( A 2 ) Tr ( A ) + Tr ( A )
with which we obtain:

III
= det ( A ) =
1
[Tr( A )]3 + 2 Tr(A 3 ) 3Tr( A 2 ) Tr(A )
6
or in indicial notation:

Ciudad Real - Spain
Draft
1 TENSORS
III
71
1
A ii A jj A kk + 2 A ij A jk A ki 3A ij A ji A kk
6
= det ( A ) =
NOTE: It is interesting to note that the principal invariants of A are formed by the three
fundamental invariants of a second-order tensor, namely Tr (A ) , Tr ( A 2 ) , Tr ( A 3 ) , i.e.:
I A = Tr ( A )
III
} {
1
1
2
2
I A Tr ( A 2 ) = [Tr ( A ) ] Tr ( A 2 )
2
2
1
3
= det ( A ) = [Tr ( A )] + 2 Tr ( A 3 ) 3 Tr ( A 2 ) Tr ( A )
6
II A =
Problem 1.82
Show that II T = III T Tr ( T 1 ) , where II T =
invariant of T , and III
1
[Tr( T )]2 Tr(T 2 )
2
is the second principal
is the third principal invariant (the determinant of T ).
Solution:
It was shown in Problem 1.78 that T 1 =
1
T 2 TI T + 1 II T , then, by applying the double
III T
scalar product with the second-order unit tensor we obtain:

T 1 : 1 =
Tr ( T 1 ) =
1
1
T 2 TI T + 1 II T : 1 =
T 2 : 1 T : 1 I T + 1 : 1 II T
III T
III T
1
Tr ( T 2 ) Tr ( T ) I T + Tr (1) II T
III T
2
III T Tr ( T 1 ) = Tr ( T 2 ) I T + 3 II T
14243
= 2 II T
III T Tr ( T 1 ) = II T
Problem 1.83
Show that:
r r
1
(1 + c b) 1 = 1
(c b )
r r
( + c b)
where c and b are vectors, 1 is the second-order unit tensor, and and are scalars.
Solution:
r
Let us consider that T = (1 + c b) , and the inverse of a tensor obtained in Problem

1.78:
T 1 =
1
T 2 TI T + 1 II T
III T
(1.66)
Next, we obtain T 2 :

Ciudad Real - Spain
Draft
72
r r
r r
T 2 = T T = (1 + c b) (1 + c b)
r r
r r
r r r r
= 2 1 1 + 1 (c b) + (c b) 1 + 2 (c b) (c b)
r r r r
r r r r
where it fulfills that (c b) (c b) = (c b)(c b) , (see Problem 1.20). Then, the above
expression can be rewritten as follows:
r r
r r r r
T 2 = 2 1 + 2 (c b) + 2 (c b)(c b)
and its trace is given by:
r r
r r r r
Tr ( T 2 ) = Tr 2 1 + 2 (c b) + 2 (c b)(c b)
r r
r r
r r
= 2 Tr (1) + 2 Tr (c b) + 2 (c b) Tr (c b)
r r
r r r r
= 3 2 + 2 (c b) + 2 (c b)(c b)
r r
r r
= 3 2 + 2 (c b) + 2 (c b) 2
Next, we calculate the principal invariants of T
r r
r r
r r
I T = Tr (1 + c b) = Tr (1) + Tr (c b) = 3 + (c b)
r r 2
r r
r r
( I T ) 2 = 3 + (c b) = 9 2 + 6 (c b) + 2 (c b) 2
} {
r r
r r
r r
r r
1
1
2
I T Tr ( T 2 ) = 9 2 + 6 (c b) + 2 (c b) 2 3 2 + 2 (c b) + 2 (c b) 2
2
2
r r
2
= 3 + 2 (c b)
II T =
]}
r r
r r
III T = det (1 + c b) = 3 + 2 c b (see Problem 1.50)
Then, the equation in (1.66) becomes:

III T T 1 = T 2 I T T + II T 1
r r
r r r r
= 2 1 + 2 (c b) + 2 (c b)(c b)
r r
r r
r r
3 + (c b) (1 + c b) + 3 2 + 2 (c b) 1
r r
r r r r
r r
r r
= 2 1 + 2 (c b) + 2 (c b)(c b) 3 2 1 3 (c b) (c b)1
r r r r
r r
2 (c b)(c b) + 3 2 1 + 2 (c b)1
r r
r r
= 1 2 + (c b)1 (c b)
r r
r r
= ( 2 + c b)1 (c b)
r r
r r
1
T
= ( 3 + 2 c b)1 (c b) = [adj( T ) ] = [cof ( T ) ]
(1.67)
r r
Taking into account that T = (1 + c b) , III T = 3 + 2 c b , the above equation

becomes:
T
r r
r r
(c b) 1
(c b)
1 III T
=
1
= 1
r r
III T
III T
( 3 + 2 c b )
(1.68)
or:

Ciudad Real - Spain
Draft
1 TENSORS
r r
1
(1 + c b) 1 = 1
r r
r r (c b)
( + c b)
Tensorial notation
(1.69)
(c b )
( + c k b k ) i j
Indicial notation
(1.70)
[{c}{b}T ]
T
( + {c} {b})
Matrix notation
(1.71)
( ij + c i b j ) 1 =
[[1] + [{c}{b} ] ]
T
[ 1]
73
ij
NOTE 1: The above equation is also valid for matrices of n-dimensions.

In the particular case when = 1 , = 1 , we obtain:
r r
r r 1
(c b)
(1 + c b) = 1
r r
1+ c b
(1.72)
NOTE 2: The equation in (1.69) or in (1.67) can be rewritten as follows:

r r
1
T 1 = (1 + c b) 1 = 1
r r
r r (c b )
( + c b )
r r
r r
2
r r ( + c b)1 (c b)
( 3 + 2 c b )
1
=
[adj( T )]
det ( T )
1
with that we conclude that:
r r
r r
r r
adj(1 + c b) = ( 2 + c b)1 (c b)
NOTE 3: We can extend the equation in (1.69) such that:

( I sym + A B ) 1 =
I sym
( A B)
( + A : B )
where we now have that I sym is the symmetric fourth-order unit tensor, A and B are secondorder tensors, and and are scalars. With that it is easy to show that ( I sym ) 1 = I sym .
Problem 1.84
r
Taking into account that (1 + c b) 1 =

r r
1
( A + a b) 1 = A 1
r r
r r (c b) , show that:
( + c b )
r
r
r
( A 1 a) (b A 1 )
r
( + b A 1 a)
(1.73)
where a and b are vectors, A is a second-order tensor, with det ( A ) 0 ( A 1 ), and and
are scalars.
Solution:
r
Note that the expression (A + a b) can be rewritten as follows:

r r
r r
r r
r r
(A + a b) = (A 1 + 1 a b) = (A 1 + ( A A 1 ) a b) = A (1 + A 1 a b)

Ciudad Real - Spain
Draft
74
Using the inverse property such as ( A B ) 1 = B 1 A 1 , we can obtain:
r r
r r 1
r r
(A + a b) 1 = A (1 + A 1 a b) = (1 + A 1 a b) 1 A 1
r
Note that the result of the algebraic operation A 1 a is a vector in which we denote
r
r
c = A 1 a , with that we rewrite the above equation as follows:
r r
r r
r r
(A + a b) 1 = (1 + A 1 a b) 1 A 1 = (1 + c b) 1 A 1
1
r r
1
= 1
r r (c b) A
( + c b )
r r
1
1
= 1 A 1
r r (c b) A
( + c b)
r r
1
1
= A 1
r r c bA
( + c b )
r
1
1 r
1
= A 1
r r ( A a) (b A )
( + c b )
r
1
1 r
1
r
= A 1
r ( A a) (b A )
( + b A 1 a)
The above equation in indicial notation becomes:

(A ij + a i b j ) 1 =
A ij1
( + b p A 1 a q )
pq
( A ik1 a k )(b s A 1 )
sj
The reader should be aware here with the algebraic operation ( A 1 a) b
r r
A 1 ( a b )
14243
Invalid Expression
the latter has no consistency, since we can not have a scalar product (contraction) with the
r r
scalar (a b) . We can check this fact by means of indicial notation
r r
r
then,
the
possible
c b = c i b i = ( A 1 a) i b i = A ik1 a k b i ,
r r
1
1
1
1
1
1
( A a) b = b i A ik a k = a k A ik b i = A ik b i a k = A ik a k b i = A ik a k b i .
1 24 1 24
4 3
4 3 1 24
4r3
1 24r
4 3
1 24
4 3
r
r
r
r
r
r
r r
bA 1a
aA T b
A 1:(ba)
A T :( ab )
expressions
are
A 1:( ab )T
For the particular case when = 1 , = 1 , we fall back on the Sherman-Morrison formula:
r
r
r r 1
( A 1 a) (b A 1 )
1
r
( A + a b) = A
r
1 + b A 1 a
Sherman-Morrison formula
(1.74)
(tensorial notation)
The above equation in matrix notation becomes
[[ A] + [{a}{b} ] ]
T
= [ A]
}{
[ A] 1 {a} {b}T [ A] 1
1 + {b}T [ A] 1 {a}
Sherman-Morrison formula
(1.75)
(matrix notation)
r
NOTE 1: Note that if (A + a b) = A (1 + A 1 a b) , the determinant is defined

as follows:
r r
r r
r r
det (A + a b) = det A (1 + A 1 a b) = det [A ]det (1 + A 1 a b)
r
r
= det [A ]( 3 + 2 b A 1 a)

Ciudad Real - Spain
Draft
1 TENSORS
75
with that, the equation in (1.73) can be rewritten as follows:
r
r
r r
r
r
1
( A + a b) 1 =
A ( 2 + b A 1 a) A 1 A ( A 1 a) (b A 1 )
]}
with = det (A + a b) = A ( 3 + 2 b A 1 a) .
(1.76)
r
r
r r
r
r
adj(A + a b) = A ( 2 + b A 1 a) A 1 A ( A 1 a) (b A 1 )
]}
NOTE 2: We can extend the equation in (1.73) such that:

( D + A B ) 1 =
D 1
( + B : D
[(D
: A)
: A ) (B : D 1 )
(1.77)
where we now have that D is a fourth-order tensor, A and B are second-order tensors, and
and are scalars.
Note that:
( D + A B ) 1 =
{D (
+ B : D 1 : A )D 1 D (D 1 : A ) (B : D 1 )
]}
with = det (D + A B ) = D ( 3 + 2 B : D 1 : A ) .
(1.78)
where we can conclude that:

det (D + A B ) = det (D )( 3 + 2 B : D 1 : A )
adj(D + A B ) = D ( 2 + B : D 1 : A )D 1 D (D 1 : A ) (B : D 1 )
Problem 1.85
(1.79)
]}
(1.80)
a) Let C = (1 + a b + c d) be a second-order tensor. Show that:

r
r r
[r
r r
r r r
r r r r
1 + a b + c d = 3 + 2 (c d) + 2 (a b) + (a b)(c d) (a d)(b c )
(1.81)
r
where 1 + a b + c d = det (1 + a b + c d) represents the determinant of

r
the tensor C . b) For the particular case when = 1 , d = a , c = b , show that:

r r
r r
r r
r r
det (1 + a b + b a) = 1 + ( + )(a b) a b
Solution:
r
(1.82)
r
We define an auxiliary tensor D = 1 + a b and in turn we have C = (D + c d) .

According to Problem 1.84, (see equation (1.73)), it holds that:
r
r r
r
det (D + c d) = D (1 + d D 1 c ) , where:
r r
r r
det (D) D = det (1 + a b ) = 3 + 2 (a b) and

Ciudad Real - Spain
Draft
76
r r
1
(D ) 1 = (1 + a b) 1 = 1
r r
r r (a b)
( + a b )
With that, we can say that:
r
r r
r
det (D + c d) = D (1 + d D 1 c )
r 1
r r
r r r
= 3 + 2 (a b ) (1 + d 1
r r (a b) c
( + a b )
r r r r
1 r
r r
r
= 3 + 2 (a b ) 1 + d 1 c
r r d (a b) c
( + a b )
r r
r r
1 r r
dc
r r (d a) (b c )
( + a b )
r r
r r r r
r r
= 3 + 2 (a b ) 1 + (c d)
r r (a d)(b c )
2 ( + a b )
r r
r r
r r
r r
r r r r
Note that (d a) (b c ) = (a2d) (b2c ) = (a d)(b c ) .
1 3 1 3
r r
= 3 + 2 (a b ) 1 +
scalar
scalar
[
[
r r
r r
r r
r r r r
r r
det (D + c d) = 3 + 2 (a b) 1 + (c d) 3 + 2 (a b)
r r (a d)(b c )
2

( + a b )
r
r
r r r
r
r
r
= 3 + 2 (a b) 1 + (c d) (a d)(b c )

r r
r r
r r r r
r r r r
= 3 + 2 (c d) + 2 (a b) + (a b)(c d) (a d)(b c )
Then:
r r
r r
r r
r r
r r r r
r r r r
det (1 + a b + c d) = 3 + 2 (c d) + 2 (a b ) + (a b )( c d) (a d)(b c )
thereby demonstrating the equation (1.81).

r
For the particular case when d = a , c = b , we have:
[
[
r r
r r
r r
r r
r r r r
r r r r
det (1 + a b + c d) = 3 + 2 (c d) + 2 (a b) (a d)(b c ) (a b)(c d)
r r
r r
r r
r r
r r r r
r r r r
det (1 + a b + b a) = 3 + 2 (b a) + 2 (a b) (a a)(b b) (a b)(b a)
r r
r r r r
r r r r
= 3 + 2 ( + )(a b) (a a)(b b) (a b)(a b)
r r 2 r 2 r 2 r r 2
In Problem 1.1 we have shown that a b = a b a b holds, thus:
]
]
( )
r r
r r
r r
r r
det (1 + a b + b a) = 3 + 2 ( + )(a b) a b
For the particular case when = 1 we get:

r r
r r
r r
r r
det (1 + a b + b a) = 1 + ( + )(a b) a b

Ciudad Real - Spain
Draft
1 TENSORS
Problem 1.86
r
77
a) Obtain the inverse of the tensor C = (1 + a b + c d) .

r r
p p
pq
b.1) Given a second-order tensor D = B + r r
r
r
(B q) (B q)
where B = B T and B 1 ,
r
r
q B q
show that:
D 1 = B 1 +
r r r
r
r
r
( p q + p B 1 p ) r r
[q q ] r 2 r q ( B 1 p)
r r 2
( p q)
( p q)
sym
(1.83)
b.2) If B is a positive definite tensor, obtain the conditions under which D is a non-singular
tensor.
Solution:
Denoting by A = (1 + a b) we obtain C = ( A + c d) , and by taking into account

r r
1
( A + a b) 1 = A 1
r
r
( A 1 a) (b A 1 )
r
r
( + b A 1 a)
(1.84)
which was obtained in Problem 1.84, (see equation (1.73), thus

r r
( A + c d) 1 = A 1
r
r
r
( A 1 c ) (d A 1 )
r
(1 + d A 1 c )
(1.85)
It was shown in Problem 1.83 that:

r r
1
(1 + c b) 1 = 1

A
r r
1
= (1 + a b) 1 = 1
(c b )
r r
( + c b)
(1.86)
(a b)
r r
( + a b )
Furthermore, we have
r r
r r
r r
r 1
(a b ) r 1 r
(a b) r 1 r
(b c ) r
A c = 1
r r c = 1c
r r c = c
r r a
( + a b )
( + a b )
( + a b )
r r
r r
r r
r
r 1
r
(a b) 1 r
(a b)
( d a) r
1 r
1
d A = d 1
r r = d 1 d
r r = d
r r b
( + a b )
( + a b )
( + a b )
With that we conclude that
] [
r r r
r
r r
r r
r r
r
r r r
(1 + a b + c d) 1 = (1) 1 + ( 2 ) (a b) + ( 3) (1) c + ( 2 ) (b c )a (1) d + ( 2 ) (a d)b
(1.87)
where
(1) =
( 2) =
r r
( + a b )

Ciudad Real - Spain
Draft
78
( 3) =
r
r
(1 + d A 1 c )
r
r 1 r r
d A 1 c = (d c )
r r r r
r r (d a )(b c )
( + a b )
NOTE: The equation in (1.87) is also valid for matrices of n-dimensions.

b.1) We can rewrite the tensor D as follows:
r r
r
r
r r
r
r
p p
p p
(B q) (B q)
(B q) (B q)
1
1
D = B 1 + 1 r r 1
= B 1 + (B B ) r r (B B )
r
r
r
r
pq
q B q
pq
q Bq
r r
r
r
r
r
r
r
(B q) (B q)
( B 1 p) p ( B 1 B q ) ( B q )
p p
= B 1 +
= B 1 + B 1 r r B 1
r
r
r r
r
r
pq
q B q
pq
q Bq
r r
r
1 r
( B p) p q ( B q )
= B 1 +
r
r r
r
pq
q Bq
and by denoting by
r
r
a = ( B 1 p)
r r
b=p
r r
c =q
r
r
d = (B q)
= r r
pq
1
= r
r
q B q
we get
r r
r r
D = B 1 + a b + c d
= B C
D 1 = ( B C ) 1 = C 1 B 1
where C = 1 + a b + c d . The inverse of C can be obtained via subsection (a) with

= 1 . Moreover, we have:
(1) = 1 ,
( 2) =
1
1
1
= r r r
r
r r =
r r = r r
r r
( + a b) (1 + a b) p q (1 + r 1 r ( B 1 p) p) ( p q + p B 1 p)
pq
r
r 1 r r
d A 1 c = (d c )
r r r r
r r (a d )(b c )
( + a b )
r r
r
r r r
1
= (( B q ) q ) + r r r
(( B 1 p) ( B q ) )( p q )
1 r
( p q + p B p)
r
r r r
r
r ( p ( B T B ) q ) ( p q )
= q B q +
r r r
r
( p q + p B 1 p)
1
1
r
= r
r
r
r r r
1 r
r ( p ( B T B ) q ) ( p q )
1 r
(1 + d A c ) q B q
1 + r
q Bq +
r
r r r
r
q Bq
( p q + p B 1 p )
r r r
1 r
( p q + p B p)
= r
r r r
( p ( B T B ) q ) ( p q )
r r r
r
( p q + p B 1 p )
1
1
( 2 ) (3) = r r r 1 r r
r r r = r
r r r
T
T
( p q + p B p) ( p ( B B ) q ) ( p q ) ( p ( B B ) q ) ( p q )
( 3) =

Ciudad Real - Spain
Draft
1 TENSORS
r r
( 2 ) ( 3) (a d ) = r
( p (B
= r
( p ( B T
r r
( 2 ) ( 3) (b c ) = r
( p (B
r r r r
79
r
1
1 r
r r r (( B p) ( B q ))
B) q) ( p q )
r
r
1
1
T
r r r ( p (B B) q ) = r r
( p q)
B) q) ( p q )
r r
1
1
r r r ( p q) = r
r
T
B) q ) ( p q )
( p (B B) q )
( 2 ) (3) (b c )(a d ) = r
( p (B
r r r
1
1 r
r r r (( B p) ( B q ))( p q ) = 1
B) q ) ( p q )
The equation in (1.87) becomes:
[
[
] [
r r r
r
r r
r
r r r
C 1 = 1 + ( 2 ) (a b ) + ( 3) c + ( 2 ) (b c )a d + ( 2 ) (a d )b
r r r r
r r
r r
r r r r
C 1 = 1 + ( 2 ) a b + ( 3) c d + ( 3) ( 2 ) (a d ) c b + ( 3) ( 2 ) (b c ) a d + L
r r r r r r
2
L + ( 3) ( 2 ) (b c )(a d ) a b
r r r r r r
r r
r r r r
2
C 1 = 1 + ( 2 ) + ( 3) ( 2 ) (b c )(a d ) a b + ( 3) c d + ( 3) ( 2 ) (a d ) c b + L
r r r r
L + ( 3) ( 2 ) ( b c ) a d
}[
Note that: { ( 2 ) + (3) (22 ) (b c )(a d )}= ( 2 ) { + (3) ( 2 ) (b c )(a d )}= ( 2 ) { 1} = 0 , thus
1
1
r r r r
r r r r
r r r r
r r
r r r r
C 1 = 1 + ( 3) c d + ( 3) ( 2 ) (a d ) c b + ( 3) ( 2 ) (b c ) a d
r r r r
r r
r r r r
C 1 = 1 + ( 3) c d + ( 3) ( 2 ) (a d ) c b + ( 3) ( 2 ) (b c ) a d
r r r
r
r
r
( p q + p B 1 p)
1 r r
=1 + r
r r r [q ( B q ) ] + r r [q p ] + L
T
( p q)
( p (B B) q ) ( p q )
r
1
1 r
L+ r
r ( B p) ( B q )
T
( p (B B) q )
With that, we can obtain:

D 1 = C 1 B 1
r r r
r
r
r
1 r r
( p q + p B 1 p )
= 1 + r
r r r [q ( B q ) ] + r r [q p ] + L
T
( p q)
( p (B B) q ) ( p q )
[
[
= B 1
]
]
r 1
1
1 r
L+ r
r ( B p) ( B q ) B
T
( p (B B) q )
r r r
r
1
r
r
( p q + p B p)
1 r r
1
1
+ r
r r r [q ( B q )] B + r r [q p ] B + L
T
( p q)
( p (B B) q ) ( p q )
r
1
1 r
1
L+ r
r ( B p) ( B q ) B
T
( p (B B) q )
Note that:
r
r
r
r
r
{[q ( B q )] B } = [q ( B q )] B = [q ( B q ) ]B
r
r
= [q ( B B ) q ) ]
r
r
r
r
[( B p) ( B q )] B = ( B p) ( B B ) q
1
ij
ik
1
kj
1
kj
= q i ( B kp q p ) B kj1 = q i ( B kp B kj1 q p )
ij
If now we consider the symmetry of B , i.e. B = B T , we obtain:

Ciudad Real - Spain
Draft
80
r r r
r
( p q + p B 1 p ) r r
[q q ] + r 1
r r r r
r
( p q) ( p q)
( p q)
r r r
r
( p q + p B 1 p ) r r
[q q ] + r 1
+
r r r r
r
( p q) ( p q)
( p q)
r r r
r
( p q + p B 1 p ) r r
+
[q q ] r 2 r
r r 2
( p q)
( p q)
D 1 = B 1 +
= B 1
= B 1
r
r
[q ( p B )] + ( p1q ) [( B
r r
1
r
{[qr ( p B )] + [( B
1
r
[q ( B
p) q ]
r
p) q ]
r
p)]
r
sym
Note that, due to the symmetry of B , it holds that p B 1 = B 1 p = s , and B T B = 1 .

b.2) A tensor is non-singular if det (D ) 0 . By using the equation obtained previously we get:
r r
r r
D = B 1 + a b + c d
( [
])
r r
r r
r r
r r
det (D) = det B 1 + a b + c d = det ( B )det 1 + a b + c d
Note that det ( B ) > 0 , since B is a positive definite tensor. Then, the condition under which
r r
r r
D is a non-singular tensor if det 1 + a b + c d
0 . By using the determinant
expression obtained in Problem 1.85 we get:
r r
r r
r r
r r r r
r r r r
det (1 + a b + c d ) = 3 + 2 (c d ) (a d )( b c ) (a b )(c d )
r r
r r r
r r r
r
r
r r r r r r
where = 1 , a b = ( B 1 p ) p = p B 1 p , a d = ( B 1 p ) ( B q ) = p q , b c = p q
r r r
r
r
r
r r
r
1 r
c d = q ( B q ) = q B q , (c d ) = r
r q B q = 1 ,
q B q
[(a d )(b c ) (a b )(c d ) ] = r r r

r r r r
r r r r
r r r r
r
r
1
1 r r
r ( p q )( p q ) ( p B p )( q B q )
pq q B q
1
Thus:
r r
r r
r r r r
r
r
1
1 r r
det 1 + a b + c d = r r r
r ( p q )( p q ) ( p B p )(q B q ) 0
( p q )(q B q )
r r r
r
r r
r
r
Then, the conditions are: p 0 , q 0 , ( p q ) 0 , i.e. p and q can not be orthogonal
vectors. Another condition that must be met is:

r r r r
r
r r
r
( p q )( p q ) ( p B 1 p )(q B q ) 0
14243 144424443
4
4
>0
>0
Note that by the fact that B is positive definite tensor, the scalar (q B q ) > 0 is always
r r
r
r
positive for any vector q 0 . The same apply to ( p B 1 p ) > 0 , since, if the tensor is
positive definite so is its inverse. Note also that D is a positive definite tensor if
r r
r
r r
r
r r
( p q ) 2 > ( p B 1 p )(q B q ) and ( p q ) > 0 . These two conditions can be replaced by
r r
r
r r
r
( p q ) > ( p B 1 p )(q B q ) .
Problem 1.87
Let A = A () and be a second-order tensor and a scalar respectively, show that:
dA
dA
A 1
= A Tr
d
d

Ciudad Real - Spain
Draft
(1.88)
1 TENSORS
81
Solution:
In Problem 1.81 and in Problem 1.78, we have demonstrated, respectively, that:
III
= det ( A ) = A =
1
[Tr( A )]3 + 2 Tr(A 3 ) 3Tr( A 2 ) Tr(A )
6
(1.89)
III A A 1 = A 2 AI A + II A 1
where I A = Tr (A ) , II A =
(1.90)
1
[Tr ( A )]2 Tr ( A 2 ) .
2
Note also that the following derivatives are true:

d [I A ] d [Tr ( A )] d [A kk ] d [A ik ik ] d [A ik ]
dA
dA
=
=
=
=
ik =
: 1 = Tr
d
d
d
d
d
d
d
d (A 2 )
d Tr ( A 2 )
dA
dA
= Tr
= 2 Tr A
= Tr 2A
d
d
d
d
d Tr ( A 3 )
dA
= 3Tr A 2
d
d
Taking the derivative of (1.89) with respect to we obtain:
}
[
d ( III A ) 1 d
[Tr( A )]3 + 2 Tr(A 3 ) 3Tr( A 2 ) Tr(A )
=
d
d
6
d Tr ( A 3 )
d Tr ( A 2 )
d [Tr ( A )]
1
2 d [Tr ( A )]
Tr ( A ) 3 Tr ( A 2 )
= 3[Tr ( A )]
+2
3
dt
d
d
d
6
1
dA
dA
2 dA
dA
2
2
3[Tr ( A )] Tr + 6 Tr A
6 Tr A d Tr ( A ) 3Tr ( A ) Tr d
d
6

dA
1
dA
dA
2
2
= Tr A 2
Tr A d Tr ( A ) + 2 [Tr ( A )] Tr ( A ) Tr d
d

or
d ( III A )
dA
dA
dA
= Tr A 2
Tr A
I A + II A Tr
d
d
d
Taking the scalar product of the equation in (1.90) with

III A A 1
(1.91)
dA
, we can obtain:
d
dA
dA
dA
dA
dA
= A 2 AI A + II A 1
= A2
A
I A + II A
d
d
d
d
d
and the trace of the above equation is given by:

dA
dA
dA
2 dA
Tr A 1
A
I A + II A
III A = Tr A
d
d
d
d
dA
dA
dA
= Tr A 2
Tr A
I A + Tr
II A
d
d
(1.92)
Comparing equations (1.91) and (1.92), we can conclude that:

d ( III A )
dA
dA
A 1
= III A Tr A 1
= III A Tr
d
d
d

Ciudad Real - Spain
Draft
82
1.1.11 Isotropic and Anisotropic Tensors

Problem 1.88
Let C be a fourth-order tensor, whose components are
C ijkl = ij kl + ik jl + il jk
(1.93)
where ij are the second-order unit tensor components, and , , are scalar.
a) What kind of symmetry has the tensor C ? b) What conditions must be met to guarantee
the symmetry of C ?
Solution:
The tensor has major symmetry if C ijkl = C klij holds. Taking into account the equation in
(1.93), we conclude that the tensor has major symmetry since
C klij = kl ij + ki lj + kj li = C ijkl
We check now if the tensor has minor symmetry, e.g. C ijkl = C ijlk
C ijlk = ij lk + il jk + ik jl C ijkl
We can easily verify this fact by adopting i = 2 , j = 1 , k = 1 , l = 2 , thus

C ijkl = C 2112 = 21 12 + 21 12 + 22 11 =
C ijlk = C 2121 = 21 21 + 22 11 + 21 12 =
Then, the tensor C has minor symmetry if and only if = , with that we obtain:
C ijkl = ij kl + ( ik jl + il jk )
Note that ij kl has major and minor symmetry, while the tensors ik jl , il jk are not
sym
symmetric. Note also that ( ik jl + il jk ) = 2I ijkl .
Problem 1.89
Let C be a fourth-order tensor, whose components are given by:
C ijkl = ij kl + ( ik jl + il jk )
(1.94)
where , are constant real numbers. Show that C is isotropic.

Solution:
Applying the transformation law for the fourth-order tensor components:
C = a im a jn a kp a lq C mnpq
ijkl
(1.95)
and by replacing the relation C mnpq = mn pq + ( mp nq + mq np ) into the above equation,

we obtain:

Ciudad Real - Spain
Draft
1 TENSORS
(
+ (a
83
C = a im a jn a kp a lq mn pq + mp nq + mq np
ijkl
= a im a jn a kp a lq mn pq
im a jn a kp a lq
)]
mp nq + a im a jn a kp a lq mq np )
= a in a jn a kq a lq + a ip a jq a kp a lq + a iq a jn a kn a lq
= ij kl + ik
jl
+ il jk
(1.96)
= C ijkl
which proves that C is an isotropic tensor, i.e. the C -components do not change for any
transformation basis.
Problem 1.90
Let C be a symmetric isotropic fourth-order tensor which is represented by its components
as follows:
C ijkl = ij kl + ik jl + il jk
C = 1 1 + 2 I
(indicial notation)
where and are scalars, 1 is the second-order unit tensor, I is the symmetric fourthorder unit tensor, i.e. I I sym .
a) Given a symmetric second-order tensor , obtain which is given by the following
algebraic operation = C : . Express the result in indicial and tensorial notation.
b) Show that and have the same eigenvectors, i.e. the same principal directions.
c) If are the eigenvalues (principal values) of , obtain the eigenvalues of .
Solution:
a)
Tensorial notation:
Indicial notation
ij = C ijkl kl
= C :
)]
= ij kl kl + ik jl kl + il jk kl
= 1 1 : + 2 I :
{
{
Tr ( )
= ij kl + ik jl + il jk kl
= (1 1 + 2 I ) :
= ij kk + ij + ji
sym
( )
sym
= ij kk + 2 ij
= Tr ( )1 + 2
= ij kk + 2 ij
where we have considered the symmetry of the tensor = T .

b) and c) Starting from the definition of eigenvalue-eigenvector of a tensor :
n = n
and by substituting the value of obtained previously we can obtain:

Ciudad Real - Spain
Draft
84
(Tr( )1 + 2 ) n = n
Tr ( )1 n + 2 n = n
2 n = n Tr ( )n
Tr ( )n + 2 n = n
2 n = ( Tr ( ) )n
Tr ( )
n
n =
= n
Note that the last equation is the definition of eigenvalue-eigenvector of . With that we
conclude that and have the same eigenvectors (they are coaxial). And the eigenvalues of
can be obtained as follows:
=
Tr ( )
2
(
(
(1
(
(
If we denote by 1) = 1 , (2) = 2 , 3) = 3 and ) = 1 , 2) = 2 , 3) = 3 . The explicit
form of the above relationship is given by:
1
0
0
2
0
1
where it is also true that 0
0
1
= 1 0
0
2
0
3
0
2
0
1 0 0
Tr ( ) 0 1 0
2
0 0 1
3
0
0
1 0 0
1
= Tr ( ) 0 1 0 + 2 0
0 0 1
0
3
0
2
0
0
0
2
0
0
0
Problem 1.91
a) Obtain the inverse of the fourth-order tensor C = 2 I + 1 1 where I I sym is the
symmetric fourth-order unit tensor, 1 is the second-order unit tensor, and > 0 and are
scalars. b) Obtain the determinant of C . In addition, if we consider that =
E
,
(1 + )(1 2)
E
, find the possible values of E and that guarantee that the tensor C is positive
2(1 + )
definite. c) Obtain also the reciprocal of the expression = C : in function of > 0 , ,
where and are symmetric second-order tensors.
Solution:
a) We use the equation obtained in (1.77):
( D + A B ) 1 =
D 1
( + B : D
[(D
: A)
: A ) (B : D 1 )
Denoting by D = I , A = B = 1 , = 2 , = , we obtain:
C 1 = ( 2 I + 1 1) 1 =
1 1
I
(I 1 : 1) (1 : I 1 )
1
2
2 ( 2 + 1 : I : 1)
Remember that it holds that I 1 = I , (I 1 : 1) = I : 1 = 1 . Then we obtain the scalar value of

1 : I 1 : 1 = 1 : I : 1 = 1 : 1 = Tr (1) = 3 . We also express in indicial notation:

Ciudad Real - Spain
Draft
1 TENSORS
sym
1 : I 1 : 1 = 1 : I : 1 = ij I ijkl kl
= ij
85
1
( ik
2
jl
+ il
jk
) kl
1
( ij ik jl kl + ij il jk kl )
2
1
= ( jj + jj ) = 3
2
=
Resulting that:
C 1 = ( 2 I + 1 1) 1 =
I
(1 1)
2
2 ( 2 + 3 )
Let us check whether C : C 1 = I sym I holds or not:

1
C : C 1 = ( 2 I + 1 1) :
2 I 2 ( 2 + 3 ) (1 1)
2
2
2
C : C 1 = (
I:I
I : (1 1) +
(1 1) : I
(1 1) : (1 1)
2
2 ( 2 + 3 )
2
2 ( 2 + 3 )
According to Problem 1.27 it fulfills that I : I = I , I : (1 1) = (1 1) : I = 1 1 , and

(1 1) : (1 1) = 3(1 1) . With that we get:
2
3 2
C : C 1 = I +
2 ( 2 + 3 ) 2 2 ( 2 + 3 ) (1 1) = I
444444 24444444
1
4
3
=0
b) We can directly use the equation (1.37), (see Problem 1.50):
det I sym + A B = 3 + 2 A : B
and by denoting by = 2 , = , A = B = 1 we conclude that:

det (2 I + 1 1) = (2 ) 3 + (2 ) 2 1 : 1 = (2 ) 3 + (2 ) 2 3 = (2 ) 2 (2 + 3 )
The tensor C is definite positive if the eigenvalues are positive numbers, i.e.:
E
>0
2(1 + )
E
E
E
2 + 3 > 0 2
+3
=
>0
2(1 + )
(1 + )(1 2 ) (1 2)
>0 =
Denoting by y1 = (1 + ) 0 , y 2 = (1 2 ) 0 , we can conclude that:
E
E
=
>0
2(1 + ) 2 y1
2 + 3 =
E
E
=
>0
(1 2 ) y 2
E > 0
y1 > 0
E < 0
y < 0
1
E > 0
y 2 > 0
E < 0
y < 0
2
The above conditions must fulfill simultaneously. Then, by means of Figure 1.4 we can
conclude that E > 0 and 1 < < 0.5 .
Ciudad Real - Spain
Draft
86
y ()
y2 = (1 2 ) 0
zone not feasible
y1 = (1 + ) 0
zone not feasible

1
( y 2 > 0 E > 0)
E >0
( y 1 < 0 E < 0)
( y 1 > 0 E > 0)
( y 2 < 0 E < 0)
0 .5
Figure 1.4
c)
= C:
C 1 : = C 1 : C :
C 1 : = I sym : = sym =
= C 1 :
1
1
=
I
1 1 : =
I:
1 1:
2 ( 2 + 3 )
2
2 ( 2 + 3 )
2
1
=
Tr ( )1
2
2 ( 2 + 3 )
(see Figure 1.5). It is interesting to compare this last equation with Problem 1.41.
x3
13
33
23
x1
13
32
22
31
11
x3
21
12
33
= C:
23
32
22
31
x2
11
21
12
x2
x1
= C 1 :
C 1
Figure 1.5
Ciudad Real - Spain
Draft
1 TENSORS
87
Problem 1.92
Let Q e (N) be a second-order tensor, which is known as the elastic acoustic tensor, and is
defined as follows:
Q e (N) = N C e N
where N is the unit vector and C e is the isotropic symmetric fourth-order tensor and given
e
by C e = (1 1) + 2 I , whose components are: C ijkl = ij kl + ( ik jl + il jk ) . Obtain
the components of the elastic acoustic tensor.
Solution:
Using symbolic notation we obtain:
( ) (C

Q e (N) = N C e N = N i e i
= N i C epqrs N j
pqrs e p
ip sj (e q e r )
eq er es
= N p C epqrs N s
(e

) (N e )
er
Then, the components of Q e (N) are:
Q e qr = N p C epqrs N s
= N p pq rs + ( pr qs + ps qr ) N s

= pq rs N p N s + (N p pr qs N s + N p ps qr N s )

= N q N r + (N r N q + N s qr N s )

Note that N is the unit vector, then N s N s = 1 holds. With that we obtain:

Q e qr = qr + ( + )N q N r
tensorial
Q e (N) = 1 + ( + )N N
Problem 1.93
Let Q be a symmetric second-order tensor and given by:
Q (N) = 1 + ( + )N N
where , are scalars, and N is the unit vector.
a) Obtain the eigenvalues of Q (N) and determine the restrictions on and so that there is
inverse of Q (N) , i.e. Q 1 .
b) Taking into account that =
E
E
, =
, determine the possible values of
(1 + )(1 2)
2(1 + )
( E , ) for which Q (N) is a positive definite tensor.
c) Obtain the inverse of Q (N) .
Solution:
r
a) It was shown in Problem 1.50 that, given the vectors a and b it holds that:
r r
r r
det 1 + a b = 3 + 2 a b

Ciudad Real - Spain
Draft
88
The eigenvalues can be determined by means of the characteristic determinant

det (Q 1) = 0 , where i are the eigenvalues of Q . Then:
(
(
)
)
det 1 + ( + )N N 1 = 0
det ( )1 + ( + )N N = 0
Denoting by = ( ) and = ( + ) we conclude that:
det ( )1 + ( + )N N = 0

( ) 3 + ( ) 2 ( + )N N = 0
{
=1
( ) [( ) + ( + )] = 0
2
( ) 2 [( + 2 ) ] = 0
The above characteristic equation has the following solutions:

( ) [( + 2 ) ] = 0
2
solution
1 =
2
( ) = 0
2 =
[( + 2 ) ] = 0 = ( + 2 )
3
In the principal space of Q , the components of Q are:
Qij = 0
( + 2 )
0
0
The inverse of Q exits if the determinant of Q is non-zero:

Q = 2 ( + 2 ) 0
+ 2 0
b) A tensor is definite positive if its eigenvalues are greater than zero, then:
E
= 2(1 + ) > 0
E
E
E (1 )
+ 2 =
+2
=
>0
(1 + )(1 2)
2(1 + ) (2 2 + 1)
1
(1 + ) 0
We check that
1
2
(2 + 1) 0 0.5
Denoting by y1 = (1 + ) 0 , y 2 = (1 ) 0 , y 3 = ( 2 2 + 1) 0 , we can rewrite the

restrictions as follows:

Ciudad Real - Spain
Draft
1 TENSORS
89
E > 0
E
y1 > 0
= 2 y > 0 E < 0
1
y < 0
y2 , y3 > 0
E > 0
y2 , y3 < 0
+ 2 = Ey 2 > 0
y3
y 2 > 0, y 3 < 0
E < 0 y < 0, y > 0
3
2
with which we obtain:

] 1 ; 0.5[ ] 1 ; [
] ; 1[
y ()
y 2 = (1 ) 0
= 1
E<0
zone not feasible
E > 0
E < 0
E >0
y1 = (1 + ) 0
=1
E >0
= 0 .5
y 3 = ( 2 2 + 1) 0
c) The inverse of the Q (N) -components in the principal space of Q (N) are given by:
Qij = 0
( + 2 )
inverse
Qij1
=0
Then, the eigenvalues of Q (N) 1 are Q1 1 = Q21 =
0
1
0
1
( + 2 )
, Q3 1 =
Q 1 =
1
( + 2 )
2
1
. Recall that a tensor
( + 2 )
and its inverse share the same principal space, i.e. they are coaxial tensors. Moreover, we can
express the spectral representation of Q (N) 1 as follows:

Ciudad Real - Spain
Draft
90
Q 1 =
1 ( a )
a N
N( a )
a =1
= Q11N (1) N (1) + Q2 1N ( 2 ) N( 2 ) + Q3 1N (3) N (3)
= Q 1 (N (1) N (1) + N( 2 ) N ( 2 ) ) + Q 1N (3) N(3)

1
Q11 (1 N (3)
=
N(3) ) + Q3 1N (3) N (3)

= Q11 (1 N N) + Q3 1N N
where we have considered that N(3) = N . It is interesting to see Problem 1.76. Then:

Q 1 = Q11 (1 N N) + Q3 1N N
=
=
(1 N N) +
NN
( + 2 )
NN+
NN
( + 2 )
=
=
1
1
( + 2 ) N N
( + 2 ) N N
1
1
1
Note that Q 1 = N C e N N C e N , where C e =

I
evaluate the tensor Qinv = N C
e 1
(1 1) . We
2 ( 2 + 3 )
N :
Qinv = N C 1 N
1
1 1 ( + )
ij kl Nl
(Qinv ) jk = N i C ijkl N l = N i
il jk
2 2 ik jl
2 ( 2 + 3 )
1 1
N i ij kl N l
(N i ik jl N l + N i il jk N l )
(Qinv ) jk =
2 2
2 (2 + 3 )
(Qinv ) jk =
1

N jNk
(N k N j + N l N l jk )
4
2 (2 + 3 )
1
jk +
4 2 (2 + 3 ) N j N k
2 +

1
jk +
=
4 (2 + 3 ) N j N k
(Qinv ) jk =
(Qinv ) jk
Thus:
Qinv =
2 +
1+
4 (2 + 3 ) N N
Note that 0 and (2 + 3 ) 0 , and moreover, these conditions are the same as those to
guarantee that C 1 (see Problem 1.91).

Ciudad Real - Spain
Draft
1 TENSORS
91
1.1.12 Polar Decomposition

Problem 1.94
Let us consider that F has inverse ( det ( F ) 0 ), and that can be decomposed as:
F = Q U = V Q
If U has the eigenvalues a associated with the eigenvectors N( a ) , and V has the eigenvalues
a associated with the eigenvectors n ( a ) , show that:
a = a
Obtain also the relationship between the eigenvectors N( a ) and n ( a ) .
Solution:
Based on the definition of eigenvalue and eigenvector of the tensor U :
U N( a) = a N( a)
(here, the index does not indicate summation)
and using the definition of F we can obtain the following relationship:

QT F = QT Q U = QT V Q
QT F = U = QT V Q
and considering the definition of eigenvalue-eigenvector of U :
U N(a ) = aN( a)
Q T V Q N(a ) = aN( a)
Q Q T V Q N ( a ) = a Q N( a )
123
1
we obtain
V Q N( a ) = a Q N ( a )
V n(a) = a n(a)
where we have assumed that n ( a ) = Q N ( a ) . Furthermore, by comparing the two definitions

of eigenvalue-eigenvector of the tensors U and V , we can verify the they have the same
eigenvalues and different eigenvectors and they are related to each other by the orthogonal
transformation n ( a ) = Q N ( a ) .
1.1.13 Spherical and Deviatoric Tensors

Problem 1.95
Let be a symmetric second-order tensor, and s dev be a deviatoric tensor. Prove that
s:
s
= s . Also show that and dev are coaxial tensors.

Ciudad Real - Spain
Draft
92
Solution: First, we make use of the definition of a deviatoric tensor:

I
1+s
3
= sph + dev = sph + s =
s=
I
1.
3
Afterwards we calculate:
I
1
3 [ ] 1 [I ]
s
=
1
=
which in indicial notation is:

s ij
kl
ij
kl
1 [I ]
1
ij = ik jl kl ij
3 kl
3
Therefore
s ij
s ij
kl
= s ij ik jl kl ij = s ij ik
3
jl
1
1
s ij kl ij = s kl kl s ii
{
3
3
=0
= s kl
s:
s
=s
To show that two tensors are coaxial, we must prove that dev = dev :
dev = ( sph ) = sph =
I
1
3
I
I
1 = 1
3
3
I
= 1 = dev
3
Therefore, we have shown that and dev are coaxial tensors. In other words, they have the
same principal directions (eigenvectors).
1.1.14 Miscellaneous Problems

Problem 1.96
1
Consider that J = [det (b )] 2 = ( III b ) 2 , where b is a symmetric second-order tensor, i.e. b = b T .

Obtain the partial derivatives of J and ln(J ) with respect to b .
Solution:

Ciudad Real - Spain
( III b ) 2
J
=
b
b
1 III
1
1
1
b
= ( III b ) 2
= ( III b ) 2 III b b T
b
2
2
1
1
1
= ( III b ) 2 b 1 = J b 1
2
2
Draft
1 TENSORS
1

ln III b 2
[ln( J )]
=
b
b
93
1 III b 1 1
= b
2 III b b
2
1.1.15 Voigt Notation

Problem 1.97
a) Write the equation = C : in Voigt notation, where C = 1 1 + 2 I is the isotropic
symmetric fourth-order tensor, and the tensors and are structured according to Voigt
notation as follows:
11

22

{} = 33 ;
12
23
13
11

22

{} = 33
2 12
2 23
2 13
b) Write the equation = C 1 : in Voigt notation, where the tensor C 1 (see Problem 1.91)
is given by.
C 1 =
I
11
2
2 ( 2 + 3 )
Solution:
We write the equation = (1 1 + 2 I ) : in indicial notation:
)]
ij = ij kl + 2 ik jl + il jk kl = ij kl + ik jl + il jk kl
2
The second-order unit tensor in Voigt notation is:

1
1

1 0 0
1
ij = 0 1 0 Voigt {} =

0
0 0 1
0

0

Then, the term (1 1)ij = ij kl in Voigt notation becomes:
I ijkl = ij kl
1
1
1
1

1
1
Voigt I = [1 1 1 0 0 0] =
0
0
0
0

0
0

Ciudad Real - Spain
[]
Draft
1 1 0 0 0
1 1 0 0 0
1 1 0 0 0
T
= {}{}
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
94
The symmetric fourth-order unit tensor I ijkl =
I ijkl
I1111
I
2211
I 3311
Voigt [I ] =
I1211
I 2311
I1311
1
ik jl + il jk in Voigt notation is:
2
I1122
I 2222
I1133
I 2233
I1112
I 2212
I1123
I 2223
I 3322
I1222
I 3333
I1233
I 3312
I1212
I 3323
I1223
I 2322
I1322
I 2333
I1333
I 2312
I1312
I 2323
I1323
I1113 1
I 2213 0
I 3313 0
=
I1213 0
I 2313 0

I1313 0

0 0 0 0 0
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
2
0 0 0 1 0
2
0 0 0 0 1
2
With these, we can conclude that C = 1 1 + 2 I in Voigt notation becomes:

1
1
1
[C ] =
0
0
1 1 0 0 0
1
0
1 1 0 0 0
0
1 1 0 0 0
+ 2
0 0 0 0 0
0
0
0 0 0 0 0
0 0 0 0 0
0
0 0 0 0 0 + 2

1 0 0 0 0
+ 2

0 1 0 0 0
+ 2
=
1
0 0 2 0 0 0
0
0
1
0 0 0 2 0 0
0
0

1
0 0 0 0 2 0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
thus
= (1 1 + 2 I ) :
Voigt
0 0 0 11
11 + 2

+ 2
0 0 0 22
22
33
+ 2 0 0 0 33
=
0 0 212
0
0
12 0
23 0
0
0
0 0 2 23
0
0
0 0 2 13
13 0

14444444444 244444444444
4
3
{ } = [C ]{ }
b)
= C 1 :
=
I
1 1 : =
I:
1 1:
2 ( 2 + 3 )
2
2 ( 2 + 3 )
2
Tr ( )1
=
2
2 ( 2 + 3 )
ij =
1
ij
kk ij
2
2 ( 2 + 3 )
Note that:

Ciudad Real - Spain
Draft
1 TENSORS
95
+
1
11
( 11 + 22 + 33 ) 11 =
( 2 + 3 ) 11 2 ( 2 + 3 ) ( 22 + 33 )
2
2 ( 2 + 3 )
1
=
22
( 11 + 22 + 33 ) 22 =
( 2 + 3 ) 22 2 ( 2 + 3 ) ( 11 + 33 )
2
2 ( 2 + 3 )
11 =
22
1
33
( 11 + 22 + 33 ) 33 =
( 2 + 3 ) 33 2 ( 2 + 3 ) ( 11 + 22 )
2
2 ( 2 + 3 )
1
1
1
=
12
( kk ) 12 =
2 12 = 12
{ 2 12
2
2 ( 2 + 3 )
=0
33 =
12
2 23 =
2 13 =
23
13
Restructuring the above in Voigt notation we obtain:
( 2 + 3 )
11
2 ( 2 + 3 )
22
33 2 ( 2 + 3 )
=
2 12
2 23
2 13
0
2 ( 2 + 3 )
2 ( 2 + 3 )
+
2 ( 2 + 3 )
( 2 + 3 )
+
2 ( 2 + 3 )
( 2 + 3 )
{ } = [C ] 1 { }
0
1
0 11
22
0
33

0 12
23
0 13
1.1.16 Tensor-Valued Tensor Function. Tensor Fields.

Problem 1.98
r
Let T ( x , t ) be a symmetric second-order tensor, which is expressed in terms of the position

r
( x ) and time (t ) . Also, bear in mind that the tensor components, along direction x3 , are
equal to zero, i.e. T13 = T23 = T33 = 0 .
r
NOTE: We define T ( x , t ) as a field tensor, i.e. the value of T depends on position and time.
r
r
If the tensor is independent of any one direction at all points ( x ) , e.g. if T ( x , t ) is
independent of the x3 -direction, (see Figure 1.6), the problem becomes a two-dimensional
problem (plane state) so that the problem is greatly simplified.

Ciudad Real - Spain
Draft
96
2D
x2
T
Tij = 11
T12
x2
T12
T22
T22
T22
T12
T12
T12
T11
T11
T11
x1
T12
T22
x3
x1
Figure 1.6: A two-dimensional problem (2D).
a) Obtain T11 , T22 , T12 in the new reference system ( x1 x ) defined in Figure 1.7.
2
b) Obtain the value of so that corresponds to the principal direction of T , and also find
an equation for the principal values of T .
c) Evaluate the values of Tij , (i, j = 1,2) , when T11 = 1 , T22 = 2 , T12 = 4 and = 45 . Also,
obtain the principal values and principal directions.
d) Draw a graph that shows the relationship between and components T11 , T22 and T12 ,
and in which the angle varies from 0 to 360 .
Hint: Use the Voigt Notation, and express the results in terms of 2 .
x2
a11
a ij = a 21
x1
x2
a12
a 22
0 cos sin 0
0 = sin cos 0

1 0
0
1

x1
Figure 1.7: A two-dimensional problem (2D).

Solution:
a) Here we can apply the transformation law in Voigt notation {T } = [M] {T } , where
T11
T
22
T
{T } = 33 ;
T12
T23

T13


Ciudad Real - Spain
Draft
T11
T
22
T
{T } = 33
T12
T23

T13

1 TENSORS
a11 2
2
a 21
2
[M] = a 31
a 21 a11
a a
31 21
a 31 a11
97
a12 2
a 22 2
a13 2
a 23 2
2a11 a12
2a 21 a 22
2a12 a13
2a 22 a 23
a 32 2
a 33 2
2a 31 a 32
2a 32 a 33
a 22 a12
a 32 a 22
a13 a 23
a 33 a 23
a 32 a12
a 33 a13
2a 31 a 33
(a13 a 21 + a11 a 23 )
(a 33 a 21 + a 31 a 23 )
(a 33 a11 + a 31 a13 )
2a11 a13
2a 21 a 23
(a11 a 22 + a12 a 21 ) (a13 a 22 + a12 a 23 )

(a 31 a 22 + a 32 a 21 ) (a 33 a 22 + a 32 a 23 )
(a 31 a12 + a 32 a11 ) (a 33 a12 + a 32 a13 )
For the particular case shown in Figure 1.7, the transformation matrix [M] , after eliminate the
role and column associated with the x3 -direction, becomes:
T11 a11
T = a 2
22 21
T12 a 21 a11

2
a12
a 22
a 22 a12
T11

2a 21 a 22
T22
a11 a 22 + a12 a 21 T12

2a11 a12
(1.97)
T = A T AT
x2
T22
x2
T22
T12
T11
T12
T11
T11
T11
T12
T12
x1
T22
x1
T22
x1
T = AT T A
Figure 1.8: Transformation law for (2D) tensor components.

The transformation matrix, a ij , in the plane, can be evaluated in terms of a single parameter,
:
a11
a ij = a 21
a 31
a12
a 22
a 32
a13 cos
a 23 = sin

a 33 0

sin
cos
0
0
0
(1.98)
By substituting the matrix components a ij given in (1.98) into (1.97) we obtain:

2
T11 cos
T =
2
22 sin
T12 sin cos


Ciudad Real - Spain
sin 2
cos 2
cos sin
Draft
2 cos sin T11
2 sin cos T22

cos 2 sin 2 T12

(1.99)
98
trigonometric identities,
2 cos sin = sin 2 ,
1 cos 2
1 + cos 2
cos 2 sin 2 = cos 2 , sin 2 =
, cos 2 =
, (1.99) becomes:
2
2
Making
use
of
the
following
1 + cos 2
T11
T = 1 cos 2
22
2
T12

sin 2

1 cos 2
sin 2
2
T11
1 + cos 2

sin 2 T22
2
T12

sin 2
cos 2
Explicitly, the above components are given by:
1 + cos 2
1 cos 2
T11 +
T22 + T12 sin 2
T11 =
2
2
1 cos 2
1 + cos 2
T22 T12 sin 2

T11 +
T22 =
2
2
sin 2
sin 2
T12 =
T11 +
T22 + T12 cos 2
Reordering the previous equation, we obtain:
T11 + T22 T11 T22
+
cos 2 + T12 sin 2
T11 =
2
2

T11 + T22 T11 T22
cos 2 T12 sin 2

T22 =
2
2
T T22
T12 = 11
sin 2 + T12 cos 2
(1.100)
b) Recalling that the principal directions are characterized by the lack of any tangential
components, i.e. Tij = 0 if i j , in order to find the principal directions for the plane case, we
let T12 = 0 , hence:

T T22
T T22
T12 = 11
sin 2 = T12 cos 2
sin 2 + T12 cos 2 = 0 11
2
2
2 T12
2 T12
sin 2
=
tg(2 ) =
cos 2 T11 T22
T11 T22
Then, the angle corresponding to the principal direction is:
2 T12
T11 T22
1
2
= arctg
(1.101)
To find the principal values (eigenvalues) we must solve the following characteristic equation:
T11 T
T12
T12
=0
T22 T
2
T 2 T ( T11 + T22 ) + T11 T22 T12 = 0
And by evaluating the quadratic equation we obtain:

Ciudad Real - Spain
Draft
1 TENSORS
T(1, 2 )
[ ( T11 + T22 )]
99
2(1)
T11 + T22
[ (T11 + T22 )]2
[(T11 + T22 )]2
2
4(1) T11 T22 T12
2
4 T11 T22 T12
4
By rearranging the above equation we obtain the principal values for the two-dimensional case
as:
T(1, 2 ) =
T11 + T22
T T22
11
2
2
(1.102)
2
+ T12
c) We directly apply equation (1.100) to evaluate the values of the components Tij , (i, j = 1,2) ,
where T11 = 1 , T22 = 2 , T12 = 4 and = 45 , i.e.:
1 + 2 1 2
T11 = 2 + 2 cos 90 4 sin 90 = 2.5
1 + 2 1 2
cos 90 +4 sin 90 = 5.5

T22 =
2 2
1 2
T12 =
sin 90 4 cos 90 = 0.5
And the angle corresponding to the principal direction is:
2 T12
T11 T22
1
2
= arctg
2 (4)
=
1 2
( = 41.4375 )
The principal values of T ( x , t ) can be evaluated as follows:

T(1, 2 ) =
T11 + T22
T T22
11
2
2
2
2
+ T12
T1 = 5.5311
T2 = 2.5311
d) By referring to equation in (1.100) and by varying from 0 to 360 , we can obtain
different values of T11 , T22 , T12 , which are illustrated in the following graph:

Ciudad Real - Spain
Draft
100
x1
T1
= 41.437
T2
x1
= 131.437
2
Components
T22
T1 = 5.5311
T22
T12
T11
0
0
50
100
-2
-4
200
250
T2 = 2.5311
300
350
T11
x1
45
T12
150
= 86.437
-6
TS max = 4.0311
Problem 1.99
Obtain the principal values (eigenvalues) and the principal directions (eigenvectors) of the
symmetric part of T , whose components in the Cartesian system are given by:
5 1
Tij =
3 4
(i, j = 1,2)
Solution:
The symmetric part of the tensor is given by:
sym
Tij =
5 2
1
Tij + T ji =
2
2 4
The principal values:

5
=0
2 9 + 16 = 0
The solution of the quadratic equation is given by:

(1, 2 ) =
9 (9) 2 4 (1) (16)

2 1
1 T1 = 6.5615
2 T2 = 2.4385
We can draw the Mohr circle (2D) of the tensor T sym :

Ciudad Real - Spain
Draft
1 TENSORS
101
TSsym
sym
sym
(T11 , T12 )
TII = 2.4385
TI = 6.5615
sym
TN
For the plane case, the principal direction can be obtained by means of the equation:
tan(2) =
sym
2 T12
sym
T11
sym
T22
22
=4
54
= 37.982
Problem 1.100
Find the gradient of the function f ( x1 , x 2 ) = cos( x1 ) + exp x1x2 at the point ( x1 = 0, x 2 = 1) .
Solution: By definition, the gradient of a scalar function is given by:
r
x f =
where:
f
= sin( x1 ) + x 2 exp x1x2
x1
f
f
e1 +
e2
x1
x 2
f
= x1 exp x1x2
x 2
r
r
x f ( x1 , x 2 ) = sin( x1 ) + x 2 exp x1x2 e 1 + x1 exp x1x2 e 2 x f (0,1) = [1] e1 + [0] e 2 = 2e1
Problem 1.101
r
Let v and be, respectively, vector and scalar, and twice continuously differentiable. Using
indicial notation, show that:
r
r
r
a) x ( x v ) = 0
r
r
r
b) x ( x ) = x 2
r
r
r
c) x ( ) = x + x
r
r
r
d) x (v ) = x v + x v
r
r
r
e) x ( A B) = x A : B + A ( x B)
( A and B are second-order tensors)
Solution:
a) Considering that
r
r
r
x v = ijk v k , j e i

Ciudad Real - Spain
Draft
(1.103)
102
r
x () =
()
el
xl
(1.104)
then
r
r
r
r
x ( x v ) =
( ijk v k , j e i ) e l =
( ijk v k , j il )
xl
x l
=
( ljk v k , j )
xl
(1.105)
= ljk v k , jl
Note that ljk is an antisymmetric tensor in lj and vk , jl is a symmetric tensor in lj , thus:

ljk v k , jl = 0
(1.106)
b)
r
r
x ( x ) =
=
=
( ,i e i ) e j =
( ,i ij )
x j
x j
, j
x j
x j
= , jj
x j
(1.107)
2
2
=
r
=x
x 2
j
c)
[ xr ( )]i = ( ) ,i = ,i + ,i
r
r
= [ x ]i + [ x ]i
(1.108)
r
d) The result of x (v ) is a scalar which can be expressed as follows:
r
r
x (v ) = (vi ) ,i
= ,i v i + v i , i
r
r
r
r
= x v + x v
r
e) Considering that ( A B) ij = A ik B kj , [ x ( A B)]i = ( A B) ij , j = ( A ik B kj ) , j , thus
r
r
( A ik B kj ) , j = A ik , j B kj + A ik B kj , j = [ x A : B ]i + [A ( x B)]i
Problem 1.102
r
r
r
r
Let a and b be vectors. Show that the following identity x (a + b) = x a + x b holds.
Solution:
r
Observing that a = a j e j , b = b k e k , x = e i
(a j e j + b k e k )
x i
ei
a j
x i

e j ei +
r r
r
, we can express x (a + b) as follows:
x i
r
r
b k
a
b
r
r

ek ei = i + i = x a + x b
x i
x i x i
Working directly with indicial notation we obtain:
r
r r
r
r
r
r
x (a + b) = (a i + b i ), i = a i , i +b i , i = x a + x b

Ciudad Real - Spain
Draft
1 TENSORS
Problem 1.103
103
r
Find the components of ( x a) b .
Solution: Bearing in mind that a = a j e j , b = b k e k , x = e i
( i = 1,2,3 ), the following is

x i
true:
a j
a j
r r (a j e j )
a j
r
e i (b k e k ) =
( x a) b =
x e j e i (b k e k ) = b k ik x e j = b k x e j
x i
i
k
i
Expanding the dummy index k , we obtain:

a j
bk
x k
a j
= b1
x1
+ b2
a j
x 2
+ b3
a j
x 3
Thus,
j =1
b1
a1
a
a
+ b2 1 + b3 1
x1
x 2
x 3
j = 2 b1
a 2
a
a
+ b2 2 + b3 2
x1
x 2
x 3
j = 3 b1
a 3
a
a
+ b 2 3 + b3 3
x1
x 2
x 3
Problem 1.104
Prove that the following relationship is valid:
r
q 1 r r
1 r
r
= x q 2 q xT
T T
T

r r
r
where q( x , t ) is an arbitrary vector field, and T ( x , t ) is a scalar field.
r
x
Solution:
r
q
1
1
qi qi
r
x =
T x T T = T q i ,i 2 q i T,i
T
,i

i
1 r r
1 r
r
= x q 2 q x T (scalar)
T
T
Problem 1.105
Show that:
r
r
r
r
a) rot (a) = x (a) = ( x a) + ( x a)
r r
r r
(1.109)
r
r
r
r
r
r
b) x (a b) = ( x b)a ( x a)b + ( x a) b ( x b) a
(1.110)
2
r
r
r
r
r
c) x ( x a) = x ( x a) x a
(1.111)
2
r
r
r
r
r
d) x ( x ) = x + ( x ) ( x )
(1.112)

Ciudad Real - Spain
Draft
104
Solution:
r
a) The result of the algebraic operation x (a) is a vector, whose components are given
by:
r
[
r
x
r
(a) i
= ijk (a k ) , j
= ijk ( , j a k + a k , j )
= ijk a k , j ijk , j a k
(1.113)
r
r
r
r
= ( x a) i ijk ( x ) j a k
r
r
r
r
r
= ( x a) i ( x a) i
r
r
r
r
r
r
r
r
r
with that we check the identity: rot (a) = x (a) = ( x a) + ( x a) .
r r
r r
The components of the vector product (a b) are given by (a b) k = kij a i b j . Then:
r
[
r
x
r r
(a b) l = lpk ( kij a i b j ) , p
(1.114)
= kij lpk (a i , p b j + a i b j , p )
b) Considering that kij = ijk , the result of ijk lpk = il jp ip jl and by substituting in the
above equation we obtain:
r
[
r
x
r r
(a b) l = kij lpk (a i , p b j + a i b j , p )
= ( il jp ip jl )(a i , p b j + a i b j , p )
(1.115)
= il jp a i , p b j ip jl a i , p b j + il jp a i b j , p ip jl a i b j , p
= al , p b p a p, p b l + al b p, p a p b l , p
Note
that
r r
r
( x b) a l = a p b l , p .
r r
[( a) b] = a
r
x
l, pb p
[(
r
x
a)b]l
r r
= a p, p b l ,
[(
r
x
b)a]l
r r
= al b p, p ,
r
r
c) The components of ( x a) are given by ( x a) i = ijk a k , j . Then:
123
ci
r
r
r
r
r
x ( x a) q = qli c i ,l
= qli ( ijk a k , j ) ,l
(1.116)
= qli ijk a k , jl
Considering that qli ijk = qli jki = qj lk qk lj , the above equation becomes:
r
[
r
x
r
r
r
( x a) q = qli ijk a k , jl = ( qj lk qk lj )a k , jl
(1.117)
= qj lk a k , jl qk lj a k , jl
= a k ,kq a q ,ll
r
r
r
Note that [ x ( x a)]q = a k , kq and x 2 a q = a q ,ll .
d)
r
r
r
r
r
x ( x ) = ( ,i ),i = ,ii + ,i ,i = x + ( x ) ( x )
2
(1.118)
where and are scalar functions.

Ciudad Real - Spain
Draft
1 TENSORS
105
Another interesting identity originating from the above is:

r
r
r
r
r
x ( x ) = x + ( x ) ( x )
2
(1.119)
r
r
r
r
r
x ( x ) = x + ( x ) ( x )
2
Subtracting the two previous identities we obtain:

r
r
r
r
r
r
x ( x ) x ( x ) = x x
2
(1.120)
r
r
r
r
r
x ( x x ) = x x
2
Problem 1.106
r
r
Let be a scalar field, and u be a vector field. a) Show that x ( x v ) = 0 and
r
r
r
r
x ( x ) = 0 .
r
r
r
r
r r
r rr r
r r
r
r
r
r
r
r
r
r
r
b) Show that x x v v = ( x v )( x v ) + x ( x v ) v ( x v ) ( x v ) ;
r
r
r
r
r
r
r
r
r
r
r
r
r
c) Referring = x v , show that x ( x 2 v ) = x 2 ( x v ) = x 2 .
[(
Solution:
r
Regarding that: x v = ijk v k , j e i
r
r
r
r

x x v =
ijk vk , j e i e l = ijk
vk , j il = ijk
vk , j = ijk vk , ji
xl
xl
xi
r
The second derivative of v is symmetrical with ij , i.e. v k , ji = v k ,ij , while ijk is antisymmetric
( )
( )
with ij , i.e., ijk = jik , thus:

ijk v k , ji = ij1v1, ji + ij 2 v 2, ji + ij 3 v3, ji = 0
Note that ij1v1, ji = 0 since the double scalar product between a symmetric and an
antisymmetric tensor is zero.
Likewise, we can show that:
r
r
r
r
r
x ( x ) = ijk , kj e i = 0 i e i = 0
r
b) Denoting by = x v we obtain:
[(
) ]
r
r
r r
r
r
x x v v
r
r r
r
= x ( v )
Observing the equation in (1.110), it holds that:
r
r r
r r
r r
r r
r r
r
r
r
r
r
x ( v ) = ( x v ) ( x )v + ( x ) v ( x v )
r
r
r
r
r
r
Note that x = x ( x v ) = 0 . Then, we can draw the conclusion that:
r
r r
r r
r r
r r
r
r
r
r
x ( v ) = ( x v ) + ( x ) v ( x v )
r
r
r rr r
r r
r
r
r
r
r
r
r
= ( x v )( x v ) + x ( x v ) v ( x v ) ( x v )
c) Observing the equation in (1.111) we obtain:
r
r
r
r
2r
r
r
r
r
r
x v = x ( x v ) x ( x v )
r
r
r
r
r
r
= x ( x v ) x

Ciudad Real - Spain
Draft
106
Applying the curl to the above equation we obtain:
r
r
r
r
r
r
2r
r
r
r
r
r
r
r
x ( x v ) = x [ x ( x v )] x ( x )
144 2444
4r
3
=0
r
r
Referring once again to the equation in (1.111) to express the term x ( x ) :
r
r
r
r
r
r
r
2r
2r
2r
r
r
r
r
r
r
r
r
r
r
r
x ( x v ) = x ( x ) = x ( x ) + x = x x ( x v ) + x
14 244
4
3
=0
r
r
2
r
r
= x ( x v )
Problem 1.107
Show that:
r
r
r
r
a) x (a b) = ( x a) b + a ( x b) rot (a) b + a rot (b)
Solution:
r
(1.121)
r
The expression x (a b) is a scalar which can be expressed as follows:
r r
r
x (a b) = (ijk a j b k ) ,i
= ijk a j ,i b k + ijk b k ,i a j
1 r
3
1 r
2
3
r2
r
( a)k
( b ) j
r
r
r r r r
r
r
= ( x a) b + a ( x b)
Problem 1.108
r
a) Let T be an arbitrary second-order tensor, obtain the symbolic notation of a.1) ( x T ) ,
r
r
r
r
r
r
r
a.2) ( x T )T , a.3) ( x T T ) , and a.4) ( x T T )T . a.5) Considering c a constant vector,
show that:
r
r
r
r r r
T
r
r
r
x ( T c) = ( x T ) c = c x T
r
r
r
r
b) Obtain the symbolic notation of x ( x )T .
r
r
r
u
r
r
c.1) Consider the second-order tensor F = r + 1 , prove that x ( x F )T = 0 , c.2)
x
r
r
r
r
x F T = 0 . c.3) Obtain the explicit expression for x F components.
Solution:
r
r
a.1) ( x T ) =
Tqj

e p Tqj (e q e j ) =
e p e q e j = Tqj , p ipq ei e j = ipq Tqj , p ei e j
x p
x p
a.2) ( x T )T = ipq Tqj , p e j ei = jpq Tqi , p e i e j
r
a.3) ( x T T ) =
T jq

e p T jq (e q e j ) =
e p e q e j = ipq T jq , p e i e j
x p
x p
r
a.4) ( x T T )T = jpq Tiq , p ei e j

Ciudad Real - Spain
Draft
1 TENSORS
107
where we have considered the definition e j e k = ijk ei .

r
a.5) Let us make that a = T c = ( Tqj c j )e q = aq e q , thus:
r
r
r
aq
a

r
r
e p aq e q =
x (T c) = x a =
ipq ei = ipq q e i = ipq aq , p ei
x p
x p
x p
ipq aq , p ei = ipq ( Tqj c j ), p ei = ipq Tqj , p c j ei + ipq Tqj c j , p ei = ipq Tqj , p c j e i

{
=0
where we have considered that c is constant, i.e. c j , p =
c j
x p
= 0 jp .
r
Note that ipq Tqj , p are the components of ( x T )ij (see (a.1)), thus
r
r
r
r
r
r r
r
r
r
r
x ( T c ) = ipq Tqj , p c j ei = ( x T )ij c j ei = ( x T ) c i e i = c ( x T )T i e i
r
r
r
( x T ) c = ipq Tqj , p ei e j c k e k = ipq Tqj , p c k ei jk = ipq Tqj , p c j ei

r
r
b) We have already shown that ( x ) = ipq qj , p e i e j , thus
r
r
qj , p

r
r
e s ipq qj , p e j ei = ipq
e s e j e i = ipq tsj qj , ps et e i
x ( x )T =
xs
xs
= ( iqp )( tjs ) qj , ps et ei = iqp tjs qj , ps et e i = qpi jst qj , ps et ei
Note that:
r
r
qj , p

r
r
e s ipq qj , p ei e j = ipq
e s e i e j = ipq qj , ps tsi et e j
x ( x ) =
xs
xs
= its ipq qj , ps et e j = ( tp sq tq sp ) qj , ps et e j
= ( tp sq qj , ps tq sp qj , ps )e t e j = ( sj , ts tj , ss )e t e j
r
r
r
r
r
r
u
r r u r r
u r r
r
r
r
c.1) Note that x F = x r + 1 = x r + x (1) = x r = x J , where
x
x
x
r
r
r
u
u
r
r
we have denoted by J = r . Taking into account qj = J qj = q = uq , j into x ( x )T
x
x j
we get:
r
r
r
r
x ( x J )T = iqp tjs J qj , ps e t ei = iqp tjsuq , jps et ei
Note that uq , jps = uq , pjs = uq , psj , i.e. it is symmetric in js , and the tensor tjs = tsj is
r
r
antisymmetric in js , so tjsuq , jps = 0tqp , and x ( x F )T = 0ti et ei = 0 .
Taking into account that
it ij is
iqp tjs = qt qj qs = it qj ps + ij qs pt + is pj qt is qj pt qs pj it ps qt ij
pt pj ps
then

Ciudad Real - Spain
Draft
108
iqp tjs Fqj , ps = ( it qj
ps
+ ij qs
pt
+ is pj qt is qj
pt
qs pj it ps qt ij u q , jps
= it qj psu q , jps + ij qs pt uq , jps + is pj qt u q , jps is qj pt u q , jps qs pj it u q , jps ps qt ij u q , jps

= it u j , jss + u s ,its + ut , ppi u j , jti it u s , pps ut ,ipp = 0ti
Note that it u j , jss = it u p , pss = it u p ,ssp = it us , pps , u s ,its = u j ,itj = u j , jti , ut , ppi = ut ,ipp .
r
r
c.2) We express x J T in indicial notation:
r
J qj

r
x J T =
e p J qj (e j e q ) =
e p e j e q = J qj , p ipj ei e q
x p
x p
= ipj J qj , p ei e q = ipj uq , jp ei e q = 0 ip ei e q
Note that uq , jp = uq , pj is symmetric in jp meanwhile ipj = ijp is antisymmetric in jp .

r
r
c.3) We express x J in indicial notation (see (a.1)):
r
r
x J = ipq J qj , p e i e j = ipquq , jp e i e j
Expanding the term ipquq , jp we get:

ipquq, jp = ip1u1, jp
123
4 4
+ ip 2u2, jp
123
4 4
+ ip 3u3, jp
123
4 4
i11u1, j1 + i12u2, j1 + i13u3, j1

+
i 21u1, j 2 + i 22u2, j 2 + i 23u3, j 2

+
i 31u1, j 3
+
+ i 32u2, j 3
+
+ ip 3u3, j 3
thus
u3,12 u2,13 u3, 22 u2, 23 u3,32 u2,33 J 31, 2 J 21,3
r

r J) = u
( x
ij
1,13 u3,11 u1, 23 u3, 21 u1,33 u3,31 = J 11,3 J 31,1
u2,11 u1,12 u2, 21 u1, 22 u2,31 u1,32 J 21,1 J 11, 2
J 32, 2 J 22,3 J 33, 2 J 23,3
J 12,3 J 32,1 J 13,3 J 33,1

J 22,1 J 12, 2 J 23,1 J 13, 2
Note that
J 13, 2 J 12,3
r
T
r
( x J )ij = J 11,3 J 13,1
J 12,1 J 11, 2
= 0ij
J 23, 2 J 22,3 J 33, 2 J 32,3 u1,32 u1, 23 u2,32 u2, 23 u3,32 u3, 23

J 21,3 J 23,1 J 31,3 J 33,1 = u1,13 u1,31 u2,13 u2,31 u3,13 u3,31
J 22,1 J 21, 2 J 32,1 J 31, 2 u1, 21 u1,12 u2, 21 u2,12 u3, 21 u3,12
Note that, if
=
1
(J + J T )
2
r
r
r
r
1rr
1rr
r
r
r
r
x ( x )T = x ( x J )T + x ( x J T )T = 0
1 24
4 3
3 2
2 144244
=0
r
u
where J = r .
x

Ciudad Real - Spain
Draft
=0
1 TENSORS
109
Problem 1.109
r
Let a and v be vectors, show that
r
r r
r
r
r
r
r
( x v ) a = x v ( x v )T
Solution:
r
[r
r
] a
r
r
If we consider ( x v ) i = ijk vk , j , then ( x v ) a s = sip ijk vk , j a p . Note also that the
relationship sip ijk = psi jki = pj sk pk sj holds, then
r
[(
r
x
r r
v ) a s = sip ijk v k , j a p = ( pj
= (v s , p v p , s ) a p =
pk sj )v k , j a p = ( pj
r
r
r
r
r
x v ( x v ) T a
{[
sk
sk v k , j
pk
sj v k , j )a p
r
r
r
r
If we denote by l = x v , then x v ( x v ) T = 2( x v ) skew = 2l skew . Note that the axial
r
r r
r
r
vector associated with the antisymmetric tensor ( x v ) skew = (v x ) skew is the vector
r 1 rr r
= ( x v ) (see Problem 1.37). If we recall the property of an antisymmetric tensor
2
r skew r r r
r
( x v )
a = a , we conclude that
r skew r r r
r
a = a
( x v )
r 1 r
r
r r
1 rr
r
r
x v ( x v ) T a = ( x v ) a
2
2
r
r
r
r
r r
r v ( r v ) T a = ( r v ) a
x
x
x
[
[
]
]
Problem 1.110
r
r r
Let u = u( x ) be a vector field. By means of components of u , a) show that

r
r
r
r
r
r
r
r
2r
2r
r
r
r
r
r
r
r
r
x u = x ( x u) when x ( x u) = 0 , b) show that x u = x ( x u) when
r
r
r
r
x ( x u) = 0 .
Solution:
We have proven in Problem 1.105 that the following is true:
r
r
r
r
2r
r
r
r
r
r
x ( x a) = x ( x a) x a
indicial
ilq qjk ak , jl = a j , ji ai , jj
Then, we can obtain
r
r
r r
r
2r
r
r
r
r
r
r
r
x ( xu) x u = x ( x u) x ( x u)
indicial
ui , jj = u j , ji ilq qjk uk , jl
Then, it is easy to verify that:
r
r
r r
r
2r
r
r
r
r
r
r
r
x ( x u) x u = x ( x u) x ( x u)
14 244
4r 3
a)
r
2r
r
r
r
x u = x ( x u)
=0
Components:
ui , jj = u j , ji
ui ,11 + ui , 22 + ui ,33 = u1,1i + u 2, 2i + u3,3i
u1,11 + u1, 22 + u1,33 = u1,11 + u 2, 21 + u3,31
u 2,11 + u 2, 22 + u2,33 = u1,12 + u2, 22 + u3,32

u + u + u = u + u + u
3, 22
3,33
1,13
2 , 23
3,33
3,11

Ciudad Real - Spain
Draft
u1, 22 + u1,33 = u 2, 21 + u3,31
u 2,11 + u 2,33 = u1,12 + u3,32

u + u
3, 22 = u1,13 + u 2 , 23
3,11
(1.122)
110
Note that in the Cartesian System we have:

r
u = ui e i = u1e1 + u2 e 2 + u3e3
r
r
r
r
u u
u u
u
u
r
( x u) rot (u) = (rot (u) )i ei = 3 2 e1 + 1 3 e 2 + 2 1 e3
x
x
x3
x1
x2
2
3
1
4243
4243
4243
1 4 r 4
1 4 r 4
1 4 r 4
= (rot (u) )3
= (rot (u) )1
= (rot (u) )2
r
r
r
r
r
r
r
r
r (rot (u) )3 (rot (u) )2
(rot (u) )1 (rot (u) )3
(rot (u) )2 (rot(u) )1
r
r
e1 +
e 2 +
e 3
x ( x u) =
x2
x3
x3
x1
x1
x2
r
r
(rot (u) )3 (rot (u) )2 u2 u1 u1 u3
x x
x2
x3
2 1 x2 x3 x3 x1
r
r
r
r
(rot (u) )1 (rot (u) )3 u3 u2 u2 u1
r

r
r
x ( x u) i =
=
x3 x2 x3 x1 x1 x2
x3
x1
r
r
(rot (u) ) (rot (u) )
2
1
u1 u3 u3 u2
x x
x1
x2
1 3 x1 x2 x2 x3
u2,12 u1, 22 u1,33 + u3,13
= u3, 23 u2,33 u2,11 + u1, 21

u u u + u
2, 32
1,31 3,11 3, 22
r
r
r
r
r
r
If we are considering that x ( x u) = 0 then:
u2,12 u1, 22 u1,33 + u3,13 0 u2,12 + u3,13 = u1, 22 + u1,33

r
r
r

r
r
x ( x u) i = u3, 23 u2,33 u2,11 + u1, 21 = 0 u3, 23 + u1, 21 = u2,33 + u2,11
u u u + u 0 u + u
1,31
2 ,32
2, 32 = u3,11 + u3, 22
1,31 3,11 3, 22
which are the same conditions as those presented in equation (1.122).

b)
r
r
r
r
r
2r
r
r
r
r
r
r
r
x ( x u) x u = x ( x u) x ( x u)
14 r 4
2 3
r
r
r
2r
r
r
r
x u = x ( x u)
=0
Components
u1,11 + u1, 22 + u1,33 = (u2,12 u1, 22 u1,33 + u3,13 )
ui , jj = ilq qjk uk , jl
u2,11 + u2, 22 + u2,33 = (u3, 23 u2,33 u2,11 + u1, 21 )

u + u + u = (u u u + u )
3, 33
1,31
3,11
3, 22
2 , 32
3,11 3, 22
r
r
r
r
And if we consider x ( x u) = 0 we get:
r
[ xr ( xr u)]i = u1,1i + u2, 2i + u3,3i = 0i
u1,11 + u2, 21 + u3,31 = 0
u1,12 + u2, 22 + u3,32 = 0

u + u + u = 0
2 , 23
3, 33
1,13
(1.123)
u3,31 + u2, 21 = u1,11
u1,12 + u3,32 = u2, 22

u + u = u
2 , 23
3, 33
1,13
If we replace the above equations into (1.123) we prove that the equality holds.

Ciudad Real - Spain
Draft
1 TENSORS
Problem 1.111
r
111
r r
Let v be a vector field in function of x , i.e. v = v ( x ) , whose components are given by:
v1 = x1 5 x 2 + 2 x3
v 2 = 5 x1 + x 2 3 x3
v = 2 x + 3 x + x
1
2
3
3
r
r
r
a) Obtain the gradient of v ; b) Obtain x v : 1 ; c) Apply the additive decomposition of the
r
r
tensor x v into a symmetric and antisymmetric parts; d) Obtain the axial vector associated
r
r
with the antisymmetric tensor ( x v ) skew .
Solution: a)
r
xv
v1
x1
vi v 2
r
components
r
( x v ) ij =
=
x j x1
v3
x1
r
v
= r
x
v1
x 3 1 5 2
v 2
1 3
= 5
x 3
2 3
1
v 3
x 3
v1
x 2
v 2
x 2
v3
x 2
r
r
b) x v : 1 = Tr ( x v ) = 1 + 1 + 1 = 3
r
r
r
c) x v = ( x v ) sym + ( x v ) skew =
] [
r
r
r
r
1
1
r
r
r
r
( x v ) + ( x v ) T + ( x v ) ( x v ) T
2
2
144 2444
4
3 144 2444
4
3
r
r
=( x v ) sym
r
r
=( x v ) skew
r
r
Then, the components of ( x v ) sym and ( x v ) skew are given, respectively, by:
r sym
r
( x v ) ij
1 0 0
1 vi v j
=
+
= 0 1 0
2 x j xi
0 0 1
r skew
r
( x v ) ij
0 5 2
1 vi v j
0 3
=
= 5
2 x j x i
2 3
0
d) Remember that
r
r
(W) ij ( x v ) skew
0
= W21
W31
ij
W12
0
W32
viskew
,j
1 v
= 2
2 x1
1 v
3
2 x1
W13 0
W23 = W12

0 W13

v1
x 2
v1
x3
W12
0
W23
1 v1 v 2
2 x 2 x1
0
1 v 3 v 2
2 x 2 x3
W13 0
W23 = w3

0 w2

1 v1 v 3
2 x 3 x1
1 v 2 v 3
2 x3 x 2
(1.124)
w3
0
w1
w2
w1
r
where w1 , w2 , w3 are the components of the axial vector w associated with the
r
r
antisymmetric tensor W ( x v ) skew , then, to the proposed problem we have:
0
w
3
w2

Ciudad Real - Spain
w3
0
w1
w2 0 5 2
w1 = 5
0 3

0
0 2 3
Draft
w1 = 3
w2 = 2
w = 5
3
112
The axial vector, in the Cartesian basis, is w = 3e1 + 2e 2 + 5e 3 .

1 r r
(a x ) is the
2
Alternative solution d) Resorting to Problem 1.37 where we have shown that

r
axial vector associated with the antisymmetric tensor ( x a ) skew . Then, the axial vector
r skew
r
r
r
r
( x v ) skew = (v ) ( x )
associated with the antisymmetric tensor
is the vector
r 1 rr r
w = x v .
2
e1
r 1
w=
2 x1
v1
=
e2
x 2
v2
e3
v
1 v
= 3 2
x
x3 2 2 x3
v3
v
v
v
e 1 3 1 e 2 + 2 1
x
x
1 x 2
1 x 3
e 3
(3 (3) )e1 ((2) (2) )e 2 + (5 (5))e 3 = 3e1 + 2e 2 + 5e 3

2
Problem 1.112
r
r
r
r
Let l = x v be a second-order tensor. Considering that D = ( x v ) sym and W = ( x v ) skew ,
show that
r
r
r
r
W D + D W = 2(D W) skew = ( x v x v ) skew = ( l l ) skew
Solution:
In Problem 1.35 we have shown that: given an arbitrary second-order tensor
l
skew
sym
+l
sym
skew
= 2( l
skew
it fulfills that
sym skew
Then, W D + D W = 2(D W ) skew holds. Taking into account the definition of symmetry and
Antisymmetry, i.e. D =
1
l +l
2
] , W = 1 [l l ] , we can conclude that:

2
T
W D + D W = 2(D W) skew =
[
[
skew
2
(l + l T ) (l l T )
4
1
T
T
T
T
l l + l l l l l l
2
skew
1
1
T
T
l l
= 1442444 + l l l
4l l 3 2
2
=
=0
[
[
skew
l T ]
skew
skew
1
= l l ( l l )T
2
skew
1
= 2( l l ) skew
2
r rr
r
= ( l l ) skew = ( x v x v ) skew
OBS.: Note that the resulting tensor

(l l

Ciudad Real - Spain
l T
l )T
Draft
l is symmetric, since:
= l l
l .
1 TENSORS
113
Problem 1.113
Consider the scalar J = F det(F ) and an arbitrary second-order tensor given by
l
r &
& dF
r
= x v = F F 1 , where F
dt
represents the time derivative of F . Show that the
following is true:
r
d(J ) &
r
J = J ( x v )
dt
(1.125)
Solution:
dA
dA
= A Tr
A 1 holds, where A = A () is an
d
d
arbitrary second-order tensor and a scalar. Making A = F and = t , we can obtain:
In Problem 1.87 we have shown that
dF
dt
( )
dJ
dF
&
= F Tr
F 1 = J Tr F F 1 = J Tr l = J Tr l
dt
dt
sym
) = J Tr(
r
xv
r
r
) = J ( x v )
In Problem 1.46 we have shown that given a second-order tensor F the relationship
F tpq = rjk Frt F jp Fkq holds, and if we take the time derivative of it we get:
DF
D
&
&
&
tpq =
( rjk Frt F jp Fkq ) = rjk Frt F jp Fkq + rjk Frt F jp Fkq + rjk Frt F jp Fkq
Dt
Dt
(1.126)
&
&
According to the problem statement we have l = F F 1 F = l F , with that the
&
&
&
following relations Frt = l rs Fst , F jp = l js Fsp and Fkq = l ks Fsq hold, and the equation in (1.126)
can be rewritten as follows:
DF
&
&
&
tpq = rjk Frt F jp Fkq + rjk Frt F jp Fkq + rjk Frt F jp Fkq
Dt
= rjk l rs Fst F jp Fkq + rjk Frt l js Fsp Fkq + rjk Frt F jp l ks Fsq
We multiply both sides of the above equation by ut v p w q we get:

DF
Dt
tpq u t v p w q = rjk l rs Fst F jp Fkq u t v p w q + rjk Frt l js Fsp Fkq u t v p w q + rjk Frt F jp l ks Fsq u t v p w q
= rjk ( l rs Fst u t )( F jp v p )( Fkq w q ) + rjk ( Frt u t )( l js Fsp v p )( Fkq w q )
+ rjk ( Frt u t )( F jp v p )( l ks Fsq w q )
= rjk ( l rs a s )(b j )(c k ) + rjk (a r )( l js b s )(c k ) + rjk (a r )(b j )( l ks c s )
where we have denoted by a s = Fst ut , b j = F jp v p , c s = Fsq w q . The above equation in tensorial

notation becomes:
] [
r
r r r r
r r r
r
DF r r r
u ( v w ) = ( l a) (b c ) + a ( l b) c + a b ( l c )
Dt
r r r
= Tr ( l ) a (b c )
where we have used the property of trace (see Problem 1.48). The above equation can also be
written as follows:
Ciudad Real - Spain
Draft
114
r r r
DF r r r
u ( v w ) = Tr ( l ) a (b c )
Dt
r
r
r
= Tr ( l ){( F u) [( F v ) ( F w ) ]}
r r r
= Tr ( l ) F u ( v w )
where we have used the property of determinant (see Problem 1.49), with that we conclude
that
DF
= Tr ( l ) F .
Dt
Problem 1.114
r
Let us consider a vector field represented by the unit vector field b( x ) , (see Figure 1.9).
r
r
r
Obtain the second-order projection tensor P such that p = P u holds, where u is an
r
r
arbitrary vector and p is orthogonal to the field defined by b( x ) .
r
b( x )
Figure 1.9: Vector field.

Solution:
The proposed problem can be represented by the following figure:
r
u
r r
r
u b = p = P u
r r
a = u // b
r
b( x )
And, by considering the vector summation we obtain u = a + p . In addition, the vector a can
r
r
r r
be obtained by means of the projection of u onto the direction b : a = a b = (u b) b , note
r

also that a = (u b) b = u (b b) . With that we can obtain:

Ciudad Real - Spain
Draft
1 TENSORS
115
r r r
p =ua
r r r r
= u (u b) b = u (u b) b
r r
= 1 u (b b) u
r
= 1 (b b) u
r
= P u
p i = ui ai

= u i (u k b k )b i

=u u b b
k
ik
= ( ik

b k b i )u k
= Pik u k
Thus, we conclude that the projection second-order tensor is given by:

P =1 b b
The same result could have been obtained by means of vector product. We draw the proposed
problem from another perspective for better visualization:
r
ub
r
b( x )
r
u
r
b (u b)
r
Taking into account that a (b a) = [(a a)1 a a] b , (see Problem 1.17), we can obtain
r

b (u b) = (b b)1 b b
r r
r
r r

] u = [1 b b] u = p .
Then we can present a vector as follows:
r r r
r
r
u = u // b + ub = (b b) u + 1 (b b) u
r
r
r
r
where u// b = (b b) u is the vector parallel to b -direction and ub = 1 (b b) u is the
perpendicular one.
Problem 1.115
r r
Given a vector field v ( x ) , show that the following relationship holds:

r
r r 1 r
r
r
r
r
x v v = x (v 2 ) v ( x v )
2
r
r r
where v is the module of v , so v 2 = v v .
Solution:
r
r
1 r 2
1 r r r
1
1
r
x (v ) i = [ x (v v )]i = (v k v k ) ,i = (v k ,i v k + v k v k ,i ) = v k v k ,i = (v x v ) i .
2
2
2
2
r
r
At one point of the vector field v , we consider a plane normal to v and recalling that the
r
projection of a second-order tensor onto a direction ( v ) is a vector which does not necessary
r
r r
r
have the same direction as ( v ), with that we represent the following vectors x v v and
r
r
r
v xv :
Note that

Ciudad Real - Spain
Draft
116
r r
r
xv v
r
r
r
( x v )
r
v
r r
cv
r
r
r
r
c ( x v )
r
r
r
v xv
r
r r
r
r
c = v ( x v )
Note that, by means of summation of vectors we can obtain:
r r r r
r
r
r
xv v + c = v xv
r r
r
r r
r
r
c = v xv xv v
r r
r r
r
r
r
c = v xv v xv T
r r
r
r
r
r
r
r
r
c = v ( x v x v T ) = v 2( x v ) skew
r
r
If we consider that w is the axial vector associated with the antisymmetric tensor ( x v ) skew , it
r
r r r
r
r
r r
r
r
fulfills that: ( x v ) skew v = w v v ( x v ) skew = v w . In addition, the relationship
r
r
r
r
r
rot (v ) x v = 2 w holds. Then,
r
r r
r
r
r r
r
r
r
c = v 2( x v ) skew = v 2 w = v ( x v )
(1.127)
r r r r
r
r
r
( x v ) v + c = v ( x v )
r r r
r r
r
r
( x v ) v = v ( x v ) c
r
r r 1 r
r
r
r
r
( x v ) v = x (v 2 ) v ( x v )
2
r
r
r
r
r
r
It is interesting to note that: when ( x v ) is a symmetric tensor, i.e. ( x v ) = ( x v ) sym , the
r
r r
r
r r
r r r
r
r
r
r
r
following is fulfilled ( x v ) skew = 0 , c = 0 , ( x v ) = 0 , and ( x v ) v = v ( x v ) has the
r
same direction as v .
r r
r
r
r
r
r
r
r
r
r
When ( x v ) = ( x v ) skew we have that c = v 2( x v ) skew = 2v ( x v ) , (see equation (1.127)).
r
r
r
r r
r
r
r
r
r
With that, v x v = x v v holds, and the vector v is perpendicular to the vector ( x v ) ,
(see Figure 1.10).

Ciudad Real - Spain
Draft
1 TENSORS
r
r
r
r
( x v ) = ( x v ) skew
117
r r
r
xv v
r r
cv
r
r
r
r
c ( x v )
r
v
r
r
r
( x v )
r rr
v xv
r
r r
r
r
r
r
r
c = v ( x v ) = 2v ( x v )
Figure 1.10
r r
r
r
r
r
r
r
r
r
r
r
r
r
x v v = (( x v ) sym + ( x v ) skew ) v = ( x v ) sym v + ( x v ) skew v
r
r
r
r
r
r
r
r
r
r
r
r
= ( x v ) sym v + ( x v ) skew v + (( x v ) skew v ( x v ) skew v )
r
r
r
r
r
r
r
r
r
= (( x v ) sym v ( x v ) skew v ) + 2( x v ) skew v
r
r
r
r
r
r
r
1
r
r
r
r
r
= ( x v + ( x v ) T ) ( x v ( x v ) T ) v + 2( x v ) skew v
2
r
r
r
r
1
r
r
= (2( x v ) T ) v + 2( x v ) skew v
2
r
r
r
r
r
r
= v ( x v ) + 2( x v ) skew v
r
r
r
1 r
r
= x (v 2 ) v ( x v )
2
r
r skew T
r
r
r
r
r
r
r
r
r
r
r
r
Remember that ( x v ) = ( x v ) skew , thus 2( x v ) skew v = v 2( x v ) skew = v ( x v )
Problem 1.116
r r
Let u( x ) be a stationary vector field. a) Obtain the components of the differential du . b)

r r
Now, consider that u( x ) represents a displacement field, and is independent of x3 . With
these conditions, graphically illustrate the displacement field in the differential area element
dx1 dx 2 .

Ciudad Real - Spain
Draft
118
Solution: According to the differential and gradient definitions, it holds that:

r r
u( x )
r r r
r r r
du u( x + dx ) u( x )
r
r r
r
du = x u dx
x2
r
x
r
dx
r r
r
u( x + dx )
r
r
x + dx
x1
Thus, the components are defined as:
du i =
u i
dx j
x j
x3
u1
du1 x1
du = u 2
2 x
du 3 1
u 3
x1
u1
x 2
u 2
x 2
u 3
x 2
u1
x3 dx
1
u 2
dx 2
x3
dx3
u 3
x3
or:
u1
u
u
dx1 + 1 dx 2 + 1 dx3
du1 =
x1
x 2
x3
u 2
u
u
dx1 + 2 dx 2 + 2 dx3
du 2 =
x3
x1
x 2
u
u
u
du 3 = 3 dx1 + 3 dx 2 + 3 dx 3
x1
x 2
x3
with
du1 = u1 ( x1 + dx1 , x 2 + dx 2 , x3 + dx3 ) u1 ( x1 , x 2 , x3 )
du 2 = u 2 ( x1 + dx1 , x 2 + dx 2 , x3 + dx3 ) u 2 ( x1 , x 2 , x3 )
du = u ( x + dx , x + dx , x + dx ) u ( x , x , x )
3
1
1
2
2
3
3
3
1
2
3
3
As the field is independent of x3 , the displacement field in the differential area element is
defined as:
u1
u1
du1 = u1 ( x1 + dx1 , x 2 + dx 2 ) u1 ( x1 , x 2 ) = x dx1 + x dx 2
2
1
u 2
u 2
du = u ( x + dx , x + dx ) u ( x , x ) =
dx1 +
dx 2
2
1
1
2
2
2
1
2
2
x1
x 2
or:
u1
u1
u1 ( x1 + dx1 , x 2 + dx 2 ) = u1 ( x1 , x 2 ) + x dx1 + x dx 2
2
1
u ( x + dx , x + dx ) = u ( x , x ) + u 2 dx + u 2 dx
2
1
1
2
2
2
1
2
2 1
x1
x 2

Ciudad Real - Spain
Draft
1 TENSORS
119
Note that the above equation is equivalent to the Taylor series expansion taking into account
only up to linear terms. The representation of the displacement field in the differential area
element is shown in Figure 1.11.
u2 +
u 2
dx 2
x 2
u2 +
( x1 + dx1 , x 2 + dx 2 )
( x1 , x 2 + dx 2 )
u1 +
u 2
u
dx1 + 2 dx 2
x1
x 2
u1
dx 2
x 2
u1 +
r
du
dx 2
u1
u
dx1 + 1 dx 2
x1
x 2
u2 +
(u 2 )
( x1 + dx1 , x 2 )
( x1 , x 2 )
x2
u 2
dx1
x1
u1 +
(u1 )
u1
dx1
x1
dx1
x1
144444444444444444424444444444444444443
=
644444444444444444474444444444444444448
x 2 ,u 2
u2 +
u1
dx2
x2
u 2
dx2
x2
dx 2
O
u2
O
u1
u1 +
dx 2
A
O
dx1
u 2
dx1
x1
dx1
u1
dx1
x1
x1 ,u1
Figure 1.11: Displacement field in the differential area element.

Ciudad Real - Spain
Draft
120
Problem 1.117
Given a second-order tensor field T ( x ) . Show that: if there is no source of the field T ( x ) it
r
r
fulfills that the divergence of T ( x ) is equal to zero, i.e. x T = 0 . For the demonstration,
consider the tensor field in a differential volume element dV = dx1 dx 2 dx 3 in the Cartesian
system.
Solution:
Let us set the tensor field T ( x ) in the differential volume element. For this purpose, we start
r
from the definition of the differential of T ( x ) which is defined by means of gradient as
follows:
r
r
r
dT T ( x + dx ) T ( x )
r
r
r
r
r
r
r
r
r
r
r
T ( x + dx ) T ( x ) = x T dx T ( x + dx ) = T ( x ) + x T dx
r
dT = x T dx
The above equation in indicial notation becomes:
r
r
r
Tij ( x + dx ) = Tij ( x ) + Tij ,k dx k
r
= Tij ( x ) + Tij ,1 dx1 + Tij , 2 dx 2 + Tij ,3 dx3
Tij
Tij
r Tij
= Tij ( x ) +
dx1 +
dx 2 +
dx3
x1
x 2
x 3
r
r
The representation of the field components Tij ( x + dx ) can be appreciated in Figure 1.12.
r
Note that on the face normal to x1 + dx1 act the components Ti1 ( x ) +
Ti1
dx1 , since
x1
according our convention, the first index indicate the direction in which points out and the
second index indicates the normal plane.
r
Once established the field Tij ( x + dx ) in the differential volume element, we apply the total
r
r
balance of the field components Tij ( x + dx ) according to the directions x1 , x 2 , x3 .
r
Total balance of Tij ( x + dx ) in dV according to x1 -direction is equal to zero (there is no

source):
T
T
T
T11 + 11 dx1 dx 2 dx3 + T13 + 13 dx 3 dx1 dx 2 + T12 + 12 dx 2 dx1 dx3 T11 dx 2 dx3
x1
x3
x 2
T13 dx1 dx 2 T12 dx1 dx3 = 0
By simplifying the above equation we obtain:

T
T11
T
dx1 dx 2 dx3 + 13 dx3 dx1 dx 2 + 12 dx 2 dx1 dx3 = 0
x1
x3
x 2
T11 T12 T13

=0
+
+
x3
x 2
x1
Similarly, according to the directions x 2 and x3 we will obtain, respectively:

T21 T22 T23
+
+
=0
x1
x 2
x3

Ciudad Real - Spain
and
Draft
T31 T32 T33

+
+
=0
x1
x 2
x3
1 TENSORS
121
x3
Rear face
T11
T33 +
Rear face
T23 +
T13 +
T21
T33
dx3
x3
T13
dx3
x3
T23
dx3
x3
T32 +
T12
T22
T31 +
T31
dx1
x1
T32
T12 +
T21 +
T
T11 + 11 dx1
x1
T21
dx1
x1
T31
dx 3
T32
dx2
x2
T12
dx2
x2
T22 +
T22
dx2
x2
x2
dx1
T13
T23
x1
Rear face
T33
dx 2
Figure 1.12: Tensor field components in the differential volume element.

Then, we have the following set of equations that must be met simultaneously:
T11 T12 T13
+
+
=0
x1
x 2
x3
T21 T22 T23
+
+
=0
x1
x 2
x3
T31 T32 T33

+
+
=0
x1
x 2
x3
T11,1 + T12, 2 + T13,3 = 0
T21,1 + T22, 2 + T23,3 = 0
T31,1 + T32, 2 + T33,3 = 0
T1 j , j = 0
T2 j , j = 0
T3 j , j = 0
Tij , j = 0 i
Thus demonstrating that in the absence of source, the divergence is zero:

Tij , j = 0 i
r
( x T ) i = 0 i
tensorial

r
r
x T =0
NOTE 1: If we have a tensor field, the order of the source is a minor order of the tensor, e.g.
the source of a vector field is represented by a scalar field.
NOTE 2: If the divergence of a tensor field is positive we have a source, on the contrary if
the divergence is negative we have a sink.

Ciudad Real - Spain
Draft
122
Problem 1.118
Show that:
r r
r r
[( xr T ) u] a = [ xr ( T a )] u
(1.128)
r
r r r
r
where T = T (x ) is a second-order tensor field, u = u( x ) is a vector field, and a an arbitrary
r
vector (independent of ( x ) ).
Solution:
r r
r
Note that the term [( x T ) u] a is a vector, which in indicial notation becomes:
r r
r r
r
r
{[( x T ) u] a}i = [( x T ) u]ik (a ) k
r
= ( x T ) ikp u p a k = Tik , p u p a k = Tik , p u p a k
(1.129)
Note also that:

r
r
r
gradient
r
( T a )i = Tik ak
[ x ( T a )]ij = ( T a ) i , j = ( Tik ak ), j
r
r
[ x ( T a )]ij = ( Tik ak ), j = Tik , j ak + Tik ak , j = Tik , j ak
{
=0
r
r
where we have considered that a is independent of (x ) . With that, we can conclude that:
r r
r
r
r
{[ x ( T a )] u}i = [ x ( T a )]ij u j = Tik , j a k u j = Tik , p u p a k
(1.130)
If we compare (1.129) with (1.130) we show (1.128).

Problem 1.119
r
Show that if the magnitude of a vector, = (t ) , is constant with time, this implies that is
r
d
at any time t .
orthogonal to
dt
Solution:
r
We start from the definition of the magnitude of a vector, where = holds, thus:
r r
r
r
r
( ) = d ( ) = d () + d () = 2 d () = 0
r r
r
r
d
dt
dt
dt
dt
dt
r
r d
dt
NOTE: A particular case of this problem is the circular motion in which:

r
v
r
x = constant
r
r dx
r
x
r dx
dt
v=
dt
r
x

Ciudad Real - Spain
Draft
1 TENSORS
123
1.1.17 Theorems Involving Integrals

Problem 1.120
Check the divergence theorem (Gauss theorem) for the vector field F whose Cartesian
2
components are given by Fi = xi + ( x3 x 3 ) i 3 . Consider the boundary defined by the
2
cylinder x12 + x 2 1 , 0 x3 1 .
Solution:
The divergence theorem states that:
r
r
r
x F dV = F ndS
where n is the normal to the surface and points outwards.
x3
2
x12 + x 2 1
S (2)
n ( 2)
r
r
r
r =1
h =1
S (1)
n (1)
r
x
x2
n ( 3)
x1
S ( 3)
Figure 1.13
r
r
Calculation of x F dV :
V
r
2
r
x F = Fi ,i = x i + ( x3 x3 ) i 3
,i
2
2
= xi ,i + ( x3 x3 ) ,i i 3 = ii + ( x3 x3 ) ,3
= 3 + ( 2 x3 1) = 2 x3 + 2
Thus
r
r
x F dV = (2 x3 + 2) dV =
x3 =1
x =(02 x
A
+ 2)dx3 dA = 3 dA = 3(r 2 ) = 3
2
where A is the area defined by the circle x12 + x 2 1 .
Calculation of
F ndS

Ciudad Real - Spain
Draft
124
We decompose the boundary in three areas, namely: S (1) , S ( 2) , S (3) , (see Figure 1.13), then
r
F ndS = F n
(1)
S ( 1)
F n
dS (1) +
( 2)
dS ( 2 ) +
S ( 2)
F n
( 3)
dS (3)
S ( 3)
r
2
2
2
The components of F are: F1 = x1 + ( x3 x3 ) 13 = x1 , F2 = x 2 , F3 = x3 + ( x3 x 3 ) 33 = x3 .
r
r
2
The representation of F in the Cartesian basis is given by: F = x1e1 + x 2 e 2 + x3 e 3 . The
normal for each surface are defined as follows:

r
n (1) // r
n (1) =
x12
2
x2
( x1 e1 + x 2 e 2 ) ; n ( 2 ) = e 3 ; n (3) = e 3
On the surface S (1) it holds that:

r
F n
(1)
(x e
dS (1) =
S ( 1)
+ x2 e 2 + x3 e 3 )
S ( 1)
2
x12 + x 2
x12
( 1)
2
x2
dS (1) =
1dS
1
x12
(1)
2
x2
( x1 e1 + x 2 e 2 )dS (1)
= 2rh = 2
(1 )
where we have considered the cylinder area ( 2rh = 2 ).

On the surface S ( 2) it holds that x3 = 1 :
F n ( 2) dS ( 2 ) =
(x e
1
S (2)
+ x 2 e 2 + 1e 3 ) (e 3 ) dS ( 2 ) =
S (2)
1dS
(2)
= r 2 =
S (2)
where we have considered the circle area ( r 2 = ).

On the surface S (3) , it holds that x3 = 0 :
F n (3) dS (3) =
S ( 3)
(x e
1
+ x 2 e 2 + 0e 3 ) (e 3 )dS (3) =
S ( 3)
with that: F ndS =

S
0dS
( 3)
=0
S (3)
F n (1) dS (1) +
S (1 )
F n ( 2 ) dS ( 2 ) +
S (2)
F n (3) dS (3) = 3
S ( 3)
Then, proving the divergence theorem: x F dV = F ndS = 3 .

V
Problem 1.121
Let be a domain bounded by as shown in Figure 1.14. Further consider that m is a
second-order tensor field and is a scalar field. Show that the following relationship holds:
[m :
r
r
x ( x
r
r
r
)]d = [( x ) m] nd [( x m) x ]d
[m
, ij ] d = ( , i m ij )n j d [m ij , j , i ] d

Ciudad Real - Spain
ij
Draft
1 TENSORS
125
x2
x1
Figure 1.14
Solution: We could directly apply the definition of integration by parts to demonstrate the
above relationship. But, here we will start with the definition of the divergence theorem. That
r
is, given a tensor field v , it is true that:
r
x
v d = v n d indicial v j , j d = v j n j d
r
Observing that the tensor v can be represented by the result of the algebraic operation
r
r
v = x m and the equivalent in indicial notation is v j = , i m ij , and by substituting it in
the above equation we obtain:
j, j
d = v j n j d
[,
[,
ij
[,
m ij
,j
d = , i m ij n j d
m ij + , i m ij , j d = , i m ij n j d
ij
m ij d = , i m ij n j d
[ ,
m ij , j d
The above equation in tensorial notation becomes:
[m :
r
r
x ( x
r
r
r
)]d = [( x ) m] nd [ x ( x m)]d
NOTE: Consider now the domain defined by the volume V , which is bounded by the
r
surface S with the outward unit normal to the surface n . If N is a vector field and T is a
scalar field, it is also true that:
N T,
i
ij
dV = N i T , i n j dS N i , j T , i dV
r
r
r
r
r
r
r
r
N x ( x T )dV = ( x T N ) ndS x T x N dV
where we have directly applied the definition of integration by parts.

Ciudad Real - Spain
Draft
126
Problem 1.122
r
r
Let b be a vector field, which is defined as b = x v . Show that:
b n
i
d S = , i b i dV
where = ( x ) represents a scalar field.

r
r
Solution 1: The Cartesian components of b = x v are b i = ijk v k , j and by substituting them
in the above surface integral we obtain:
b n
i
dS = ijk v k , j n i dS
Applying the divergence theorem we obtain:
b n
i
dS = ijk v k , j n i dS
= ( ijk v k , j ), i dV
= ( ijk , i v k , j + ijk v k , ji ) dV
V
= (, i ijk v k , j + ijk v k , ji ) dV = , i b i dV
123
4 4
123
4 4
V
bi
Solution 2:
b n
i
dS = (b i ), i dV = (, i b i + b i , i ) dV
note that b i = ijk v k , j b i ,i = ijk v k , ji = ijk v k ,ij = 0
b n
i
dS = , i b i dV = , i ijk v k , j dV
Problem 1.123
Let V be a volume domain which is delimited by surface S . a) Show that:
r

( x n + n x) dS = 2V 1
S
where n is the outward unit vector to surface S . b) Show also that:
r
x
ik , k
) x dV = ( n) x
S
dS dV
V
x j dV = ik n k x j dS ij dV
S
and

Ciudad Real - Spain
Draft
1 TENSORS
127
) dV = x ( n)
x (
r
x
dV = xi jk n k dS ji dV
jk , k
dS T dV
where is an arbitrary second-order tensor field.

r
dS = ndS
x2
dS
B
r
x
x1
x3
Solution:
a) Considering only the first term of the integrand, we can obtain:
r
( x n) dS = ( x 1 n) dS = ( x 1) n dS
S
By applying the divergence theorem we obtain:

r
( x n) dS = ( x 1) n dS =
S
r
x
( x 1) dV
We will continue the development in indicial notation:
xn
i
dS = x i jk n k dS = ( jk x i ) ,k dV
= ( jk ,k xi + jk x i ,k ) dV
V
Taking into account that jk ,k = 0 j , xi ,k = ik , we conclude that:
xn
i
dS = ji dV = ji dV = jiV
( x n) dS = V 1
V
T
= V1

Ciudad Real - Spain
Draft
128
Similarly, we conclude that (n x ) dS = V 1 . With that the following is true:

S

( x n + n x) dS = 2V 1
S
b) Note that it holds

( x j ik ) ,k = x j ,k ik + x j ik ,k
{
= jk
x j ik ,k = ( x j ik ) ,k ij
r
r
r
r
( x ) x = x ( x )
with that we can obtain:
r
x
r
x
) x dV = ( x ) n dS dV
S
ik , k
dV = x j ik n k dS ij dV
S
= ( ik n k ) x j dS ij dV
= ( n) x dS dV
dV = ( x j ik ) ,k dV ij dV
ik , k
r
) x dV = x ( x ) dV dV
where we have applied the divergence theorem to the first integral on the right side of
equation.
Taking into account that
r
r
[( xr ) x ]T = [ xr ( x ) ]T
r
r T
r
r
x ( x ) = [ x ( x )] T
In indicial notation
xi jk ,k = ( xi jk ) ,k ji
we can obtain:
r
x (
r
x
x (
r
) dV = [ x ( x )]T
dV T dV
r
x
) dV = ( x ) n dS T
S
dV

Ciudad Real - Spain
jk , k
jk , k
= x ( n) dS T dV
dV = ( xi jk )n k dS ji dV
S
= xi ( jk n k ) dS ji dV
S
Draft
dV = ( xi jk ) , k dV ji dV
1 TENSORS
129
Problem 1.124
Let be a scalar field which is given by:
GM
a
= r
where G and M are scalars and constants, and a is the magnitude of the vector a 0 . a)
r r
Obtain the gradient of . b) Obtain the gradient of for the particular case when a = x and
r
draw the field x in the Cartesian space.
Solution:
GM
,i = r
a
x ,i
( xr ),i
r
= GM 1 ( a ) ,i
r
a 2
,i
(1.131)
Note that:
1
1
r r 1
r
1 r r 2 r r
1 r r 2
2
( a ) ,i = ( a a ) = (a a ) ( a a ) ,i = ( a a ) ( a k a k ) ,i
2
,i 2
r r
1 r r 2
1
(a a ) ( a k ,i a k + a k a k ,i ) = (a a ) 2 ( a k ,i a k ) = r ( a k , i a k )
a
2
r
r
1 r
r
r
x ( a ) = r (a x a )
a
(1.132)
Then, the equation (1.131) becomes:

1 r
1 1
GM
,i = GM r 2 ( a ),i = GM r 2 r ( ak ,i ak ) = r 3 (ak ,i ak )
a
a a
x ,i
a
GM r r r
= r 3 (a x a ) i
a
r
( x ),i
r
(1.133)
r
r
a
Moreover, considering that the unit vector according to the direction a is given by a = r ,
a
we can obtain:
r
r
r
r
r
( x ),i = GM (a x a ) i = GM (a xr a ) i
r3
r2
a
r r
b) For the particular case when a = x we have:
(1.134)
r
r
1
1
1
2
2
( x ) ,i = r ( x k ,i x k ) = r ( ki x k ) = r ( xi ) where r = x = x12 + x 2 + x3
x
x
x
or in tensorial notation:
r
r
1 r
1 r
1 r
r
r
x ( x ) = r ( x x x ) = r ( x 1) = r ( x ) = x
x
x
x
whereupon

Ciudad Real - Spain
Draft
130
GM
,i = r
x
x i
( xr )i
r
= GM 1
r
x2
,i
r
r
( x ) = GM ( x )
i
,i
r3
(1.135)
GM
r
r
x = x r
x
GM r GM
=
r x= r 2 x
x3
x
(1.136)
r
r
Note that the vector field x is radial, i.e. it is normal to the spheres defined by x and
decreases with x = r 2 . The equation (1.136) can also be written as follows:

GM
=
r
GM GM
= 2 r=
r r
r
(r )
r = (r )r
r =
r
(1.137)
x3
Spheres
r
x
x =1
r
x
r
b
x1
x2
r
x
r
x
Figure 1.15
GM
x
NOTE: The function = r
represents the gravitational potential which has the
r
following property b = x , (see Figure 1.15), where G = 6.67384 10 11
m3
kg s 2
is the
gravitational constant, M is the total mass of the planet. We check the units:
[] = GM =
r
m 3 kg kg m m N m J
( Unit of energy per unit mass)
= 2
=
=
kg
kg
kg s 2 m
s kg
(specific energy)
r
Nm
[b]= [ ] = = mJkg = m kg = skgm = sm (Unitunit force per uniton) )
of
mass
r
x
kg
(
of accelerati
r
x

Ciudad Real - Spain
Draft
1 TENSORS
131
r
r
r
It is interesting to check that x b = x [ x ] = 0 , (see Problem 1.106).
We can obtain b on the Earth surface by means of

r
GM
r
b = x = r 2 x
x
where the total mass of Earth is M 5.98 10 24 kg and the approximate radius is
R 6.37 10 6 m , with that we obtain
r
GM
GM
b = r 2 x = 2 x 9.82 x
R
x
r
and its module is denoted by g = b 9.82
m
.
s2
Adopting that the system x has its origin at the center of mass of the body M , and invoking
r
r
the Newtons second law ( F = ma ), we can obtain the force that act in a body ( m ) due to the
r
r
gravitational field b = x :
r
r
r
GMm
F = ma = mb = r 2 x
x
(1.138)
r
We can express the above equation in a generic system ( x ) as shown below:
x2
x1
r
x
r
F (Mm)
x3
r
x (M )
x2
r
F ( mM )
r
x (m)
r
r r
x ( M ) + x = x ( m)
r r
r
x = x (m) x (M )
x1
x3
Then, for the system x the force is given by:

r
F ( mM ) =
GMm
r ( m) r ( M )
x x

Ciudad Real - Spain
r
r
( x ( m) x ( M ) )
r
r
x ( m) x ( M )
Newtons law of universal gravitation
Draft
(1.139)
132
where we use the nomenclature F (mM ) to indicate the force in m due to the influence of M .
Note also that in M we have the same force in direction and magnitude, but of opposite
r
sense to F (mM ) .
Problem 1.125
Consider that =
r
1
2
2
where r = x = x12 + x 2 + x3 :
r
a) Show that:
r r
2 2 2
r
r
x x ( x 0) 2 2 + 2 + 2 = 0 Laplace equation
(1.140)
x1
x 2
x3
r r
r
for r 0 . We use the nomenclature x ( x 0) to indicate that the origin is not included.
b) Given a closed surface S containing the origin, show that:
( ) ndS = 4
r
x
(1.141)
where n is the outward unit vector to surface.
Solution:
It was obtained in Problem 1.124 that
GM
r
r
x = x r
x
GM r GM
=
r x= r 2 x
x3
x
(1.142)
Denoting by GM = 1 we obtain:
1 1 r 1
r
r
x = x r = r 3 x = r 2 x
x x
x
(1.143)
1 r
1
r
( x ) i = r 3 x = r 3 xi
x
x
i
(1.144)
Calculating the divergence of the previous relationship we obtain:

r
x
r
( x ) = ,ii
x
1
x i ,i
i
= r 3 = r 3 xi r 3
x
x
x
,i
,i
3 r
x i ,i
= r 3 x i r 4 ( x ) ,i
x
1
x
(1.145)
r
In Problem 1.124 it was shown that x ( x ) = r ( x ) , in addition, note that xi ,i = ii = 3 ,

Ciudad Real - Spain
Draft
1 TENSORS
r
( x ) =
r
x
133
3 r
3
x i r 4 ( x ) ,i
r3
x
3 x
3
= r 3 xi r 4 ri
x
x
x
3 xi xi
3
= r 3 + r 5
x
x
r 2
3x
3
= r 3 + r 5 =0
x
x
(1.146)
c) We adopt an arbitrary sphere of radius r , whose surface area is 4r 2 . Then:
1
1

x ndS = r 2 x ndS = r 2
2
x S
x
S x
1
1
= 2 ( Area) = 2 (4r 2 ) = 4
r
r
( ) ndS = r
r
x
dS
(1.147)

Note that x n = 1 since for the sphere it holds that x // n .
It is interesting to note that by means of the divergence theorem it fulfills that:
[ ]dV = ( ) ndS
r
x
r
x
r
x
,ii dV
= ,i ni dS
(1.148)
r
r
We have demonstrated that x x ( x 0) = 0 , but that only apply to x 0 (the origin is
not included). That is, taking into account the result in (1.147), the result in (1.148) has
r r
consistency if at the point x = 0 there is a sink and equal to ( 4 ). With that, it is very
intuitive to conclude that any closed surface that does not contain the origin the following
r
holds x ndS = 0 .
Problem 1.126
a) Show that:
( ) ndS = 4GM (r )
S
(1.149)
GM
is the gravitational potential, and M (r ) is the total mass contained into
r
the sphere whose radius is r , and S -surface represents the sphere boundary.
where =
b) Consider a sphere of radius r = a which represents a planet. Obtain the total mass of the
planet in function of the mass density = (r ) .
c) Obtain the gravitational potential for r < a and r a . In this section, consider that the
mass density is constant in the planet = 0
Solution:
a) In Problem 1.125 we showed that:

Ciudad Real - Spain
Draft
134
( ) ndS = r ndS = 4

1
(1.150)
By multiply both sides of the equation by GM (r ) we obtain:

1
GM (r ) ndS = 4GM (r )
r
S
GM (r )

ndS = 4GM (r )
r
(1.151)
[] ndS = 4GM (r )
S
b)
Spherical planet
r=a
(r )
The total mass is obtained as follows:
M = (r )dV
(1.152)
Note that V = 4 r 3 dV = 4 3r 2 dr = 4r 2 dr . Then:

3
3
M = (r )dV =
V
r =a
(r )4r
dr
(1.153)
r =0
c) Remember that in Problem 1.124 (see equation (1.137)) we have obtained that
r
GM GM GM (r )
= b =
r = (r )r
= 2 r =
r =
r r
r
r r
(1.154)
By using the equation in (1.151) we get:

Ciudad Real - Spain
Draft
1 TENSORS
135
[] ndS = 4GM (r )
S

b ndS = (r ) r n dS = (r ) dS = (r )(4r 2 ) = 4GM (r )
123
=1
(1.155)
(r )r = GM (r )
GM (r )
(r ) =
r2
2
where M (r ) = V 0 = 4 r 3 0 . Then:
3
GM (r ) 4G 0
r
=
3
r2
d(r ) 4G 0
r
=
dr
3
4G 0
rdr
d ( r ) =
3
(r ) =
(1.156)
By integrating the above equation we obtain:

4G 0
rdr =
3
4G 0 r 2
(r ) =
+C
3
2
2G 0 2
(1) (r ) =
r +C
3
d =
(1.157)
where we have denoted (1) ( r ) = (r ) for r < a . For values r a the gravitational potential is
given by
GM 4Ga 3 0
=
= (2)
r
3r
for
(1.158)
ra
where M is the total mass of the planet whose value is M = V 0 = 4 a 3 0 . Note that the
3
potential has to be continuous in r = a , (see Parker (2003)), thus:
(1) (r = a) = ( 2) (r = a)
4Ga 3 0
2G 0 2
a +C =
3a
3
2Ga 0 2Ga 3 0 4 3 2GM 3 3MG
C =
=
=
=
a
a
34
a
4
2a
(1.159)
With that the equation (1.157) becomes
(1) (r ) =
2G 0 2
2G 0 2 3MG MG 2 3MG MG r 2
3
r +C =
r
= 3r
= 2 2
2a
3
3
2a
2a
2
2a
2a
(1.160)
We summarize the gravitational potential as follows:
(r ) =
(r ) =

Ciudad Real - Spain
MG r 2
3

for
2
2
2
2a 2a
MG
for
ra
r
Draft
r<a
(1.161)
136
(r )
Planet surface
a
MG
a
inflection point
3MG
2a
Figure 1.16: Gravitational potential vs. radius.
Figure 1.17: Gravitational potential (Ref. Wikipedia: Gravitational potential).
Problem 1.127
a) Show that the orbit of a planet takes place on a plane. b) Prove the Keplers laws of
planetary motion:
b.1) First Law: The orbit of a planet is an ellipse, with the Sun at one of the foci of the ellipse;
b.2) Second Law: The vector position from the Sun to the planet describes one area at a
constant rate;

Ciudad Real - Spain
Draft
1 TENSORS
137
b.3) Third Law: If T (orbital period) represents the time required for the planet to perform a
full elliptical orbit, whose major axis of the ellipse is 2a , the relationship T 2 = a 3 holds,
where is a constant.
Reminder: Expressions related to the ellipse:
x2
r
x
b
f2
f1
x1
Equation of the ellipse: x = r =

Eccentricity: e =
a 2 b2
a2
p
1 + e cos
0 < e < 1 , where a 2 =
p2
holds.
(1 e 2 ) 2
Area enclosed by an ellipse: A = ab

Solution:
M - Mass of Sun
m - Mass of the planet
r
x
x= r
x
r
r dx
v=
,
dt
r
a // x
x2
x3
r
x
Sun
r r r
c = xv
r
dx r
=v
dt
r
r
r d 2 x dv
a= 2 =
dt
dt
r r
F // a
t=0
r
h
x1
Figure 1.18: Orbit of the planet.

Ciudad Real - Spain
Draft
138
a) To show that the orbit takes place on a plane, we must prove that the vector ( c ) normal to
r
r
the plane which is defined by the vectors x and v does not change with time, i.e. c is
constant.
We recall the equation (1.138) of Problem 1.127:
r
r
r
r
GMm
GM
F = ma = mb = r 2 x
; a= r 2 x
(1.162)
x
x
r r r
Next, we obtain the rate of change of c = x v :
r
r r d r r r r r r
dc d r r
d r
= ( x v ) = ( x ) v + x (v ) = v 2v + 1 a = 0
1 3 x2
3
r
r
dt dt
dt
dt
=0
=0
r r r
Thus we have shown that the vector c = x v does not change with time, which implies that
the orbit takes place on a plane.

b.1) First Law
Since the planets orbit is performed on a plane, we take x1 x 2 as the plane of the orbit, then
r
the vector c has the same direction as x3 , (see Figure 1.18).
r
We express c in term of x :
r
r
d( x )
r dx d r
r dx
= ( x x) =
v=
x+ x
dt dt
dt
dt
and
r
d( x )
r r r
r
r dx
c = x v = ( x x)
x+ x
dt
dt
r d( x )
r 2
dx

= x
x2 x
13 + x x dt
r
dt
=0
r 2
dx
= x x
dt
GM
Taking into account that a = r

r
r r r
direction as v , i.e. (a c ) // v :
r r
x , we calculate the vector a c which has the same
r r GM r 2
dx
dx
dx
dx

a c = r 2 x x x
= GM x x
= GM ( x ) x ( x x )
x

dt
dt
dt
dt
dx
= GM
dt
r r r
r r r r r r
where we have used the property a b c = (a c )b (a b)c , (see Problem 1.17). Note
dx
dx
2

also that it holds that x = 0 since x , and x x = x = 1 . Considering that GM is a

dt
dt
constant, the following is true:
r r
dx d (GM x )
a c = GM
=
dt
dt
Since the vector c does not change with time, the following is true:

Ciudad Real - Spain
Draft
1 TENSORS
139
r r
r
r r dv r d ( v c )
ac =
c =
dt
dt
Thus
r r
d (v c ) d (GM x )
=
dt
dt
Integrating over time the above equation we obtain:
r
r r
v c = GM x + h
where h is constant vector of integration and is not dependent of time. Note that h is located
r r
on the plane x1 x 2 , since (v c ) and x are also on the plane x1 x 2 , (see Figure 1.8).
We calculate:
r
r
h x = h x cos = h cos
r
where we have denoted by h = h . Then:

r
c2 = c
r r
r r r r r r
= c c = ( x v ) c = (v c ) x
r
r
r
r r
r
r
= GM x + h ( x x ) = x GM x x + x h x = x GM + x h cos
r
= x (GM + h cos )
2
= r (GM + h cos )
where we have considered that r = x . Then, we can obtain the following equation of the
ellipse:
c2
p
c
GM
r=
=
=
(GM + h cos ) (GM + h cos ) 1 + e cos
GM
2
where we have considered that:

p=
c2
GM
and
e=
h
GM
(1.163)
b.2) Second Law

r 1 r
A = x S
2
S 0
r 1 r
r
dA = x ds
2
x2
S
A
r
x
x1

Ciudad Real - Spain
Draft
140
The rate of change of dA becomes:
r
r
r
D(dA) 1 D( x ds )
=
=
2
Dt
Dt
r
r
1 D( x )
=
ds +
2 14 4
Dt2 3
r
=0
1
2
1
2
r
r
r 1 r D ( ds )
D( x )
ds + x
2
Dt
Dt
r r
xv
1r
c (constant)
2
and its magnitude:

r
D (dA) D(dA) 1 r 1
=
= c = c
Dt
Dt
2
2
NOTE: As a consequence of second law it follows that if the areas of two sectors are equal,
the time required to perform their paths are equal, that is, according to Figure 1.19 as the areas
of the sectors OCD and EFO are equal the times to perform C D and E F are equal.
As result, when the planet is closer to the Sun its velocity is greater than when it is far.
sector EFO
sector OCD
E
D
A
C
Figure 1.19: Orbit of the planet.

b.3) Third Law
If T is the total time for a complete orbit (orbital period), we can obtain:
T
A=
D (dA)
1
1
dt =
c dt = cT
2
2
Dt
0
Taking into account the area enclosed by the ellipse: A = ab , we conclude that
1
cT = ab ,
2
thus:
T=
2ab
c
T2 =
4 2 a 2 b 2
c2
(1.164)
Considering the equation of the eccentricity, we can obtain:

e=
a 2 b2
a2
b2 = a 2 a 2e2

Ciudad Real - Spain
Draft
b 2 = a 2 (1 e 2 )
1 TENSORS
and taking into account a 2 =
141
p2
p
a=
(1 e 2 )a = p into the above equation,
2 2
2
(1 e )
(1 e )
we can obtain:
b 2 = a 2 (1 e 2 )
b 2 = ap
p=
b2
a
Whereby the equation (1.164) can be rewritten as follows:

T2 =
4 2 a 2 b 2 4 2 a 2 ab 2 4 2 a 3 p 4 2 3
=
=
=
a = a3
GM
c2
c 2a
c2
where we have considered that

Ciudad Real - Spain
(1.165)
p
1
=
, (see equation (1.163)).
2
GM
c
Draft
142

Ciudad Real - Spain
Draft
2 Continuum Kinematics
2.1 Solved Problems
2.1.1
Description of the Motion, Material Time Derivative,

Velocity, Acceleration
Problem 2.1
A continuum is defined by a square with sides b , subjected to rigid body motion which is
defined by rotating the continuum counterclockwise by an angle of 30 to the origin. Find
the equations of motion. Also obtain the new position of particle D .
r
Hint: Consider the systems x and X to be superimposed.

X 2 , x2
x2
30
b
A = A
x1
D
B
30
X 1 , x1
Solution: We apply the rigid body motion equations x = c + Q X = Q X , to c = 0 . The

components of Q are the same as the components of the transformation matrix from the
r
r
x -system to the x -system, i.e.:
cos sin 0
Q ij = sin cos 0
0
0
1
So, the continuum particles are governed by the equations of motion:
142
x1 cos 30 sin 30 0 X 1

x 2 = sin 30 cos 30 0 X 2
x 0
0
1 X 3
3

A particle which initially was at point D ( X 1 = 0 , X 2 = b , X 3 = 0 ) moves into the following

position:
x1D cos 30 sin 30 0 0 b sin 30
D

x 2 = sin 30 cos 30 0 b = b cos 30
x D 0
0
1 0
0
Problem 2.2
A continuum medium motion, in the material description, is given by:
x1 = exp t X 1 exp t X 2
t
t
x 2 = exp X 1 + exp X 2
x = X
3
3
(2.1)
for t > 0 . Find velocity, acceleration in material and spatial descriptions.

Solution:
Velocity:
V1 = exp t X 1 + exp t X 2
r r
r r
Dx ( X , t ) componentes
V ( X , t) =
V2 = exp t X 1 exp t X 2
Dt
V = 0
3
(2.2)
Acceleration:
A1 = exp t X 1 exp t X 2
t
t
A2 = exp X 1 + exp X 2
A = 0
3
(2.3)
To find the velocity and acceleration components in the spatial description we substitute
the equations of motion:
Eulerian velocity (spatial description)
v1 = x2
v2 = x1
v = 0
3
(2.4)
Eulerian acceleration (spatial description)

a1 = x1 = v 2
a 2 = x 2 = v1
a = 0
3

Ciudad Real - Spain
Draft
(2.5)
2 CONTINUUM KINEMATICS
143
Problem 2.3
The velocity field of a fluid is given by:
r
v = x1e1 + x2 e 2 + x3e 3
(2.6)
r
T ( x , t ) = 3 x 2 + x3 t
(2.7)
and the temperature field is:

Find the rate of change of temperature.
Solution:
The rate of change of any property is given by the material time derivative:
r
r
T T
T
T
DT T ( x , t ) T ( x , t )
=
+
+
vj =
x v1 + x v 2 + x v 3
t
x j
t 1
Dt
2
3
(2.8)
DT
= x3 + (0 x1 + 3 x2 + tx3 )
Dt
= x3 + (3x2 + tx3 )
(2.9)
Problem 2.4
Given the following motion:
xi = X i + 0.2tX 2 1i
(2.10)
and the temperature field (steady):

r
2
T ( x ) = 2 x1 + x 2
(2.11)
a) Find the temperature field in material description;

b) Find the rate of change of temperature for one particle that in the reference
configuration was at the position (0,1,0) .
Solution:
According to the equations of motion we have:
x1 = X 1 + 0.2tX 2 11 = X 1 + 0.2tX 2
x 2 = X 2 + 0.2tX 2 12 = X 2
x3 = X 3 + 0.2tX 2 13 = X 3
Then:
r
r 2
r r
T ( x ( X , t )) = 2 x1 ( X , t ) + x 2 ( X , t )
= 2( X 1 + 0.2tX 2 ) + ( X 2 )
r
= 2 X 1 + ( X 2 + 0.4t )X 2 = T ( X , t )
b) The material time derivative of temperature is given by:
r
DT ( X , t ) & r
T ( X , t ) = 0 .4 X 2
Dt
For the particle ( X 1 = 0; X 2 = 1; X 3 = 0) we have:

Ciudad Real - Spain
Draft
144
&
T (( X 1 = 0; X 2 = 1; X 3 = 0), t ) = 0.4 X 2 = 0.4
Problem 2.5
r r
r r
Find the velocity field V ( X , t ) in the material description and the acceleration field A( X , t )
r r
of the particle at time t in function of the rate of change of displacement U ( X , t )
Solution:
r r
r
D r r
&
V ( X , t) =
U ( X , t) = U
Dt
r r
r
D r r
&
A( X , t ) =
V ( X , t) = V =
Dt
r
D2 r r
&&
= 2 U ( X , t) = U
Dt
r &&
& r
A =V = U
(2.12)
(2.13)
(2.14)
Problem 2.6
Consider the following equations of motion in the Lagrangian description:
r
2
x1 ( X , t ) = X 2 t 2 + X 1
x1 1 t
r

Matrix form
x 2 ( X , t ) = X 3 t + X 2 x 2 = 0 1
r
x 0 0
3
x3 ( X , t ) = X 3
0 X 1

t X 2
1 X 3

(2.15)
Is the motion above possible? If so, find the displacement, velocity and acceleration fields
in Lagrangian and Eulerian descriptions. Consider a particle P that at time t = 0 was at the
point defined by the triple equation X 1 = 2, X 2 = 1, X 3 = 3 . Find the velocity of P at time
t = 1s and t = 2 s .
Solution:
Motion is possible if J 0 , thus

x1
X 1
x 2
=
X 1
x3
X 1
x1
X 2
x 2
X 2
x3
X 2
x1
X 3 1 t 2
xi
x 2
J=
=0 1
X j
X 3
0 0
x3
X 3
r
The displacement vector field is given by the definition u =
motion (2.82) we obtain:

u1
u 2
u 3
r
r
(Xr , t ) = x ( Xr, t ) X = [X t + X ] X
(X , t ) = x ( X , t ) X = [X t + X ] X
r
r
(X , t ) = x ( X , t ) X = [ X ] X = 0
1
0
t =1 0
1
r r
x X . Using the equations of
= X 2t 2
= X 3t
(2.16)
which are the components of the displacement vector in the Lagrangian description. Here,
velocity and acceleration can be evaluated as follows:

Ciudad Real - Spain
Draft
( )
r
r
d u1 X , t
d
=
X 2t 2 = 2 X 2t
V1 v1 ( X , t ) =
dt
dt
r
r
du 2 X , t
d
= ( X 3t ) = X 3
V 2 v 2 ( X , t ) =
dt
dt
r
r
du 3 X , t
d
= (X 2t ) = 0
V3 v 3 ( X , t ) =
dt
dt
( )
145
r
dV1
A1 a1 ( X , t ) = dt = 2 X 2
r
dV 2
=0
A2 a 2 ( X , t ) =
dt
r
dV
A3 a 3 ( X , t ) = 3 = 0
dt
( )
(2.17)
The inverse form of (2.15) provides us the inverse equations of motion (Eulerian
description):
X 1 1 t 2

1
X 2 = 0
X 0
0
3
r
t 3 x1 X 1 ( x , t ) = x1 t 2 x 2 + t 3 x 3
r

t x 2 X 2 ( x , t ) = x 2 tx 3
r
1 x3 X 3 ( x, t ) = x3

(2.18)
Then, the displacement, velocity and acceleration fields in Eulerian description can be
evaluated by substituting equation (2.18) into the equations (2.16) and (2.17), i.e.:
r
r
r
(Xr ( x, t ), t ) = X ( x, t )t = ( x tx )t = u ( x, t )
r
r
r
(X ( x, t ), t ) = X ( x, t )t = x t = u ( x, t )
r r
r
(X ( x, t ), t ) = u ( x, t ) = 0
r r
r
r
V (X ( x , t ), t ) = 2 X ( x , t )t = 2( x tx )t = v ( x , t )
r r
r
r
V (X ( x , t ), t ) = X ( x , t ) = x = v ( x , t )
r r
r
V (X ( x , t ), t ) = v ( x , t ) = 0
r r
r
r
A (X ( x , t ), t ) = 2 X ( x , t ) = 2( x tx ) = a ( x , t )
r r
A (X ( x , t ), t ) = a ( x , t ) = 0
r r
r
A (X ( x , t ), t ) = a ( x , t ) = 0
u1
u 2
u 3
(2.19)
(2.20)
(2.21)
Taking into account the Lagrangian description of velocity given in (2.17), the velocity of
particle P ( X 1 = 2, X 2 = 1, X 3 = 3 ) at time t = 1s is given by:
r
r
r
v1 ( X , t ) = 2 X 2 t = 2 m / s ; v 2 ( X , t ) = X 3 = 3m / s ; v 3 ( X , t ) = 0
We can also observe that at time t = 1s the particle P occupies the position:
x1 = X 2 t 2 + X 1 = 3 ;
x 2 = X 3t + X 2 = 4 ;
x3 = X 3 = 3
So, the velocity of the particle P can also be evaluated by (2.20) as:
r
v1 ( x , t ) = 2( x 2 tx 3 )t = 2( 4 + 1 3) 1 = 2m / s
r
v 2 ( x , t ) = x 3 = 3m / s
r
v ( x , t ) = 0
3
Note that, the velocities obtained via the Lagrangian or Eulerian description are the same,
since velocity is an intrinsic property of the particle.
We can also provide the velocity of the particle P at time t = 2 s :
( )
( )
( )
r
V1 v1 X , t = 2 X 2 t = 2 2 1 = 4m / s
r
V 2 v 2 X , t = X 3 = 3m / s
r
V3 v 3 X , t = 0

Ciudad Real - Spain
Draft
146
At time t = 2 s the new position of P is:

r
x1 ( X , t ) = X 2 t 2 + X 1 = 6
r
x2 ( X , t ) = X 3t + X 2 = 7
r
x3 ( X , t ) = X 3 = 3
r r
As we can verify the Lagrangian description of motion x ( X , t ) describes the trajectory of
P.
Trajectory of particle P
r
viP ( x , t = 1s) = [2;3;0]
r
Vi P ( X P , t = 1s) = [2;3;0]
t0
X iP = [2;1;3]
t = 1s
r
Vi ( X P , t = 2s ) = [4;3;0]
xiP = [3;4;3]
P
xiP = [6;7;3]
t = 2s
r
viP ( x , t = 2s ) = [4;3;0]
NOTE: Note that, the Eulerian velocity can not be obtained by means of
r r
DX ( x , t ) r r r
= 0 v ( x , t ) . We can verify this by means of the proposed problem:
Dt
r
r
r
r
r
DX i ( x , t ) X i ( x , t ) X i ( x , t )
X i ( x , t )
X i ( x , t )
r
r
r
v1 ( x , t ) +
v 2 ( x, t ) +
v3 ( x , t )
=
+
Dt
t
x 2
x 3
x1
thus:
r
r
r
r
r
r
r
X 1 ( x , t )
X 1 ( x , t )
DX 1 ( x , t ) X 1 ( x , t ) X 1 ( x , t )
r
=
+
v1 ( x , t ) +
v 2 ( x, t ) +
v3 ( x , t )
t
x 2
x 3
Dt
x1
) [
= 2tx 2 + 3t 2 x3 + 1 2( x 2 tx3 )t t 2 x3 + t 3 0 = 0
r
r
r
r
r
X 2 ( x, t )
X 2 ( x , t )
DX 2 ( x , t ) X 2 ( x, t ) X 2 ( x , t )
r
r
r
+
v 3 ( x, t )
=
v 2 ( x, t ) +
v1 ( x , t ) +
x3
x 2
t
Dt
x1
= ( x3 ) + [0 2( x 2 tx 3 )t + 1 x3 t 0] = 0
r
r
r
r
r
X 3 ( x , t )
X 3 ( x , t )
DX 3 ( x , t ) X 3 ( x , t ) X 3 ( x, t )
r
r
r
=
+
v1 ( x , t ) +
v 2 ( x, t ) +
v3 ( x, t )
t
x 2
x3
Dt
x1
= (0 ) + [0 2( x 2 tx 3 )t + 0 x3 + 1 0] = 0
r r r
Remind that u = x X , then:

Ciudad Real - Spain
Draft
147
r r
r r
r r
r r
Dx ( X , t ) D r r
Du( X , t ) r r
&
=
u( X , t ) X ( x , t ) =
u( X , t )
v ( X , t) =
Dt
Dt
Dt
Also, it fulfills that:

r r
r r
r r
r r
r r
Du( x , t ) u( x , t ) u( x , t ) r r
&
v ( x , t ) = u( x , t )
=
+
r v ( x, t )
Dt
t
x
Problem 2.7
The velocity field of the continuum, in Eulerian description, is given by:
x1
1+ t
v1 =
; v2 =
2 x2
1+ t
; v3 =
3 x3
1+ t
(2.22)
r
a) Obtain the relationship between material and spatial coordinates xi = xi ( X , t ) ;

b) Obtain the acceleration components by means of the spatial motion description.
c) Obtain the acceleration components by means of the Lagrangian motion.
Solution:
a) Considering that vi =
dxi
dt
v1 =
1
dx1
x
dx
dt
= 1 1 =
dt 1 + t
x1 1 + t
1 + t dt Lnx
dx1 =
(2.23)
= Ln(1 + t ) + Ln(C1 )
(2.24)
x1 = C1 (1 + t )
The initial condition t = 0 x1 = X 1 imply that C1 = X 1

x1 = X 1 (1 + t )
v2 =
1
dx 2 =
(2.25)
dx 2 2 x 2
dx
2dt
=
2 =
dt 1 + t
x2 1 + t
1 + t dt Lnx
(2.26)
= 2Ln(1 + t ) + LnC 2
x2 = C 2 (1 + t )
(2.27)
for t = 0 x 2 = X 2 C 2 = X 2
x2 = X 2 (1 + t ) 2
v3 =
1
dx3 =
dx3 3 x3
dx
3dt
=
3 =
dt 1 + t
x3 1 + t
1 + t dt Lnx
Ciudad Real - Spain
(2.29)
= 3Ln(1 + t ) + LnC 3
x3 = C3 (1 + t )
(2.28)
Draft
(2.30)
148
and t = 0 x3 = X 3 C3 = X 3
x3 = X 3 (1 + t ) 3
(2.31)
x1 = X 1 (1 + t )
2
x2 = X 2 (1 + t )
3
x3 = X 3 (1 + t )
(2.32)
The equations of motion are:
r r
b) Knowing v ( x , t ) in the spatial description (Eulerian), we can apply the material

time derivative to obtain:
r r
r r
r r
r r
Dv ( x , t ) r r
v ( x , t )
+ v ( x , t ) v ( x , t )
= a ( x, t ) =
Dt
t
(2.33)
vi
+ (v i , k )v k
t
v
a i = i + (vi ,1 v1 + vi , 2 v 2 + vi ,3 v3 )
t
(2.34)
ai =
thus,
1
x
+ 1
+ 0 + 0 = 0
(1 + t )
1 + t 1 + t
2x2
2x 2
2 x2
a2 =
+ 0 + 2
+ 0 =
2
2
1+ t 1+ t
(1 + t )
(1 + t )
x1
a1 =
a3 =
(2.35)
3x 3
6 x3
+ 0 + 0 + 3
=
1 + t 1 + t (1 + t ) 2
(1 + t )
3 x3
c) The Lagrangian velocity components are given by:

V1 = X 1
V2 = 2 X 2 (1 + t )
2
V3 = 3 X 3 (1 + t )
(2.36)
dV1
=0
dt
dV
a2 = 2 = 2 X 2
dt
dV
a3 = 3 = 6 X 3 (1 + t )
dt
(2.37)
thus,
a1 =

Ciudad Real - Spain
Draft
149
Problem 2.8
Consider the equations of motion:
x1 = X 1
x2 = X 2 + AX 3
x = X + AX
3
2
3
(2.38)
where A is constant. Find the displacement vector components in the material and spatial
descriptions.
Solution:
Displacement vector:
r r r
u= xX
(2.39)
u1 = x1 X 1 = 0
u 2 = x 2 X 2 = X 2 + AX 3 X 2 = AX 3
u = x X = X + AX X = AX
3
3
3
2
3
2
3
(2.40)
The equations of motion are obtained as follows:

0 X1
A X 2

A 1 X3

(2.41)
0
A = 1 A 2
A 1
(2.42)
x1 1
x = 0
2
x 3 0

1
det 0
0
1
0
1
the inverse:
1 A 2
1
0
1 A2
0
1 A
A 1
(2.43)
thus,
X1
X = 1
2 1 A2
X3

1 A 2
0
0
0 x1
1 A x2

A 1 x3

0
(2.44)
X 1 = x1
( x 2 Ax3 )
X 2 =
1 A2
1
( x3 Ax 2 )
X 3 =
1 A2
(2.45)
The displacement vector components, in the spatial description, become:

Ciudad Real - Spain
Draft
150
u1 = x1 X 1 = 0
A( x3 Ax 2 )
1
( x 2 Ax3 ) =
u 2 = x 2 X 2 = x 2
2
1 A
1 A2
A( x 2 Ax3 )
1
( x3 Ax 2 ) =
u1 = x3 X 3 = x3
2
1 A
1 A2
(2.46)
Problem 2.9
x1 = X 1
x2 = X 2 + X 3t
x = X + X t
3
3
3
(2.47)
Obtain the velocity of the particles that are passing at point (0,1,2) at time t1 = 0 s and
t2 = 1 s
Solution:
The velocity field is given by:
r r
r r
Dx ( X , t )
V ( X ,t) =
Dt
(2.48)
V1 = 0
V2 = X 3
V = X
3
3
(2.49)
in components:
For t = 0 s we have x = X , then, ( X 1 = 0, X 2 = 1, X 3 = 2)

V1 = 0
V2 = 2
V = 2
3
(2.50)
The particle that is passing at point ( x1 = 0, x 2 = 1, x3 = 2) at time t = 1 s , was at the point

(reference configuration):
x1 = 0 = X 1
x 2 = 1 = X 2 + X 3 ( X 1 = 0; X 2 = 0; X 3 = 1)
x3 = 2 = X 3 + X 3
(2.51)
V1 = 0
V2 = 1
V = 1
3
(2.52)
thus,

Ciudad Real - Spain
Draft
151
Problem 2.10
Given the Cartesian system e i , the particle motion is defined as follows:
ct
ct
+ X 2 cos
x1 = X 1 sin 2
2
2
2
X +X
X +X
2
2
1
1
ct
ct
x 2 = X 1 cos 2
X + X 2 + X 2 sin X 2 + X 2
2
2
1
1
x3 = X 3
(2.53)
where c is a constant.
Obtain the velocity components in spatial and material descriptions.
Solution:
The velocity components in the material (Lagrangian) description are:
r
V1 ( X , t ) =
ct
X 1 cos 2
2
2
X + X2
X1 + X 2
2
1
ct
X 2 sin
X2 +X2
2
1
r
V2 ( X , t ) =
ct
X 1 sin 2
2
2
X +X2
X1 + X 2
2
1
ct
+ X 2 cos
X2 +X2
2
1
r
V3 ( X , t ) = 0
(2.54)
Taking into account (2.53), we can note that the following relationship holds:
2
2
x12 + x2 = X 12 + X 2
(2.55)
Then, the velocity components in the spatial (Eulerian) description are:

cx
r
v1 ( x , t ) = 2 2 2
x1 + x 2
r
v 2 ( x, t ) =
(2.56)
c x1
x12
2
x2
+
r
v3 ( x , t ) = 0
r r r
The inverse equations of motion, X = X ( x , t ) , are:
ct
ct
sin 2
x + x 2 cos x 2 + x 2 0
1
2
2
1
X1
x1
ct
ct
X = cos

2 x 2 + x 2 sin x 2 + x 2 0 x 2
2
2
1
X3 1
x

0
0
1 3

Ciudad Real - Spain
Draft
(2.57)
152
Problem 2.11
The Eulerian velocity field components are:
v1 = x1
x2
v2 =
2t + 3
v3 = 0
(2.58)
Find the parametric equations of the trajectory of the particle which was at ( X 1 , X 2 , X 3 ) in
the reference configuration.
Solution:
To find the path line (trajectory) we must solve the system:
dx1
dt = x1
x
dx 2
= 2
2t + 3
dt
dx3
=0
dt
(2.59)
x1 (t = 0) = X 1
x2 (t = 0) = X 2
x (t = 0) = X
3
3
(2.60)
with the initial conditions
x1
X1
x2
X2
dx1
= dt
x1 0
x
Ln 1 = t
X
1
dx 2
dt
=
x2
2t + 3
0
x
Ln 2
X
2
x1 = X 1 exp t
( )
= Ln 2t + 3 Ln 3
x2 = X 2
2
t +1
3
(2.61)
x3 = X 3
Then, the equations of motion are given by:

x1 = X 1exp t
; x2 = X 2
2
t + 1 ; x3 = X 3
3
(2.62)
Problem 2.12
Consider the following equations of motion:
x1 = X 1
x2 = 2 t X 3 + X 2
x = X
3
3
(2.63)
r
and a physical quantity represented by the scalar field q ( x , t ) in the Eulerian description:
r
q ( x , t ) = 2 x1 + x 2 x3 + 1

Ciudad Real - Spain
Draft
(2.64)
153
a) Obtain the Lagrangian description of the physical quantity;

b) Obtain the Lagrangian and Eulerian velocities;
c) Obtain the rate of change of the physical quantity.
d) Obtain the local rate of change of q at the spatial point (1,3,2) .
Solution:
r
r r
a) To obtain the Lagrangian description we have, q ( x , t ) = q ( x ( X , t ), t ) = Q( X , t ) , i.e., we

r
substitute the equations of motion (2.63) into the equation of the variable q ( x , t ) given by
(2.64), i.e.:
r
Q( X , t ) = 2 X 1 + X 2 + ( 2t 1) X 3 + 1
(2.65)
r r
r r
Dx ( X , t )
V ( X , t) =
Dt
(2.66)
V1 = 0
V2 = 2 X 3
V = 0
3
(2.67)
b) The velocity vector
Lagrangian description
The inverse equations of motion:

x1 = X 1
x2 = 2 t X 3 + X 2
x = X
3
3
inverse
X 1 = x1
X 2 = x 2 2 t x3
X = x
3
3
Then, the Eulerian velocity components are given by:

v1 = 0
v 2 = 2 x3
v = 0
3
(2.68)
c) The rate of change of the variable is obtain by applying the material time derivative
r
& D Q( X , t ) = 2 X
Q=
3
Dt
(2.69)
or
&
q=
r
r
q ( x , t )
r
+ xq v
123
4t4
(2.70)
= 0 ( steady )
&
q = 0 + q, i v i
q
q
q
=0+
v1 +
v2 +
v3 = [(2)(0) + (1)(2 x 3 ) + ( 1)(0)]
x 2
x 3
x1
= 2 x3

Ciudad Real - Spain
Draft
(2.71)
154
&
We could have obtained the same result by starting from Q = 2X 3 and substituting
X 3 = x3 , thus
r r
r
&
&
q ( x , t ) = Q( X ( x , t ), t )
r
&
q ( x , t ) = 2 x3
(2.72)
r
d) Note that the physical quantity field is stationary, i.e. q = q ( x ) , then the local rate of
change is
r
q ( x )
= 0 at any spatial point.
t
Problem 2.13
Given the Lagrangian displacement field:
u1 = ktX 2
u2 = 0
u3 = 0
and the Eulerian temperature field:

r
T ( x , t ) = ( x1 + x 2 ) t
a) Find the rate of change of temperature for a particle that at time t = 1s is passing at point
(1,1,1) .
Solution:
r
r
r
r
& ( x , t ) = T + T x or T ( X , t ) = dT ( X , t )
&
We can apply the definition T
r
t
t x t
By means of the equation u i = xi X i we can obtain the equations of motion:

u1 = x1 X 1
x1 = X 1 + ktX 2
u 2 = x2 X 2
x2 = X 2
u 3 = x3 X 3
x3 = X 3
The Lagrangian temperature field (material description) can be obtained as follows:
r
r r
T ( x ( X , t ), t ) = ( x1 + x 2 ) t = (( X 1 + ktX 2 ) + ( X 2 ) ) t = X 1t + kX 2 t 2 + X 2 t = T ( X , t )
Then, the material time derivative becomes:
r
&
T ( X , t ) = X 1 + 2kX 2 t + X 2
If we want to find the rate of change of temperature of a particle which is passing through
the point x1 = 1, x 2 = 1, x3 = 1 at t = 1s , we have two possibilities, namely: 1) Finding the
position of said particle in the reference configuration and replacing in the above equation.
2) obtaining the expression of the rate of change of temperature in the spatial (Eulerian)
r r
description. To do this, we need the equations of motion X ( x , t ) :
x1 = X 1 + ktX 2
x2 = X 2
x = X
3
3
X 1 = x1 ktx 2
X 2 = x2
X = x
3
3
r r
&
& r
T ( X ( x , t ), t ) = X 1 + 2kX 2 t + X 2 = ( x1 ktx 2 ) + 2kt ( x 2 ) + ( x 2 ) = T ( x , t )
& r
by simplifying the above we obtain T ( x , t ) = x1 + ktx 2 + x 2 . Then:

Ciudad Real - Spain
Draft
155
&
T ( x1 = 1, x 2 = 1, x3 = 1, t = 1) = (1 k ) + 2k + 1 = k + 2
r
r
& ( x , t ) = T + T x
T
r
t x t
T x1 T x 2 T x3
= ( x1 + x 2 ) +
x t + x t + x t
2
3
1
= ( x1 + x 2 ) + (tkX 2 + t (0) + (0)(0) )

= x1 + x 2 + tkX 2
Note that x 2 = X 2 , then:

& r
T ( x , t ) = x1 + x 2 + tkx 2
Problem 2.14
Let us consider the following equations of motion:
x1 = X 1
x2 = X 2 + X 3
2
x3 = X 3 + 2 X 2
a)
(2.73)
Is this motion possible? Justify your answer;
b) Obtain the velocity components in the Lagrangian and Eulerian descriptions;

c)
Obtain the path line (trajectory equation).
Solution:
a) Obtaining the Jacobian determinant:
1 0 0
xi
t2
t
= 0 1 2 =1
J=F =
4
X j
t
0 2 1
(2.74)
so that motion is possible for:

J =1
t2
>0t <2 s
4
(2.75)
b) The Lagrangian velocity components:
V = 0
1
D
t
X2 + X3 =
V2 =
2
Dt
t
D
V3 =
X3 + X2 =
2
Dt
X3
2
(2.76)
X2
2
The inverse equations of motion:

Ciudad Real - Spain
Draft
156
J
x1 1 0 0 X 1
X1
x = 0 1 t X inverse X = 1 0
2
2 2
2
J
t
0
x 3 0 2 1 X 3
X3

0
1
t
2
0 x1
t
2 x2

1 x3

(2.77)
By replacing X i into the Lagrangian velocity expression we obtain the velocity in the
spatial description:
t
x2
2 x tx
2
= 3 22
v1 = 0 ; v 2 =
2
t
4t
2
2
x3
t
x3
2 = 2 x 2 tx 3
; v3 =
t2
4 t2
2
2
x2
(2.78)
c) The trajectory can be obtained by eliminating t of the equations of motion (2.73):

x1 = X 1
( x3 X 3 ) X 3 = ( x 2 X 2 ) X 2
Problem 2.15
x3 =
X2
X2
x2 2 + X 3
X3
X3
(2.79)
The velocity field at point x of a steady fluid is given by:

2
b 2 ( x12 x 2 )
b2 x x
r
v =U
e1 + 2U 2 1 22 2 e 2 + Ve 3
2
( x12 + x 2 ) 2
( x1 + x 2 )
(2.80)
where U and V are constants.

r
r
Show that x v = 0 and find the Eulerian acceleration field.
Solution:
v
r
v
v
r
x v = v i ,i = 1 + 2 + 3
x1 x 2 x3
= 2Ub 2
2
x1 ( x12 3 x 2 )
2
( x12 + x 2 ) 3
+ 2Ub 2
2
x1 ( x12 3 x 2 )
2
( x12 + x 2 ) 3
=0
The acceleration:
r
r v
r r
r
a=
+ xv v
t
r r
r
= xv v
The components of the spatial velocity gradient are given by:

2
2
x1 (3 x 2 x12 ) x 2 (3 x12 x 2 ) 0
r
2Ub
2
2
r
( x v ) ij = 2
x 2 (3 x12 x 2 ) x1 (3 x 2 x12 ) 0
2 3
( x1 + x 2 )
0
0
0
r r
r
The acceleration components are given by a i = ( x v ) ij (v ) j :
2

Ciudad Real - Spain
Draft
157
2 x1U 2 b 4
2
2 3
( x1 + x 2 )
a i = 2 x 2U 2 b 4
2
2 3
( x1 + x 2 )
Problem 2.16
r
Let ( X , t ) be a scalar field in Lagrangian (material) description. Find the relationship

r
r
r
between the material gradient of ( X , t ) , i.e. X ( X , t ) , and the spatial gradient of
r
r
r
( x , t ) , i.e. x ( x , t ) .
r
Solution:
Remember that a Lagrangian variable ( X , t ) can be expressed in the Eulerian
(current) configuration by means of the equations of motion, i.e.:
r
r r
( X , t ) = ( X ( x , t ), t ) = ( x , t ) .
Then, from the scalar gradient definition we obtain:
r
r r
r
r
r
r
( X , t ) ( X ( x , t ), t ) x ( x , t )
r
r
X ( X , t ) =
r
=
r=
r
r F = x ( x, t ) F
x
x
X
X
In addition we have the inverse form:
r
r
r
r r
r
r
( x , t ) ( x ( X , t ), t ) X ( X , t )
r
r
x ( x , t ) =
r
r=
r F 1 = X ( X , t ) F 1
r =
x
x
X
X
Problem 2.17
Given the following Eulerian velocity field components:
v1 = 0 ;
v2 = 0 ;
v3 = f ( x1 , x 2 ) x3
a) Find the particle trajectory;

b) Obtain the mass density ( ), knowing that at t = 0 we have = f ( x1 , x 2 ) .
Solution:
dx1
= v1 = 0 x1 (t ) = C1 at t = 0 x1 = X 1 x1 (t = 0) = C1 = X 1 ;
dt
dx 2
= v 2 = 0 x 2 (t ) = C 2 at t = 0 x 2 = X 2 x 2 (t = 0) = C 2 = X 2
dt
dx3
dx3
= v3 = f ( x1 , x 2 ) x3 = f (C1 , C 2 ) x3
=
dt
x3
f (C , C
1
2 ) dt
Ln( x 3 ) = f (C1 , C 2 )t + k
denoting by k = Ln(C 3 ) we obtain:

Ln( x3 ) Ln(C 3 ) = f ( X 1 , X 2 )t
x
ln 3
C
3
x
= f ( X 1 , X 2 )t 3 = exp f ( X 1 , X 2 )t x3 = C 3 exp f ( X 1 , X 2 )t
C3

Ciudad Real - Spain
Draft
158
at t = 0 x3 = X 3 x3 (t = 0) = C 3 = X 3
Summarizing:
x1 = X 1
x2 = X 2
(2.81)
x3 = X 3 exp
f ( X 1 , X 2 )t
The mass density:
with
1 0
Fij = 0 1
Fij =
0
0
xi
X j
= exp f ( X 1 , X 2 )t
? ? exp f ( X 1 , X 2 )t
The values marked by ( ? ) are not necessary to obtain the determinant, then:
0
F
f (X1, X 2 )
exp f ( X 1 , X 2 )t
Note that according to the problem statement, t = 0 , = f ( x1 , x 2 ) , and according to the

equations in (2.81) we conclude that 0 = f ( X 1 , X 2 ) .
Problem 2.18
Calculate the material time derivative
described as follows:
D
for the property when said property is
Dt
Material description: ( X , t ) = X 1t 2 ;
r
Spatial description: ( x , t ) =
Solution:
x1t 2
.
(1 + t )
a) Material time derivative of ( X , t ) = X 1t 2 :

r
r
D
&
( X , t ) ( X , t ) = 2 X 1t
Dt
r
b) Material time derivative of ( x , t ) =
x1t 2
:
(1 + t )
r
r
r
r
r ( x , t ) ( x , t )
( x , t )
D
r v =
( x, t ) =
+x
+
vi
t
t
Dt
xi
r
r
r
r
( x , t )
( x , t )
( x , t ) ( x , t )
v3
v2 +
v1 +
+
=
x3
x 2
t
x1
r
2
x t ( x , t )
= 1 +
v1 + 0 + 0
(1 + t )
t
x1

Ciudad Real - Spain
Draft
159
We need to know the velocity component v1 . We start from the principle that a property is
intrinsic to the particle, then:
r
( X , t ) = X 1t 2
r r
x1t 2
r
( X ( x, t ), t ) = ( x , t ) =
(1 + t )
X1 =
x1
(1 + t )
The velocity becomes:
r
D
X 1t 2 = 2 X 1t e 1
v( X , t) =
Dt
x
r
v ( x, t ) = 2 1 t e1
(1 + t )
Returning to the material time derivative we obtain:

r
r
D
x1t 2 ( x , t )
+
( x , t ) =
v1
Dt
t (1 + t ) x1
2
2
2 x1t
xt t
X1
=
1 2 +
(1 + t ) (1 + t )
(1 + t )
2 x1t
x t2
=
1 2
(1 + t ) (1 + t )
2 x1t
=
(1 + t )
t2
x
+
2 1 t
(1 + t ) (1 + t )
We could also have obtained the same result by starting from

and also knowing the equations of motion X 1 =
r
r
D
&
( X , t ) ( X , t ) = 2 X 1t
Dt
x1
, i.e.:
(1 + t )
r
r
D
&
( X , t ) ( X , t ) = 2 X 1t
Dt
r r
r r
x
D
&
& r
( X ( x, t ), t ) ( X ( x, t ), t ) ( x, t ) = 2 1 t
Dt
(1 + t )
Problem 2.19
Consider the following equations of motion in the Lagrangian description:
x1 = X 1t 2 + 2 X 2 t + X 1
x1
Matrix

form
2
x 2 = 2 X 1t + X 2 t + X 2 x 2 =
x
x = 1 X t + X
3
3
3 2 3
t 2 + 1 2t
2
t +1
2t
0
0
0 X 1

0 X 2
1
t + 1 X 3
2

(2.82)
Find the components of the displacement vector in Lagrangian and Eulerian descriptions.
Solution:
r
The displacement vector can be obtained as follows u = x X . By replacing the equations

of motion (2.82) we obtain:
u1 = x1 X 1 = X 1t 2 + 2 X 2 t
2
u 2 = x 2 X 2 = 2 X 1t + X 2 t
1
u 3 = x 3 X 3 = 2 X 3 t

Ciudad Real - Spain
Draft
160
which are the displacement components in the Lagrangian description (material).

To obtain the Eulerian displacement we need to obtain the inverse equations of motion
(2.82), which the result is:
2tx 2 x1 (1 + t )
X1 = 3
3t 1 t t 2
x
0
1

2 x1t 2 x 2 (1 + t 2 )
0
x2 X 2 =
3t 3 1 t t 2
3t 3 1 t t 2 x 3

2 x3
1
(t + 2)
X 3 =
2
(t + 2)
(1 + t )
2t
X1
1

2
(1 + t 2 )
2t
X 2 = 3
2
X 3t 1 t t
3
0
0
We can also use the definition u = x X , but now we replace the material coordinate to
obtain the displacement vector components in the Eulerian description:
2tx 2 x1 (1 + t )
u1 = x1 X 1 = x1 3
3t 1 t t 2
2 x t 2 x 2 (1 + t 2 )
u 2 = x2 X 2 = x2 1 3
3t 1 t t 2
2 x3
u 3 = x 3 X 3 = x 3
(t + 2)
Problem 2.20
The following equations describe
the motion of a body (continuum
medium):
x1 = X 1 + 0.2 X 2 t
x2 = X 2
x = X
3
3
X 2 , x2
Reference configuration
B
1
O
D
X 1 , x1
X 3 , x3
Figure 2.1: Reference configuration t = 0 .

At time t = 0 , the cubic body (with sides 1) has one vertex at the origin of the system
which is indicated by point O, (see Figure 2.1). Obtain the configuration of the body at
time t = 2 s .
Solution:
To obtain the current configuration of the body at time t = 2 s , we will analyze the particle
motion. The particle which occupies position O (origin) at t = 0 has material coordinate:

Ciudad Real - Spain
Draft
161
X1 = 0
X 2 = 0
X = 0
3
Substituting in the equations of motion:

x1 = 0
x i ( X 1 = 0, X 2 = 0, X 3 = 0, t ) x 2 = 0
x = 0
3
Then, we conclude that the particle does not change its position during motion.
The particles occupying the line OA , in the initial configuration, have the reference
coordinate ( X 1 , X 2 = 0, X 3 = 0) . In spatial coordinates:
x1 = X 1 + 0.2 X 2 t = X 1
x2 = X 2 = 0
x = X = 0
3
3
That is, all particles belonging to line OA do not move during motion. Similarly, we can
verify that the line ( X 1 , X 2 = 0, X 3 = 1) in the reference configuration ( X 1 , X 2 = 0, X 3 = 1)
does not move:
x1 = X 1 + 0.2 0 2 = X 1
x2 = X 2 = 0
x = X = 0
3
3
The particles belonging to line CB ( X 1 , X 2 = 1, X 3 = 0) at time t = 2 s will move according

to:
x1 = X 1 + 0.2 1 2 = X 1 + 0.4
x2 = X 2 = 1
x = X = 0
3
3
Then, all particles belonging to line CB will move 0.4 according to x1 -direction.
The particles belonging to line OC at t = 0 , will move to positions:
x1 = X 1 + 0.2 X 2 t = 0 + 0.2 2 X 2 = 0.4 X 2
x2 = X 2
x = X = 0
3
3
Following the same procedure for the remaining particles, we obtain the final configuration
of the body at time t = 2 s , (see Figure 2.2).

Ciudad Real - Spain
Draft
162
x2
0.4
0.4
C 1
C
E
Current configuration at
t = 2s
E 1
A=A
x1
G=G
x3
Figure 2.2: Current configuration at time t .

Problem 2.21
x1 = X 1 + t 2 X 2
x1
Matrix

form
2
x 2 = t X 1 + X 2 x 2 =
x
x = X
3
3
3
1
2
t
0
0 X 1

0 X 2
1 X 3

t2
1
0
1) Obtain the trajectory of particle Q which originally ( t 0 ) was at X i = (1,2,1) ;

2) Consider the current configuration at t = 0.5 s . Obtain the velocity and acceleration
4
components of the particle P that was originally at X i = (16 ; 15 ;1) ;
15
3) Obtain the equations of motion in the Eulerian description;
4) Obtain the velocity and acceleration components of one particle that at time ( t = 0.5 s ) is
passing at point xi = (1,0,1) .
NOTE: Consider the International System of Units (SI-Units).
Solution:
1)Using the equations of motion and by replacing the material coordinates of the point
X i = (1,2,1) , we obtain:
x1 = 1 + 2t 2
2
x2 = 2 + t
x = 1
3
The above equations represent the motion of the particle. To obtain the trajectory, we
eliminate the time of the equations of motion, i.e.:
x1 2 x 2 = 3
x3 = 1
which indicates that the particle moves in a straight line defined by ( x1 2 x 2 = 3) on the
plane x3 = 1 . The graphical representation follows

Ciudad Real - Spain
Draft
163
Particle trajectory
X 3 , x3
( x1 2 x 2 = 3)
x3 = 1
X 2 , x2
X 1 , x1
2) The velocity and acceleration components of the particle P are given by:
V1 = 2tX 2
r
r r
Dx
components
V ( X , t) =
V 2 = 2tX 1
Dt
V = 0
3
A1 = 2 X 2
r
r r
Dv
components
A2 = 2 X 1
A( X , t ) =
Dt
A = 0
3
4
Then, to the particle originally located at the point X i = (16 ; 15 ;1) , at t = 0.5 s , we have:
15
4
4
V1 = 2 0.5 15 = 15 m / s
16
16
V 2 = 2 0.5 (15 ) = 15 m / s
V = 0
3
and
8
4
A1 = 2 15 = 15 m / s 2
2
16
32
A2 = 2 (15 ) = 15 m / s
A = 0
3
3) The inverse equations of motion can be obtained as follows:

x1 = X 1 + t 2 X 2 X 1 = x1 t 2 X 2
2
2
x2 = t X 1 + X 2 X 2 = x2 t X 1
x = X X = x
3
3
3
3
x t 2 x2
X1 = 1
1 t4
x t 2 x1
X 2 = 2
1 t4
X 3 = x3
4) The velocity and acceleration of the particle that at time ( t = 0.5 s ) is passing through the
point xi = (1,0,1) can be obtained by means of velocity and acceleration in Eulerian
description:
Velocity:

Ciudad Real - Spain
Draft
164
x t 2 x1
v1 = 2t 2
1 t4
V1 = 2tX 2
x t 2 x2
substituti ng
t = .5 s
V 2 = 2 X 1t v 2 = 2t 1
0
4
X1 , X 2
x (1, 01)
1 t
V = 0
3
v 3 = 0
v1 = 15 m / s
16
m/s
v 2 =
15
v 3 = 0
Acceleration:
x t 2 x1
a1 = 2 2
1 t4
A1 = 2 X 2
x t 2 x2
substituti ng
t = .5 s
A2 = 2 X 1 a 2 = 2 1
4
X1 , X 2
x (1, 01)
1 t
A = 0
3
a 3 = 0
2
a1 = 15 m / s
32
m / s2
a 2 =
15
a 3 = 0
We can obtain the initial position of the particle by using the inverse equations of motion
which was obtained in paragraph (3), xi (1,0,1) :
x t 2 x 2 1 (0.5 2 )(0) 16
=
=
X1 = 1
15
1 t4
1 (0.5) 4
x 2 t 2 x1 0 (0.5 2 )(1)
4
=
=
X 2 =
4
4
15
1 t
1 (0.5)
X = x = 1
3
3
We can verify that it is the same particle P referred to in paragraph 2. It is logical that we
have obtained the same velocity and acceleration using either the material or spatial
description, since the velocity and acceleration are intrinsic of the particle.
Problem 2.22
The acceleration vector field is described by:
r
r
r r
r r
D v v
r
=
a ( x , t) =
+ xv v
Dt t
Show that acceleration can also be written as:

r
r
v2
Dv v
r
=
+x
2
Dt t
r
r
r
v2
r v
r
r
v ( x v ) =
+x
2
t
r
r
v2
r v
r
v rot v =
+x
2
t
r r
+ rot v v
Solution:
To prove the above relationship one need only demonstrate that:
v2
r r
r
r
xv v = x
2
r
r
r
r
v ( x v )
Expressing the terms on the right of the equation in symbolic notation we obtain:

Ciudad Real - Spain
Draft
v2
r
x
2
165
r
r
r
1
r
v ( x v ) = e i
v j v j (v i e i )
e r (v s e s )
2 x i
x r
Using the definition of the permutation symbol (see Chapter 1) we can express the vector
product as:
v2
r
x
2
r
r
v
r
1
r
v ( x v ) = e i
v j v j (vi e i ) rst s e t
2 x i
x r
v j
v s
1
ek
= e i 2v j
rst itk v i
2
x r
x i
where we have used the equation e i e t = itk e k . In Chapter 1 we also proved that
rst itk = rst kit = rk si ri sk , then:
v2
r
x
2
r
v j
r
r
r
v ( x v ) = v j
x i
v j
=vj
x i
v j
=vj
x i
v2
r
x
2
v s
ek
x r
v
v
e i rk si v i s ri sk v i s
x r
x r
v
v
e i v s s v i k e k
x
x i
k
e i ( rk si ri sk )v i
e k
r
v j
r
v
v
r
r
v ( x v ) = v j
e i v s s e k + vi k e k
x i
x k
x i
v j
v
v
e i v s s ik e i + v i k e k
= sj v s
x i
x k
x i
v
v
v
= v s s e i v s s e i + vi k e k
x i
x i
x i
r
v e
(v )
vi
= k k vi =
x i
x i
r r
r
= xv v
NOTE: It is interesting to see Problem 1.115 (Chapter 1).

Problem 2.23
r r
Consider the equations of motion x ( X , t ) and the temperature field T ( x , t ) given by:
x1 = X 1 (1 + t )
x 2 = X 2 (1 + t )
x = X
3
3
r
2
T ( x ) = x12 + x 2
Find the rate of change of temperature for the particle P at time t = 1s given that particle
P was at point ( X 1 = 3, X 2 = 1, X 3 = 0) at time t = 0 .
Solution 1:
In this first solution we first obtain the material time derivative of the Lagrangian
r
temperature, so, we have to obtain the temperature in Lagrangian description T ( X , t )
(Lagrangian temperature):

Ciudad Real - Spain
Draft
166
r
2
T ( x ) = x12 + x 2
By substi tuting
the equations of motion
r
2
T ( X , t ) = X 12 (1 + t ) 2 + X 2 (1 + t ) 2
The material time derivative of the Lagrangian temperature is given by:

r
r
DT dT ( X , t )
2
&
T ( X , t)
=
= 2 X 12 (1 + t ) + 2 X 2 (1 + t )
Dt
dt
By substituting t = 1s , ( X 1 = 3, X 2 = 1, X 3 = 0) , into the above equation we obtain:

r
2
&
T ( X , t ) = 2 X 12 (1 + t ) + 2 X 2 (1 + t ) = 2(3) 2 (1 + 1) + 2(1) 2 (1 + 1) = 40
Solution 2:
In this alternative solution we directly use the definition of material time derivative of the
r
r
r
r
& ( x , t ) = DT = T ( x ) + T ( x ) v ( x , t ) .
Eulerian variable, i.e. T
k
t
x k
Dt
From the equations of motion we obtain:

r
v1 ( X , t ) = X 1
x1 = X 1 (1 + t )
r
velocity
x 2 = X 2 (1 + t ) v 2 ( X , t ) = X 2
r
x = X
3
3
v 3 ( X , t ) = 0
The equations of motion in Eulerian description are given by:

x1
X 1 = (1 + t )
x1 = X 1 (1 + t )
x2
inverse of the motion

x 2 = X 2 (1 + t ) X 2 =
(1 + t )
x = X
3
3
X 3 = x3
So, it is possible to obtain the Eulerian velocity as follows:

r r
r
r
x1
V1 ( X ( x , t ), t ) = X 1 ( x , t ) = (1 + t ) = v1 ( x , t )
r r
x2
r
r
= v 2 ( x, t )
V 2 = ( X ( x , t ), t ) = X 2 ( x , t ) =
(1 + t )
r
V3 = v 3 ( x , t ) = 0
Afterwards, the material time derivative of the Eulerian temperature, T ( x , t ) , is given by:
r
DT ( x , t ) & r
T ( x, t ) =
Dt
r
T ( x )
t3
12
= 0 (Stationar y field)
x
x
& r
T ( x , t ) = 2 x1 1 + 2 x 2 2 + 0
1+ t
1+ t
T
T
T
+
v1 +
v2 +
v3
x 2
x 3
x1
2x 2 2x 2
2
2
& r
T ( x, t ) = 1 + 2 =
( x12 + x 2 )
1+ t 1+ t 1+ t
The position of particle P at time t = 1s is evaluated as follows:

Ciudad Real - Spain
Draft
167
x1 = X 1 (1 + t ) = 3(1 + 1) = 6
x 2 = X 2 (1 + t ) = 1(1 + 1) = 2
x = X = 0
3
3
Then, by substituting the spatial coordinates in the expression of the material time
derivative of temperature we obtain:
2
2
2
& r
&
( x12 + x 2 ) =
T ( x , t ) = T ( x1 = 6, x 2 = 2, x 3 = 0, t = 1) =
(6 2 + 2 2 ) = 40
1+ t
1+1
& r
Alternatively, the expression T ( x , t ) could also have been obtained as:
r
2
&
T ( X , t ) = 2 X 12 (1 + t ) + 2 X 2 (1 + t )
2
2
r r
x
x
r 2
r 2
&
T ( X ( x , t ), t ) = 2[X 1 ( x , t )] (1 + t ) + 2[X 2 ( x , t )] (1 + t ) = 2 1 (1 + t ) + 2 2 (1 + t )
(1 + t )
(1 + t )
2
2
& r
=
( x12 + x 2 ) = T ( x , t )
(1 + t )
Problem 2.24
Consider the motion:
x i = X i (1 + t )
t > 0
Obtain the velocity field in the spatial description.

Solution:
The velocity is obtained by means of the material time derivative of the equations of
motion:
&
Vi = x i =
d
[X i (1 + t )] = X i
dt
(2.83)
To find the velocity in the spatial description we need to obtain the inverse equations of
motion and replace it into the equation (2.83):
xi
x i = X i (1 + t ) X i = (1 + t )
r
v = X ( x , t ) = x i
i
i
1+ t
Problem 2.25
Consider the temperature field T ( x ) in the spatial description and the equations of motion:
2
T = 2( x12 + x 2 )
i {1,2}
xi = X i (1 + t )
Find the rate of change of temperature at time t = 1s for one particle that was at position
(1,1) in the reference configuration.
r
NOTE: We can observe that the temperature field is a steady field, i.e. T = T ( x ) .

Ciudad Real - Spain
Draft
168
Solution 1:
In this first solution we obtain the equation for temperature in the material description:
r
2
T ( x ) = 2( x12 + x 2 )
by replacing the equations

of motion
r
T ( X , t ) = 2 X 2 (1 + t ) 2 + X 2 (1 + t ) 2
1
2
The material time derivative is given by:
r
r
DT dT ( X , t )
2
&
T ( X , t) =
=
= 2 2 X 12 (1 + t ) + 2 X 2 (1 + t )
Dt
dt
By replacing t = 1s and material coordinate ( X 1 = 1; X 2 = 1) we obtain:

&
T ( X 1 = 1; X 2 = 1; t = 1) = 16
Solution 2:
In this alternative solution we use directly the definition of the material time derivative of
Eulerian property:
r
T ( x ) = 2( x12 + x12 )
x i = (1 + t ) X i
r
r
r
& ( x , t ) = DT = T ( x ) + T ( x ) x k
T
Dt
t
x k t
r
r
T ( x )
Note that T ( x ) is not a function of time, so
=0:
t
i {1,2}
i {1,2}
T x1
T x 2
& r
T ( x, t ) = 0 +
+
x1 { x 2 {
t
t
V1 = X 1
V2 = X 2
x
x
& r
T ( x , t ) = 0 + 4 x1 1 + 4 x 2 2
1+ t
1+ t
2
2
4x
4x
& r
T ( x, t ) = 1 + 2
1+ t 1+ t
The particle that at reference configuration was at position (1,1) , at time t = 1s will be at
position xi = (1 + t ) X i = 2 X i , i.e. ( x1 = 2; x 2 = 2 ):
&
T ( x1 = 2; x 2 = 2; t = 1) =
4( 2) 2
1+1
4( 2) 2
1+1
= 16
Problem 2.26
x1 = X 1 exp t + X 3 (exp t 1)
t
t
x 2 = X 2 + X 3 (exp exp )
x = X
3
3
Obtain the velocity and acceleration components in Lagrangian and Eulerian descriptions.
Solution:

Ciudad Real - Spain
Draft
169
First we obtain the inverse equations of motion:

x1 = X 1 exp t + X 3 (exp t 1)
t
t
x 2 = X 2 + X 3 (exp exp )
x = X X = x
3
3
3
3
x1 X 1 exp t = x 3 (exp t 1)
x 2 X 2 = x 3 (exp t exp t )
x = X X = x
3
3
3
3
thus:
X 1 = x1 exp t exp t (exp t 1)
2t
t
X 2 = x 2 x 3 (exp 1)exp
X = x
3
3
(2.84)
or
x1

x2 =
x
3
exp t
0
0
X 1
X1
inverse
t
t
1 (exp exp ) X 2 X 2 =
X
X
0
1
3
3
0
(exp t 1)
exp t
0
0
exp t (exp t 1) x1

1 (exp 2t 1)exp t x 2
x
0
1
3
0
a) The velocity components in the material description are given by:

V1 = X 1exp t + X 3 exp t
r
Vi =
x j ( X , t ) V 2 = X 3 exp t + X 3 exp t = X 3 (exp t + exp t )
Dt
V = 0
3
(2.85)
b) The acceleration components in the material description are given by:

A1 = X 1exp t + X 3 exp t
r
r
DV i ( X , t )
Ai ( X , t ) =
A2 = X 3 (exp t exp t )
Dt
A = 0
3
(2.86)
To obtain the velocity and acceleration in the spatial description it is sufficient to replace
the values of X 1 , X 2 , X 3 , given by the equation (2.84), into the equations (2.85) and (2.86),
i.e.:
v1 = x1 + x 3
t
t
v 2 = x 3 (exp + exp )
v = 0
3
Velocity in the
spatial descriptio n
Problem 2.27
r
a1 = x1 + x 3
t
t
a 2 = x 3 (exp exp )
a = 0
3
Accelerati on in the
spatial descriptio n
r r
The motion of the continuum, x = x ( X , t ) , is given by the following equations:

x1 = 1 ( X 1 + X 2 )exp t + 1 ( X 1 X 2 )exp t
2
2
t
t
1
1
x 2 = 2 ( X 1 + X 2 )exp 2 ( X 1 X 2 )exp
x = X
3
3
0 t constant

Ciudad Real - Spain
Draft
170
Express the velocity components in the material and spatial descriptions.

Solution:
The velocity components using material description are:
r
Dx1 ( X , t ) 1
1
= ( X 1 + X 2 )exp t ( X 1 X 2 )exp t
V1 =
Dt
2
2
Dx 2 ( X , t ) 1
1
= ( X 1 + X 2 )exp t + ( X 1 X 2 )exp t
V 2 =
Dt
2
2
V3 = 0
(2.87)
To express the velocity components in the spatial description we need the inverse
r r r
equations of motion, i.e. we need to find X = X ( x , t ) :
(exp t
x1
t
(exp
x2 =
x
3
+ exp t )
2
exp t )
2
0
0
X 1

0 X 2

1 X 3
(exp t exp t )
2
(exp t + exp t )
2
0
(exp 2t + 1)exp t
X1
1

inverse
X 2 = (exp 2t 1)exp t
X 2
0
3
(exp 2t 1)exp t
(exp 2t + 1)exp t
0
0 x1

0 x 2
2 x 3

Then, to obtain the Eulerian velocity we replace the above equations into (2.87), which the
result is:
v1 = x 2
v 2 = x1
v = 0
3
Problem 2.28
Given the motion:
x i = ( X 1 + ktX 2 ) i1 + X 2 i 2 + X 3 i 3
i {1,2,3}
and the temperature field

T = x1 + x 2
Obtain the rate of change of T of a particle that in the current configuration is located at
point (1,1,1) .
Solution:
Considering the equations of motion:
x1 = X 1 + ktX 2
x2 = X 2
x = X
3
3

Ciudad Real - Spain
Draft
171
we replace xi into the temperature field, with that we obtain the temperature field in the
material description:
r
r
T ( x ) = x1 + x 2 T ( X , t ) = X 1 + ktX 2 + X 2
The material time derivative is given by:

r
DT D ( X 1 + ktX 2 + X 2 )
1,1
&
T ( X , t) =
= kX 2 = k x 2 (,1) T = k
&
=
Dt
Dt
The material time derivative for a property expressed in the spatial description is given by:
DT T T x k
&
T ( x1 , x 2 , x 3 , t ) =
=
+
Dt
t x k t
Considering T = x1 + x 2 , we obtain:
T T x1 T x 2 T x 3
&
+
+
+
T ( x1 , x 2 , x 3 , t ) =
t
x 2 { x 3 {
t
t
{ x1 t
{
=0
=0
=0
=0
&
T ( x1 , x 2 , x 3 , t ) = kX 2
we obtain the inverse equations of motion:

x1 = X 1 + ktX 2
X 1 = x1 ktx 2
inverse
X 2 = x 2
x2 = X 2
x = X
X = x
3
3
3
3
&
T ( x1 , x 2 , x 3 , t ) = kX 2 = kx 2
For one particle in the current configuration at the position (1,1,1) we have:
&
T ( x1 = 1, x 2 = 1, x 3 = 1, t ) = k
Problem 2.29
Given a steady velocity field: it asks readers to give their opinion on whether particle
velocities are constant or not. If not, in which situation is met. Justify the answer.
Solution:
A field ( x , t ) is said to be steady if the local rate of change does not vary over time, so:
r
( x , t )
=0
t
= ( x ) Steady state (stationary) field
(2.88)
For example, let us consider a stationary (steady state) velocity field as shown in Figure 2.3.
Then, as we can verify, the field representation for any time, e.g. t1 and t 2 , does not
change. However, that does not mean that the velocities of the particles do not change
over time. In light of Figure 2.3, we can now focus our attention on the fixed spatial point
r
r
r
x * . At time t1 the particle Q is passing through point x * with velocity v * . Let us also
consider another particle P , which is passing through another point with velocity
r
r
r
v P (t1 ) v * . At time t 2 the particle P is now passing through the point x * . It follows that
r
if we are dealing with a steady state velocity field, then the velocity of particle P at x *

Ciudad Real - Spain
Draft
172
must be v * , i.e. v P (t 2 ) = v * . We can easily contrast this with the material time derivative of
velocity, which is always associated with the same particle, i.e.:
r r
r r
r r r
r r r
r r
Dv ( x , t ) r r
v ( x , t )
r
r
a ( x, t ) =
+ x v v ( x) = x v v ( x) = a ( x)
Dt
123
4t4
r
(2.89)
= 0 (Stationay )
The rate of change of velocity (acceleration) will be zero if the velocity field is stationary
r r
r
v ( x , t) r
r
= 0 and homogeneous ( x v = 0 ).
We can also verify that, although spatial velocity is independent of time, that does not
mean material velocity is also, since:
r r r r r
r r
v ( x ) = v ( x ( X , t )) = v ( X , t )
t1
(2.90)
r r
v ( x)
r r
r
r
v ( x * , t1 ) = v * = v Q
Particle - Q
Particle- P
r
r
v P v*
r
x*
t2
r r
v ( x)
r r
r
r
v ( x * , t2 ) = v * = v P
Particle - P
r
x*
Figure 2.3: Steady velocity field.

Ciudad Real - Spain
Draft
2.1.2
Deformation/strain
Deformation
173
Tensors,
Homogeneous
Problem 2.30
A rod, which is considered as a one-dimensional solid, undergoes a uniform stretching and
is given by:
= exp at
where a = constant .
(2.91)
r r
a) Obtain the equations of motion x = x ( X , t ) ;

b) Obtain the rate-of-deformation tensor components, i.e. D -components.
x1
= exp at
Figure 2.4.
Solution:
Using the approach in 1D we have:
=
ds dx
=
= exp at dx = exp at dX
dS dX
dx = exp
at
(2.92)
dX
(2.93)
x1 = exp at X 1 + C
(2.94)
x = exp 0 X 1 + C X = X + C C = 0
(2.95)
at t = 0 x = X , thus
with that we obtain the equations of motion:

x1 = exp at X 1
x2 = X 2
x = X
3
3
(2.96)
The velocity field components become:

dx1
at
v1 = dt = a X 1 exp = a x1
v 2 = 0
v = 0
3

Ciudad Real - Spain
Draft
(2.97)
174
And the rate-of-deformation can be obtained as follows:

v j
1 v
D ij = i +
2 x j xi
a 0 0
Dij = 0 0 0
0 0 0
(2.98)
Problem 2.31
x1 = X 1 + 2 X 3
x2 = X 2 2 X 3
x = X 2 X + 2 X
3
1
2
3
Obtain the Green-Lagrange strain tensor components, i.e. E -components.

Solution 1:
The displacement field components are given by
u1 = x1 X 1 = 2 X 3
u 2 = x 2 X 2 = 2 X 3
u = x X = 2 X + 2 X
3
3
1
2
3
The Green-Lagrange strain tensor can be expressed in function of Lagrangian displacement

as follows:
E ij =
=
1 u i u j u k u k
+
+
2 X j X i X i X j
1 u i u j
+
2 X j X i
u
= i
X j
sym
1 u k u k
+
2 X i X j
1 u k u k
2 X i X j
where the material (Lagrangian) displacement gradient is given by:

u1
X 1
u i u 2
=
X j X 1
u 3
X 1
u1
X 2
u 2
X 2
u 3
X 2
u1
X 3 0 0 2
u 2
= 0 0 2
X 3

u 3 2 2 0
X 3
Note that, in this case, the displacement gradient is an antisymmetric tensor. That is, the
symmetric part is the null tensor, and the remaining term is:
0 0 2 T 0 0 2 2 2 0
1 u u k 1
= 0 0 2 0 0 2 = 2 2 0
E ij = k

2 X i X j 2
2 2 0 2 2 0 0
0 4

Solution 2:

Ciudad Real - Spain
Draft
175
We can directly apply the definition:

E ij =
1
1
C ij ij = Fki Fkj ij
2
2
where:
x1
X 1
x
x
Fij = i = 2
X j
X
1
x3
X 1
x1
X 2
x 2
X 2
x3
X 2
x1
X 3 1 0 2
x 2
= 0 1 2
X 3

x3 2 2 1
X 3
Thus
1 0 2 T 1 0 2 1 0 0 2 2 0
1
E ij = 0 1 2 0 1 2 0 1 0 = 2 2 0

2
2 2 1 2 2 1 0 0 1 0
0 4
Problem 2.32
Consider a homogeneous transformation defined by the following equations:
x1 = X 1 + 2 X 2 + X 3
x2 = 2 X 2
x = X + 2 X
1
3
3
(2.99)
Show that, for a homogeneous transformation, vectors that are parallel in the reference
configuration remain parallel after deformation.
For the demonstration consider two vectors defined by the vector position of two particles
A and B in the reference configuration:
r
X A = e1 + e 2
rB
X = 2e1 + 2e 2 + e 3
(2.100)
Solution:
The vector connecting the two particles in the reference configuration is given by:
r r r
V = B A = e1 + e 2 + e 3
(2.101)
and the deformation gradient is:

1 2 1
xi
= 0 2 0
Fij =
X j
1 0 2
(2.102)
We can obtain the vector position of the particle in the current configuration by means of:
r
r
dx = F dX
Homogeneous transformation
r
r
x=FX
(2.103)
thus,

Ciudad Real - Spain
Draft
176
xiA
1 2 1 1 3
= 0 2 0 1 = 2

1 0 2 0 1

xiB
1 2 1 2 7
= 0 2 0 2 = 4

1 0 2 1 4

(2.104)
and the vector that connect these two points is:

r r
r
v = x B x A = 4e1 + 2e 2 + 3e 3
(2.105)
then any vector parallel to V , for example the vector 2e1 + 2e 2 + 2e 3 , after transformation
r
becomes: 8e1 + 4e 2 + 6e 3 , which is parallel to v .
Problem 2.33
Consider a pure shear deformation represented by homogenous deformation:
r r
x = X + k t X 2e1
(2.106)
where e i is the Cartesian basis, and the components of the above equation are:
x1 = X 1 + k t X 2
x2 = X 2
x = X
3
3
(2.107)
Obtain the geometry in the current configuration of Figure 2.5 which is represented by a
rectangle in the reference configuration.
X2
B
X1
Figure 2.5
Solution:
The deformation gradient:
1 k t 0
xi
Fij =
= 0 1 0
X j
0 0 1
(2.108)
r
Note that this is a case of homogenous deformation, i.e. x = F X + c with c = 0 .

The Jacobian determinant:
J = F =1

Ciudad Real - Spain
Draft
(2.109)
177
Note that, in this case there is no dilatancy.

The particles which are on the line BC have coordinates ( X 1 , X 2 ,0) in the reference
configuration, with that we obtain:
x1 = X 1 + k t X 2
x2 = X 2
x = 0
3
(2.110)
The particles which are on the line OA , coordinates ( X 1 ,0,0) , in the current configuration
assume the position:
x1 = X 1
x2 = 0
x = 0
3
(2.111)
then, the line OA does not change its position during motion, (see Figure 2.6).
x2
B
x1
Figure 2.6
Problem 2.34
x1 = X 1 +
2
X2
2
x2 =
2
X1 + X 2
2
x3 = X 3
(2.112)
a) Show that this deformation is characterized by a homogeneous transformation;

b) Obtain the displacement field components in material and spatial descriptions;
c) Consider the particles that are located according to the equation:
2
X 12 + X 2 = 2
X3 = 0
Obtain the new configuration of these particles in the current configuration;

d) Obtain the right Cauchy-Green deformation tensor components ( C ) and the GreenLagrange strain tensor ( E ).
e) Obtain the principal values of C and E .
Solution:
a) The equation of a homogeneous deformation is described by:

Ciudad Real - Spain
Draft
178
xi = Fij X j
(2.113)
where
xi
Fij =
=
X j
2
2
1
2
2
0
1
0
0
1
(2.114)
Note that the equation (2.113) is in accordance with (2.112):
x1
x =
2
x3

2
2
1
2
2
0
1
0
0
X1
0 X 2

X3
1
(2.115)
Hence, we are dealing with homogeneous deformation. The inverse of (2.115) is obtained
as follows:
X1 2
X = 2
2
X3 0

2
2
0
0 x1
0 x 2

1 x3

X 1 = 2 x1 2 x 2
X 2 = 2 x1 + 2 x 2
X = x
3
3
(2.116)
b) The displacement field is given by:
2
2
X 2 X1 =
X2
u1 = x1 X 1 = X 1 +
2
2
r r r
2
2
X1 + X 2 X 2 =
X1
u = x X u 2 = x 2 X 2 =
2
2
u 3 = x3 X 3 = 0
which, in spatial coordinates, becomes:
(2.117)
u1 = x1 X 1 = x1 2 x1 2 x 2 = x1 + 2 x 2
u 2 = x 2 X 2 = x 2 2 x1 + 2 x 2 = 2 x1 x 2
(2.118)
u 3 = x 3 X 3 = x3 x3 = 0
c) The particles describing a circle in the reference configuration:

2
X 12 + X 2 = 2
X3 = 0
(2.119)
in the current configuration becomes:
(2 x
2 x2
) + (
2
2 x1 + 2 x 2
=2
(2.120)

Ciudad Real - Spain
Draft
179
2
3x12 + 3 x 2 4 2 x1 x 2 = 1 (an ellipse equation)
(2.121)
(see Figure 2.7).

2
Current configuration
1,5
1
0,5
Reference configuration
x2
0
-2
-1
-0,5
-1
-1,5
-2
x1
Figure 2.7
d) The right Cauchy-Green deformation tensor and the Green-Lagrange strain tensor are
given, respectively, by:
C = FT F
E=
1
(C 1)
2
(2.122)
Then, the C -components are:
C ij =
1
2
2
0
2
2
1
0
0
1
2
2
1
2
2
0
1
0
0 =
3
2
2
0
2
3
2
0
(2.123)
And the eigenvalues of (2.123) are:

C1 =
3
3
+ 2 2.914 ; C 2 = 2 0.086 ; C 3 = 1
2
2
(2.124)
The E -components are:

Ciudad Real - Spain
Draft
180
1
C ij ij
2
3
2
2
1
3
= 2
2
2
0
0
E ij =
0
1
1 0 0
1
0 0 1 0 = 2 2
4

0
0 0 1
1
2 2
1
0
0
0
(2.125)
The eigenvalues of E can be obtained as follows:
1+ 2 2
1 =
7
4
2
=0
2 16
1 2 2
2 =
4
(2.126)
1+ 2 2
1 2 2
0.957 ; E 2 =
0.457 ; E 3 = 0
4
4
(2.127)
4
2
2
2
2 =0
1
Then, the three eigenvalues of E are:

E1 =
Problem 2.35
Let us consider the following equations of motion:
x1 = X 1 +
1
X2
2
x2 =
;
r
1
X1 + X 2
2
x3 = X 3
(2.128)
a) Obtain the displacement field ( u ) in the Lagrangian and Eulerian descriptions;

b) Determine the material curve in the current configuration for a material circle defined in
the reference configuration as:
2
X 12 + X 2 = 2
X3 = 0
c) Obtain the components of the right Cauchy-Green deformation tensor and the GreenLagrange strain tensor;
d) Obtain the principal stretches.
Solution:
The deformation gradient is given by:
2 1 0
xi
1
= 1 2 0
Fij =
X j 2
0 0 2
J = F = 0.75
And by comparing this with the equations of motion in (2.128) we have:

2 1 0 X 1
x1
x = 1 1 2 0 X
2
2 2
0 0 2 X 3
x3

x i = Fij X j
So, we can verify that the proposed example is a case of homogeneous deformation in
r r
which c = 0 . The inverse form of the above equation is given by:

Ciudad Real - Spain
Draft
4 2 0 x1
X1
X = 1 2 4 0 x
2
2 3
0
X3
0 3 x 3

The displacement field is defined by

Lagrangian displacement become:
181
4
2
X 1 = 3 x1 3 x 2
2
4
(2.129)
X 2 = x1 + x 2
3
3
X 3 = x3
r
r r
u = x X , after which the components of the
r
1
1
u1 ( X , t ) = x1 X 1 = X 1 + X 2 X 1 = X 2
2
2
r
1
1
u i = xi X i u 2 ( X , t ) = x 2 X 2 = X 1 + X 2 X 2 = X 1
2
2
r
u 3 ( X , t ) = x3 X 3 = 0
(2.130)
The components of the Eulerian displacement can be obtained by substituting the Eulerian
description of motion (2.129) into (2.130), the result of which is:
r r
r
1
u1 ( X ( x, t ), t ) = 2 X 2 ( x , t ) =
r r
r
1
u 2 ( X ( x , t ), t ) = X 1 ( x , t ) =
2
r r
r
u ( X ( x , t ), t ) = u ( x , t ) = 0
3
3
r
1 2
4
3 x1 + 3 x 2 = u1 ( x , t )
2
r
1 2
4
3 x1 + 3 x 2 = u 2 ( x , t )
2
(2.131)
2
The particles belonging to the circle X 12 + X 2 = 2 in the reference configuration will form
a new curve in the current configuration which is defined by:
2
X 12
2
X2
2
4
4
2
2
= 2 x1 x 2 + x1 + x 2 = 2 20 x12 32 x1 x 2 + 20 x 2 = 18
3
3
3
3
which is an ellipse equation (Figure 2.7 shows the material curve in different
configurations).
The components of C and E can be obtained by using the definitions C = F T F and
E=
1
(C 1) :
2
C ij = Fki Fkj
1
E ij = C ij ij
2
2 1 0 2 1 0 1.25 1 0
1
C ij = 1 2 0 1 2 0 = 1 1.25 0
4
0 0 2 0 0 2 0
0 1
1.25 1 0 1 0 0 0.125 0.5 0
1
E ij = 1 1.25 0 0 1 0 = 0.5 0.125 0

2
0
0 1 0 0 1 0
0
0

In the principal space of C its components are given by:

Ciudad Real - Spain
Draft
182
2
1
C ij = 0
0
0
2
3
0
22
0
C ij = 0
0
0
3
0
2
0
where i show the principal stretches. Therefore, to calculate these we need to obtain the
C eigenvalues:
1.25 C
1
C1 = 2.25
= 0 C 2 2.5C + 0.5625 = 0
1
1.25 C
C 2 = 0.25
2
1
C ij = 0
0
0
22
0
2.25
0
0

0.25 0
= 0
2 0
0
1
3
0
0
0
0
0
2
0
1.5 0 0

= 0 0.5 0
0 1
3 0

0
0
2.0
material curve
1.5
Reference Conf.
Current Conf.
1.0
0.5
x2
0.0
-2
-1
-0.5
-1.0
-1.5
-2.0
x1
Figure 2.8: Material curve.

Problem 2.36
Show that
r
X
[(detF ) F T ] = 0
r
(2.132)
r
r
Hint: The Nansons formula da = J F T dA , or da = da n = J F T N dA .
Solution:

Considering the Nansons formula in indicial notation da n i = J Fki 1N k dA , with J = det (F )

we apply the surface integral:

Ciudad Real - Spain
Draft
183
1
ki N k dA
n da = J F
i
(2.133)
S0
Note that, if we consider a function f , it holds that:
f da =
f ,i dV =
f
dV
xi
Denoting by f = 1 , we obtain:
da = 0 i
Returning to equation (2.133), and applying the divergence theorem to the integral on the
right of the equation we obtain:
n da = 0 = J F
i
1
ki N k dA =
1
ki
k
dV0 =
V0
S0
(J F ),
r
X
V0
J Fki 1 dV0 = 0 i
X k
(2.134)
[(detF ) F T ] dV0 = 0
r
V0
Then, if the above volume integral is valid for the entire volume we can guarantee that is
also valid locally, i.e.:
[(detF ) F T ] = 0
r
r
X
Problem 2.37
r r
sym
(2.135)
r r
sym
&
r &
r&
Show that E = F T X u( X , t)
and b) D = x u( x , t) , where E is the GreenLagrange strain tensor and D is the rate-of-deformation tensor.
Solution:
) [
] [
& D E = D 1 F T F 1 = 1 F T F + F T F = 1 (F T F )T + (F T F ) = F T F
&
&
&
&
&
E
2
2
Dt
Dt 2
sym
Note that:
&
r r
D xi ( X , t )
&
&
= Dxi ( X , t ) = [u i ( X , t )] = u i ( X , t ) = r u( X , t )
&
Fij =
X
ij
Dt X j X j
Dt
X j
X j
with that we demonstrate that:
& D E = FT F
&
E
Dt
sym
r r sym
r &
= F T X u( X , t )
b)
] [
r r sym
r T
r r
1
1 rr
T
sym
r
r
r&
l + ( l ) = x v + ( x v ) = ( x v ( x , t ))
= x u( x , t )
2
2
r r
r r
&
where we have considered v ( x , t ) = u( x , t ) .
D=l
sym

Ciudad Real - Spain
Draft
184
Problem 2.38
Consider the velocity field:
v1 = 5 x 2 + 2 x3
v 2 = 5 x1 3 x3
v = 2 x + 3 x
1
2
3
Show that this motion corresponds to a rigid body motion.

Solution:
At first we obtain the spatial velocity gradient
v1
x1
r
vi ( x, t ) v 2
l ij =
=
x j
x
1
v3
x1
v1
x 2
v 2
x 2
v3
x 2
(l ), whose components are given by:
v1
x3 0 5 2
v 2
= 5
0 3
x3
2 3
0
v3
x3
(2.136)
Remember that ( l ) can be decomposed into a symmetric ( D ) and antisymmetric ( W )

part, i.e. l = D + W = W . Since D = 0 , there is no strain during motion, i.e. a rigid body
motion.
Problem 2.39
Let us consider the following velocity field:
v1 = 3 x 2 + 1x 3
v 2 = 3 x1 5 x3
v = 1x + 5 x
1
2
3
Show that this motion corresponds to rigid body motion.

Solution: First we obtain the components of the spatial velocity gradient
v1
x1
r
vi ( x , t ) v 2
l ij =
=
x j
x
1
v3
x1
v1
x 2
v 2
x 2
v 3
x 2
(l ):
v1
x3 0 3 1
v 2
0 5 = l ijskew
= 3
x3
1 5
0
v3
x3
Taking into account that l can be decomposed into a symmetric ( l sym D ) and an
antisymmetric ( l skew W ) part, i.e. l = D + W , we can thus conclude that D = 0 , which is
a characteristic of rigid body motion.
Problem 2.40
The displacement field components are given by:

Ciudad Real - Spain
Draft
185
u1 = 3 X 12 + X 2
2
u 2 = 2 X 2 + X 3
2
u 3 = 4 X 3 + X 1
Obtain the vector dx (current configuration) correspondent to the vector in the reference
r
configuration represented by dX at the point P(1,1,1) .
X 3 , x3
dX 1
dX k = dX 2
dX 3
r
dX
X 2 , x2
X 1 , x1
Solution:
To determine the vector dx we need to obtain the deformation gradient F . The

deformation gradient components can be obtained by using the relationship:
Fij = ij +
u i
X j
1
1 + 6 X 1
0
Fij =
1 + 4X 2
1
0
1 + 8X 3
0
1
The deformation gradient components evaluated at the point P(1,1,1) are:

Fij
7 1 0
= 0 5 1
1 0 9
Then, the vector components dx i are given by:

dx i = Fij dX j
dx1 7 1 0 dX 1 7 dX 1 + dX 2
dx = 0 5 1 dX = 5dX + dX
2
3
2
2
dx 3 1 0 9 dX 3 dX 1 + 9 dX 3
Problem 2.41
Consider a continuum in which the displacement field is described by the following
equations:
u1 = 2 X 12 + X 1 X 2
2
u 2 = X 2
u = 0
3

Ciudad Real - Spain
Draft
186
By definition, a material curve is always formed by the same particles. Let OP and OT be
material lines in the reference configuration, where O( X 1 = 0, X 2 = 0, X 3 = 0) ,
P ( X 1 = 1, X 2 = 1, X 3 = 0) and T ( X 1 = 1, X 2 = 0, X 3 = 0) . Find the material curves in the
current configuration. Also find the deformation gradient.
Solution:
a) The equations of motion can be obtained by means of the displacement field, i.e.:
u i = xi X i
x1 = X 1 + 2 X 12 + X 1 X 2
2
substituti x 2 = X 2 + X 2
ng
the values of u1 ,u 2 ,u 3
x = X
3
3
x1 = u1 + X 1
x2 = u 2 + X 2
x = u + X
3
3
3
Then, to obtain the material curve, one need only substitute the material coordinates with
the particles belonging to the line OP in the equations of motion, (see Figure 2.9). Notice
that the material curve OP in the current configuration is no longer a straight line, but the
line OT is still a straight line in the current configuration (see Figure 2.10).
The components of the deformation gradient can be obtained as follows:
F jk
x1
X 1
x
= 2
X
1
x 3
X 1
x1
X 2
x 2
X 2
x 3
X 2
x1
X 3 (1 + 4 X + X )
X1
1
2
x 2
0
1 + 2X 2
=
X 3

0
0
x 3
X 3
0
0
2.5
material curve
x2
1.5
1
0.5
Current Conf.
Reference Conf.
O0
0.5
1.5
2.5
3.5
4.5
x1
Figure 2.9: Deformation of the material curve OP .

Ciudad Real - Spain
Draft
187
Reference Conf.
x2
0.1
0.08
Reference Conf.
0.06
0.04
0.02
0
O0
0.5
1.5
2.5
3.5
x2
Current Conf.
0.1
0.08
0.06
0.04
0.02
0
Current Conf.
O0
0.5
1.5
2.5
3.5
x1
Figure 2.10: Deformation of the material curve OT .

Problem 2.42
Starting from the definition
r
&
r
J = J x v , is valid.
Solution: Considering that Fij =
DF
D
[det (F )] = ij cof Fij , show that the equation
Dt
Dt
( )
x i
, the material time derivative of F det (F ) is given
X j
by:
D
[det ( F )] = D x i ( X , t ) cof Fij = D x i ( X , t ) cof Fij = D (v i )cof Fij
X j
X j
Dt
Dt
Dt
X j
r
and considering that v i ( x ( X , t ), t ) , we can state that:
( )
( )
( )
D
[det (F )] = vi x k cof Fij
x k X j
Dt
( )
By referring to the definition of the cofactor: [cof (Fij )]T = (Fij )1 det (Fij ) , we can also state
the following is valid:
D
[det (F )] = vi xk Fij
Dt
xk X j
( )
( )
det Fij =
vi
Fkj F ji
x k
( )
( )
det Fij =
vi
v
ki det Fij = i det Fij
x k
xi
( )
( )
= Jvi ,i
An alternative solution is presented in Problem 1.113 in Chapter 1.

Ciudad Real - Spain
Draft
188
Problem 2.43
r
Let dx be a differential line element in the current configuration. Find the material time
r
derivative of dx .
Solution:
r
r
r
r
D
D
D r D
( F dX ) =
( F ) dX + F
( dX ) = l 123
F dX
dx =
r
Dt 3
Dt
Dt
Dt
1 24
4
dx
r
0
r
r r
r
= l dx x v dx
And, whose components are represented by:

r
v ( x , t )
D r
dx = v i , k dx k = i
dx k
x k
Dt i
Problem 2.44
Let us consider the equations of motion:
x1 = X 1 + 4 X 1 X 2
2
x2 = X 2 + X 2
2
x3 = X 3 + X 3
Find the Green-Lagrange strain tensor ( E ).

Solution:
Referring to the E equation:
E=
1
( F T F 1)
2
E ij =
1
( Fki Fkj ij )
2
(2.137)
where the components of F are derived as:

x1
X 1
x k x 2
Fkj =
=
X j X 1
x 3
X 1
x1
X 2
x 2
X 2
x 3
X 2
x1
X 3 (1 + 4 X )
4X1
2
x 2
0
1 + 2X 2
=
X 3

0
0
x 3
X 3
1+ 2X 3
And,
Fki Fkj
0
0 (1 + 4 X 2 )
4X1
(1 + 4 X 2 )
4X
1 + 2X 2
0
0
1 + 2X 2
=
1
0
0
1 + 2X 3
0
0
2
(1 + 4 X 2 )
(1 + 4 X 2 ) 4 X 1
0
2
2
0
= (1 + 4 X 2 ) 4 X 1 ( 4 X 1 ) + (1 + 2 X 2 )
0
0
(1 + 2 X 3 ) 2
0
0
1 + 2X 3
Then substituting the above into the equation in (2.137) we obtain:

Ciudad Real - Spain
Draft
189
(1 + 4 X 2 ) 2 1
(1 + 4 X 2 ) 4 X 1
0
1
2
2
0
E ij = (1 + 4 X 2 ) 4 X 1 ( 4 X 1 ) + (1 + 2 X 2 ) 1
2
2
0
0
(1 + 2 X 3 ) 1
Problem 2.45
Obtain the principal invariants of E in terms of the principal invariants of C and b .
Solution:
The principal invariants of E are given by:
I E = Tr ( E )
II E =
1 2
I E Tr ( E 2 )
2
III E = det ( E )
1
2
Considering E = (C 1) , the principal invariants can also be expressed as follows:

The First Invariant:
1
1
1
1
I E = Tr ( E ) = Tr (C 1) = Tr (C 1) = [Tr (C ) Tr (1) ] = (I C 3)
2
2
2
2
The Second Invariant:

II E =
1 2
I E Tr ( E 2 )
2
where
2
2
IE
1 2
1
= (I C 3) = I C 6 I C + 9
4
2
) [ ( )
1
1
1
1
Tr ( E ) = Tr (C 1) = Tr (C 1) 2 = Tr C 2 2C + 1 = Tr C 2 2 Tr (C ) + Tr (1)
4
4
4
2
1
= Tr C 2 2 I C + 3
4
2
[ ( )
]
The term Tr (C ) can be obtained as follows:
2
C C = C
C12
C ij2 = 0
0
0
2
C2
0
It is also true that:
2
2
0 Tr C 2 = C12 + C 2 + C 3
2
C3
( )
2
2
2
I C = (C1 + C 2 + C 3 ) = C12 + C 2 + C 3 + 2 C1 C 2 + C1 C 3 + C 2 C 3
1444 24444
4
3
2
II C
2
2
2
C12 + C 2 + C 3 = I C 2 II C
Therefore we have:
Tr ( E 2 ) =
1 2
I C 2 II C 2 I C + 3
4
Whereupon, the second invariant can also be expressed as:

Ciudad Real - Spain
Draft
190
II E =
1 1 2
1 2
1
4 I C 6 I C + 9 4 I C 2 II C 2 I C + 3 = 4 ( 2 I C + II C + 3)
2
The Third Invariant:

3
1
1
III E = det ( E ) = det (C 1) = det [(C 1)]
2
2
The term det[(C 1)] can also be expressed as:

C1 1
0
det (C 1) = 0
C2 1
0
0
0
= (C1 1)(C 2 1)(C 3 1)
C3 1
= C1C 2 C 3 C1C 2 C1C 3 C 2 C 3 + C1 + C 2 + C 3 1 = III C II C + I C 1
Then:
III E =
1
( III C II C + I C 1)
8
In short we have:
1
(I C 3 )
2
1
II E = ( 2 I C + II C + 3)
4
1
III E = ( III C II C + I C 1)
8
I C = 2I E + 3
IE =
II C = 4 II E + 4 I E + 3
III C = 8 III E + 4 II E + 2 I E + 1
Problem 2.46
Let = (I C , II C , III C ) be a scalar-valued tensor function, where I C , II C , III C are the
principal invariants of the right Cauchy-Green deformation tensor C . Obtain the
derivative of with respect to C and with respect to b . Check whether the following
equation is valid F ,C F T = ,b b or not.
Solution:
Using the chain rule of derivative we obtain:
,C =
(I C , II C , III C ) I C
II C
III C
=
+
+
C
I C C II C C
III C C
(2.138)
Considering the partial derivative of the invariants:

I C
II C
III C
=1 ,
= IC 1 C T = IC 1 C ,
= III C C T = III C C 1 , we can obtain:
C
C
C
,C =
,C
(I C 1 C ) + III C C 1
1+
I C
II C
III C
1
=
I + II I C 1 II C + III III C C
C
C
C
C
(2.139)
It is also true that:

Ciudad Real - Spain
Draft
191
1
,b =
I + II I b 1 II b + III III b b
b
b
b
b
(2.140)
We apply the dot product of the above equation with F on the left and with F T on the
right, i.e.:

T
T
1
T
F ,C F T =
I + II I C F 1 F II F C F + III III C F C F
C
C
C
C
(2.141)
And considering the following relationships:
F 1 F T = F F T = b
C = F T F F C F T = F F T F F T = b b = b2
And considering the relationship C 1 = F 1 b 1 F we conclude that:

C 1 = F 1 b 1 F F C 1 F T = F F 1 b 1 F F T = b 1 b
Then, the equation in (2.141) can be rewritten as follows:

1
F ,C F T =
I + II I C b II b + III III C b b
C
C
C
C
+
F , C F T =
I C 1
b+
III C b 1 b
II C
III C
I C II C
It is also valid that:
F , C F T =
I b 1
B+
III b b 1 b
+
III b
II b
I b II b
F ,C F T = , b b
Taking into account the equation (2.140) we can conclude that the equation ,b b = b ,b
is valid, indicating that the tensors ,b and b are coaxial.
Problem 2.47
Show that the Green-Lagrange strain tensor ( E ) and the right Cauchy-Green deformation
tensor ( C ) are coaxial tensors.
Solution:
Two tensors are coaxial if they have the same principal directions. Coaxiality can also be
demonstrated if the relation C E = E C holds.
Starting with the definition C = 1 + 2 E , we can conclude that:
C E = (1 + 2 E ) E = 1 E + 2 E E = E (1 + 2 E ) = E C
Thus, we can prove that E and C are coaxial tensors.

Ciudad Real - Spain
Draft
192
Problem 2.48
Obtain
the
&
E = F T D F
relationship
starting
from
the
definition
r
r
D
(ds ) 2 (dS ) 2 = dX 2 E dX . Get also the relationship between
(ds ) 2 and D .
Dt
Solution:
Taking the material time derivative of (ds ) 2 (dS ) 2 = dX 2 E dX we obtain:
D
(ds ) 2 (dS ) 2
Dt
The term
r
D
D r
dX 2 E dX
(ds ) 2
=
Dt
Dt
r
r
r
r
r
r
D r r
&
&
&
[dx dx ] = 2dX E dX + 2dX E dX + 2dX E dX
=
{
{
Dt
=0
=0
r
r
r D r
&
= 2 dx
[dx ] = 2dX E dX
Dt
=
D r
[dx ] can be expressed as follows:
Dt
D r
Dt [dx ]
r
D
=
F dX
Dt r
&
= F dX
r
= l F dX
D
D x k
[dx k ] =
Dt X i
Dt
D x k
Indicial
=
Dt X i
v k
dX i
=
X i
dX i
D x k
dX i =
dX i
DX i t

r
r
&
2 dX E dX
r D r
[dx ]
= 2dx
r
r Dt
= 2dx l F dX
r
r
= 2 F dX l F dX
r
r
= 2dX F T l F dX
We can apply the additive decomposition of the spatial velocity gradient ( l ) into a
symmetric ( D ) and an antisymmetric ( W ) part:
r
r
&
2 dX E dX
r
r
= 2 dX F T l F dX
r
r
= 2dX F T (D + W) F dX
r
r
r
r
= 2 dX F T D F dX + 2 dX F T W F dX
r
r
= 2 dX F T D F dX
r
r
r
r
r
r
Note that dX F T W F dX = dx W dx = W : (dx dx ) = 0 , since W is
r
r
antisymmetric tensor and (dx dx ) is a symmetric tensor. Then, we conclude that:
an
&
E = F T D F
D
(ds ) 2 and D as follows:
Dt
r
r
r
r
D
(ds ) 2 = 2dX F T D F dX = 2dx D dx
Dt
With that it is possible to relate

Ciudad Real - Spain
Draft
193
Problem 2.49
&
&
&
Obtain the material time derivative of the Jacobian determinant ( J ) in terms of ( E ), ( C ),
&
( F ).
Solution:
&
We starting from the relationship J = J Tr (D ) , where D is the rate-of-deformation tensor
&
&
which is related to E by means of the relationship D = F T E F 1 , then:
) (
&
&
&
J = J Tr (D) = J Tr F T E F 1 = J F T E F 1 : 1
In indicial notation we have:

&
&
&
&
&
& J
J = J Fki 1 E kp F pj1 ij = J Fki 1 F pi1 E kp = J ( F 1 F T ) : E = J C 1 : E = C 1 : C
2
&
&
The J can still be expressed in terms of F . To this end let us consider the following
1 &
&
&
&
equation E kp = (Fsk Fsp + Fsk Fsp ) . Then, J can also be expressed by:
2
J &
&
1 &
&
&
&
J = J Fki 1 F pi1 E kp = J Fki 1 F pi1 Fsk Fsp + Fsk Fsp = Fki 1 F pi1 Fsk Fsp + Fki 1 F pi1 Fsk Fsp
2
2
J
J &
&
&
&
&
&
= si Fki 1 Fsk + si F pi1 Fsp = Fks1 Fsk + F ps1 Fsp = JFts1 Fst = JFst Fts1
2
2
&
&
= JF T : F = JF : F T
In short, there are various different ways to express the material time derivative of the
Jacobian determinant:
J 1 &
&
C :C
= J F : F T
2
J
&
&
&
= J Tr (C 1 E ) = Tr (C 1 C ) = J Tr ( F F 1 )
2
&
&
J = J Tr (D) = J C 1 : E
where we have used the trace property: A : B = Tr ( A B T ) = Tr ( A T B ) in which A and B

are arbitrary second-order tensors.
Problem 2.50
The displacement field components are given by:
2
u1 = 0.1 X 2
u 2 = 0
u = 0
3
a) Is this motion possible? Justify the answer;

b) Obtain the right Cauchy-Green deformation tensor;
c) Find the current vectors related to the material vectors b = 0.01e 1 and c = 0.015e 2 , said
vectors are at the point P (1,1,0) in the reference configuration;
r
d) Obtain the stretches of the vectors b and c , at the point P (1,1,0) ;

r
e) Find the angle variation defined by the two vectors b and c .

Ciudad Real - Spain
Draft
194
Solution:
a) A motion is possible if the Jacobian determinant is positive. The deformation gradient is
given by:
1 0 0 0 0 .2 X 2
u i
Fij = ij +
= 0 1 0 + 0
0

X j
0 0 1 0
0
0 1 0 .2 X 2
0 = 0
1

0 0
0

0
0
The determinant is Fij = J = 1 > 0 . Then, the motion is possible.

b) The right Cauchy-Green deformation tensor is defined as C = F T F , then the
0 0 1 0.2 X 2 0 1
0 .2 X 2
0
1
0 .2 X
0
= 0 .2 X
2
2
1 0
1
0
0 .2 X 2 + 1 0
C ij =
2
2
0
0 1 0
0
1 0
0
1
c) The vector b = 0.01e 1 at the point P(1,1,0) deforms according to:

r
r
b = F P b
b1 1 0.2 1 0 0.01 0.01

b = 0
1
0 0 = 0
2
b 0
0
1 0 0
3
and the vector c = 0.015e 2 in the current configuration becomes:
c 1 1 0.2 1 0 0 0.003
c = 0
1
0 0.015 = 0.015
2
c 0
0
1 0 0
3
d) The stretch can be obtained as follows:

r
b =
r
b
r
b
0.012
=1
0.01
and the stretch of c is given as:

r
c
r
c
r
c
0.003 2 + 0.015 2
= 1.0198 1.02
0.015
Alternative solution: Taking into account that M = M C M and by evaluating C at the
point P we obtain:
1
C ij ( X 1 = 1, X 2 = 1, X 3 = 0) = 0.2 X 2
0
0 .2
+ 1 0
0
1
0 .2 X 2
2
2
X2
0 .2 0
1
0.2 1.04 0
=
0
0 1
Then, by applying b = b C b and c = c C c we can obtain:

Ciudad Real - Spain
Draft
0 . 2 0 1
1
0.2 1.04 0 0 = 1
= [1 0 0 ]

0
0
1 0

0.2 0 0
1
0.2 1.04 0 1 = 1.04
= [0 1 0 ]

0
0
1 0

195
b = 1
c = 1.0198
e) In the current configuration the angle between the vectors b and c can be obtained
according to the relation:
r r
b c
cos = r r
b c
cos =
(0.01e 1 + 0e 2 + 0e 3 ) (0.003e 1 + 0.015 e 2 + 0e 3 )

0.01
0.003 + 0.015
0.00003
0.01 0.000234
= 0.196116135
= arccos(0.196116135) 78.69
In the reference configuration the angle between these two vectors is 90 , then angle
variation is:
= 90 78 .69 = 11 .3
Alternative solution: Given two directions in the reference configuration represented by their
unit vectors M and N , the angle formed by these unit vectors in the current configuration
(after motion) is given by:
cos =
M C N
M C N
=
MN
M C M N C N

Denoting by M = b and N = c it fulfills that:
0 .2 0 0
1
C c = [1 0 0] 0.2 1.04 0 1 = 0.2

0
0
1 0
Then,
cos =
0 .2
bC c
bC c
=
=
= 0.196116135
b c
C b c C c

1 1.04

Problem 2.51
Obtain an equation for mass density in terms of the third invariant of the right CauchyGreen deformation tensor, i.e. 0 = 0 ( III C ) .
Solution:
Starting by the definition:
0 ( X ) = ( x, t) J
and considering that the third invariant III C = det (C ) = det ( F T F ) = J 2 , we obtain
J = III C , then:
Ciudad Real - Spain
Draft
196
0 =
III C
(2.142)
Problem 2.52
At a certain moment, the displacement field of a continuous medium is:
u1 = (a1 1) X 1
u 2 = (a 2 1) X 2 + a1X 1
u 3 = (a 3 1) X 3
where is a constant. Determine a1 , a 2 and a 3 knowing that the solid is incompressible,

that a segment parallel to the X 3 -axis does not stretch and that any element area defined in
the plane X 1 X 3 remains unchanged.
Solution:
r
Based on the definition of the displacement field, i.e. u = x X , we obtain:

u1 = x1 X 1 = (a1 1) X 1
x1 = a1 X 1
u 2 = x 2 X 2 = (a 2 1) X 2 + a1X 1
u 3 = x 3 X 3 = (a 3 1) X 3
x 2 = a 2 X 2 + a1X 1
x3 = a3 X 3
Then, the equations of motion are:

x1 = a1 X 1
x 2 = a 2 X 2 + a1X 1
x = a X
3 3
3
0
x1 a1

x 2 = a1 a 2
x 0
0
3
0 X 1

0 X 2 (homogeneous deformation
a3 X 3

which is possible to establish that F = a1 a 2 a 3 > 0 .

By means of the incompressibility condition dV = F dV0 F J = 1 , the following
relationship is true:
a1 a 2 a 3 = 1
By the fact that a segment parallel to the X 3 -axis, e.g. Mi = [0 0 1], does not stretch that
implies that the stretching according to this direction is unitary, i.e. M = 1 , thus
M = 1 + 2M E M = 1 + 2 E 33 = 1
The components of the Green-Lagrange strain tensor ( E =

a1
a2
E 33 = 0
1 T
F F 1 ) are given by:
2
0 1 0 0
0 0 1 0

0
a 3 0 0 1

2
2 2
a1 + a1 1 a1 a 2
0
1
2
0
=
a1 a 2
a2 1
2
2
0
0
a 3 1
a1
1
E ij = 0
2
0 a1
0
a a
0 1
2
0
a3 0
thus:
2
E 33 = a 3 1 = 0
a 3 = 1
Any element area on the plane X 1 X 3 does not change

Ciudad Real - Spain
Draft
0
x1 a1

x 2 = a1 a 2
x 0
0
3
197
0 X 1

0 X 2
a3 X 3

with Ni(1) = [1 0 0] and Ni(3) = [0 0 1] we obtain:
n i(1)
0
a1
a a
= 1
2
0
0
0 1 a1

0 0 = a1
a 3 0 0
n i(3)
0
a1
a a
= 1
2
0
0
0 0 0

0 0 = 0
a 3 1 a 3

Then, the area in the current configuration is obtained as follows:
e1
r (1) r (3)
n n = a1
0
e2 e3
a1 0 = a1e1 a1 a 3 e 2 + 0e 3
0
a3
r
and its module does not change N (1) N (3) = n (1) n (3) = 1 :
r
r
2
2
n (1) n (3) = 1 = (a1 ) 2 + (a1 a 3 ) 2 a12 a 3 2 + a12 a 3 = 1
2
We have previously obtained that a 3 = 1 , with that we obtain:
2
2
a12 a 3 2 + a12 a 3 = 1
a12 2 + a12 = 1 a12 =
1
1
a1 =
2
(1 + )
(1 + 2 )

a1 =
1
(1 + )
2
a 2 = (1 + 2 )
a3 = 1
Problem 2.53
Consider the solid shown in Figure 2.11 which is subjected to a homogenous deformation.
r r
a) Obtain the general expression of the material displacement field U ( X , t ) in

function of the material displacement gradient tensor J .
r r
b) Obtain U ( X , t ) knowing that also holds the following boundary conditions:
r
r
u 2 ( X , t ) = u3 ( X , t ) = 0
X 1 , X 2 , X 3
u1 ( X 1 = 0, X 2 , X 3 , t ) = 0
u1 ( X 1 = L, X 2 , X 3 , t ) =
c) Justify the possible values (positive and negative) that can take .
d) Calculate the material and spatial strain tensors and the infinitesimal strain tensor.

Ciudad Real - Spain
Draft
198
x3
x1
x2
Figure 2.11
Solution:
A homogeneous deformation is characterized by F ( X , t ) = F (t ) . In addition, we know

that:
r
r
Homogeneous
F ( X , t ) = 1 + J ( X , t ) deformation F (t ) = 1 + J (t )
where J is the material displacement gradient tensor. Note that the homogenous
deformation is not dependent on the vector position, with that we can obtain:
r r
r
r r
u( X , t )
J (t ) =
r
J (t ) dX = du( X , t )
X
r
where c (t ) is the constant of integration. Then:
r r
r r
u( X , t ) = J (t ) X + c (t )
r r
r r
u( X , t ) = J (t ) X + c (t )
In components:
u1 J 11 X 1 + J 12 X 2 + J 13 X 3 c1

u 2 = J 21 X 1 + J 22 X 2 + J 23 X 3 + c 2
u J X + J X + J X c
32 2
33 3
3
3 31 1
b) From the conditions in paragraph b) we can conclude that:

r
condition 1) u 2 ( X , t ) = u 3 ( X , t ) = 0
X 1 , X 2 , X 3 :
u1 J 11 X 1 + J 12 X 2 + J 13 X 3 c1
J 21 = 0; J 22 = 0; J 23 = 0, c 2 = 0

u 2 = 0 = J 21 X 1 + J 22 X 2 + J 23 X 3 + c 2
u = 0 J X + J X + J X c J 31 = 0; J 32 = 0; J 33 = 0, c3 = 0
32 2
33 3
3
31 1
3
condition 2) u1 ( X 1 = 0, X 2 , X 3 , t ) = 0 :
u1 = 0 J 11 X 1 + J 12 X 2 + J 13 X 3 c1

0
+ 0 {J 12 = 0; J 13 = 0, c1 = 0
u2 =
0
u
0

3
condition 3) u1 ( X 1 = L, X 2 , X 3 , t ) =

Ciudad Real - Spain
Draft
199
u1 J 11 L 0

u 2 = 0 + 0 J 11 =
L
u = 0 0
3

Hence, the components of the displacement gradient are:
L
J ij = 0
0 0
0 0
0 0
And the displacement field components are:
L X1
r r
r
r r
components
u( X , t ) = J (t ) X + c (t ) u i ( X , t ) = 0
c) The motion is possible and has physical meaning if F > 0 :

1 + L
components
F (t ) = 1 + J (t ) Fij = 0
0 0
1 0 F = 1 + > 0 > L
L
0 1
d)
The material strain tensor (the Green-Lagrange strain tensor):
E=
1 T
F F 1
2
1 2
+
2
L 2 L
0
E ij =
components

0 0
0 0
0 0
The spatial strain tensor (the Almansi strain tensor):
e=
1
1 F FT
2
components

1 2
1 0 0
+
2
L 2 L
eij =
2
0 0 0
1 + 0 0 0
The infinitesimal strain tensor:
L
ij = 0

Ciudad Real - Spain
Draft
0 0
0 0
0 0
200
Problem 2.54
The tetrahedron shown in Figure 2.12 undergoes homogeneous deformation ( F = const. )
with the following consequences:
1. The points O , A and B do not move;
2. The solid volume becomes " p" times the initial volume;
p
3. The length of the segment AC becomes
times the initial length;
4. The angle AOC becomes 45 .

Justify why we can not use the infinitesimal deformation theory;
Obtain the deformation gradient, and the possible values of " p" and displacement
field in material and spatial descriptions;
c)
Draw the deformed solid.
a)
b)
x3
C
x2
A
x1
Figure 2.12.
Solution:
a) The angle AOC = 90 becomes 45 , so we are not dealing with a small deformation,
since in the case of small deformation << 1 , and in this problem we have
<<
0.7854 ;
4
b) We have a case of homogeneous deformation. Then, the equations of motion are given
by:
r r
r
x = F (t ) X + c (t )
x1 F11

x 2 = F21
x F
3 31
F12
F22
F32
F13 X 1 c1

F23 X 2 + c 2
F33 X 3 c3

The point O( X 1 = 0, X 2 = 0, X 3 = 0) does not move:

0 F11

0 = F21
0 F
31
F12
F22
F32
F13 0 c1

F23 0 + c 2

F33 0 c3
c1 0

c 2 = 0
c 0
3
The point A( X 1 = a, X 2 = 0, X 3 = 0) does not move:

Ciudad Real - Spain
Draft
a F11

0 = F21
0 F
31
F12
F22
F32
F13 a

F23 0
F33 0

201
a aF11

0 = aF21
0 aF
31
F11 = 1
F21 = 0
F = 0
31
The point B( X 1 = 0, X 2 = a, X 3 = 0) does not move:

0 1 F12

a = 0 F22
0 0 F
32

F13 0

F23 a
F33 0

0 aF12

a = aF22
0 aF
32
F12 = 0
F22 = 1
F = 0
32
Gathering the above information, we have:

1 0 F13
Fij = 0 1 F23
0 0 F33
F = F33 > 0
The volume of the solid becomes " p" times the initial volume. The relationship between the initial
(reference) volume and the current (final) volume is given by:
dV = F dV0
dV = F dV
V final = F Vinitial = F33Vinitial
where we have considered the homogeneous deformation case. With this, we conclude that
F33 = p
(The length of segment AC becomes
p
2
times the initial length). As we are dealing with a
homogeneous deformation, a line in the reference configuration will remain a line in the
current configuration.
The point C ( X 1 = 0, X 2 = 0, X 3 = a ) moves to:
C
x1 1 0 F13 0
C

x 2 = 0 1 F23 0
x C 0 0 p a

3
C
x1 aF13
C
x 2 = aF23
x C ap
The length of segment AC in the reference configuration is L AC = a 2 . The vector that

connect the points A A( x1 = a, x 2 = 0, x3 = 0) and C ( x1 = aF13 , x 2 = aF23 , x3 = ap) in the
current configuration is given by:
AC = (aF13 a )e1 + (aF23 )e 2 + (ap)e 3
and its magnitude is:

AC = l AC = (a ( F13 1)) 2 + (aF23 ) 2 + (ap) 2 = a ( F13 1) 2 + ( F23 ) 2 + ( p) 2
Using the information provided by the problem l AC =

l AC =
a ( F13 1) 2 + ( F23 ) 2 + ( p ) 2 =
p
2
L AC , we obtain:
p
2
p
2
L AC
a 2
( F13 1) 2 + ( F23 ) 2 + ( p ) 2 = p
Ciudad Real - Spain
Draft
202
thus
( F13 1) 2 + ( F23 ) 2 + p 2 = p 2
( F13 1) 2 + ( F23 ) 2 = 0
F13 = 1
F23 = 0
Then, the deformation gradient components are:

1 0
Fij = 0 1
0 0
1
0
The angle AOC changes to 45 .

dX i(1) = [1 0 0]
dX i( 2 )
= [0 0 1]
dxi(1) = Fij dX (j1)
dx1( 2) 1
( 2)
dxi( 2 ) = Fij dX (j 2 )
dx 2 = 0
dx ( 2) 0
3
r
r
dx (1) dx ( 2 )
cos( AOC ) = cos(45 ) = r (1) r ( 2 ) =
dx
dx
dx1(1) 1 0 1 1 1
(1)

dx 2 = 0 1 0 0 = 0
dx (1) 0 0 p 0 0

3
0 1 0 1

1 0 0 = 0
0 p 1 p

2
2
where dx (1) = 1 , dx ( 2) = 1 + p 2 , dx (1) dx ( 2) = 1 . Then:

1
1+ p
2
2
p = 1
As the Jacobian determinant must be greater than zero F = p > 0 , this implies that p = 1 :
1 0 1
Fij = 0 1 0
0 0 1
The equations of motion become:

x1 1 0 1 X 1 X 1 + X 3

x 2 = 0 1 0 X 2 = X 2
x 0 0 1 X X
3
3
3
The material displacement field becomes:

r r
r r
u( X , t ) = x X
u1 X 1 + X 3 X 1 X 3

u 2 = X 2 X 2 = 0
u X
X 0
3
3
3
The spatial displacement field becomes:

Ciudad Real - Spain
Draft
203
u1 x3

u 2 = 0
u 0
3
c)
x3
C
C
x1 aF13 a
C

x 2 = aF23 = 0
x C ap a

3
a
a
B = B
x2
A = A
x1
Problem 2.55
Consider the following equations of motion:
x1 = X 1
x 2 = X 2 X 3
x 3 = X 3 + X 2
a) Obtain the deformation gradient, the right Cauchy-Green deformation tensor, the left
Cauchy-Green deformation tensor, the Green-Lagrange strain tensor and the Almansi
strain tensor. Check whether this case represents a homogeneous deformation.
b) Obtain the right stretch tensor, the spin tensor of polar decomposition and the principal
space of the left Cauchy-Green deformation tensor of the polar decomposition.
c) Obtain the final length of an initial length element equal to 2 which is in the X 3 direction, and the angular distortion of an initial angle 30 which is in the plane X 1 X 2 .
d) Obtain the strain tensor by considering the small deformation regime.
Solution:
a) The deformation gradient ( F )
x1
X 1
xi x 2
Fij =
=
X j X 1
x3
X 1

Ciudad Real - Spain
x1
X 2
x 2
X 2
x3
X 2
Draft
x1
X 3 1 0
0
x 2
= 0 1
X 3
0 1
x3
X 3
204
In general we have dx = F dX , and if we are dealing with a homogeneous deformation (a

r r
r
particular case of motion) the relationship x = F X + c holds, a fact that can be checked
r r
by means of the equations of motion in matrix form with c = 0 :
0 X1
x1 1 0
x = 0 1 X
2
2
x 3 0 1 X 3

The right Cauchy-Green deformation tensor ( C = F T F ):

0 1 0
0 1
0
0
1 0
0 1 0 1 = 0 1 + 2
0
C ij = Fki Fkj =
0 1 0 1 0
0
1+ 2
The left Cauchy-Green deformation tensor ( b = F F T )

0 1 0
0 1
0
0
1 0
0 1 0 1 = 0 1 + 2
0
bij = Fik F jk =
0 1 0 1 0
0
1+ 2
1
2
The Green-Lagrange strain tensor ( E = (C 1) ) and the Almansi strain tensor

1
2
( e = (1 b 1 ) ) are defined by their components as follows:

1
0
0 1 0 0
0 0

1
1
1
2
E ij = (C ij ij ) = 0 1 +
0 0 1 0 = 0 2
2
2
2
0
0 0
0
1 + 2 0 0 1

0
1 0 0 1

1
1
1
1
eij = ( ij bij ) = 0 1 0 0
2
2
1+ 2
0 0 1
0
0
0
0
0
2
= 1 0
0
2 1 + 2
0
0
2
1 +
0
0
1+ 2
We can check the results by the relationship E = F T e F :
0
0 0
1 0
1
2
0
E ij = 0 1
1+ 2
2
0 1
0
0
0 1 0
0
0 0
0 1 = 1 0 2
0
2
0 0
1
2 0
1+ 2
0
0
b) According to the format of the Cartesian components of C , we can verify that the
original space is already the principal space of C , i.e. the principal directions are
N i(1) = [1 0 0] , N i(1) = [0 1 0] , N i(1) = [0 0 1] . By definition, the right stretch tensor is

given by U = C , and its components are:

Ciudad Real - Spain
Draft
U ij = 0
0
inverse
0 U ij1
1 + 2
1+
0
205
= 0
0
1
1+ 2
0
2
1+
By means of the right polar decomposition ( F = R U R = F U 1 ), we obtain:
R ij = Fik U 1
kj
0 1
1 0
= 0 1 0
0 1
0
1
1+ 2
0
1
0
=
1+ 2
2
1+
1+ 2
0
0
1
1
Note that by means of the format of the Cartesian components of b indicate that the
principal directions are [1 0 0] , [0 1 0] , [0 0 1] , but this is not the principal
directions of b related to the polar decomposition. Note that there are two equal
eigenvalues related to the directions [0 1 0] , [0 0 1] , then any direction in the plane
x 2 x 3 is a principal direction.
X 2 , x2
Any direction on the plane

x 2 x 3 is a principal
direction of b
The principal direction n (1) is

unique, associated with the
eigenvalue b1 = 1 .
X 1 , x1
X 3 , x3
Figure 2.13: Principal space of b .

Remember that the polar decomposition is unique, i.e. there is one principal base b for the
polar decomposition associated with N( a ) . By means of the relation n ( a ) = R N ( a ) we can

obtain the principal base of b for the polar decomposition:
i
n ( 2) = R N(2)
i
n ( 3)
= R N ( 3)

Ciudad Real - Spain
1+ 2
=
0
1+ 2 0
1+ 2
=
0
1+ 2 0
Draft
0 0
1
1 =

2
1 0 1 +

0
1
0 0
1
0 =

2
1 1 1 +

0
1
0
1

0

206
In addition, we can check that the relation R = n ( a ) N ( a ) holds:

a =1

R ij = n i(1)N (j1) + n i( 2) N (j2 ) + n i(3) N (j3)
1
1
= 0[1 0 0] +

1+ 2
0

0
1 [0

1 0 0
0 0
1
0 0 0 +
0 1
=
2
1 + 0
0 0 0
1+ 2
1
=
0
1+ 2 0
1 0] +
1
1+ 2
0
1
0 +
1+ 2
0
0
0
0
[0 0 1]
0 0
0
0 1
1
1
c) By means of the stretch definition according to the direction M , i.e. M
r
dx
ds
,
= r =
dS
dX
and considering that the stretch is not dependent on line integral (homogeneous
deformation), it holds that:
L final = ds = M dS = M dS = M Linitial
The stretch according to X 3 -direction is given by:

X = C 33 = 1 + 2 E 33 = 1 + 2
3
Then:
2
L final = M dX 2 = 1 + 2 ( Linitial ) = 2 1 + 2
As we are dealing with a homogeneous deformation, a line in the reference configuration

remains a line in the current configuration, (see Figure 2.14).
X 3 , x3
A
L final
2
A
Linitial = 2
X 1 , x1

Ciudad Real - Spain
x1A 1 0 0 X 1A
A
A
x 2 = 0 1 X 2
x3A 0 1 X 3A

1 0 0 0 0
= 0 1 0 = 2
0 1 2 2
X 2 , x2
Figure 2.14.
Draft
207
According to Figure 2.14 we can check that:

Linitial = 2 2 + (2 ) 2 = 4(1 + 2 )
2
Linitial = 2 1 + 2
To obtain the angle in the current configuration formed by two unit vectors, we can use
the equation:
cos =
cos + 2 M E N
MN
(2.143)
where is the angle between the unit vectors M and N in the reference configuration,
and is the angle between the to new unit vectors in the current configuration.
r
Considering that the Green-Lagrange strain tensor is independent of X , we adopt two unit
vectors in the plane X 1 X 2 forming an angle = 30 , e.g. N i = [1 0 0] and
M i = [cos 30 sin 30 0] . With these data we have:

0 0
E N = 1 [1 0 0]0 2
2
0 0
0 cos 30
0 sin 30 = 0
2 0
The stretches:
2M
0
0 1
1
C M = [1 0 0] 0 1 + 2
=M
0 0 = 1

0
0
1 + 2 0
2N
M =1
0
0 cos 30
1
0 1 + 2
= N C N = [cos 30 sin 30 0]
0 sin 30
0
0
1+ 2 0
= cos 2 30 + (1 + 2 ) sin 2 30
= 1 + 2 sin 2 30
Then, N = 1 + 2 sin 2 30 . Then, we obtain:
cos =
cos + 2 M E N
cos 30
=
MN
1 + 2 sin 2 30
As we are dealing with a homogeneous deformation, we adopt two lines in the reference
configuration and we obtain the angle formed by these lines in the current configuration.
For example, adopting the lines OB = [cos 30 0 0] and OC = [cos 30 sin 30 0]. And
according to the equations of motion, the point O does not move. Then, we obtain the
new position of the points B and C , (see Figure 2.15):
x1B 1 0
0 X 1B 1 0
0 cos 30 cos 30
B
X B = 0 1 0 = 0
x 2 = 0 1 2
x B 0 1 X B 0 1 0 0
3
3

Ciudad Real - Spain
Draft
208
C
x1 1 0
0 X 1C 1 0
0 cos 30 cos 30
C
X C = 0 1 sin 30 = sin 30
x 2 = 0 1 2
x C 0 1 X C 0 1 0 sin 30
3
3
X 3 , x3
sin 30
sin 30
X 2 , x2
cos 30
30
B = B
X 1 , x1
Figure 2.15.
Then the angle formed by the new unit vectors O B and O C is:
O B O C = O B O C cos
cos 2 30 = cos 2 30 cos 2 30 + sin 2 30 + 2 sin 2 30 cos
cos =
cos 30
1 + 2 sin 2 30
d)
0 0 0
ij = 0 0 0
0 0 0
Problem 2.56
A rigid body motion is characterized by the following equation:
r
r r
x = c(t ) + Q(t ) X
(2.144)
r
Find the velocity and the acceleration fields as a function of , where is the axial vector
&
associated with the antisymmetric tensor ( = Q Q T ).
Solution:
r
The material time derivative of x = c(t ) + Q(t ) X is given by

r D r r r & r
& &
v=
x x =c + Q X
Dt

Ciudad Real - Spain
Draft
209
&
&
Let us consider that = Q Q T Q = Q . The above equation can also be expressed as:
r
r r
&
v = c + Q X
r r
r r
&
v = c + ( x c)
r r r
r
If is an antisymmetric tensor, it holds that a = a , where (angular velocity vector)
is the axial vector associated with the antisymmetric tensor . Then, the associated
velocity can be expressed as:
r r
r r
&
v = c + ( x c)
r r
r r
&
= c + ( x c)
(2.145)
Note that Q(t ) is only dependent on time, hence the axial vector (angular velocity)
r r
associated with is also time-dependent, i.e. = (t ) .
Then, its acceleration is given by:
r r && r && r
& r c
a = v = x = && + Q X
&& &
&
By referring to Q = Q + Q , the above equation can also be expressed as:
r
r
r &&
&
&
a = c + ( Q + Q) X
r
r
r
&
= && + Q X + Q X
c &
r
r
r
= && + Q X + Q X
c &
r
r r
r r
= && + ( x c ) + ( x c )
c &
we can state that:

r &
r && r
r r
r
r
r r
a = c + ( x c) + [ ( x c)]
(2.146)
r r
&
where shows the angular acceleration.
r
For a rigid body motion where c = 0 , the velocity becomes v = x whose components
are vi = ipq p x q , and the rate-of-deformation tensor D becomes:
D ij =
=
1 vi v j
+
2 x j xi
1
ipq p qj
2
1 ( ipq p x q ) ( jpq p x q ) 1
x
x
=
= ipq p q + jpq p q
+
2
2
x j
xi
x j
xi
1
1
+ jpq p qi = ipj p + jpi p = ipj p ipj p = 0 ij
2
2
So, once again we have proved that D = 0 for a rigid body motion.
Problem 2.57
r
a) A continuum is rotating as a rigid body with a constant angular velocity = 3 e 3 :
a.1) Obtain the velocity components in the spatial and material descriptions;
a.2) Obtain the acceleration in the spatial (Eulerian) description;
a.3) If 3 = 3rad / s obtain the vector position, velocity and acceleration at time t = 2.5s
of the particle that in the reference configuration was at (1,1,0) .

Ciudad Real - Spain
Draft
210
b) Taking into account Problem 1.124 where we have obtained the body force vector (per
r
GM
x
unit mass) b = r x where g = b is the acceleration of gravity caused by gravitational
field. Now, if we consider the Earth as a sphere that rotates around its axis with angular
r
velocity = 3 e 3 , obtaining the acceleration of gravity ( g ) at sea level in terms of the

latitude .
Solution:
X 3 , x3
r
= 3e 3
r r
r = r e r = re r
3
r
r
e3
er
r
x
X 2 , x2
X 1 , x1
Figure 2.16.
r r
a.1) By means of the previous problem we conclude that v ( x , t ) = x , or in indicial

notation:
vi = ijk j x k = i1k 1 x k + i 2 k 2 x k + i 3k 3 x k = i 3k 3 x k
{
{
=0
=0
= i 313 x1 + i 32 3 x 2 + i 33 3 x3 = i 31 3 x1 + i 32 3 x 2
{
=0
Then:
v1 = 132 3 x 2 = 3 x 2
v 2 = 2313 x1 = 3 x1
v = 0
3
r r
r r r
Note that the field v ( x , t ) is stationary, i.e. v = v ( x ) .
(2.147)
For a rigid body motion, the equations of motion are governed by:
r
r
x = Q(t ) X
where the orthogonal matrix components are given by the transformation matrix from the
r
r
system x to x , thus:

Ciudad Real - Spain
Draft
211
x1 cos (t ) sin (t ) 0 X 1 cos (t ) X 1 sin (t ) X 2

x 2 = sin (t ) cos (t ) 0 X 2 = sin (t ) X 1 + cos (t ) X 2
x 0
0
1 X 3
X3
3

Considering that =
d(t )
and by integrating we obtain:
dt
d(t ) = dt
(t ) = t
Then, we can rewrite the equations of motion as follows:

x1 cos (t ) sin (t ) 0 X 1 X 1 cos(t ) X 2 sin(t )

x 2 = sin (t ) cos (t ) 0 X 2 = X 1 sin(t ) + X 2 cos(t )
x 0
0
1 X 3
X3
3

(2.148)
To obtain the expression of velocity in the material (Lagrangian) description, we replace the
equations of motion (2.148) into the equations (2.147):
r
v1 ( X , t ) = 3 ( X 1 sin(t ) + X 2 cos(t ))
r
v 2 ( X , t ) = 3 ( X 1 cos(t ) X 2 sin(t ))
r
v3 ( X , t ) = 0
(2.149)
a.2) The Eulerian acceleration can be obtained by means of the definition of material time
r r
derivative of v ( x , t ) , i.e.:
r r
r r
r r
r r
r r r
v ( x , t ) v ( x , t ) x ( X , t )
r
a ( x, t ) =
+
= x v v ( x, t )
r
x
t
123
4t4
r
0
where the spatial velocity gradient components are given by:
r r
r
v ( x , t )
r
r = ( x v )ij
x ij
v1
x1
v
= 2
x
1
v 3
x1
v1
x 2
v 2
x 2
v 3
x 2
v1
x 3 0
v 2
= 3
x 3

v 3 0
x 3
3
0
0
0
0 (antisymmetric)
With that, we check that we are dealing with a rigid body motion. Then, the Eulerian
acceleration components are given by:
0
r
r r r
r v v ( x , t ) ] =
a i ( x , t ) = [ x
i
3
0
3
0
0
2
0 3 x 2 3 x1

2
0 3 x1 = 3 x 2
0 0 0

r r
2
2
We can express the acceleration a ( x , t ) = 3 x1 e 1 3 x 2 e 2 in the cylindrical coordinate,

(see Figure 2.16). Note that:
x1 = r cos ,
x1 = r cos ,
e 1 = e r cos e sin ,
e 2 = e r sin + e cos . Then, the
acceleration in the cylindrical coordinate system becomes:

Ciudad Real - Spain
Draft
212
r
2
2
a = 3 x1e 1 3 x 2 e 2
2
2
= 3 ( r cos )(e r cos e sin ) 3 ( r sin )(e r sin + e cos )

2
= 3 r (cos 2 + sin 2 )e r
2
= 3 re r
2r
= 3 r
The latter is known as the centripetal acceleration.

a.3) The particle at position (1,1,0) in the reference configuration describes a circular path
of radius r = 2 on the x1 x 2 -plane, (see Figure 2.17).
X 2 , x2
Particle P at t = 2.5s
r r
v P ( x , t = 2 . 5)
1
r
x
r r
v ( X , t = 0)
Particle P
r
X
X 1 , x1
Trajectory of particle P
Figure 2.17.
r
In the reference configuration ( t = 0 ) it fulfills that X = x . For the particle P we have:

r
v1P ( x , t = 0) = 3 x 2 = 3 X 2 = (3)(1) = 3
P r
v 2 ( x , t = 0) = 3 x1 = 3 X 1 = (3)(1) = 3
P
v3 = 0
2
3 X 1 9
r

2
a iP ( x , t = 0) = 3 X 2 = 9
0 0
At time t = 2.5s the position, velocity, and acceleration of the particle P are given by:
x1P X 1 cos(t ) X 2 sin(t ) cos(3 2.5) sin(3 2.5) 0.59136
P

x 2 = X 1 sin(t ) + X 2 cos(t ) = sin(3 2.5) + cos(3 2.5) = 1.28464

x P

X3
0
0
r
v1P ( x , t = 2.5) = 3 x 2 = (3)(0.59136) = 3.85391
P r
v 2 ( x , t = 2.5) = 3 x1 = (3)(1.28464) = 1.77409

P
v 3 = 0

Ciudad Real - Spain
Draft
213
2
3 x1 5.322
r
2
a iP ( x , t = 2.5) = 3 x 2 = 11.562
0
b) For a particle located on the surface of the Earth, due to rotation, this particle will feel as
if being projected outward according to r -direction, (see figure below). Keep in mind that
the real force is the Centripetal due to the centripetal acceleration. For convenience, we
adopt a fictitious force, centrifugal force, which would be the cause of this apparent
v
outward projection. Associated with this force we have the centrifugal acceleration ( a ctfu )
v
which is equal but opposite to the centripetal acceleration ( a ctpe ).
x3 , z
x3 , z
3
3
r
x3
r
b
v
a ctfu
x2 , y
x1 , x
Acceleration of gravity for defined latitude is given by:

r
v
g = a ctfu + b
Remember
v
a ctfu
r
+b =
we have
that
v
a ctfu
r
b =g,
given
v
+ 2 a ctfu
two
r
r
b cos + b
vectors
it
holds
that
, (see Problem 1.02). For this particular case,
v
v
2r
2
a ctpe = a ctfu = 3 r = 3 r . Also check that
r = R cos
and
cos = cos( ) = cos . With that, we obtain:

r
v
g = a ctfu + b =
v
a ctpe
r
r
v
2 a ctpe b cos + b
2
2
= ( 3 r ) 2 2( 3 r ) g cos + g 2
2
2
= ( 3 R cos ) 2 2( 3 R cos ) g cos + g 2
thus
2
4
g = g 2 2 g 3 R cos 2 + 3 R 2 cos 2

Ciudad Real - Spain
Draft
214
Pol
Note that at the poles ( = 90 ) we have g = g , and in the line of Ecuador it holds that
Ecu
2
4
2
2
g = g 2 2 g 3 R + 3 R 2 = ( g 3 R ) 2 = g 3 R .
Problem 2.58
Consider a rod subjected to successive displacements as shown in figure below
B0
L0
L0
L(1)
L( 2 )
L(f1)
L L( 2 )
L(1)
L
( 2)
Show that the engineering strain (Cauchy strain) is not additive to successive increments of
strain, i.e. (1) + ( 2) .
Solution:
The Cauchy strain was obtained as:
C =
L L L 0
=
= 1
L0
L0
Then, the total strain experienced by the body, i.e. from the B0 -configuration to the B configuration is:
C =
L( 2 ) L0 L( 2 )
=
1
L0
L0
In the B -configuration the engineering strain is:

(1
C) =
L(1) L0 L(1)
=
1
L0
L0
In the B -configuration considering only the displacement increment u ( 2) , we obtain:

(
C2) =
L( 2) L(1) L( 2)
= (1) 1
L(1)
L
thus
L(1)
L( 2)
L( 2 )
(1
(
C ) + C2 ) =
1 + (1) 1
L
L 1 = C
L
0
0
An essential requirement for any strain is to that it can be possible to characterize the real
displacement. In this case the final length is:

Ciudad Real - Spain
Draft
215
L(1)
=
1dx = L(1) L0 = L(1)
L
0
0
(1)
( 2)
L + L = L
L1
L1
L( 2 )
( 2 ) dx = (1) 1 dx = L( 2 ) L(1) = L( 2 )
C
L
0
0
L0
L0
(1
C ) dx
L0
L
C dx =
1 dx = L L0 = L
0
0 0
The Green-Lagrange strain tensor

Note that the Green-Lagrange strain tensor in the B -configuration is given by:
G =
We
could
E=E
(1)
+F
have
(1)T
obtained
the
L2 L2
0
2 L2
0
same
1 2
1
2
expression
by
using
E F , where for the uniaxial case we have

( 2)
(1)
(
E ( 2 ) G2) , F (1) (1) =
the
relationship
(1
E G , E (1) G ) ,
L(1)
. Then:
L0
E = E (1) + F (1)
E ( 2) F (1)
(1
(
G = G ) + (1) G2) (1)
1 L(1)
=
2 L0
2
L(1)
1 +
L0
1 L( 2)

2 L(1)

L( 2) 2 L2
2
2
0
L L0
=
=
2 L2
2 L2
0
0
2.1.3
2
L(1)
L0
Polar Decomposition of the Deformation Gradient
Problem 2.59
Let us consider the Cartesian components of the deformation gradient:
5 3 3
Fij = 2 6 3
2 2 4
obtain the tensors U (right stretch tensor), V (left stretch tensor), and R (rotation tensor).
Solution:
Before obtaining the tensors U , V , R , we analyze the deformation gradient F .
The motion is possible if the determinant of F is greater than zero, det ( F ) = 60 > 0 . The
eigenvalues and eigenvectors of F are given by:
F11 = 10 associated with eigenvector mi(1) = [0.6396021491; 0.6396021491; 0.4264014327]

Ciudad Real - Spain
Draft
216
F22 = 3 associated with mi( 2 ) = [ 0.5570860145; 0.7427813527; 0.3713906764]
F33 = 2 associated with mi(3) = [ 0.4082482905; 0.4082482905; 0.8164965809]
It is easy to check that the basis formed by these eigenvectors does not form an orthogonal

basis, i.e. mi(1) mi( 2) 0 , mi(1) mi(3) 0 , mi( 2) mi(3) 0 . We can also verify that if D is the
matrix containing the eigenvectors of F :
mi(1) 0.6396021491;
0.6396021491;
0.4264014327
( 2)
D = mi = 0.5570860145; 0.7427813527; 0.3713906764
m (3) 0.4082482905; 0.4082482905; 0.8164965809

i
we find that det (D ) = 0.905 1 , and D 1 D T . However, it holds that:

10 0 0
5 2 2
D = 3 6 2 = ( F T )
D 0 3 0
ij
0 0 2
3 3 4
and
5 2 2
10 0 0
3 6 2 D 1 = 0 3 0
D
3 3 4
0 0 2
The right Cauchy-Green deformation tensor components, C = F T F , are given by:

33 31 29
C ij = Fki Fkj = 31 49 35
29 35 34
Then the eigenvalues and eigenvectors of C are given by:
C11 = 9.274739
eigenvector
N i(1) = [0.6861511933; 0.7023576528; 0.1894472683]
C 22 = 3.770098
eigenvector
N i( 2 ) = [0.5105143234; 0.2793856273; 0.8132215099]
C 33 = 102.955163
eigenvector
N i(3) = [ 0.518239; 0.65470405; 0.550264423]
T
These eigenvectors constitute an orthogonal basis, so, it holds that AC 1 = AC , and
det (AC ) = 1 (improper orthogonal tensor):
N (1) 0.6861511933 0.7023576528 0.1894472683

(i 2)
AC = Ni = 0.5105143234 0.2793856273 0.8132215099
N (3) 0.518239
0.65470405
0.550264423
i
Furthermore, it holds that:
C11
0
A
0
T
C
C 22
0
0
33 31 29
33 31 29
C11
A = 31 49 35 = C ; A 31 49 35 A T = 0
0 C
ij
C
C
29 35 34
29 35 34
0

C 33
C 22
0
0
0

C 33
In the C principal space we obtain the components of the right stretch tensor, U , as:
1
U = Uij = 0
0
2
0
0 C11
= 0
0
3 0

C 22
0
0 3.0454455
0
0
0 =
0
1.9416741
0
C 33
0
0
10.1466824
and its inverse:

Ciudad Real - Spain
Draft
U 1 = Uij1
1
= 0
0
1
2
0
217
0
0
0
3.0454455

0
0
0 =

1.9416741
1
0
0
10.1466824
We can evaluate the components of the tensor U in the original space by means of the
transformation law:
4.66496626 2.25196988 2.48328843
A U AC = 2.25196988 6.00314487 2.80907159 = U ij
2.48328843 2.80907159 4.46569091
T
C
and
0.31528844 0.05134777 0.14302659
0.24442627 0.12519889 = U ij1

A U AC = 2.25196988
0.14302659 0.12519889 0.38221833
T
C
Then, the rotation tensor of the polar decomposition is given by the equation R = F U 1 ,
which is a proper orthogonal tensor, i.e. det (R ) = 1 .
R ij =
Fik U 1
kj
0.10094326 0.05592536
0.9933191
0.10658955 0.98826538 0.10940847
=
0.04422505 0.11463858 0.9924224
The left Cauchy-Green deformation tensor components, b = F F T , are given by:

43 37 28
bij = Fik F jk = 37 49 28
28 28 24
Next, the eigenvalues and eigenvectors of b are given by:
b11 = 9.274739
eigenvector
n i(1) = [0.6212637156 0.7465251613 0.238183919]
b22 = 3.770098
eigenvector
n i( 2 ) = [0.4898263742 0.1327190337 0.8616587383]
b33 = 102.95516
eigenvector
i
n (3) = [ 0.611638389 0.6519860747 0.448121233]
Note that, the tensors b and C have the same eigenvalues but different eigenvectors. If
the eigenvectors of b constitute an orthogonal basis then it holds that Ab1 = AbT , and
det (Ab ) = 1 :
i
n (1) 0.6212637156 0.7465251613
0.238183919
( 2)
Ab = n i = 0.4898263742 0.1327190337 0.8616587383
n (3) 0.611638389 0.6519860747 0.448121233

i
and, it also holds that:
b11
0
A
0
T
b
b22
0
0
b11
43 37 28
43 37 28
A = 37 49 28 = b ; A 37 49 28 A T = 0
0 b
b
b
ij
0
28 28 24
28 28 24

b33

Ciudad Real - Spain
Draft
b22
0
0
0

b33
218
Since C and b have the same eigenvalues, it follows that Uij = Vij , i.e. they have the same
components in their respectively principal space. Additionally, it holds that Uij1 = Vij 1 .
The components of the tensor V in the original space can be evaluated by:
5.3720129 2.76007379 2.41222612
A V Ab = A U Ab = 2.76007379 6.04463857 2.20098553 = Vij
2.41222612 2.20098553 3.6519622
T
b
T
b
and
0.28717424 0.07950684 0.14176921
A V Ab = A U Ab = 0.07950684 0.23396031 0.08848799 = Vij1
0.14176921 0.08848799 0.42079849
T
b
T
b
The polar decomposition rotation tensor obtained previously has to be the same as the one
obtained by R = V 1 F .
We could also have obtained the tensors U , V , R , by means of their spectral
representation. That is, if we know the principal stretches, i , and the eigenvectors of C
( N (i ) ), and the eigenvectors of b ( n (i ) ), it is easy to show that:

U ij = a N ( a ) N ( a ) = 1 Ni(1)N (j1) + 2 N i( 2 ) N (j2 ) + 3 Ni(3) N (j3)
ij
a =1
i

Vij = a n ( a ) n ( a ) = 1 n (1) n (j1) + 2 n i( 2) n (j2) + 3 n i(3) n (j3)
ij
a =1
3 (a) (a )

i
R ij =
n N = n (1) N (j1) + n i( 2 )N (j2 ) + n i(3) N (j3)
ij
a =1

Fij = a n ( a ) N( a ) = 1 n i(1) N (j1) + 2 n i( 2 ) N (j2 ) + 3 n i(3) N (j3)
ij
a =1
F=
a R N( a ) N ( a ) =
a =1
n(a) n (a) R
a =1
= R a N ( a ) N( a ) = a n ( a ) n ( a ) R
a =1
a =1
= R U = V R
3
As we can verify, the representations of the tensors R and F are not the spectral
representations in the strict sense of the word, i.e., i are not eigenvalues of F , and
neither n (i ) nor N (i ) are eigenvectors of F .
Problem 2.60
The deformation gradient at one point of the body is given by:
F = 0.2e1 e1 0.1e1 e 2 + 0.3e 2 e1 + 0.4e 2 e 2 + 0.1e 3 e 3
where e i
(i = 1,2,3) represents the Cartesian basis.

Ciudad Real - Spain
Draft
219
a) Obtain the deformation tensors b and C ;

b) Obtain the eigenvalues and eigenvectors of b and C ;
c) Write the spectral representation of F in function of the eigenvalues of C ( C a )
3
and check if F = a n ( a ) N ( a ) holds, where a are the principal stretches, n

a =1
are the eigenvectors of b , and N are the eigenvectors of C ;

d) Obtain the spectral representation and components of: ( R ) spin tensor of the polar
decomposition; the stretch tensors U and V ;
Solution
The deformation gradient components can be represented as follows:
F = Fij e i e j = 0.2e1 e1 0.1e1 e 2 + 0.3e 2 e1 + 0.4e 2 e 2 + 0.1e 3 e 3
0 .2 0 .1 0
Fij = 0.3 0.4
0
0
0
0.1
a) The left Cauchy-Green deformation tensor ( b = F F T ) components:

T
bij = Fik F jk
0
0.2 0.1 0 0.2 0.1 0
0.05 0.02
0 .3 0 .4
0 .3 0 .4
= 0.02 0.25
=
0
0
0
0
0
0
0.1 0
0
0.1
0
0.01
(2.150)
The right Cauchy-Green deformation tensor ( C = F T F ) components are given by:

0.2 0.1 0
0
C ij = Fki Fkj = 0.3 0.4
0
0
0.1
0
0.2 0.1 0 0.13 0.1
0.3 0.4
= 0.1 0.17
0
0
0
0
0.1 0
0
0.01
b) The eigenvalues and eigenvectors of b and C are obtained as follows;
C N = C ( a ) N( a )
(2.151)
C C1 = 0
where the index (a ) does not indicate summation. Note that we already know one
eigenvalue of C , i.e. C (3) = 0.01 , (see C -components in (2.151)). Then, the characteristic
determinant becomes:
0.13 C
0 .1
=0
0 .1
0.17 C
(0.13 C )(0.17 C ) 0.01 = 0
The solution of the quadratic equation is:

C (1) = 0.25198
C ( 2 ) = 0.04802
Then:

Ciudad Real - Spain
Draft
220
Cc (1) = 0.25198
C (3) = 0.01
N i(3)
0.633399
(1) = 0.77334
Ni
0
= 0

1

C ( 2 ) = 0.04802
0.77334
( 2 ) = 0.63399
Ni
Similarly for the eigenvectors of the tensor b :
b n = b( a ) n ( a )
where the index (a ) does not indicate summation. Then

b(1) = 0.25198
b(3) = 0.01
0.098538
= 0.995133
0
= 0

1

n i(1)
n i(3)
b( 2 ) = 0.04802
i
n ( 2)
0.995133
= 0.098538
As expected, the tensors C and b have the same eigenvalues but different eigenvectors.
0
0
0.252
0
0.048 0
C ij =
0
0
0.01
0
0
0.252
0
0.048 0
bij =
0
0
0.01
In addition, the spectral representations of the tensors C and b are given respectively by:
C=
2a N ( a ) N ( a )
a =1
b=
n
2
a
(a)
n(a)
a =1
where a > 0 are the principal stretches. Considering that 2a = C a are the eigenvalues of
C and of b , the principal stretches are:
(1) = 0.25198 0.501976
c) To check if
F=
( 2 ) = 0.04802 0.219134
n
a
(a)
N(a )
(3) = 0.01 = 0.1
holds we calculate the components of
a =1
a n ( a ) N ( a ) with the results obtained previously, i.e.:
ij
a =1

Ciudad Real - Spain
Draft
221
a n ( a ) N ( a ) = 1n i(1) N (j1) + 2 n i( 2 ) N (j2 ) + 3n i(3) N (j3)
a =1
ij
0.06247 0.0762 0
0.76958 0.6309 0
0.0762 0.06247 0 + 0.219134 0.6309 0.76958 0 +
= 0.50197
0
0
0
0
0
0 0 0
+ 0.10 0 0
0 0 1
0.2 0.1 0
0 = Fij
= 0 .3 0 .4
0
0
0.1
Checking that F = a n ( a ) N ( a ) .
a =1
d)
R=
(a)
N( a)
components
a =1
0.832 0.554 0
(R )ij = 0.554 0.832 0
0
0
1
which can be verified with:

0.76958 0.6309 0 0.06247 0.0762 0 0 0 0 0.832 0.5547 0
0.832
0
R ij = 0.0762 0.06247 0 + 0.6309 0.76958 0 + 0 0 0 = 0.5547
0
0
0 0
0
0 0 0 1 0
0
1
U=
a N ( a ) N( a )
components
0.333 0.139 0
(U)ij 0.139 0.388 0
0
0
0.1
components
0.222 0.028 0
(V )ij 0.028 0.5 0
0
0
0.1
a =1
V=
n
a
(a )
(a)
a =1
Problem 2.61
For a given motion (shear deformation):
x1 = X 1 + kX 2
x2 = X 2
x = X
3
3
k constant
Obtain the tensors: F (deformation gradient), C (the right Cauchy-Green deformation

tensor), b (the left Cauchy-Green deformation tensor), E (the Green-Lagrange strain
tensor), U (the right stretch tensor), V (the left stretch tensor) and R (the spin tensor of
the polar decomposition).

Ciudad Real - Spain
Draft
222
Solution:
The deformation gradient components:
x1
X 1
xi x 2
Fij =
=
X j X 1
x3
X 1
x1
X 2
x 2
X 2
x3
X 2
x1
X 3 1 k 0
x 2
= 0 1 0
X 3
0 0 1
x3
X 3
The right Cauchy-Green deformation tensor ( C = F T F ), whose components are:

k
1 0 0 1 k 0 1
k 1 0 0 1 0 = k 1 + k 2
C ij = Fki Fkj =

0 0 1 0 0 1 0
0

0
0
The left Cauchy-Green deformation tensor ( b = F F T ), whose components are:

2
1 k 0 1 0 0 1 + k
bij = Fik F jk = 0 1 0 k 1 0 = k
0 0 1 0 0 1 0
The Green-Lagrange strain tensor, E =

1
k
1
E ij = k 1 + k 2
2
0
1 0
0 1
1
(C 1) , whose components are:
2
0 1 0 0
0 k
0 1 0 = 1 k k 2
0
2
0 0 1
0 0
1
0
0
Note that there is only deformation on the x1 x 2 -plane.

Considering the polar decomposition F = R U = V R , we obtain:
C = (V R)T (V R)
= RT VT V R
= RT V V R
= RT V 2 R
= RT b R
For simplicity, we will work on the x1 x 2 -plane, with that we represent the rotation
tensor components as follows:
cos sin c s
R ij =
=
sin cos s c
(i, j = 1,2)
where cos 2 + sin 2 = c 2 + s 2 = 1 holds. The relationship C = R T b R becomes:

k c s 1 + k 2 k c s
1
k 1 + k 2 = s c
1 s c

k
(c 2 + c 2 k 2 + 2 sck + s 2 )
( sck 2 s 2 k + c 2 k )
=
2
2
2
(c 2 + s 2 k 2 2 sck + s 2 )
( sck s k + c k )

Ciudad Real - Spain
Draft
223
(c 2 + c 2 k 2 + 2 sck + s 2 ) = 1 (c 2 k 2 + 2 sck + 1) = 1
From the relationship
we obtain
k
c . Then, starting from the relation ( sck 2 s 2 k + c 2 k ) = k and by considering that
2
k
s=
c , we obtain:
2
s=
c=
1
k2
+1
4
s=
k2 + 4
k
2
k2
+1
4
k
k2 + 4
thus:
2
2
k +4
k
R ij =
2
k +4
0
k
2
k +4
2
k2 + 4
0
From the polar decomposition F = R U = V R , we obtain U = R T F and V = F R T ,

whose components are:
2
2
k +4
k
U ij = R ki Fkj =
2
k +4
0
Vij = Fik R jk
k
k2 + 4
2
k2 + 4
0
0
2
1 k 0 k + 4
k
0 0 1 0 =
2
0 0 1 k + 4

1
0
2
1 k 0 k + 4
k
= 0 1 0
2
0 0 1 k + 4
k
k +4
2+ k2
2
2+k
0
2
k +4
k
0 =
2
k +4
1
0
k
k2 + 4
2
k2 + 4
0
k2 + 4
0
k
k2 + 4
2
k2 + 4
0
Problem 2.62
A deformable parallelepiped of dimensions 2 2 1 is in the reference configuration as
indicates in Figure 2.18. This body is subjected to motion:
r r
x ( X , t ) = exp X 2t e1 + tX 12 e 2 + X 3 e 3
(2.152)
where ( X 1 , X 2 , X 3 ) are the material coordinates, and t represents time.

r
a) Obtain the deformation gradient F , for all X and time t .

b) Obtain the right Cauchy-Green deformation tensor C , and the principal stretches.
c) Obtain the right stretch tensor U and the rotation tensor R . Check that the latter
is a proper orthogonal tensor.
Ciudad Real - Spain
Draft
224
d) Find the volume of the deformed parallelepiped at time t = 1s .

X2
X3
X1
Figure 2.18.
Solution:
a) According to the equation (2.152), the vector position components are x1 = exp X 2t ,
x 2 = tX 12 , x3 = X 3 , then the deformation gradient ( F ) components are given by:
0
xi
Fij =
= 2tX 1
X j
0
0
1
t exp X 2t
0
0
b) The right Cauchy-Green deformation tensor are defined by C = F T F , whose

components are C ij = Fki Fkj :
0
t exp X 2t
C ij =
0 0
0 2tX 1
1 0
2tX 1
0
0
0 4t 2 X 12

0 = 0
1 0

t exp X 2t
0
0
0
2
t exp
0
2 X 2t
0
1
Note that this space is the principal space (principal directions) of C . Considering that i
are the principal stretches, the following relationship is fulfilled:
C = U2 =
2
a
N ( a ) N( a )
U=
a =1
N( a) N( a)
a =1
As we are working in the principal space of C , we can obtain the principal stretches as
follows:
1 = + 4t 2 X 12
2 = + t 2 exp 2 X 2t
3 = + 1
which are positive numbers, since U = a N ( a ) N( a ) is a positive definite tensor by

a =1
definition, thus:
1 = 2tX 1
2 = t exp X 2t
3 = 1
c)

Ciudad Real - Spain
Draft
2tX 1
U ij = 0
0
0
0
t exp
0
X 2t
1
2tX
1
0
=
U ij1
225
0
1
t exp X 2t
0
According to the polar decomposition F = R U R = F U 1 , we can obtain the rotation

tensor ( R ) components as follows:
0
R ij = 2tX 1
0
t exp
0 2tX 1
0 0
1
0
X 2t
0
0
0
1
t exp X 2t
0
0
0 1 0
0 = 1 0 0

0 0 1
Note that the orthogonality condition R R 1 = R R T = 1 holds:

R ik R jk
0 1 0 0 1 0 1 0 0
= 1 0 0 1 0 0 = 0 1 0
0 0 1 0 0 1 0 0 1
and the proper condition det (R ) = 1 .

d) To calculate the final volume we use the relationship dV = JdV0 , where J = F is the
Jacobian determinant and is given by:
t exp X 2t
0
J = 2tX 1
0
0
0 = 2t 2 X 1exp X 2t
1
0
0
At time t = 1s we have J = 2 X 1exp X 2 . Then, the volume at time t = 1s is given by:
dV = JdV
V0
(2 X exp )dX
= 4(exp 1) 25.556
=
X2
3 dX 2 dX 1
X 1 = 0 X 2 = 0 X 3 =0
2
NOTE: We can not use the equation V = JV0 because we are not dealing with
homogeneous deformation case.
Problem 2.63
A body is subjected to motion:
x1 = X 1
x 2 = X 2 + kX 3
x3 = X 3 + kX 2
where k is a constant.
a) Obtain the deformation gradient ( F ); the right Cauchy-Green deformation tensor
( C ); the Green-Lagrange strain tensor ( E ).
Ciudad Real - Spain
Draft
226
b) Calculate the displacement field, the magnitude (dx) 2 of sides OA and OB , and
diagonal OC after deformation of figure below.
X3
dX 2
C
dX 3
X2
X1
c) Consider now a square as figure below

x3
C
23
B
B
x2
c.1) Obtain the stretches according to directions OC and BA ; c.2) Obtain the angle 23 in
the current configuration in function of k .
c.3) Apply the polar decomposition of the tensor F in order to obtain U and R .
Solution:
a) C = F T F . The deformation gradient components are:
1 0 0
xi
Fij =
= 0 1 k
X j
0 k 1
0
0
1 0 0 1 0 0 1
0 1 k 0 1 k = 0 1 + k 2
C ij = Fki Fkj =
2k
0 k 1 0 k 1 0
2k
1+ k2
The Green-Lagrange strain tensor, E =
1
(C 1) , whose components are:
2
1
0
0 1 0 0
0 0
1
0 1 0 = 1 0 k 2
2
2k
E ij = 0 1 + k
2
2
2
0 2 k
2k
1 + k 0 0 1
0
Ciudad Real - Spain
Draft
0
2k
k2
227
b.1) The displacement field, u = x X , whose components are:

u1 = x1 X 1 = 0 ;
r
b.2) Calculation of (dx ) 2 = dx 2
u 2 = x 2 X 2 = kX 3
u 3 = x 3 X 3 = kX 2
r
r r
(dx )2 = dx dx
r
r
= F dX F dX
r
r
= dX F T F dX
r
r
= dX C dX
Explicitly:
(dx )
= [dX 1
0
0 dX 1
1
0 1 + k 2
dX 3 ]
2k dX 2
0
2k
1 + k 2 dX 3
dX 2
= (dX 1 ) 2 + (dX 2 ) 2 (1 + k 2 ) + (dX 3 ) 2 (1 + k 2 ) + 4k (dX 2 )(dX 3 )
Then, for the diagonal OC we have [0 dX 2 dX 3 ] , with that we obtain:
(dx )2 = (dX 2 ) 2 (1 + k 2 ) + (dX 3 ) 2 (1 + k 2 ) + 4k (dX 2 )(dX 3 )

For the side OA we have [0 dX 2 0] , with that we obtain:
(dx )2 = (dX 2 ) 2 (1 + k 2 )
For the side OB we have [0 0 dX 3 ] , with that we obtain:
(dx )2 = (dX 3 ) 2 (1 + k 2 )
c) The stretch according to the N -direction (reference configuration) is given by the
equation ( N )2 = N C N .
c.1) The stretch according to the OC -direction: Ni = 0
( )
OC
= 0
1
2
0
1
1
2
0 1 + k
2
0
2k
The stretch according to the BA -direction: Ni = 0
( )
BA
= 0

Ciudad Real - Spain
1
2
0
1
1
2
0 1 + k
2
0
2k
Draft
1
2
0
2k
1+ k2
1
2
1
, is:
2
1
= (1 + k ) 2
2
1
1
, with that we obtain:
2
0
0
1
2
2k
2 = (1 k )
1+ k 2 1
228
c.2) The variation of the angle. We can directly use the equation:
cos =
MC N
M C N
=
M N
M C M N C N
where the unit vector according to the OB -direction is Mi = [0 0 1], and according to the
OA -direction is N i = [0 1 0] . With that we obtain:
( )
0
0 0
1
0 1 + k 2
2k 0 = 1 + k 2
= [0 0 1]

0
2k
1 + k 2 1

( )
0
0 0
1
0 1 + k 2
2k 1 = 1 + k 2
= [0 1 0]

0
2k
1 + k 2 0

OB
OA
0
0 0
1
C N = [0 0 1] 0 1 + k 2
Mi ij j
2k 1 = 2k

0
2k
1 + k 2 0
Then:
cos 23 =
2k
M C N
=
MN
1+ k2
c.3) The polar decomposition of F = R U = V R , where:

C = U2 =
a N(a ) N(a )
U= C =
a =1
a N( a ) N ( a )
a =1
Calculation of the eigenvalues of C . Note that according to the format of C -components,

there is only deformation according to x 2 x 3 -plane. In addition, we know one eigenvalue
1 = 1 associated with the direction N i(1) = [1 0 0] . By means of the characteristic
determinant we obtain:
(1 + k 2 )
2k
=0
2k
(1 + k 2 )
( ) (
2(1 + k ) + (1 k )
2 2 1 + k 2 + 1 2k 2 + k 4 = 0
2
2 2
The roots are: 2 = 1 + k 2 + 2k = (1 + k ) 2
=0
3 = 1 + k 2 2k = (1 k ) 2
Then, in the principal space of C we have:

0
1
0 (1 + k ) 2
C ij =
0
0
The principal directions are 2 Ni( 2) = 0
1
2
(1 k ) 2
0
0
1
( 3)
, 3 N i = 0
2
1
2
1
. Then,
2
the transformation matrix between the original space and the principal space is:

Ciudad Real - Spain
Draft
1
a ij = A = 0
0
1
2
1
2
229
0
1
2
1
That is, it must meet:

C = A C AT
0
1
0 (1 + k ) 2
0
0
0 1
0 = 0

(1 k ) 2
0 1
0
1
0 1+ k2
2
2k
1 0
0
1
2
1
2
0 1
2k 0
1+ k 2
0
1
2
1
0
1
2
1
2
Then, in the principal space of C , we have:

0
1
0 (1 + k ) 2
C ij =
0
0
(1 k ) 2
0
0
+ 1
0
+ (1 + k ) 2
U ij = 0
0
0
2
+ (1 k )
0
0
0
0
1
0 (1 + k )
0
U ij =
0
0
(1 k )
The inverse tensor in the principal space can be obtained as follows:
U 1
ij
1
= 0
0
0
(1 k )
0
1
(1 + k )
0
The components of U in the original space are given by:

U 1 = A T U 1 A
1
U ij = 0
0
1
2
1
2
0
1
2
1
1
0
0
1
(1 + k )
0
0 1
0 0
1
0
(1 k )
0
1
2
1
2

0 1
1
= 0
2
1
0

2
0
1
(1 k 2 )
k
(1 k 2 )
0
k
(1 k 2 )
1
(1 k 2 )
From the polar decomposition we have F = R U R = F U 1 , thus

Ciudad Real - Spain
Draft
230
1 0 0 1
R ij = 0 1 k 0
0 k 1
0
1
(1 k 2 )
k
(1 k 2 )
0 1 0 0
k
= 0 1 0
2
(1 k )
0 0 1
1
(1 k 2 )
Problem 2.64
Given the following motion:
x1 = 1 X 1
x 2 = 3 X 3
x = X
2 2
3
a) Obtain the final volume to a unit cube;

b) Obtain the deformed area to a unit square defined in the X 1 X 2 -plane, and draw the
deformed area;
c) Apply the polar decomposition and obtain the tensors U , V and R
Solution:
a)
x1 1

x2 = 0
x 0
3
0
0
2
0 X 1

3 X 2
0 X 3

1
Fij = 0
0
3 (homogenous deformation)
0
0
2
The determinant of F is given by F J = 1 2 3 , and the deformed volume:

integrating
dV = F dV0 V final = F Vinitial = 1 2 3
b) Applying the Nansons formula and by considering the particular case (homogeneous
deformation):
r integrating r
r
r
da = JF T dA a final = JF T Ainitial
where
e1
r
Ainitial = 1
e2
0
e3
0 = e 3 ; Fij1 =
0
2 3
1
0
1 2 3
0
0
0
1 2
0 1
1 3 = 0

0
0
0
1
3
1
2
With that the deformed area vector is:

Ciudad Real - Spain
Draft
a1
1

a 2 = 1 2 3 0
a
3
0
0
0
1
2
231
0
0 0
1
0 =
1 2
3
1 0
and its magnitude is:

r
a final = ( 1 2 ) 2 = 1 2
X 3 , x3
B (0,0, 2 )
r
a final = 1 2
C (1 ,0, 2 )
O(0,0,0)
B (0,1,0)
r
Ainitial = 1
C (1,1,0)
A(1,0,0)
X 2 , x2
A( 1 ,0,0)
X 1 , x1
where the points A(1,0,0) , B(0,1,0) and C (1,1,0) move according to the equations of
motion:
x1A 1
A
x2 = 0
x A 0
3
C
x1 1
C
x2 = 0
xC 0
3
c)
0
0
2
According
U= C = F
0 1 1

3 0 = 0
0 0 0

0
0
2
0
0
2
0 0 0

3 1 = 0
0 0 2

0 1 1

3 1 = 0
0 0 2

to
the
F and
polar
decomposition
V = b = F F

Ciudad Real - Spain
x1B 1
B
x2 = 0
x B 0
3
definition
F = R U = V R
where
we obtain:
Draft
232
1
C ij = 0
1
bij = 0
0
0
3
0
0
2
0 1
2 0
0 0
0 1
3 0
0 0
0
0
2
0
0
3
0 2
1
=0
3
0 0

0 2
1
=0
2
0 0

0
2
3
0
22
0
0
2
3
0
0
22
1
U ij = 0
1
Vij = 0
0
2
0
0
3
0
0
0
0
0
Note that the original space coincides with the principal space of C . Note also that C and
b have the same eigenvalues but different principal directions. To obtain the spin tensor of
the polar decomposition we apply R = F U 1 = V 1 F , thus:
1
R ij = 0
1
R ij = 0
0
0
2
0
1
3
0
0 1
3 0
0
1
0 0
1 0
2
0
1
2
0
0
0
2
0
1 0 0
0 = 0 0 1

0 1 0
1
3
0 1 0 0
3 = 0 0 1

0 0 1 0

Problem 2.65
x1 = 3 X 1
x2 = 2 X 2
x3 = 3 X 3 X 2
Obtain the material ellipsoid associated with the material sphere defined in the reference
2
configuration by X 12 + X 2 + X 32 = 1 (see Figure 2.19). Check that the ellipsoid in the
principal space of the left stretch tensor V has the shape:
x1 2
2
1
x 22
22
x32
2
3
=1
where 1 , 2 , 3 are the principal stretches.

Ciudad Real - Spain
Draft
233
X 2 , x2
X 3 , x3
Material surface
(always constituted by the
same particles)
X 1 , x1
Figure 2.19: Material sphere.

Solution:
The law of motion and its inverse are given by:
x1

x2 =
x
3
0
0
0
2
1
0 X 1
X1
inverse
0 X 2 X 2 =
X
X
3 3
3
3
3
0
1
2
3
6
0
0
0
x1

0 x2

3 x3
The equations of motion in the spatial description are given by:
X1 =
X 2 =
X 3 =
3
x1
3
x2
2
3
3
x2 +
x3
6
3
Replacing the above into the equation of the sphere we obtain:

2
2
X 12 + X 2 + X 3 = 1
2
2
3
3
3
x2
x + +
x +
x3 = 1
3 1
6 2
3
2
By simplifying the above equation we obtain:

2
2
x12 + x 2 + x3 + x 2 x3 = 3
which is the equation of an ellipsoid. We now represent the ellipsoid equation in the
principal space of the left stretch tensor V . Recall that the tensor V and b are coaxial, i.e.
the have the same principal directions), and is also true that:
V = b = F FT

Ciudad Real - Spain
Draft
234
The components of b are

3
bij = 0
0
0 3 0
0 0
2
0 1
3
0
2
1
0
0
0
3
0
0 =
5
3
0 3
3
3
Note that we know already one eigenvalue b1 = 3 associated with the eigenvector
n i(1) = [1 0 0] . Then, the other principal directions are in the plane x 2 x 3 , with that we
obtain
eigenvecto
b2 = 6 r n i( 2) = 0

2
2
eigenvecto
b3 = 2 r n i(3) = 0

2
2
thus:
1
3 0 0
Transforma
matrix
bij = 0 6 0 tion a ij = 0
0 0 2
1 = 3
0
Vij = 0
2 = 6
0
0
2
2
2
2
0
2
2
2
0
3 = 2

Then, applying the transformation law from x1 , x 2 , x3 -system to the x1 , x 2 , x3 -system we
obtain:
x1 1

x2 = 0
x
3
0
0
2
2
2
2
0
2
2
2
x1

x2
x
3
x1 = x1
2
2
x2 +
x3
x2 =
2
2
2
2
x +
x3
x3 =
2
2
2
with that, the equation of the ellipsoid in the principal space of V is represented by:
2
2
x12 + x 2 + x3 + x 2 x3 = 3
(x1 )
2
2
2
2
2 2
2
+
x2 +
x3 +
x2 +
x3 +
x2 +
x3
x2 +
x3 = 3
2
2
2
2
2
2
2
Simplifying the above equation we obtain:
x1 2 x 22 x32
x 2
x2
x 2
x2 x2 x2
+
+
= 1 2 + 2 2 + 3 2 = 12 + 2 + 32 = 1
3
6
2
3
22
1
( 3)
( 6)
( 2)

Ciudad Real - Spain
Draft
235
X 3 , x3
X 2 , x2
x3
3 = 2
2 = 6
1 = 3
x2
x1
X 1 , x1
Figure 2.20: The material ellipsoid (deformed configuration).
x2
x1
x3
x2
X2
x2
x1
X1
x3
X3
x1
x3
Figure 2.21: The left polar decomposition.

Ciudad Real - Spain
Draft
236
2.1.4
Infinitesimal Deformation Regime
Problem 2.66
Given the equations of motion
x1 = X 1 + 4 X 1 X 2 t
2
x2 = X 2 + X 2 t
2
x3 = X 3 + X 3 t
(2.153)
a) Obtain the velocity field;

b) Obtain the infinitesimal strain field;
c) At time t = 1 s , obtain the infinitesimal strain tensor.
Solution:
a) Velocity:
V1 = 4 X 1 X 2
r
r r
dx
2
V 2 = X 2
V ( X , t) =
dt
2
V3 = X 3
(2.154)
r
A1 = 0
r r
dV
A( X , t ) =
A2 = 0
dt
A = 0
3
(2.155)
u1 = x1 X 1 = X 1 + 4 X 1 X 2 X 1 = 4 X 1 X 2
2
2
u 2 = x2 X 2 = X 2 + X 2 X 2 = X 2
2
2
u3 = x3 X 3 = X 3 + X 3 X 3 = X 3
(2.156)
b) Acceleration:
c) Displacement field:
Then, the infinitesimal strain tensor components are given by:

ij =
u1
X 1
u i u 2
=
x j X 1
u 3
X 1
u1
X 2
u 2
X 2
u 3
X 2
1 u i u j
+
2 x j xi
u1
X 3 4 X
2
u 2
= 0
X 3

u 3 0
X 3
(2.157)
4X1
2X 2
0
0
0
2X 3
(2.158)
thus:
4 X 2
ij = 2 X 1

Ciudad Real - Spain
2 X1
2X 2
0
Draft
0
0
2X3
(2.159)
237
Problem 2.67
Consider the infinitesimal strain tensor:
0
2
X3
2
l
X2X3
l2
0
0
X2X3
ij = 0
l2
X2
0 3
l2
(2.160)
and the infinitesimal spin tensor:
0
0
ij = 0
0
2
2
0 2 X 2 X 3
2l
2l 2
2
2
X2 X3
(2.161)
Obtain the displacement field components.

Solution:
The displacement gradient is related to the infinitesimal strain tensor and the infinitesimal
spin tensor as follows:
u i , j = ij + ij
ij =
1
ui, j + ui, j
2
; ij =
1
ui, j ui, j
2
(2.162)
thus:
0
u i , j = 2 0
2l
2
3X 3
2X 2 X 3
0
2X 2 X 3
2
X2
2
X3
2
X2
(2.163)
u1
= 0 u1 = 0
x1
u 2
= 2 (2 X 2 X 3 )
x 2 2l
u 2 =
2l (2 X
2
2X3
)x 2 u 2 =
2l 2
u 3
= 2 (2 X 2 X 3 )
x3
2l
u 3 =
2l 2

Ciudad Real - Spain
(2 X 2 X 3 )x3 u 3 =
Draft
(2.164)
2l 2
2
X2 X3
+ C1 ( X 3 )
2
X3 X2
(2.165)
+ C2 ( X 2 )
(2.166)
238
To determine the constant C1 ( X 3 ) from the result (2.165) we take the derivative of u 3
with respect to X 3 :
u 2
= 2
X 3 2l
2 C1 ( X 3 )
C1 ( X 3 )
2
2
2
= 3 X 3
X 2 +
= 2 X 2 3X 3
X 3 2l
X 3
C1 ( X 3 ) =
(2.167)
3
X 3
Similarly we find the constant C2 ( X 2 ) :

u 3
= 2
X 2
2l
2 C 2 ( X 2 )
C 2 ( X 2 )
2
2
2
= X2
X 3 +
= 2 X2 + X3
X 2
X 2
2l
X3
C2 ( X 2 ) = 2
3
(2.168)
Then, the displacement field is given by:

u1 = 0 ; u 2 =
2l 2
[X
2
2 X3
3
X3
; u3 =
3
X2
2
X 3 X 2 +
3
2l 2
(2.169)
Problem 2.68
Show that, in the infinitesimal strain regime, the rate of change of the infinitesimal strain
&
tensor ( ) is equal to the rate-of-deformation tensor ( D ).
Solution:
Consider the relationship between the rate of change of he Green-Lagrange deformation
&
tensor ( E ) and the rate-of-deformation tensor ( D ):
&
E = F T D F
(2.170)
& & &

In the case of small deformation F 1 holds, in addition it fulfills that E e then:
& &
E = =D
(2.171)
Problem 2.69
Given the equations of motion
x1 = X 1
2t
x 2 = X 2 + X 1 exp 1
3t
x3 = X 3 + X 1 exp 1
(
(
)
)
(2.172)
Obtain the rate-of-deformation ( D ) and compare with the rate of change of the
&
infinitesimal strain tensor ( ).
Solution:
By definition, the rate-of-deformation tensor ( D ) is the symmetric part of the spatial
velocity gradient, i.e.:

Ciudad Real - Spain
Draft
239
1
(l + l T )
r
r
r
and ( x , t ) = sym u (u) sym
2
r
l = xv
D=
(2.173)
By definition, the infinitesimal strain tensor is equal to the symmetric part of the
displacement gradient:
r
& D
= (u) sym
(2.174)
Dt
r r r
The displacement field is given by u = x X . Considering the equations of motion, the
displacement field components become:
u1 = x1 X 1 = X 1 X 1 = 0
2t
2t
u 2 = x 2 X 2 = X 2 + X 1 exp 1 X 2 = X 1 exp 1
3t
3t
u 3 = x 3 X 3 = X 3 + X 1 exp 1 X 3 = X 1 exp 1
r
r Du
The velocity field is given by v =
. Then, the velocity field components, in material
Dt
(
(
)
)
(
(
)
)
coordinates, are:
V1 = 0
2t
V2 = X 1 (2exp )
3t
V3 = X 1 ( 3exp )
(2.175)
Given the inverse equations of motion:

x1 = X 1
2t
x 2 = X 2 + X 1 (exp 1)
3t
x3 = X 3 + X 1 (exp 1)
inverse
X 1 = x1
2t
X 2 = x 2 x1 (exp 1)
3t
X 3 = x3 x1 (exp 1)
(2.176)
we can obtain the velocity field in spatial coordinates:

v1 = 0
2t
v 2 = 2 x1 exp
3t
v3 = 3x1exp
(2.177)
The spatial velocity gradient ( l ) components are given by:

0
vi
r
= 2exp 2t
( l ) ij = ( x v ) ij =
x j
3exp 3t
0 0
0 0
0 0
(2.178)
and

Ciudad Real - Spain
Draft
240
0
1
1
2exp 2t
(D) ij = ( l ij + l ji ) =
2
2
3exp 3t
= exp 2t
3 exp 3t
2
exp 2t
0
0
We also obtain the spin tensor W = l
Wij =
1
l ij l ji
2
0 0
0
+ 2exp 2t
0 0
0 0 3exp 3t

T
0 0
0 0

0 0

(2.179)
exp 3t
2
skew
components
= exp 2t
3 exp 3t
2
exp 2t
0
0
exp 3t
2
(2.180)
The infinitesimal strain tensor ( )

Starting from the displacement field:
u1 = 0
2t
u 2 = x1 (exp 1)
3t
u 3 = x1 (exp 1)
(2.181)
the displacement gradient components can be obtained as follows:

0
r
u i
2t
(u)ij = = exp 1
x j
exp 3t 1
(
(
)
)
0 0
0 0
0 0
(2.182)
We can decompose (u) into a symmetric and an antisymmetric part:
(u)ij = ( symu)ij + ( skewu)ij

r
(2.183)
= ( )ij + ()ij
The symmetric part:
0
1
sym r
2t
u ij = exp 1
2
exp 3t 1
(
(
)
)
0 0
0
+ exp 2t 1
0 0
0 0 exp 3t 1

(
(
)
)
0 0
0 0
0 0
0
exp 2t 1 exp 3t 1
1
= exp 2t 1
0
0
= ij
2
3t
0
0
exp 1
(2.184)
We also provide the infinitesimal spin tensor:

Ciudad Real - Spain
Draft
( )ij
241
0
(exp 2t 1) (exp 3t 1)0
1
0
0
= (exp 2t 1)
(exp 3t 1)
0
0
(2.185)
Then, the rate of change of is:

&
( )ij

exp 2t 1 exp 3t 1
0
D
D 1
( )ij = exp 2t 1
0
0
=
Dt
Dt 2
3t
exp 1
0
0
0
exp 2t exp 3t
0
0
= exp 2t
3 exp 3t
0
0
(2.186)

&
D=
(2.187)
Problem 2.70
Consider a material body in a small deformation regime, which is subjected to the
following displacement field:
u1 = (2 x1 + 7 x 2 ) 10 3
; u 2 = (10 x 2 x1 ) 10 3
; u 3 = x3 10 3
a) Find the infinitesimal spin and strain tensor;

b) Find the principal invariants of the infinitesimal strain tensor, as well as the
correspondent characteristic equation;
c) Draw the Mohrs circle for strain, and obtain the maximum shear strain;
d) Find the dilatation and the deviatoric infinitesimal strain tensor.
Solution
a) For the displacement gradient we obtain:
(u)ij
u1
x1
u i u 2
=
=
x j x1
u 3
x1
u 1
x 2
u 2
x 2
u 3
x 2
u1
x 3 2
7
0
u 2
= 1 10 0 10 3
x 3
0
0
1
u 3
x 3
m
m

In the International System of Units the displacement gradient is dimensionless, i.e.

r
r u m
[u] = r = .
x m
As for the infinitesimal spin tensor we obtain:

Ciudad Real - Spain
Draft
242
ij =
skew r
u ij
u j
1 u
= i
2 x j x i
0 4 0

= 4 0 0 10 3

0 0 0
Then for the infinitesimal strain tensor we have:
ij =
sym r
u ij
u j
1 u
= i +
x j xi
2
3 0
2

= 3 10 0 10 3

0
0 1
b) The principal invariants are defined as I = Tr ( ) ,
II =
III = det ( ) , (see Chapter 1). Then, it follows that:
1
[Tr()]2 Tr ( 2 ) ,
2
I = Tr ( ) = (2 10 + 1) 10 3 = 11 10 3
2
3
0 2
3
0 2
3
0
1
2
2
II = [Tr ( )] Tr ( ) = 3 10 0 + 3 10 0 + 3 10 0 10 6 = 1 10 6
2
0
0
1
0
0 1
0
0
1
III = det ( ) = 11 10 9
Then, the characteristic determinant is:

2 10 3
3 10 3
0
3
10 10 3
0
3 10
=0
3
0
0
1 10
whilst the characteristic equation is:

3 I 2 + II III = 0
3 + 11 10 3 2 + 11 10 6 11 10 9 = 0
c) To draw the Mohrs circle for strain, (see Appendix A - textbook), we need to evaluate
the eigenvalues of . But, if we take a look at the components of we can verify that
= 1 is already an eigenvalue associated with the direction n i = [0 0 1] . So, to obtain

the remaining eigenvalues one only need solve the following system:
2 10 3
3 10 3
=0
3
10 10 3
3 10
+ 12 10 + 11 10
1 = 1.0 10 3
=0
2 = 11.0 10 3
Then by restructuring the eigenvalues such that I > II > III , we obtain:
I = 1.0 10 3
II = 1.0 10 3
III = 11.0 10 3
Then the maximum shear (tangential) strain is evaluated as follows:

S max =

Ciudad Real - Spain
I III
= 6 10 3
2
Draft
243
Finally, the Mohrs circle for strain can be depicted as:

S (10 3 )
S max = 1 max = 6
2
II = 1
III = 11
N (10 3 )
I = 1
d) The volumetric strain (dilatation) - V is:

V = I = Tr ( ) = 12 10 3
The additive decomposition of into a spherical and a deviatoric part is denoted by

= sph + dev , where the spherical part is given by:
sph
ij
0
4 0
Tr ( )
0 4 0 10 3
=
ij =
3
0
0 4
And, the deviatoric part is given by:

dev
ij
= ij
sph
ij
2
3 0 4 0
0
2 3 0
0 4 0 10 3 = 3 6 0 10 3
= 3 10 0
0
0 0 4
0
0 0
0 4

Problem 2.71
At one point, the displacement gradient is represented by its components as follows:
4 1 4
r
(u) ij = 1 4 2 10 3
4 0
6
(2.188)
Obtain:
a)
the infinitesimal strain and spin tensors;
b) the components of the spherical and deviatoric parts of the infinitesimal strain tensor;
c)
the principal invariants of : I , II , III ;
d) the eigenvalues and eigenvectors of the rate-of-deformation tensor.

Solution:
a) The infinitesimal strain tensor ( ) is the symmetric part of the displacement gradient:

Ciudad Real - Spain
Draft
244
r 1
r
r
= sym u = (u) + (u) T
2
(2.189)
Then:
4 1 4 4
1 4
8 0 0 4 0 0
1
+ 1 4 0 = 1 0 8 2 = 0 4 1
ij = 1 4 2

2
2
4 0
0 2 12 0 1 6
6 4 2 6

r
The infinitesimal spin tensor = skewu
4 1 4 4
1 4
0 2 8 0 1 4
1
1 4 0 = 1 2 0
ij = 1 4 2
1
2 = 1 0

2
2
8 2 0 4 1 0
4 0
6 4 2 6
[ 10 ]
3
[ 10 ]
3
b) The tensor can be additively decomposed into a spherical and deviatoric part:
(2.190)
= sph + dev
where the spherical part is given by:
sph
Tr ( )
6
=
1 = 1 = 21
3
3
sph
ij
2 0 0
= 0 2 0
0 0 2
[ 10 ]
3
(2.191)
The deviatoric part is given by:

dev
ij
4 0 0 2 0 0 2 0 0
= 0 4 1 0 2 0 = 0 6 1
0 1 6 0 0 2 0 1 4
[ 10 ]
3
(2.192)
c) The principal invariants of are:
[ 10 ]
3
I = Tr ( ) = 6
I =
I
4 1 4 0 4 0
+
+
= 17
1 6 0 6 0 4
[ 10 ]
3 2
(2.193)
[ 10 ]
3 3
III = 4 (4) 6 4 = 100
d) The infinitesimal strain tensor components:

4 0 0
ij = 0 4 1
0 1 6
[ 10 ]
3
(2.194)
Note that 1 = 4 10 3 is one eigenvalue associated with the eigenvector [ 1,0,0] . To

obtain the remaining eigenvalues, we need to solve the characteristic determinant:
4
1

Ciudad Real - Spain
1
6
Draft
=0
(2.195)
245
(4 )(6 ) 1 = 0
2 2 25 = 0
=
2 (2) 2 4 1 (25)
2 4 + 4 25
b b 2 4ac
=
=
= 1 26
2a
2 1
2
(2.196)
= 6.0990
1
2 = 4.099
thus:
1 = 4 10 3 ;
2 = 6.0990 10 3 ;
3 = 4.099 10 3
(2.197)
III = 4.099 10 3
(2.198)
Restructuring we obtain:
I = 6.0990 10 3 ;
II = 4 10 3 ;
Problem 2.72
Obtain the infinitesimal strain tensor and the infinitesimal spin tensor for the following
displacement field:
x12
u i = x1 x 2
0
Solution:
The infinitesimal strain tensor
In the small deformation regime, the strain tensor is given by:
L
L
E ij eij ij =
1 u i u j
+
2 x j
x i
We need to obtain the displacement gradient components:

u 1
x1
u j u 2
=
x k x1
u 3
x1
u1
x 2
u 2
x 2
u 3
x 2
u1
x 3 2 x
1
u 2
= x2
x 3

u 3 0
x 3
0
0
0
x1
0

L
E ij
L
eij
ij
u j
1 u
= i +
x i
2 x j
2 x
1 1
= x2
2
0
x1
0
0 2 x1
0 + 0

0 0

x2
x1
0
0 2 x1
0 = x 2
0 2
x2
2
x1
0
The infinitesimal spin tensor:

Ciudad Real - Spain
Draft
246
ij
1 u i u j
x i
2 x j
2 x1
1
= x2
2
0
x1
0
0 2 x1
0 0

0 0

x2
x1
0
0
0
x
0 = 2
2
0 0
x2
2
0
0
Problem 2.73
Figure 2.22 shows the transformation experienced by the square ABCD of unit side.
X 2 , x2
x2
D
C
45
C
B
A = A
X 1 , x1
x1
Figure 2.22: Body subjected to rotation.

a) State the equations of motion;
b) Is the theory valid for small deformation? justify the answer;
c) Is the finite deformation valid? Justify.
Solution:
The transformation law between systems x x is given by:
x1
x1 cos sin 0 x1

= 45
x 2 = sin cos 0 x 2 x 2 =
x
x 0
0
1 x3
3

3
2
2
2
2
0
2
2
2
2
0
0
x1

0 x 2

1 x 3
(2.199)
NOTE: Remember that by definition of the transformation matrix a ij from x to x is

given by:
cos( x1 , x1 ) cos( x1 , x 2 ) cos( x1 , x3 ) cos(315 ) cos(225 ) cos(90 )

cos( x , x ) cos( x , x ) cos( x , x ) = cos(405 ) cos(345 ) cos(90 )
a ij =
2
1
2
2
2
3
cos( x3 , x1 ) cos( x3 , x 2 ) cos( x3 , x3 ) cos(90 ) cos(90 ) cos(0 )

Ciudad Real - Spain
Draft
247
Considering the spatial and material coordinates are superimposed, the equations of
motion are defined by the inverse of the equation in (2.199):
x1

x 2 =
x
3
2
2
2
2
0
0
X 1

0 X 2

1 X 3
2
2
2
2
0
2
2
X1 +
X2
x1 =
2
2
2
2
x2 = 2 X 1 + 2 X 2
For example, the point C in the reference configuration has the material coordinates
X 1C = 1 ,
C
x2 =
C
X 2 =1.
After
the
motion
we
C
x1 =
have
2
2
(1) = 2 ,
(1) +
2
2
2
2
(1) = 0
(1) +
2
2
Displacement field:
u1 = x1 X 1 =
u 2 = x 2 X 2 =
2
2
2
X1
X 2 X1 = X1
1
2
2 X2
2
2
2
2
2
2
X1 +
X2 X2 =
X1 + X 2
1
2
2
2
2
Displacement material gradient:

u1
X 1
u i u 2
=
X j X 1
u 3
X 1
u1
X 2
u 2
X 2
u 3
X 2
u1 2
1

X 3 2
u 2 2
=
X 3 2

u 3
X 3 0

The infinitesimal strain tensor is given by = symu =

2
1
ij = 0
0
1 0
2
0
0
2
2
r
r
1
(u) + (u)T , thus:
2
1 0 0 ij
2
0
0
Note that, for a rigid body motion the strain tensors must be equal to zero, i.e. = 0 (the
infinitesimal strain tensor), E = 0 (the Green-Lagrange strain tensor), e = 0 (the Almansi
strain tensor). Calculating the Green-Lagrange strain tensor components we have:
u j u k u k
1 u
+
E ij = i +
2 X j X i X i X j

Ciudad Real - Spain
Draft
0 0 0

= 0 0 0

0 0 0
248
Problem 2.74
A square of side b turns counterclockwise of 30 . After turning the square is deformed
such that the base maintains its initial length and the height is doubled (see Figure 2.23).
Calculate the deformation gradient, the right Cauchy-Green deformation tensor, and the
Green-Lagrange strain tensor.
Solution:
X 2 , x2
C
x2
D
2b
30
x1
30
A = A
X 1 , x1
Figure 2.23: Body under rotation/deformation.

Note that we can apply the decomposition of motion: first we apply a pure deformation
and then a rotation is applied:
The motion is governed by the right
stretch tensor of the polar
decomposition:
X 2 , x2
D
2b
1 0 0
U ij = 0 2 0
0 0 1
where we have applied the

definition of stretch. Note that they
are principal values. We then apply a
rotation, where the components of
R
are the same as the
r
transformation matrix from the x r
system to the x -system:
B B
A = A
X 1 , x1
cos sin 0
R ij = sin cos 0
0
0
1
Then, applying the left polar decomposition F = R U we obtain:

Ciudad Real - Spain
Draft
249
cos sin 0 1 0 0 cos 2 sin 0

Fij = R ik U kj = sin cos 0 0 2 0 = sin 2 cos 0
0
0
1 0 0 1 0
0
1
For the proposed problem, we have:

cos 30 2 sin 30 0
Fij = sin 30 2 cos 30 0
0
0
1
As we are dealing with a homogenous deformation the equation x = F X + c holds, in

r r
this case with c = 0 . For example, for a particle at point D in the reference configuration
moves to the point:
x1D cos 30 2 sin 30 0 X 1D cos 30 2 sin 30 0 0 2b sin 30
D

D
x 2 = sin 30 2 cos 30 0 X 2 = sin 30 2 cos 30 0 b = 2b cos 30
x D 0
0
1 X 3D 0
0
1 0
0
a fact that can be easily checked by means of Figure 2.23.

By means of definition of the right Cauchy-Green deformation tensor, C = F T F , we can
obtain the Cartesian components:
sin 0 cos 2 sin 0 1 0 0
cos
2 sin 2 cos 0 sin 2 cos 0 = 0 4 0
C ij = Fki Fkj =
0
0
1 0
0
1 0 0 1
1
2
The Green-Lagrange strain tensor, E = (C + 1) , and its components are:

1 0 0 1 0 0 0 0 0

1
E ij = 0 4 0 0 1 0 = 0 1.5 0
2
0 0 1 0 0 1 0 0 0
Note that the original space coincides with the principal space. We could also have
obtained the components of C and E by means of its spectral representations:
C=
2a N( a ) N ( a ) , E =
a =1
2
a
1)N ( a ) N ( a ) , where a are the principal stretches.
a =1

Ciudad Real - Spain
2 (
Draft
250

Ciudad Real - Spain
Draft
3 Stress
3.1 Solved Problems
3.1.1
Force, Stress Tensor, Stress vector
Problem 3.1
Ignoring the curvature of the Earths surface, the gravitational field can be assumed to be
uniform as shown in Figure 3.1, where g is the acceleration caused by gravity (the gravity
of the Earth). Find the resultant force acting on the body B .
x3
x2
x1
Figure 3.1: Gravitational field.
Solution:
All bodies immersed in a force field are subjected to the body force b , and in the special
case presented in Figure 3.1 this is given by:
0
r
b i ( x, t ) = 0

g
m
s2

Hence, the total force acting on the body can be evaluated as follows:
252
Fi =
r
0
b i ( x , t ) dV =
g dV
[m3 ]
kg m } kg m
We can also verify the F unit: [F] = 3 2 dV = 2 = N ( Newton ) .
m s
s

V
Problem 3.2
The Cauchy stress tensor components at a
point P are given by:
8 4 1
ij = 4 3 0.5 Pa
1 0 .5 2
r
a) Calculate the traction vector ( t (n ) ) at P
x3
C (0,0,5)
which is associated with the plane ABC
defined in Figure 3.2

b) With reference to paragraph a).
r
B (0,2,0)
Obtain the normal ( N ) and tangential ( S )

traction vectors at P (see Appendix A of the
textbook, (Chaves(2013)).
x1
x2
A(3,0,0)
Figure 3.2: Plane ABC .

Solution:
First, we obtain the unit vector which is normal to the plane ABC . To do this we choose
two vectors on the plane:
BA = OA OB = 3e 1 2e 2 + 0e 3
BC = OC OB = 0e 1 2e 2 + 5e 3
Then, the normal vector associated with the plane ABC is obtained by means of the cross
product between BA and BC , i.e.:
e1
r
n = BC BA = 0
3
e2
e3
5 = 10e 1 + 15e 2 + 6e 3
0
r
Additionally, the unit vector codirectional with n is given by:
v
n 10
15
6
= v = e1 + e 2 + e 3
n
n 19
19
19
2
2
we can obtain the traction components as:

Ciudad Real - Spain
Draft
3 STRESS
t (n)
i
= ij n j
253
t1
8 4 1 10
t = 1 4 3 0.5 15 Pa
2 19

t 3
1 0 .5 2 6

t1
26
t = 1 8 Pa
2
19
t 3
29.5

b) The traction vector t (n) associated with the normal n can be broken down into a
r
r
normal ( N ) and a tangential ( S ) vector as shown in Figure 3.3. Then,
r
r
r
t (n ) = N + S
t (n) = N n + S s
or
r
where N and S are the magnitudes of N and S , respectively.

r
t (n)
x3
t (n) = n
r
r
r
t (n ) = N + S
r
N = N n
r
N
r
S
r
t ( n)
e3
e2
e1
= 2 + 2
N
S
x2
x1
Figure 3.3: Normal and tangential stress vector.

As we have seen in Appendix A (textbook), the normal component, N , can be evaluated
as follows:
r

N = t ( n ) n = ( n) n = n n = : (n n) = t i( n) ni = ( ij n j )ni = ni ij n j = ij (ni n j )
Thus:
N = t i ni
10
1
= 2 [26 8 29.5] 15 1.54 Pa

19
6

Then the tangential component, S , can be obtained by means of the Pythagorean

Theorem, i.e.:
r
t ( n)
= 2 + 2
N
S
2 = t i( n ) t i( n ) 2
S
N
where
t i( n) t i( n)
26
1
= 2 [26 8 29.5] 8 4.46
19
29.5
Thus,
S = t i( n ) t i( n ) 2 = 4.46 2.3716 2.0884 Pa

N

Ciudad Real - Spain
Draft
254
Problem 3.3
The stress state at a point in the continuum is represented by the components of the
Cauchy stress tensor as:
2 1 0
ij = 1 2 0 Pa
0 0 2
2
a) Obtain the components of in a new system x1 , x , x3 , where the transformation
matrix is given by:
x3
3 0 4
1
a ij = A = 0 5 0
5
4 0 3
x2
x3
x1
where
e3
a11 = cos 1
e2
3
e
a12 = cos 1
a13 = cos 1
e1
1
e
e2
x2
M
x1
b) Obtain the principal invariants of ;

c) Obtain the eigenvalues and eigenvectors of . Also verify if the eigenvectors form a
basis transformation between the original and the principal space;
d) Illustrate the Cauchy stress tensor graphically, i.e. with the Mohrs circle in stress, (see
Appendix A);
e) Obtain the spherical ( sph ) and the deviatoric ( dev ) part of . Also, find the principal
invariants of dev ;
f) Obtain the octahedral normal ( oct ) and tangential ( oct ) components of , (see
N
S
Appendix A).
Solution:
a) As we have seen in Chapter 1, the transformation law for the components of a secondorder tensor is given by:
= a ik a jl kl
ij
Matrix form
= A AT
Thus,
3 0 4 1 1 0 3 0 4 2 0 .6 0
1
= 2 0 5 0 2 2 0 0 5 0 = 0 .6 2 0 .8
ij
5
4 0 3 0 0 2 4 0 3 0 0 .8 2
These new components ij can be appreciated in Figure 3.4.

Ciudad Real - Spain
Draft
3 STRESS
x3
x3
255
x
2
x2
x1
x1
= A AT
x3
33
23
13
13
x3
33
23
23
13
22
12
x2
12
11
23
13
22
x
2
12
12
11
x1
x1
= AT A
Figure 3.4: Basis transformation.
b) The principal invariants of the Cauchy stress tensor can be calculated as follows:
I = Tr ( ) = ii = 11 + 22 + 33
] (
1
1
( Tr ) 2 Tr ( 2 ) = ii jj ij ij
2
2
2
2
= 11 22 + 11 33 + 33 22 12 13 2
23
II =
1
ii jj kk 3 ii jk jk + 2 ij jk ki
6
2
2
+ 212 23 13 11 2 22 13 33 12
23
III = det ( ) = ijk i1 j 2 k 3 =

= 11 22 33
By substituting the values of ij for those in the proposed problem we obtain:

I = 6
II =
2 0
0 2
2 0
0 2
2 1
1 2
= 11
III = 6
c) The principal stresses ( i ) and principal directions ( n(i ) ) are obtained by solving the
following set of equations:

Ciudad Real - Spain
Draft
256
1
0 n1 0
2
1
2
0 n 2 = 0

0
0
2 n 3 0
To obtain the nontrivial solutions of n(i ) we have to solve the characteristic determinant,
which is a cubic equation for the unknown magnitude :
ij ij = 0
3 I 2 + II III = 0
However, if we look at the format of the Cauchy stress tensor components, we can notice
that we already have one solution as in the x3 -direction the tangential components are
equal to zero, then:
(
(
3 = 2 Principal n13) = n (3) = 0 , n 33) = 1
direction
2
To obtain the other two eigenvalues, one only need solve:

2
= (2 ) 1 = 0
2
1 = 1
2 = 3
Then we can express the Cauchy stress tensor components in the principal space as:
1 0 0
= 0 3 0 Pa
ij
0 0 2
Additionally, the principal direction associated with 1 = 1 is calculated as follows:

(
1
0 n11) 0
2 1
(1)
(1)
1
n (1) = 0 n1 + n 2 = 0 n (1) = n (1)
2 1
0 2 (1)
1
2
(1)
n (1) 0 n1 + n 2 = 0
0
0
2 1 3
(
(
with n 31) = 0 and by using the condition n11) + n (21) = 1 we obtain:
(
n11) = n (1) =
2
1
2
then n i(1) =
Since is a symmetric tensor, the principal space is formed by an orthogonal basis, so, it
is valid that:
n (1) n ( 2 ) = n ( 3)
n ( 2 ) n ( 3) = n (1)
n ( 3) n (1) = n ( 2)
Thus, the second principal direction can be obtained by the cross product between n ( 3)
and n (1) , i.e.:
e2
0
1
n ( 2 ) = n ( 3) n (1)
e1
= 0
1
e3
1
1
1 =
e1 +
e2
2
2
0
which can also be checked by the following analysis:

The Principal direction associated with 2 = 3 :

Ciudad Real - Spain
Draft
3 STRESS
257
(
1
0 n12 ) 0
2 3
( 2)
( 2)
n ( 2 ) = 0 n1 + n 2 = 0 n ( 2 ) = n ( 2 )
1
23
0 2 ( 2)
1
2
( 2)
(
n 32 ) 0 n1 n 2 = 0
0
0
2 3
(
(
With n 33) = 0 and using the condition n13) + n (23) = 1 we obtain:
(
n 12 ) = n ( 2 ) =
2
then n i( 2 ) =
1
2
As we have seen in Chapter 1, the eigenvectors of a symmetric tensor form the

transformation matrix D , from the original system to the principal space, i.e.
= D D T , thus:
0
0
1 = 1
0
2 = 3
0 =

0
3 = 2
0
2
1
2
1
1
0
2 1 0 2
1
0 1 2 0
2
0 0 2
1
0
2
1
2
0
d) The graphical representation of a second-order tensor can be obtained from the

description in Appendix A. To do this we have to restructure the eigenvalues of so that
I > II > III , thus:
I = 3
II = 2
III = 1
Then the three circumferences are defined by:

Circle 1
Circle 2
Circle 3
1
( II + III ) = 1.5
2
1
(center )C 2 = ( I + III ) = 2.0
2
1
(center )C 3 = ( I + II ) = 2.5
2
(center )C1 =
;
;
;
1
( II III ) = 0.5
2
1
(radius ) R 2 = ( I III ) = 1.0
2
1
(radius ) R3 = ( I II ) = 0.5
2
(radius ) R1 =
Then, we can illustrate the Cauchy stress tensor at P by means of Mohrs circle in stress as
shown in Figure 3.5.

Ciudad Real - Spain
Draft
258
S max =
S max = 1
1
( I III ) = 1.0
2
R2
R1
C3
III = 1
C1
R3
II = 2
I = 3 = N max
Figure 3.5: Mohrs circle in stress at the point P .

e) As defined in Chapter 1 of the textbook, a second-order tensor can be broken down
additively into a spherical and a deviatoric part, i.e.:
Tensorial notation
Indicial notation
sph
dev
ij = ij + ij
= sph + dev
= m1 +
1
dev
= kk ij + ij
3
dev
= m ij + ij
dev
(3.1)
A schematic representation of these components in the Cartesian basis can be appreciated

in Figure 3.6 and the value of the scalar m is evaluated as follows:
m =
11 + 22 + 33 1 + 2 + 3 1
I
1
6
=
= kk = Tr ( ) = = = 2
3
3
3
3
3 3
Then the spherical part becomes:

sph
ij
2 0 0
= m ij = 2 ij = 0 2 0
0 0 2
And, the deviatoric part can be evaluated as follows:

dev
ij
11 12 13 m 0
= 12 22 23 0 m

13 23 33 0
0

1 (211 22 33 )
3
1
(2 22
=
12
3
13
0
0
12
11 33 )
23
23
11 22 )
13
1
3
(2 33
Thus,
dev
ij

Ciudad Real - Spain
1
0 0 1 0
2 2
1
=
22
0 = 1 0 0

0
0
2 2 0 0 0
Draft
3 STRESS
259
Now let us remember from Chapter 1 that and dev are coaxial tensors, i.e., they have
the same principal directions, so we can use this information to operate in the principal
space of to obtain the eigenvalues of dev = sph . With that we obtain:
dev
ij
1
=0
m
0

3 0

0
2
0
0
0 1 0 0
0 = 0 1 0

m 0 0 0

0
m
0
Then the invariants of dev are given by:

I dev = Tr ( dev ) = 0
II dev = 1
III dev = 0
Traditionally, in engineering, the invariants of the deviatoric stress tensor are represented
by:
J1 = I dev = 0
J 2 = II
dev
J 3 = III dev
1 2
I 3 II
3
1
3
=
2 I 9 I II + 27 III
27
=
x3
33
23
13
13
23
12
11
22
12
x2
1
14444444 244444444
4
3
x3
x3
dev
33
23
13
m
13
x2
23
12
dev
22
12
dev
11
x1
x2
x1
sph
dev
Figure 3.6: The spherical and deviatoric part of .

Ciudad Real - Spain
Draft
260
f) The octahedral normal and tangential components, (see Appendix A in Chaves (2013)),
can be expressed as:
oct =
N
oct
S
oct
1
(1 + 2 + 3 ) = 1 ii = I = m
3
3
3
1
2
2
2 I 6 II =
=
J2 =
3
3
( ) + ( ) + ( )
dev 2
1
dev 2
2
dev 2
3
Then, by substituting the values of the proposed problem we obtain:

oct = m = 6
N
oct =
2
2
J2 =
3
3
Problem 3.4
At point P the Cauchy stress tensor components are:
1 2 3
ij = 2 4 6 MPa
3 6 1
Find:
(3.2)
a) the traction vector t related to the plane which is normal to the x1 -axis;
r
b) the traction vector t associated with the plane whose normal is (1,1,2) ;
r
c) the traction vector t

2 x1 2 x 2 x3 = 0 ;
associated with the plane parallel to the plane
d) the principal stress at the point P ;

e) the principal directions of at the point P .
Solution:
a) In this case, the unit vector is (1,0,0) . Then, the traction vector is given by:
t i(n)
1 2 3 1 1
= 2 4 6 0 = 2

3 6 1 0 3
(3.3)
b) The unit vector associated with the direction (1,1,2) is:

1
1
1
ni =
6
2

(3.4)
thus,

Ciudad Real - Spain
Draft
3 STRESS
t i(n)
261
5
1 2 3 1
1
1 = 1 10
=
2 4 6
6
6
1
3 6 1 2

(3.5)
c)
5
1 2 3 2
2
1
1
2 = 1 10
(n)
n i = 2 t i = 2 4 6
3
3
3
7
3 6 1 1
1
(3.6)
d) Solving the characteristic determinant

1
2
2
4
3
3
6
=0
(3.7)
we obtain:
1 = 10 ; 2 = 0 ; 3 = 4
(3.8)
e) The principal directions are:

Associated with 1 = 10
3
9n1 + 2n 2 + 3n 3 = 0

(1)
2n1 6n 2 + 6n 3 = 0 n i = 6
3n + 6n 9n = 0
5
2
3
1

(3.9)
Similarly, we obtain:
n ( 2)
i
2
1
1 ; n ( 3) = 2
= i

0
3

(3.10)
Normalization of the principal directions:
n i(1)
3
n i(1)
1
6
= r
=
70
n (1)
5

n i( 2 )
2
n ( 2)
1
i
1
= r
=
5
n ( 2)
0

n i(3)
1
n ( 3)
1
i
2
= r
=
14
n ( 3)
3

Problem 3.5
r

Show that S = t (n) (1 n n) , where t (n) is the traction vector resulting from
projecting the second-order tensor onto the n -direction, and S is the tangential stress
vector associated with the plane.
Solution 1:
r
r
r

S = t (n) t (n ) n n
r (n) r (n)
r

S = t t n n
r (n)
r

S = t (1 n n)

Ciudad Real - Spain
Draft
262
Solution 2:
We can also solve the problem just using the components of the equation
r
r

S = t (n) [ : (n n)]n :
Si
= t i(n) (n k n l kl ) n i
(n)
(n)

= t i nink t k
(n)
k
= t k ik n i n k t (n)

= t (n) ( ik n i n k )
k
which in tensorial notation becomes

r
r

S = t (n) (1 n n)
Problem 3.6
The stress state at one point P of the continuous medium is given schematically by:
x3
1
22
4
1
x2
x1
Obtain the value of the component 22 of the Cauchy stress tensor such that there is at
least one plane passing through P in which is free of stress;
Obtain the direction of the plane.
Solution:
r
We seek to find a plane whose direction is n such that t (n ) = 0 . We can relate the Cauchy
stress tensor to the traction vector by means of the equation:
t (n) = n
thus:
(
t1n ) 0 1
(n )
t 2 = 1 22
t (n ) 4 1
3
4 n1 0

1 n 2 = 0

0 n 3 0

Resulting in the following system of equations:

Ciudad Real - Spain
Draft
3 STRESS
263
n2 + 4n3 = 0 n3 = 4 n 2
n1 + 22n 2 + n3 = 0
1
4n1 + n2 = 0 n1 = n 2
4
By combining the above equations we obtain:

n1 + 22n 2 + n3 = 0
1
1
n2 + 22n 2 n2 = 0
4
4
1
1
+ 22 n 2 = 0
4
4
r r
1
1
1
Then, for n 0 , we have: + 22 = 0 22 = .

To determine the direction of the plane we start by the restriction n i n i = 1 , then:
2
2
nini = 1 n1 + n 2 + n3 = 1
2
2
1
1
n2 + n2 + n2 = 1
2
4
4
2
6
r
r
Thus, the direction of the normal to the plane, when it meets t (n ) = 0 :
n2 =
2 2
3
n1 = n3 =
1
2
4
ni =
6
1

3.1.2
Eigenvalues and Eigenvectors of the Stress Tensor
Problem 3.7
The stress field in the medium is represented by:
1
ij = 0
2 x2
0
1
4 x1
2 x2
4 x1
(3.11)
where xi are the Cartesian coordinates.

a) Obtain the traction vector acting at the point ( x1 = 1, x2 = 2, x3 = 3) associated with
the plane x1 + x 2 + x3 = 6 ;
b) Obtain the projection of the traction vector according to the normal and tangential
direction to the plane x1 + x 2 + x3 = 6 ;
Solution:
a) The unit vector which is normal to the plane x1 + x 2 + x3 = 6 is:
Ciudad Real - Spain
Draft
264
1
i = 1 1
n
3
1
t (n) = n
(3.13)
1 0 4
ij ( x1 = 1, x 2 = 2, x3 = 3) = 0 1 4
4 4 1
(3.12)
(3.14)
The traction vector t (n) :
thus,
1 0 4
1
0 1 4 1 1
=
3
4 4 1
1
1
5
=
3
9

t i(n)
(3.15)
b) The normal stress associated with this plane is

1
r (n)
1
1 1
N = t n =
[5 5 9] 1 = (5 + 5 + 9) = 19
3
3
3
3
1
(3.16)
and the tangential stress is

r r
2 = 2 + t ( n ) t ( n )
S
N
(3.17)
5
r (n ) r ( n )
1
t t =
[5 5 9] 5 1 = 131
3
3
3
9

(3.18)
thus
2
32
19 131
2 = +
=
S
3
9
3
(3.19)
Problem 3.8
Given a continuum where the stress state is known at one point and is represented by the
Cauchy stress tensor components:
1 1 0
ij = 1 1 0 Pa
0 0 2
(3.20)
a) Find the principal stresses and the principal directions (eigenvectors).

Ciudad Real - Spain
Draft
3 STRESS
265
Solution:
To obtain the principal stresses i = i and principal directions n(i ) we must solve the
following system of equations:
1
1
1
1
0
0
0 n1 0
0 n 2 = 0

2 n 3 0

(3.21)
for nontrivial solutions of n(i ) , which is equivalent to solve:
ij ij = 0
But if we look at the format of the matrix (3.20), we can note that we have one solution,
since the tangential components in the x3 -direction are zero, then:
direction
(
(
1 = 2 n11) = n (1) = 0 , n31) = 1
2
To obtain the remaining solutions it is sufficient to solve:

1
1
= (2 ) = 0
1
1
We can easily verify that the roots of the above equations are:
2 = 2 and 3 = 0
We express the stress tensor components in the principal space as follows:

2 0 0
= 0 2 0 Pa
ij
0 0 0
b) The principal directions

b.1) To obtain the principal direction associated with the solution 2 = 2 , we substituting
this solution into the equation (3.21):
(
1
0 n12 ) 0
1 2
1
1 2
0 n ( 2 ) = 0

2
(
0
0
2 2 n 32 ) 0
(
n 12 ) + n ( 2 ) = 0
2
( 2)
( 2)
n1 n 2 = 0
(
(
Solving the system we obtain n 32) = 0 , n12) = n (22) and by using the restriction
2
(
(
n12 ) + n ( 2 ) = 1 we obtain: n12 ) = n ( 2 ) =
2
2
. n ( 2) =
b.2) For the solution 3 = 0 , we obtain:

(
n13) + n ( 3) = 0
2
( 3)
( 3)
n1 + n 2 = 0
(3)
2n 3 = 0
(
1
0 n13) 0
1 0
1
1 0
0 n ( 3) = 0

2
(
0
0
2 0 n 33) 0
(
(
Solving the system we obtain n (33) = 0 , n13) = n (23) and using the restriction n13) + n (23) = 1 ,
(
we obtain: n13) =
1
1 ( 3) 1
, n (23) =
. n =
2
2
2

Ciudad Real - Spain
Draft
1
2
0 .
266
As we have seen, the eigenvectors form a matrix transformation ( A ) between the two
systems, i.e. = A A T , thus:
0
0
1 = 2
0
2 = 2
0 =

0
0
3 = 0
0
1
2
1
2
0
1
1 1 0
1
0 1 1 0
2
0 0 2 1
0
1
2
1
0
1
2
1
Problem 3.9
A prismatic dam is subjected to water pressure. The dam has thickness equal to b and
height equal to h , (see Figure 3.7). Obtain the restrictions of the Cauchy stress tensor
Cartesian components on the faces BC , OB and AC .
x2
a - mass density of water

g - acceleration of gravity
C
a
a g (h x 2 )
h
x1
Figure 3.7
Solution:
The face BC has normal vector n (i BC ) = [0 1 0] . Considering that this face has no
traction vector, we conclude that:
t i( BC )
= 0 i = ij n j
11
21
31
12
22
32
13 0 12 0
23 1 = 22 = 0

33 0 32 0

what is the same as i 2 = 0 and because of the symmetry of we have 2i = 0 .
The face OB has as normal vector n (i BC ) = [ 1 0 0] . Considering that in this face the
traction vector components are t i(OB ) = [ a g (h x 2 ) 0 0] , we conclude that:
t i(OB )
a g ( h x 2 )
= n
=
0
ij j

Ciudad Real - Spain
11
21
31
Draft
12
22
32
13 1 11 a g (h x 2 )
0
23 0 = 21 =

0
33 0 31

3 STRESS
267
which is the same as i1 = a g (h x 2 ) i1 .
The face AC has normal vector n (i BC ) = [1 0 0] . Considering that in this face there is no
traction vector, we conclude that:
t i( AC )
= 0 i = ij n j
11
21
31
12
22
32
13 1 11 0
23 0 = 21 = 0

33 0 31 0

which is the same as i1 = 0 and because of the symmetry we have 1i = 0 .
3.1.3
Other measure of stress
Problem 3.10
Prove that the following relationship are valid:
P = J dev F T + J m F T
S = JF 1 dev F T + J m C 1
where P and S are the first and second Piola-Kirchhoff stress tensors, respectively, C is
the right Cauchy-Green deformation tensor, F is the deformation gradient, J is the
Jacobian determinant, and the scalar m is the mean normal Cauchy tress. Also prove that
the following relationships are true:
P : F = S : C = 3J m
Solution:
First of all we prove that P : F = S : C :
P:F
= Pij Fij
= ( Fik S kj ) Fij = S kj ( Fik Fij )
= S kj ( F T F ) kj = S kj (C ) kj
= S :C
Secondly, by referring to the definition P = J F T , and the different components of

by = sph + dev , we obtain:
P = J ( dev + m 1) F T
= J dev F T + J m 1 F T
= J dev F T + J m F T
Thirdly, by taking into account the definition S = JF 1 F T , and by breaking down

into = sph + dev , we obtain:
S ij = JFik 1 kp F jp1
S = JF 1 F T
= JF 1 ( dev + m 1) F T
= JFik 1 ( dev + ( m ) kp ) F jp1

kp
= JF 1 dev F T + JF 1 m 1 F T
= JFik 1 dev F jp1 + JFik 1 ( m ) kp F jp1

kp
= JF 1 dev F T + J m ( F T F ) 1
= JFik 1 dev F jp1 + J ( m ) Fik 1 F jk1

kp
= JF 1 dev F T + J m C 1
Ciudad Real - Spain
= JFik 1 dev F jp1 + J ( m ) C ij 1

kp
Draft
268
Then by applying the double scalar product between S and C we can obtain:
S :C
= ( JF 1 dev F T + J m C 1 ) : C
= JF 1 dev F T : C + J m C 1 : C
where the term JF 1 dev F T : C becomes:

JF 1 dev F T : C
= ( JF 1 dev F T ) : {
C
( JF 1 dev F T ) ij ( F T F ) ij
= ( Fip 1 dev F jk1 )( Fqi Fqj )

pk
F T F
= J qp qk dev
pk
= J dev pk = J dev
pk
kk
= J dev4
123
4 :1
=0
Tr ( dev ) =0
Thus:
S : C = J m C 1 : C = J m Tr (C 1 C ) = J m Tr (1) = 3 J m
Now, by taking the double scalar product between P and F we obtain:

P : F = J dev F T : F + J m F T : F
Then by analyzing the term J dev F T : F we can conclude that:

dev
dev
J dev F T : F = ( J dev F T ) ij ( F ) ij = J ik F jk1 Fij = J ik ik = J dev4 = 0
123
4 :1
Tr ( dev ) =0
Thus,
P : F = J m F T : F = J m Tr ( F T F T ) = J m Tr (1) = 3 J m
3.1.4
Maximum shear stress, Mohr circle in stress
Problem 3.11
Obtain the maximum shear stress at a point in which the stress state is given by:
x2
30 MPa
20 MPa
x1
x3
Figure 3.8
Ciudad Real - Spain
Draft
3 STRESS
Solution:
269
Note that the axes xi are

principal directions. We draw the
Mohrs circle with the principal
stresses:
max (MPa)
I = 30 MPa , II = 20MPa and

III = 0 .
20
30
N (MPa)
Figure 3.9
30 0
max =
(3.22)
= 15 MPa
Problem 3.12
Consider the stress state at a point represented by the infinitesimal element shown in
Figure 3.10.
a) Draw the Mohrs circle;
x2
b) Obtain the maximum normal stress,

and indicate the plane in which occurs;
20 MPa
c) Obtain the maximum shear stress.
5 MPa
x1
10 MPa
x3
Figure 3.10
Solution:
S
max = 15
N max = 10MPa
S max =
20

Ciudad Real - Spain
10
Draft
10 (20)
= 15MPa
2
N (MPa)
270
Problem 3.13
Determine for which values of * are possible
the following stress state at the point P :
Case a) N = 4
=2
Case b) N = 4
=1
Case c) N = 7
=0
P
2
Figure 3.11
Solution
The pair ( N ; ) is feasible if it belongs to the gray zone of the Mohrs circle including the
circumferences, (see Figure 3.12).
Figure 3.12: Mohrs circle in stress.

Ciudad Real - Spain
Draft
3 STRESS
271
Case a)
Case b)
Case c)
1
6
Figure 3.13: Mohrs circle.
Case a): In this case the pair ( N = 4; = 2) belongs to the circumference formed by the
principal stresses 2 and 6 , thus * can assume any value, (see Figure 3.14).
Case a)
2
1


Case b) In this case the solution is:
*2) * *1)
(
(
(3.23)
where *2) and *1) are identified in Figure 3.15.

(
(

Ciudad Real - Spain
Draft
272
Limit cases
( 4,1)
( 4,1)
1
* ( x ) 6
* ( 2)
* (1) 6

Starting from the circumference equation:
( x xC ) 2 + ( y y C ) 2 = R 2
For the case *1) , we have: x = 4; x C =

(
( *1) + 2)
(
2
(3.24)
y = 1; y C = 0; R =
( *1) 2)
(
2
Substituting these values into the circumference equation we obtain:

( x xC ) 2 + ( y y C ) 2 = R 2
2
( * 2)
( * + 2)
4 (1)
+ (1 0 )2 = (1)
2
2
(3.25)
*1) = 4.5
(
For the case
*2 )
(
, we have: x = 4; xC =
(6 + *2 ) )
(
2
; y = 1; y C = 0; R =
(6 *2 ) )
(
2
substituting these values into the circumference equation we obtain:

( x xC ) 2 + ( y y C ) 2 = R 2
2
(6 *2 ) )
(6 + *2 ) )
(
(
4
+ (1 0)2 =
2
2
(3.26)
*2 ) = 3.5
(
thus:
(3.27)
3.5 * 4.5
Case c) In this case the only possible solution is that N is a principal stress, then
* = 7

Ciudad Real - Spain
Draft
(3.28)
3 STRESS
273
* = 7
Problem 3.14
Obtain the maximum normal and tangential stress and draw the corresponding Mohrs
circle in stress for the following stress states:
a)
0
ij = 0
0 0 0
(3.29)
2 0 0
ij = 0
0
0
0
(3.30)
b)
Solution:
a) The principal values. If we check the format of the Cauchy stress tensor
components, we can observe that the value (3) = 0 is already an eigenvalue. Then,
to obtain the remaining eigenvalues, it is sufficient to solve:

= ( ) 2 2 = 0 = = 0

(3.31)
(1) = 0
( ) 2 2 = 0 2 2 + 2 2 = 0 (2 + ) = 0
( 2) = 2
(3.32)

Ciudad Real - Spain
Draft
274
N max = 2
max =
max =
Figure 3.17:
b)
Figure 3.18.
N max =
max =
(3.33)
(2) 3
=
2
2
Problem 3.15
Make the representation of the Mohrs circle for the following cases:
1) Unidimensional case (traction);
2)Unidimensional case (compression);
3) Bidimensional case (traction)
4) Triaxial case

Ciudad Real - Spain
Draft
3 STRESS
275
5) State of pure shear

Solution:
1) Unidimensional case (traction)
x
I
I
0
0 0
0 0
0 0
2) Uniaxial compression
x
0
0
0
II
0
0
0
0
II
3) Biaxial case
II
I
II
I
0

Ciudad Real - Spain
0
II
0
0
0
Draft
276
4) Triaxial case
III
I
0
II
0
II
0
0
0
III
III
II
5) State of pure shear

S
0 0
0 0
0 0 0
3.1.5
Feature of the stress tensor
Problem 3.16
The Cauchy stress tensor components at point P are given by:
5 6 7
ij = 6 8 9 GPa
7 9 2
(3.34)
Obtain:
a) the mean stress;
b) The deviatoric and volumetric part of the tensor .
Solution:
The mean stress
m =
kk 5 + 8 + 2
=
=5
3
3
(3.35)
b)

Ciudad Real - Spain
Draft
3 STRESS
sph
ij
m
= 0
0
m
0
0 5 0 0
0 = 0 5 0

m 0 0 5

sph
dev
ij = ij + ij
dev
ij
277
dev
sph
ij = ij ij
(3.36)
0 6 7
= 6 3 9
7 9 3
Problem 3.17

Consider the Cauchy stress tensor components, in the Cartesian base (e1 , e 2 , e3 ) :
5 3 2
ij = 3 1 0
2 0 3
(3.37)
Given the transformation law between the systems x and x' :

x'1
x' 2
x '3
x1
3
5
4
5
x2
3
5
x3
4
5

where the system x' is constituted by the basis (e'1 , e' 2 , e' 3 ) .
r
a) Obtain the traction vector t ( e'2 ) associated with the plane whose normal is e' 2 . Express

the result in the Cartesian system (e'1 , e' 2 , e' 3 ) according to the format:
r
3
t (e'2 ) = ( )e1 + ( )e 2 + ( )e
(3.38)
b) Obtain the spherical and deviatoric parts of the Cauchy stress tensor.
Solution:
a) As defined in the textbook (see Chaves (2013)), the first row of the transformation
matrix is formed by the direction cosines between the x'1 -axis with x1 , x2 and x3 , thus:
3 0 4
1
A = 0 5 0
5
4 0 3
(3.39)
and the transformation law for the second-order tensor components is:
' = A A T

Ciudad Real - Spain
Draft
(3.40)
278
thus:
3
3
5 0 4 5 3 2 5
5
= 0 1 0 3 1 0 0
ij
4 0 3 2 0 3 4
5
5
5
t i( e' 2 )
9 9 2
1
1 0 = 9 5 12
5
2 12 31
0 3
5
0
9
1
= 5
5
12

(3.41)
(3.42)
r
9
12
t (e'2 ) = e1 + (1)e 2 + e
3
5
since:
11

21

31
( e' )
13 t 1 1
( e' )
= t 2 1
23
( e' )
t 3 1
33
12
22
32
t1
( e' 2 )
( e' 2 )
t2
( e' )
t3 2
t1
( e'3 )
( e' 3 )
t2
( e' )
t3 3
(3.43)
b)
sph
dev
ij = ij + ij
(3.44)
I
dev
ij + ij
3
I = 5 +1+ 3 = 9
(3.45)
3 0 0
= 0 3 0
0 0 3
(3.46)
sph
ij
dev
sph
ij = ij ij
3
2 2 3 2
5 3
3
1 3
0 = 3 2 0
=

2
0
3 3 2 0 0
(3.47)
Problem 3.18
The stress state in an continuous medium is given by the Cauchy stress tensor Cartesian
components:
0
ij = Cx 3
Cx 3
0
Cx1
0
Cx1
where C is a constant. Consider that the body is free of body force.

a) Calculate the traction vector at the point P (4,4,7) associated with the plane whose
2
1
2
normal vector is given by n = e1 + e 2 e 3 .
3
b) Represent the Mohrs circle in stress at the point P .

Ciudad Real - Spain
Draft
3 STRESS
279
Solution:
a) The traction vector is given by:
r
(n)
t (n) = n ; t i = ij n j
0
ij ( x1 = 4; x 2 = 4; x3 = 7) = Cx3
Cx3
0
Cx1
(3.48)
0 0
Cx1 = 7C

0 0

7C
0
4C
0
4C
2
1
nj = 2
3
1

(3.49)
(3.50)
thus, we obtain:
0
r (n)
t i = ij n j = 7C
7C
0
4C
0 2
14C
1 2 = 1 18C
4C
3
3
8C
0 1

(3.51)
b)
0
0 7
7 0 4
ij = C
0 4 0
(3.52)
The eigenvalues (principal stresses) are obtained by solving:

0 0 0
0 7
7 0 4 0 0 = 0
C

0 4 0 0 0
C
0
0 7
C 7 0 4 C 0
0 4 0
C
0
0 =0
(3.53)
Considering that = C we obtain:

0
7
7 4 = C 3
C
0 4
3 + 16 + 49 = 0
0
7
7 4 = 0
0 4
2 + 65 = 0
= 65
(3.54)
(3.55)
With that we obtain = C 65

Ciudad Real - Spain
Draft
280
Then, the Mohrs circle is represented by:

S
C 65
III = C 65
I = C 65
Problem 3.19
The stress state at a point of the body is given by the Cauchy stress tensor components as
follows:
x3
t ( e 3 ) = 8e1
t ( e 2 ) = 6e1
e3
e1
e2
x2
t ( e 1 ) = 6 e 2 + 8e 3
x1
Figure 3.19:
a) Obtain the deviatoric part of the stress tensor;
b) Obtain the principal stresses ( I , II , III ) and the principal directions;
c) Draw the Mohrs circle in stress;
d) Obtain the maximum shear stress at the point;
e) Find the traction vector associated with the plane whose normal vector is
6
n = 0.75e1 + 0.25e 2
e3 ;
4
f) Obtain the normal and tangential stress vector associated with the plane described in
paragraph (e).
Solution:
According to Figure 3.19 we can obtain the Cauchy stress tensor components as follows:
Ciudad Real - Spain
Draft
3 STRESS
281
0 6 8
ij = 6 0 0
8 0 0
a)
sph
dev
ij = ij + ij
sph
The spherical part is ij =
I
ij = 0 ij since I = 0 . Then, the deviatoric part is given by:
3
dev
ij
sph
ij
0 6 8
ij = 6 0 0
8 0 0
b) The eigenvalues can be obtained by means of the characteristic determinant:

6
6
8
8
0 =0
3 + 100 = 0
2 + 100 = 0
whose solutions are 1 = 0 , 2 = 10 , 3 = 10 , (principal stresses). The principal directions are:

eigenvector
1 = 0 n i(1) = [0 0.8 0.6]

eigenvector
2 = 10 n i( 2 ) = [ 0.707 0.424 0.566]

eigenvector
3 = 10 n (3) = [0.707 0.424 0.566]
i
I = 10 , II = 0 , III = 10
c) The Mohrs circle can be appreciated as follows:

S
max = 10
II = 0
III = 10
I = 10
d) We can directly obtain the maximum shear stress by means of the Mohrs circle:
max =
I III
= 10
2
e) Considering t i (n) = ij n j , we can obtain the traction vector associated with the plane whose
normal vector is n = 0.75e1 + 0.25e 2
6
e3 :
4
Draft
Ciudad Real - Spain
282
t 1 (n) 0 6 8 0.75 3.39898

(n)

t 2 = 6 0 0 0.25 4.5
t (n) 8 0 0
6
3
4
f)
r
t (n)
r
S
r
N
The magnitude of N can be obtained by the projection N = t (n) n = t i (n) n i , thus:
r
N
0.75
(n )
= t i n i [ 3.39898 4.5 6] 0.25 5.09847
6
4
The vector N is given by:

r
r
N = N n = 3.82385e1 1.27462e 2 + 3.12216e 3

r
r
r
In addition, the relationship t (n) = N + S holds, with that the tangent stress vector is
obtained as follows:
r
r
r
S = t (n) N
( 3.39898 + 3.82385)e1 + (4.5 + 1.27462)e 2 + (6 3.12216)e 3
(0.42487 )e1 + (5.77462)e 2 + (2.87784)e 3
and its module as:

r
S
(0.42487 )2 + (5.77462)2 + (2.87784)2
41.808713 = 6.465966
r
NOTE: We could also have used the equation S
r
module of S .

Ciudad Real - Spain
Draft
r r
r
= t (n) t (n) N
to obtain the
3 STRESS
283
Problem 3.20
The Cauchy stress tensor field of a continuous medium is represented by:
3 x1
r
ij ( x ) = 21
31
2
5x2
3x 2
32
2 x3
0
a) Obtain the body force (per unit volume) to ensure the balance of the continuum.
b) For a particular point ( x1 = 1, x 2 = 1, x3 = 0 ):
b.1) Draw the Mohrs circle. Obtain the maximum normal and tangential stress
component.
b.2) Obtain the traction vector associated with the plane whose normal is
1
ni =
3
1
3
b.2.1) Obtain the normal and tangential in this plane.

Solution:
a) Due to the symmetry of the Cauchy stress tensor ( = T ) we have:
3 x1
r 2
ij ( x ) = 5 x 2
0
2
5x2
3x 2
2 x3
2 x3
0
+ 12, 2 + 13,3 = b1
3 + 10 x 2 + 0 = b1
r r components 11,1
r
x + b = 0 21,1 + 22, 2 + 23,3 = b1 0 + 3 + 2 = b 2
0 + 0 + 0 = b
31,1 + 32, 2 + 33,3 = b1
with that we obtain:

10 x 2 3
bi = 5
(Force per unit volume)
(3.56)
b) For the particular point ( x1 = 1, x 2 = 1, x3 = 0 ) we have:

3 5 0
ij = 5 3 0
0 0 0
where we can verify that 3 = 0 is one principal value. For the other eigenvalues, it is
sufficient to solve:
3
5
=0
5
3
(3 ) = (5) 2
3 = 5
1 = 8
2 = 2
Restructuring the eigenvalues:

I = 8 , II = 0 , III = 2

Ciudad Real - Spain
Draft
284
b.1) The Mohrs circle is drawn in figure below:

S
max = 5
II = 0
III = 2
I = 8
By means of the Mohrs circle we can obtain the maximum shear stress max = 5 and the
maximum normal stress N max = I = 10 .
e) Considering that t i (n) = ij n j , we can obtain the traction vector associated with the plane
whose normal vector is n =
1
1
1
e1 +
e2 +
e3 :
3
3
3
t 1 (n)
3 5 0 1
8
(n) 1
1 = 1 8
5 3 0
t 2 =
3
3
t (n)
0 0 0 1
6
b.2) The normal stress component is obtained as follows:

N = ti
(n)
ni =
1
3
1
[8 8 0] 1 = 16
3
3
1
To obtain the tangential component we apply directly S

r
t (n)
r r
r
= t (n) t (n) N
, where
8
r (n) r (n)
1 1
(n) (n)
[8 8 0] 8 = 128 . Then:
= t t = ti ti =

3
3 3
0

r
S
r r
r
= t (n) t (n) N
128 16
128
=
3 3
9
S =
128
3
Problem 3.21
The stress state at one point of the body is given by means of the spherical and deviatoric
part of the Cauchy stress tensor as follows:
sph
ij
1 0 0
= 0 1 0
0 0 1
dev
ij
0 6 8
= 6 0 0
8 0 0
a) Obtain the Cauchy stress tensor components;

Ciudad Real - Spain
Draft
3 STRESS
285
b) Find the principal stresses ( I , II , III ) and principal directions;

c) Obtain the maximum shear stress;
d) Draw the Mohrs circle in stress for the cases: d.1) the Cauchy stress tensor ( ij ), d.2)
sph
dev
the spherical part ( ij ) and; d.3) the deviatoric part ( ij );
Solution:
ij =
a)
sph
ij
dev
ij
1 0 0 0 6 8 1 6 8
= 0 1 0 + 6 0 0 = 6 1 0

0 0 1 8 0 0 8 0 1
dev
In Problem 3.19 we have obtained the principal values of ij whose values are the same
as for the proposed problem. As the tensor and its deviatoric part have the same principal
directions, i.e. they are coaxial, we can automatically obtain the principal stresses:
=
ij
sph
ij
dev
ij
0 9 0 0
1 0 0 10 0
0 1 0 + 0 0
0 = 0 1 0
=

0 0 1 0 0 10 0 0 11
The principal directions are the same as those provided in Problem 3.19.
d) Mohrs circle in stress
S
S
max = 10
+
dev = 0
II
dev = 10
III
dev = 10
I
dev
N
I = II = III = 1
Deviatoric part
Spherical part
144444444444444 2444444444444444
4
3
S
max = 10
III = 9

Ciudad Real - Spain
II = 1
Draft
I = 11
286
Note that the spherical part contribution deviates (translate) the Mohrs circle of the
deviatoric part according the N -axis, and does not alter the value of the maximum shear
stress.
Problem 3.22
At one point P in the continuum medium, The Cauchy stress tensor is represented by
its Cartesian components as follows:
1 1 0
ij = 1 1 0 MPa ,
0 0 2
a) Obtain the principal stresses and principal directions at the point P ;

b) Obtain the maximum shear stress;
c) Draw the Mohrs circle for the cases: c.1) the Cauchy stress tensor ( ij ), c.2) the
sph
dev
spherical part ( ij ) and; c.3) the deviatoric part ( ij );
d)
i.) Find the traction vector associated with the plane whose normal vector is
r
n = 1.0e1 + 1.0e 2 + 0e 3 ;
ii.) Obtain the normal and tangential stress on the plane.
f) Obtain the eigenvalues and eigenvectors of the deviatoric part ( dev ).

Solution:
a) (See Problem 3.8). The eigenvalues are I = 2 , II = 2 , III = 0
b) and c)
dev
ij
=
ij

Ciudad Real - Spain
sph
ij
1 0 0
1 0 0
2 0 0
0 2 0 4 0 1 0 = 2 0 1 0
=
3
3
0 0 2
0 0 1
0 0 0
Draft
3 STRESS
287
max = 1
N
III = 1.333
I , II = 0.667
I = II = III = 1.333
Deviatoric part
Spherical part
144444444444444 2444444444444444
4
3
S
max = 1
III = 0
I , II = 2
d) The traction vector is obtained by t (n) = n , we need to normalize the normal vector
r
n
1
1
e1 +
e 2 + 0e 3 . Thus:
to the plane, i.e. n = r =
n
2
2
(
t 1n) 1 1 0
2
1
(n)
1 1 = 1 2
t 2 = 1 1 0

2
t ( n ) 0 0 2 2 0
0
3

Problem 3.23
The Cauchy stress tensor components at one point are:
0
0
29
0 26 6 Pa
ij =
0
6
9
Decompose the stress tensor in a spherical and a deviatoric part, and obtain the principal
stresses and principal directions of the deviatoric part.
Solution:
Consider the additive decomposition of the stress tensor into a spherical and deviatoric
part:
Ciudad Real - Spain
Draft
288
dev
sph
ij = ij + ij
The deviatoric part is given by

dev
ij
11 m
= 12
13
12
22 m
23
13
23
33
where the mean stress is:

m =
( 29 26 + 9)
1
ii =
=4
3
3
thus:
dev
ij
0
0 25
0
0
29 4
0
= 0 30 6 Pa
=
26 4
6
0
6
9 4 0
6
5
The spherical part components are:

hyd
ij
sph
ij
4 0 0
= 0 4 0 Pa
0 0 4
To verify the above operations, the following relationship must be verified:

ij =
dev
ij
sph
ij
0
0 4 0 0 29
0
0
25
0 30 6 + 0 4 0 = 0 26 6 Pa
=

0
6
5 0 0 4 0
6
9
To obtain the eigenvalues we solve the characteristic determinant of the deviatoric part:
dev
ij ij = 0 3 J 2 J 3 = 0
By solving the above cubic equation we obtain the following principal values:
dev
1 = 25 Pa
dev
2 = 6 Pa
dev = 31Pa
3
Problem 3.24
Decompose the Cauchy stress tensor:
12
ij = 21
31
4
9
32
0
2 MPa
into a spherical and a deviatoric part.

Obtain the principal invariants of the deviatoric part
Obtain the normal octahedral stress and the mean stress at this point.
Solution:
Due to the symmetry of the Cauchy stress tensor:
Ciudad Real - Spain
Draft
3 STRESS
289
0
12 4
4 9 2 MPa
ij =
0 2 3
The mean stress m = oct =
I 12 + 9 + 3 24
=
=
= 8.
3
3
3
The spherical and deviatoric parts are:

sph
ij
8 0 0
= 0 8 0
0 0 8
dev
ij
= ij
sph
ij
0 8 0 0 4 4
0
12 4
4
0 8 0 = 4 1 2
9 2
=

0 2 3 0 0 8 0 2 5
The principal invariants of the deviatoric part are:

I dev J1 = 4 + 1 5 = 0 , as expected, since the trace of any deviatoric tensor is zero.
II dev =
2 5
0 5
or using the definition: J 2 =
4 4
4 1
= 41 = J 2
) (
1 2
1
I 3 II = 24 2 3 151 = 41
3
3
III dev J 3 = det ( dev ) = 44
Problem 3.25
The stress state at one point is represented by the Cauchy stress tensor components:
a b
ij = a c
b c
where a , b and c are constants and is the value of stress. Determine the constants a ,
b and c such that the traction vector is zero on the octahedral plane.
Solution:
A octahedral plane has the following unit vector: n i =

r
1
3
[1
1 1] . The traction vector on
this plane is defined by t (n) = n , whose components are:
(
t 1n) a b
1
+ a + b 0 a + b = 1
(n)
1 1 = 1 a + + c = 0 a + c = 1
t 2 = a c
t (n) b c 3 1
b + c + 0 b + c = 1
by solving the above system we obtain b =
1
1
1
, c=
, a= .
2
2
2
Problem 3.26
At one point P in the continuous medium the Cauchy stress tensor is represented by its
Cartesian components as follows:

Ciudad Real - Spain
Draft
290
57
ij = 21
31
a)
b)
c)
d)
0
50
32
24
0 MPa ,
43
Obtain the principal stresses and principal directions at the point P ;

Obtain the maximum tangential and normal stress at this point;
Draw the Mohrs circle in stress;
r
Obtain the traction vector t (n) on the octahedral plane of the Haigh-Westergaard
space. Obtain the normal octahedral stress and the tangential octahedral stress.
Solution:
Considering the symmetry of the Cauchy stress tensor we obtain:
57 0 24
ij = 0 50 0 MPa
24 0 43
Note that the stress 22 = 50 is already a principal stress and is associated with the principal
direction n ( 2) = [0 1 0] . To find the other principal stresses we solve the following

system:
57
24
=0
24
43
1 = 25
2 100 + 1875 = 0
3 = 75
Using the definition of eigenvalue-eigenvector, we can obtain the following eigenvectors:

1 = 25
n (1) = [m 0.6 0 0.8]
3 = 75
n (3) = [ 0.8 0 0.6]
Mohrs circle in stress:

Restructuring the principal stresses such that I > II > III we have:
I = 75 , II = 50 , III = 25
b, c) The Mohrs circle is drawn in figure below:

S
max =
max = 25
III = 25

Ciudad Real - Spain
II = 50
Draft
75 25
= 25
2
I = 75 = N max
3 STRESS
291
d) The Haigh-Westergaard space is formed by principal stress directions, then the traction
r
vector in this space is given by t (n) = n , whose normal vector to the octahedral plane is
1
given by n i =
1
:
3
t (n ) = n
(
t 1n)
(n)
t 2 =
t ( n )
3
components
75 0 0
1
75
0 50 0 1 1 = 1 50
3
3
0 0 25
1
25
and its module is given by:

r
t (n)
1
8750
75 2 + 50 2 + 25 2 =
3
3
r
The normal octahedral stress is given by oct = t (n) n :

2
oct
r
t (n) = 54.00617
1
[75 50 25] 1 = 50
=
3 3
1
We could have applied directly the definition of octahedral normal stress:

oct =
I
75 + 50 + 25
= m =
= 50
3
3
The tangential octahedral stress can be obtained by means of the Pythagorean theorem:
oct =
r
t (n )
oct
8750
50 2 = 20.4124
3
We could also have applied the definition:

oct =
1
1
2
2 I 6 II =
2 150 2 6 6875 = 20.41241
3
3
where I = 150 , II = 75 50 + 75 25 + 50 25 = 6875 .

e) The spherical part:
sph
ij
50 0 0
Tr ( )
=
ij = m ij = 0 50 0
3
0 0 50
sph
ij
57 0 24 50 0 0 7 0 24
= 0 50 0 0 50 0 = 0 0 0
24 0 43 0 0 50 24 0 7
and the deviatoric part:

dev
ij
= ij
f) Considering that the tensor and its deviatoric part are coaxial tensors, we can use the
principal space to obtain the deviatoric part:
dev
ij
=
ij
sph
ij

Ciudad Real - Spain
0
75 0 0 50 0 0 25 0
0 50 0 0 50 0 = 0 0
=
0

0 0 25 0 0 50 0 0 25
Draft
292
3.1.6
Stress state in two dimension (2D)
Problem 3.27
Consider the following stress state:
5
4
2
y
6
x
Figure 3.20:
Obtain the state of plane stress at the point ij .
Solution:
In the state of plane stress ij (i, j = 1,2) we need two planes to define the stress state at the
point:
x
ij =
xy
xy
y
(3.57)
According to Figure 3.20 we verify that:

5
x = 4
xy = 2
xy = 2
y
y = 6
x
Figure 3.21:
Ciudad Real - Spain
Draft
3 STRESS
293
Then:
4 2
ij =
2 6
(3.58)
Problem 3.28
Consider a composite material, which is made up of matrix and fiber along direction of
45 such as shows in Figure 3.22. This composite material can break if the shear stress
along the fiber exceeds the value 3.8 10 6 Pa ( N / m 2 ) .
For the normal stress x = 2.8 10 6 Pa , obtain the maximum value of y for which the
material does not break.
y
45
45
Figure 3.22: Composite material (fiber-matrix).

Solution:
We need to obtain the traction vector on the plane defined by = 45 , and the tangential
components can directly be obtained by means of:
xy ( ) =
xy ( = 45 )
x y
sin 2 + xy cos 2
2
2.8 10 6 y
=
sin( 90 ) = 3.8 10 6 Pa
2
y 4.8 10 6 Pa (compression)
(see equation (1.99) in Problem 1.98).

Problem 3.29
The stress acting on two planes passing through the point P are shown in Figure 3.23.
Obtain the value of the shear stress on the plane a a and the principal stresses at this
point.

Ciudad Real - Spain
Draft
294
y
b
80 Pa
45
x
60
60 Pa
a
b
Figure 3.23: Stress state at one point, according to the planes a and b .
Solution:
To obtain the stress state at a point in the two dimensional case, we need to determine x ,
y , xy , as indicated in Figure 3.24.
xy
80 Pa
45
xy
xy
80 Pa
x
x
60
45
xy
60
xy
60 Pa
60 Pa
b
a)
b)
Figure 3.24: Stress state at a point, according to the planes a and b .

Considering Figure 3.24, we can directly obtain x and xy by means of the projection of
the traction vector 60 Pa , (see Figure 3.24(b)), i.e.:
x = 60 cos( 30 ) = 51 .962 Pa
xy = 60 cos( 60 ) = 30 Pa
To obtain y we employ the equations:

Ciudad Real - Spain
Draft
3 STRESS
x ( ) N =
xy S ( ) =
x + y
2
x y
295
x y
2
cos 2 + xy sin 2
sin 2 xy cos 2
By substituting the numerical values in the above equations we obtain:

( = 45 ) =
( = 45 ) =
51.962 + y
2
51 .962 y
2
51 .962 y
2
cos( 90 ) + 30 sin( 90 ) = 80 Pa
sin( 90 ) 30 cos( 90 )
The first equation give us the value of y :

y = 48.038 Pa
Once y is determined, we can obtain the component (= 45) :

( = 45 ) = 1.96 Pa
The principal stresses can be obtained by means of the components x , y , xy , such as

indicated in the equations:
(1, 2 ) =
(1, 2 )
x + y
2
x y

+ 2
xy
51.962 + 48 .038
51.962 48.038
=

2
2
= 80.1Pa
+ 30 2 1
2 = 19.9 Pa
or by means of the characteristic determinant:

x
xy
=0
xy
y

Ciudad Real - Spain
Draft
51.962
30
=0
30
48.038
296
Problem 3.30
Given a stress state x = 1Pa , xy = 4 Pa and y = 2 Pa . Draw a graph of angle vs.
stresses ( x , y , xy ), where is the rotation angle of the coordinate system (see Figure
3.25).
y
y = 2 Pa
xy = 4 Pa
xy = 4 Pa
x
x = 1Pa
xy
y
Figure 3.25: Stress state at one point.

Solution:
We calculate the values x , y , xy by using the equations:
x =
x + y
xy =
y =
2
x y
2
x + y
2
x y
2
cos 2 + xy sin 2
sin 2 + xy cos 2
y x
2
cos 2 xy sin 2
We can calculate the angle corresponding to the principal direction by means of the
equation:
tan 2 =
2 xy
x y
2 ( 4 )
= 8 ( = 41.437 )
1 2
and the principal stresses:

1, 2 =
x + y
2
x y

+ 2
xy
1 = 5.5311P
2 = 2.5311Pa
Considering the transformation law, we can obtain the values of x , y , xy for different
values of . Making vary from 0 to 360 we can represent the stresses x , y , xy in
function of the angle, (see Figure 3.26). We can observe that when = 41 .437 we have a
principal direction, then the tangent stress is zero ( xy = 0 ) and the principal stresses are
I = 5.5311Pa and II = 2.5311Pa .

Ciudad Real - Spain
Draft
3 STRESS
297
x
1
= 41.437
x
= 131.437
2
Stresses
1 = 5.5311
xy
x
0
0
50
100
-2
45
xy
-4
150
200
250
350
2 = 2.5311
300
= 86.437
-6
max = 4.0311
Figure 3.26: Stress components in function of the angle .

Problem 3.31
a) Consider the stress field ij (i , j = 1,2) in the Cartesian system x1 x 2 x 3 , and the
following equations:
t
2
m11 = 11 x3 dx3
t
2
t
2
m12 = 12 x 3 dx3
t
2
t
2
m22 = 22 x 3 dx3
t
2
Obtain the component transformation law of mij (i, j = 1,2) to a new system x1 x 2 x 3
which is formed by a rotation around the x3 -axis (see Figure 3.27).

Ciudad Real - Spain
Draft
298
x3 = x3
x2
x3 =
t
2
x2
x3 =
t
2
x1
x1
Figure 3.27.
Solution:
Due to the symmetry of ij = ji , we can conclude that m12 = m 21 . The transformation
matrix from x1 x 2 x 3 to x1 x 2 x 3 is given as follows:

cos sin 0
a ij = sin cos 0
0
0
1
cos sin
A=
sin cos
2D
By using the Voigt notation, we get:

t
2
m11

{m } = m22 =
m
12
t
2
11

22 x3 dx3 =

12
t
2
t
2
11

22 x3 dx3 =

12
t
2
t
2
11
[M] 22 x3dx3 = [M]

12
t
2
11

22 x3 dx3

12
t
2
with that, we can conclude that:
m11
m11
{m } = m 22 = [M]m 22 = [M]{m}
m
m
12
12
(3.59)
where [M] is the transformation matrix for a second-order tensor when it is in the Voigt
notation, (see Problem 1.98), and is given by:
a11 2
[M] = a 21 2
a a
21 11
a12
a 22
a 22 a12
cos 2

2
2a 21 a 22
= sin
a11 a 22 + a12 a 21 sin cos

2a11 a12
sin 2
cos
2
cos sin
2 cos sin
2 sin cos
cos 2 sin 2
Also considering that [M]1 = [N ]T , we obtain {m} = [N ]T {m }, where

Ciudad Real - Spain
Draft
3 STRESS
a11 2
[N ] = a 212
2a a
21 11
a12
a 22
2a 22 a12
299
cos 2

2
a 21 a 22
= sin
a11 a 22 + a12 a 21 2 sin cos

a11 a12
sin 2
cos
2
2 cos sin
sin cos
cos 2 sin 2
cos sin
The same result (3.59) could have been obtained by consider mij as a second-order tensor
in two dimensional case (2D), and by means of the transformation law of a second-order
tensor we obtain:
mij = a ik a jl m kl
m
11
m12
(i, j = 1,2)
or
m12 cos sin m11

=
m sin cos m12
22

Ciudad Real - Spain
Draft
m = Am A T
m12 cos sin

m 22 sin cos
(3.60)
300

Ciudad Real - Spain
Draft
4 The Fundamental
Equations of Continuum
Mechanics
4.1 Solved Problems
Problem 4.1
Prove that Reynolds transport theorem is valid in the following equation:
D
D
dV =
JdV 0
Dt V
Dt V
(4.1)
where V is the volume in the current configuration, V0 is the volume in the reference
configuration, J is the Jacobian determinant and is a scalar field that describes the
physical quantity of a particle per unit volume at time t .
Solution:
D
DJ
D
JdV0 = J
+
dV 0
Dt V
Dt
Dt
V0
0
r
D
r
= J
+ J x v dV 0
Dt
V0
(4.2)
r
D
r
=
+ x v dV
Dt
Problem 4.2
Show that
r
DPijL ( x , t )
r
D
PijL ( x , t ) dV =
dV
Dt V
Dt
V
(4.3)
302
where PijL ( x , t ) is a continuum property per unit mass, which can be a scalar, a vector or
higher order tensor.
Solution:
It was proven in the textbook, (see Chaves (2013)), that:
D
r
r
r v p
D
( x , t )dV = ( x , t ) + ( x , t )
dV
Dt V
x p
Dt
Then by making = PijL , and by considering it in the above equation we obtain:

D
D
PijK dV = ( PijK ) + PijK
Dt V
Dt
V
D
=
PijK + PijK
Dt
V
v p
D
v
D
+ PijK k dV
PijK + PijK
dV =
x p
x k
Dt
Dt
D
v k
Dt + x
k
14243
dV
=0
mass continuity equation
Thus, we can conclude that:

DP
D
PijK dV = ijK dV
Dt V
Dt
V
Problem 4.3
Prove that the following relationship is valid:
r
a=
r
r r
r
( v ) + x ( v v )
t
(4.4)
Solution: Based on the Reynolds transport theorem:

r
D
dV =
dV + (v n) dS
Dt V
t
V
S
r
and if we consider that = v we obtain:
r
r
r
r
( v )
D
v dV =
dV + v (v n) dS
Dt V
t
V
S
Then, the above equation in indicial notation becomes:

( v i )
( v i )
D
D
vi dV =
dV + v i (v k n k ) dS
v i dV =
dV + ( v i v k )n k dS
Dt V
t
Dt 3
t
1
2
V
S
V
V
S
= ai
Additionally, by applying the divergence theorem to the surface integral we obtain:
dV =
( v i )
( v i )
+ ( v i v k ) ,k dV
dV + ( vi v k ) ,k dV =
t
t
V
V
which in tensorial notation is:

Ciudad Real - Spain
Draft
4 THE FUNDAMENTAL EQUATIONS OF CONTINUUM MECHANICS
r
r
r r
( v )
r
+ x ( v v ) dV
a dV =
t
V
V
303
r
r ( v )
r r
r
a=
+ x ( v v )
t
Problem 4.4
Let us consider the following velocity field:
vi =
xi
1+ t
for t 0
1) Find the mass density field;

2) Prove that this motion satisfies x1 x 2 x3 = 0 X 1 X 2 X 3 .
Solution: 1) By applying the mass continuity equation we obtain:
D
Dt
v k
=0
x k
Dt
v
d
= k
dt
x k
and by using the given velocity field, we find that:

v i
1 x i
3
=
= ii =
x i 1 + t x i 1 + t 1 + t
Thus,
d
3
=
dt
1+ t
3dt
1+ t
Then by integrating the both sides of the above equation we obtain:
3dt
1+ t
ln = 3 ln(1 + t ) + C
The constant of integration C is obtained by means of the above equation if we refer to

r
the initial condition t = 0 , in which ( x , t = 0) = 0 , thus
ln 0 = 3 ln(1 + 0) + C C = ln 0
1
ln = 3 ln(1 + t ) + ln 0 = ln
(1 + t ) 3
+ ln 0 = ln
(1 + t ) 3
Thus, we can conclude that:
(1 + t )3
2) Then by using the velocity definition we obtain:

vi =
dx i
x
= i
dt 1 + t
dx i
dt
=
xi 1 + t
Additionally, by integrating the both sides of the above equation we obtain:
dx i
dt
=
xi
1+ t
lnx i = ln(1 + t ) + K i
(4.5)
Then by applying the initial condition, i.e. at time t = 0 xi = X i , we obtain:

lnX i = ln(1 + 0) + K i
Ciudad Real - Spain
Draft
K i = lnX i
304
Additionally, by substituting the value of K i into the equation (4.5) we obtain:

lnx i = ln(1 + t ) + ln X i
ln( x i ) = ln[ X i (1 + t ) ]
Hence we can conclude that x i = X i (1 + t ) , which gives us x1 = X 1 (1 + t ) , x 2 = X 2 (1 + t ) ,

0
, we obtain:
x 3 = X 3 (1 + t ) , and if we consider that =
3
(1 + t )
(12t )(12t )(12t ) = 0

1+3 1+3 1+3
x1
X1
x2
X2
x1 x 2 x 3 = 0 X 1 X 2 X 3
x3
X3
Problem 4.5
The equations of motion of a body are given, in Lagrangian description, by:
x1 = X 1 + tX 3
x 2 = X 2 + tX 3
x = X t ( X + X )
3
1
2
3
where is a constant scalar. Find the mass density in the current configuration ( ) in
terms of the mass density of the reference configuration ( 0 ) , i.e. = ( 0 ) .
Solution:
We can apply the equation 0 = J , where J is the Jacobian determinant and is given by:
x1
X 1
x 2
=
X 1
x 3
X 1
x i
J= F =
X j
Thus, we obtain =
0
J
x1
X 2
x 2
X 2
x 3
X 2
x1
X 3
1
x 2
= 0
X 3
t
x 3
X 3
0
1
t
t
t = 1 + 2(t ) 2
1
0
1 + 2(t ) 2
Problem 4.6
Given the velocity field:
v1 = ax1 bx 2
;
where a , b and c are constants.
v 2 = bx1 ax 2
2
v3 = c x12 + x 2
a) Check if the mass continuity equation is fulfilled;

b) Is the motion isochoric?
Solution:
The mass continuity equation:
r
D
r
+ ( x v ) = 0
Dt
where:

Ciudad Real - Spain
Draft
305
r
r
x v = vi ,i = v1,1 + v 2, 2 + v3,3
=aa+0=0
r
r
The motion is isochoric (incompressible medium), since x v = 0
Problem 4.7
Consider a continuous medium and one property ( x , t ) assigned by density, i.e. unit of
the property per unit volume. Obtain the rate of change of the property that is described
by a control volume in conjunction with a control surface.
Solution:
Remember that the rate of change of a property is always associated with the same
particles. By means of the material time derivative we can obtain the rate of change of a
property when this property is in Eulerian description. Then, the total rate of change of
r
( x , t ) in the volume V that is bounded by the surface S is given by:
r
r
r
D
D
D
D
( x, t )dV =
( x, t ) + ( x, t ) (dV )
( dV ) = dV
Dt V
Dt
Dt
Dt
V
V
r
r
r
D
r
( x, t ) + ( x, t ) x v dV
= dV
Dt
(4.6)
r
r
r
D
r
= ( x , t ) + ( x , t ) x v dV
Dt
We apply the definition of the material time derivative to
r
D
( x , t ) :
Dt
r
r
r
r
D
D
r
( x, t )dV = ( x, t ) + ( x, t ) x v dV
Dt
Dt V
V
r
r
r
r
( x , t ) r r
r
= ( x, t ) +
r v ( x , t ) + ( x , t ) x v dV
x
t
V
r
r
r
r
( x , t ) r
r
= ( x , t ) dV +
r v + ( x , t ) x v dV
t
x
V
V
(4.7)
r
r
r
= ( x , t ) dV + [ x (v ) ]dV
t
V
V
We can apply the divergence theorem to the second integral on the right side of the
equation to obtain:
flux of through
suface S
64 744
4
8
r
r
r
( x , t )
D
( x, t )dV =
dV + (v ) n dS
{
Dt V
1 24
4t 3
V
S flux of
(4.8)
local

Ciudad Real - Spain
Draft
306
r
( x , t )
the term
is local, the volume integral of the right side of the equation is a control
t
r
volume and the integral surface is a control surface, since the variable (v ) is in Eulerian
r
description. The term (v ) represents the flux of the property .
When there is no source or sink of the property is true that
r
D
( x, t )dV = 0 . And, note
Dt V
also that when the property is the mass density ( = ) the equation (4.7) becomes the mass
continuity equation.
r
r
r
r
D
D
r
( x, t )dV = ( x, t ) + ( x, t ) x v dV = 0
Dt
Dt V
(4.9)
r
r
r
= ( x , t ) + x ( v ) dV = 0
t
If the above equation is valid for the entire volume then it is valid locally, so
r
r
r
D
r
( x, t ) + ( x, t ) x v = 0 Mass continuity equation
Dt
(4.10)
or
r
r
r
( x , t ) + x ( v ) = 0
t
control volume
S
V
(4.11)
r
(v )
r
( x , t )
t
r
x
Mass continuity equation
r
r

q n = [(v ) n] n
control surface
Figure 4.1: Control volume and control surface.

Ciudad Real - Spain
Draft
material volume
307
control surface
control volume
t=0
v0
XP
X*
control surface
control volume
material volume
t1
r
v( x * , t1 )
xP
x*
material volume
control surface
control volume
t2
r
v( x * , t 2 )
xP
x*
Figure 4.2: Material volume vs. control volume.

Ciudad Real - Spain
Draft
308
4.1.1
Equations of Motion. Equilibrium Equations
Problem 4.8
Find the equilibrium equations in engineering notation by means of the differential volume
element equilibrium ( dxdydz ). For this purpose consider that the Cauchy stress tensor field
in the differential volume element varies as indicated in Figure 4.3.
z
Rear face
z +
Rear face
xy
z
dz
z
yz +
xz +
xz
dz
z
xy
y
bx
xy +
x + x dx
x
yz
y
dz
dy
y +
xy +
xy
xz
dz
by
xz
dx
x
yz
yz
yz +
bz
xz +
xy
y
y
y
dy
dy
dx
dx
xz
yz
Rear face
dy
Figure 4.3: The stress field in the differential volume element.

Solution:
To obtain the equilibrium equations we apply the force equilibrium condition in the
volume element. First, we evaluate the equilibrium force according to the x -direction:
=0
xy
x
dx dydz x dydz + xy +
dy dxdz
x
y
xy dxdz + xz + xz dz dxdy xz dxdy = 0

z
b x dxdydz + x +

Ciudad Real - Spain
Draft
309
Then by simplifying the above equation we obtain:
b x dxdydz +
xy
x
dxdydz +
dxdydz + xz dxdydz = 0
x
y
z
b x +
x xy xz
+
+
=0
x
y
z
The equilibrium force according to the y -direction,
Fy = 0 , can be expressed as follows
yz
dy dxdz y dxdz + yz +
dz dxdy
z
y
xy
yz dxdy + xy +
dx dydz xy dydz = 0
b y dxdydz + 22 +
b y +
xy
x
y
y
yz
x z
Finally, the equilibrium according to the z -direction,
=0
Fz = 0 , is given by:
dz dxdy z dxdy + xz + xz dx dzdy

x
z
yz
xz dzdy + yz +
dy dxdz yz dxdz = 0
b z dxdydz + z +
Additionally, by simplifying the above equation we obtain:
b z +
xz yz z
+
+
=0
x
y
z
Then, the equilibrium equations in engineering notation become:

x xy xz
+
+
+ b x = 0
y
z
x
xy y yz
+
+
+ b y = 0
y
x z
x

z
xz + yz +
+ b z = 0
x
y
z
Problem 4.9
Let be the Cauchy stress tensor field, which is represented by its components in the
Cartesian basis as:
11 = x12 ;
2
22 = x 2 ;
12 = 21 = 2 x1 x 2 ;
2
33 = x12 + x 2
23 = 32 = 31 = 13 = 0
Considering that the body is in equilibrium, find the body forces acting on the continuum.
Ciudad Real - Spain
Draft
310
Solution:
r
By applying the equilibrium equations, x + b = 0 , we obtain:
ij , j + b i = 0 i
11 12 13
+
+ b1 = 0
+
x 2
x 3
x1
21 22 23
+
+ b2 = 0
+
x 2
x 3
x1
31 32 33
+
+
+ b3 = 0
x1
x 2
x 3
2 x1 + 2 x1 + b 1 = 0
2 x 2 + 2 x 2 + b 2 = 0
b = 0
3
Thus, to satisfy the equilibrium equations the following condition must be met:
4 x1 = b 1 b 1 = 4 x1
4 x 2 = b 2 b 2 = 4 x 2
b3 = 0
r
b = 4( x1 e 1 + x 2 e 2 )
Problem 4.10
Given the velocity field:
v1 = x1 x3
2
v2 = x2 t
v3 = x 2 x 3t
and the Cauchy stress tensor field:

x 2 x1
ij = x 2 x3
x 2 x3
2
x2
x2
0
x2
2
x3
where is a constant. Obtain the body force (per unit volume) to guarantee the principle
of conservation of the linear momentum.
Solution:
From the principle of conservation of linear momentum we obtain the equations of
motion:
r
r
r
&
r
x + b = v = a
r
r
r
b = a x
The acceleration field:
r r
r r
r v ( x , t ) v ( x , t ) r r
a=
+
r v ( x, t )
t
x
ai =
v i v i
+
vj
t
x j
where
0
v i 2
= x2
t
x 2 x3
x3
v i
= 0
x j
0
0
2x2 t
x3t
x1
0
x2t
Then

Ciudad Real - Spain
Draft
ai =
311
v i v i
+
vj
t
x j
0 x3
2
= x2 + 0

x 2 x3 0
0
2 x2t
x3t
2
x1 x1 x 3 0 x1 x 3 + x1 x 2 x 3 t
2
2
3
0 x2 t = x2 +
2x2t
2
2
x 2 t x 2 x3t x2 x3 x3 x 2 t 2 + x 2 x3t 2
x1 x 3 + x1 x 2 x 3 t
2
3
x2 + 2x2t
=
x x + x x 2t 2 + x 2 x t 2
3 2
2 3
2 3
The divergence of the Cauchy stress tensor is given by:

11 12 13
+
+
= ( x 2 x3 )
x 2
x 3
x1
21 22 23
+
= (2 x2 )
+
x 2
x 3
x1
31 32 33
= ( 2 x 3 1)
+
+
x1
x 3
x 2
with that the body force (per unit volume) becomes:

r
r
b = a x
b i = a i ij , j
2
x1 x 3 + x1 x 2 x 3 t
x 2 x3
2
3
x2 + 2 x2 t
bi =
2 x2
2
2
x2 x3 + x3 x 2 t 2 + x 2 x3t 2
2 x3 1
Problem 4.11
The stress field in the medium is represented by:
1
ij = 0
2 x2
0
1
4 x1
2 x2
4 x1
(4.12)
where xi are the Cartesian coordinates.

a) Neglecting body forces, is the body in balance?
Solution:
The equilibrium equations:
r r
r
x + b = 0
{
(4.13)
ij , j = 0 i
(4.14)
r
=0
expanding,

Ciudad Real - Spain
Draft
312
i1,1 + i 2, 2 + i 3,3 = 0 i
(4.15)
11,1 + 12, 2 + 13,3 = 0
21,1 + 22, 2 + 23,3 = 0
31,1 + 32, 2 + 33,3 = 0
(4.16)
where we have used:
11
= 0; 12, 2 = 12 = 0; 13,3 = 13 = 0
x3
x1
x 2
= 21 = 0; 22, 2 = 22 = 0; 23,3 = 23 = 0
x 2
x 2
x 3
31
32
33
=
= 0; 32, 2 =
= 0; 33,3 =
=0
x 2
x 3
x3
11,1 =
21, 2
31,3
(4.17)
Problem 4.12
Given a body in equilibrium in which the Cauchy stress tensor field is represented by its
components:
3
2
11 = 6 x1 + x 2
2
; 12 = x 3
3
22 = 12 x1 + 60
3
3
33 = 18 x 2 + 6 x3
; 23 = x 2
; 31 = x12
Obtain the body force vector (per unit volume) at the point ( x1 = 2; x 2 = 4; x3 = 2 ).
Solution:
r r
r
x + b = 0
(4.18)
11 12 13
+
+
+ b1 = 0 b1 = 11 12 13
x 2
x3
x1
x 2
x3
x1
21 22 23
+
+
+ b 2 = 0 b 2 = 21 22 23
x 2
x3
x1
x 2
x3
x1
31 32 33
31 32 33
x + x + x + b 3 = 0 b 3 = x x x
2
3
1
2
3
1
(4.19)
b1 = 18 x12 0 0
b 2 = 0 0 0
b = 2 x 1 18 x 2
1
2
3
18 x12
0
bi =
2 x 1 18 x 2
1
2
(4.20)
At the point x1 = 2; x 2 = 4; x3 = 2 we obtain:

72
bi = 0
77

Ciudad Real - Spain
(Force per unit volume)
Draft
(4.21)
313
Problem 4.13
The Cauchy stress tensor field is represented by its components as follows:
x12 x 2
2
ij = k (a 2 x 2 ) x1
1 3
( x 2 3a 2 x 2 )
0
3
2
0
2ax3
2
(a 2 x 2 ) x1
(4.22)
where k and a are constants.

r
Obtain the body force field b (per unit mass) required for the stress field is in balance.
Solution:
11 12 13
+
+
+ b1 = 0 b1 = 2 x1 x 2 k + 2 x1 x 2 k = 0
x 2
x3
x1
k
21 22 23
2
2
+
+
+ b 2 = 0 b 2 = k (a 2 x 2 ) (3 x 2 3a 2 ) = 0
3
x1
x 2
x 3
31 32 33
x + x + x + b 3 = 0 b 3 = 4kax3
2
3
1
(4.23)
Then:
0
4kax3
0
bi =

1

Problem 4.14
(Force per unit mass)
(4.24)
Let us assume that the body force is b = ge 3 , where g is a constant and consider the
Cauchy stress tensor field components:
x2
ij = x3
x3
0
x2
0
x2
(4.25)
Find p such that satisfies the equilibrium equations. Consider that is a constant and that
the mass density field is homogeneous, i.e. it is independent of the vector position.
Solution:
r r
r
x + b = 0
(4.26)
11 12 13
+
+
+ b1 = 0
x 2
x 3
x1
21 22 23
+
+
+ b2 = 0
x 2
x 3
x1
31 32 33
x + x + x + b 3 = 0
2
3
1
(4.27)

Ciudad Real - Spain
Draft
314
0 + 0 + 0 + b = 0 b = 0
1
1
0 + 0 + 0 + b 2 = 0 b 2 = 0
0 + 33 + b 3 = 0
x3
33 (p )
p
= b3
=
=
x 3
x3
x3
(4.28)
g
p
=1 +
x3
dp = 1 +
dx
3
(4.29)
g
g
p = 1 +
x
x 3 p = 1 +
Verification:
+ 1 +
g = + + g g = 0
(4.30)
Problem 4.15
Show that for the following Cauchy stress field:
2
2
11 = x 2 + ( x12 x 2 ) ; 12 = 2x1 x 2
;
2
2
2
2
2
22 = x1 + ( x 2 x1 ) ; 33 = ( x1 + x 2 )
23 = 13 = 0
satisfies the equilibrium equations with zero body forces.

Solution:
ij , j + b i = 0 i
{
=0i
ij , j = 0 i
(i , j = 1,2,3)
i1,1 + i 2, 2 + i 3,3 = 0 i
i = 1 11,1 + 12 , 2 + 13,3 = 0
i = 2 21,1 + 22 , 2 + 23, 3 = 0
i = 3 31,1 + 32 , 2 + 33,3 = 0
11 12 13
+
=0
+
x 2
x 3
x1

22 23
+
=0
21 +
x 2
x 3
x1
31 32 33
=0
+
+
x1
x 3
x 2
thus:
11,1 + 12 , 2 + 31,3 = 2 x1 2x1 = 0
12 ,1 + 22 , 2 + 23, 3 = 2 x 2 + 2x 2 = 0
13,1 + 23, 2 + 33,3 = 0
with that we prove that the body is in balance.

Ciudad Real - Spain
Draft
315
Problem 4.16
Consider a body in equilibrium in which the Cauchy stress field is:
x1 + x 2
r
ij ( x ) = 12
0
12
x1 2 x 2
0
0
0
x2
Find 12 , knowing that 12 is a function of x1 and x 2 , i.e. 12 = 12 ( x1 , x 2 ) . It is also

known that the medium is free of body forces and the traction vector associated with the
r
plane x1 = 1 is given by t (n) = (1 + x 2 )e1 + (5 x 2 )e 2 .

Solution:
As the body is in equilibrium, it must satisfy the equilibrium equations:
ij , j + b i = 0 i
{
=0i
ij , j = 0 i
(i , j = 1,2,3)
i1,1 + i 2, 2 + i 3,3 = 0 i
thus
11 12 13
+
= 1 + 12 + 0 = 0
+
x 2
x 3
x 2
x1
21 22 23 12
2+0=0
+
=
+
x1
x 2
x 3
x1
31 32 33
=0+0+0=0
+
+
x1
x 3
x 2
Now considering that for the plane x1 = 1 , t (n) = (1 + x 2 )e1 + (5 x 2 )e 2 holds, we have:
12
1 + x 2

ij ( x1 = 1, x 2 ) = 12 1 2 x 2
0
0
(n)
12
1 + x 2

= ij ( x1 = 1, x 2 )n j = 12 1 2 x 2
0
0
0
0
x2
0 1 1 + x 2
0 0 = 5 x 2

x 2 0 0

(4.31)
t (n) = ij ( x1 = 1, x 2 ) n j
1 + x2
12 ( x1 = 1, x 2 ) 0 1
1 + x2
1 + x 2
0 = ( x = 1, x ) = 5 x
( x = 1, x )
12 1
1 2 x2
0 12 1
2
2
2
x 2 0
0
0
0
By means of the equilibrium equations:

12
=2
x1
12
= 2x1
12 ( x1 , x 2 ) = 2 x1 + C ( x 2 )
Using the information given in (4.31) we can obtain the constant of integration:
12 ( x1 = 1, x 2 ) = 5 x 2 = 2 + C ( x 2 )
Ciudad Real - Spain
Draft
C ( x2 ) = 3 x2
316
thus:
12 ( x1 , x 2 ) = 2 x1 x 2 + 3
Problem 4.17
The stress state in an continuous medium is given by the Cauchy stress tensor Cartesian
components:
0
ij = Cx 3
Cx 3
0
Cx1
0
Cx1
where C is a constant. Consider that the body is free of body force.

a) Show whether the body is in balance;
Solution:
a) The continuous medium is in equilibrium if the following equation holds:
r r
+ b = 0 ; ij,j + b i = 0 i (the equilibrium equations)
0
ij = Cx 3
Cx 3
0
Cx1
0
Cx1
(4.32)
(4.33)
For the proposed problem we have b i = 0 i , thus:

ij,j
i = 1 0 + 0 + 0 = 0
i1 i 2 i 3
=
=
+
+
i = 2 0 + 0 + 0 = 0
x j
x1
x 2
x3
i = 3 0 + 0 + 0 = 0
ij
(4.34)
ij,j = 0 i then the body is in equilibrium.
Problem 4.18
Considering the principle of conservation of angular momentum, show that:
[( x (a b) (a b) x ] dV = [( x t
r
r*
r
r
t * x dS
where
r r
r
r
r r
x - is the vector position; ( x , t ) is the mass density; a ( x , t ) is the acceleration; b( x , t ) is
r r
the body force (per unit mass); t * ( x , t ) is the prescribed traction vector (surface force) on
surface S .
Solution:
The principle of conservation of angular momentum states that:
r
r
r
r
r
r r
r
D
( x t * )dS + ( x b)dV =
( x v )dV = ( x a )dV
Dt V
V
V

Ciudad Real - Spain
Draft
317
Then, we apply the cross product of the above equation with an arbitrary vector z , which
r
is independent of x , and we obtain:
r
r
r
r
z ( x a )dV = z
r
r r
r
r
( x t * )dS + z ( x b)dV
r
r
r
z ( x a )dV =
r
r r
r
r
r
z ( x t * )dS + z ( x b)dV
We have shown in Chapter 1 that given three vectors a , b , c , the relationship

r r
r r r r r
r
a (b c ) = (b c c b) a holds. Then, the above equation can be rewritten as
follows:
r
r r r
r
r
r
r r r
r r
r r
r
( x a a x ) z dV = ( x t * t * x ) z dS + ( x b b x ) z dV
r
r r r r r
r r r r r
r r
r r
( x a a x ) z dV ( x b b x ) z dV = ( x t * t * x ) z dS
r
r
r r
r r
r r
r r
r r
x (a b) (a b) x z dV = ( x t * t * x ) z dS
r
r
r r r
r
r r
r r
r
r
x (a b) (a b) x dV z = ( x t * t * x ) dS z
V
with that, we conclude that:
[r
*
*
x (a b) (a b) x dV = ( x t t x) dS
Problem 4.19
1) Considering the definition of the mean stress tensor ( ):
V = dV
V
and based on the principle that the continuum is in static equilibrium, show that:
=
1
2V
[x b + b x ] dV + 2V ( x t
r
r*
r
r
+ t * x ) dS
2) Considering that the volume can be decomposed by V = V (1) V ( 2) , (see Figure 4.4).
The continuum is subjected to pressure p (1) on surface S (1) , and to pressure p ( 2) on
surface S ( 2) . Considering the continuum is free of body forces, show that:
=

Ciudad Real - Spain
(V
(1)
1
( p (1)V (1) p ( 2)V ( 2 ) )1
( 2)
V )
Draft
318
n (1)
S (1)
p (1)
V (1)
S (2)
V ( 2)
n (2)
p (2)
Figure 4.4
Solution:
r
Taking into account the equilibrium equations x + b = a = 0 (the principle of
conservation of linear momentum) for the entire continuum, it must fulfill that:
r
r
r
r
x x + b dV = 0
(4.35)
r
r
r
r
r
x x dV + x b dV = 0
In Chapter 1 (see Problem 1.123) we have shown that the following holds:
r
r
r
r
( ) x dV = ( n) x dS dV = t * x dS dV
r
r
r r
x ( ) dV = x ( n) dS T dV = x t * dS T dV
(4.36)
(4.37)
where we have considered the prescribed traction vector t * = n . By replacing (4.37) into
the equation (4.35), we obtain:
r
r
r
r
r
x x dV + x b dV = 0
r
r
r r
r
x t * dS T dV + x b dV = 0
(4.38)
r
r r
r
T dV = x t * dS + x b dV

Ciudad Real - Spain
Draft
319
Then, the following is true:
r r
r
r
dV = t * x dS + b x dV
(4.39)
Note that the tensors x t * and x b are not symmetric. This means that the equation
in (4.35) does not take in account the principle of conservation of angular momentum, i.e.
the symmetry of the Cauchy stress tensor. To guarantee the symmetry of we do:
r r
r
1 r r
r
r
+ T
1 r
dV = t * x dS + b x dV + x t * dS + x b dV
2 S
2
2 S
V
V
V
r r r r
1
1 r r* r* r
sym dV =
x b + b x dV +
x t + t x dS
2V
2S
V
(4.40)
Considering the definition of the mean stress tensor, we conclude:

r r
r
1 r r
r
r
+ T
1 r
dV = t * x dS + b x dV + x t * dS + x b dV
2
2 S
2 S
V
V
V
r r r r
1
1 r r* r* r
x b + b x dV +
sym dV =
x t + t x dS
2V
2S
V
(4.41)
r r r r
1
1 r r* r* r
x b + b x dV +
V =
x t + t x dS
2V
2S
1
2V
[r
x b + b x dV +
1
2V
r r
[x t
r
r
+ t * x dS
In addition, if we consider that the body is free of body force, the above equation becomes:
=
1
2V
[x t
r*
r
r
+ t * x dS
(4.42)
For the particular case shown in Figure 4.4 we have V = V (1) V ( 2) , S = S (1) + S ( 2) ,
r (1)
r ( 2)
t * = p (1) n (1) , t * = p ( 2) n ( 2) . In this case, the equation in (4.42) becomes:

=
r r
r* r
r r* r* r
1
*
(1)
(2)
x t + t x dS + x t + t x dS
2(V (1) V ( 2 ) ) S (1)
S ( 2)
r
r
r
r
1
p (1) x n (1) + n (1) x dS (1) + p ( 2 ) x n ( 2) + n ( 2 ) x dS ( 2 )

(1)
( 2)
2(V V ) S (1)
S (2)
r
r
1
(1) r (1) (1) r
x n + n x dS (1) + p ( 2 ) x n ( 2 ) + n ( 2 ) x dS ( 2)
p
( 2)
2(V V )
S ( 1)
S ( 2)
(1)
We have shown in Chapter 1 that is true

( x n + n x) dS = 2V 1 ,
where n is the
outward unit normal to surface S (see Problem 1.123). For this example, n ( 2) is the

Ciudad Real - Spain
Draft
320
inward unit normal to surface S ( 2) , then, we have
[x n
r
( 2)
+ n ( 2 ) x dS ( 2 ) = 2V ( 2 ) 1 ,
S (2)

=
r
r
1
(1) r (1) (1) r
x n + n x dS (1) + p ( 2 ) x n ( 2) + n ( 2 ) x dS ( 2 )
p
( 2)
2(V V )
S (1 )
S (2)
(1)
1
p (1) 2V (1) 1 p ( 2) 2V ( 2 ) 1
( 2)
2(V V )
1
p (1)V (1) p ( 2 )V ( 2 ) 1
= (1)
( 2)
(V V )
=
(1)
4.1.2
Flux Problems
Problem 4.20
1) Consider a continuum motion in which the stress power is equal to zero. Also, consider
r
r
that the heat flux is given by q = K (T ) x T , which is known as Fouriers law of thermal
conduction, where K (T ) is a second-order tensor called the thermal conductivity tensor (the
thermal property of the material), and c =
u (T )
, where c is the specific heat capacity at a
T
constant deformation (the thermal property of the material) and is expressed in units of
J
. Taking into account all previous considerations, find the
K
energy equation for this process. Then also provide the unit of K (T ) in the International
joule per kelvin, i.e. [c] =

System of Units (SI).
2) Consider the stress power is equal to zero, and that there is a continuous medium with
no internal heat source. Also consider that there is a heterogeneous material where
r
K = K ( x ) is an arbitrary second-order tensor (not necessarily symmetrical). a) Show that
the thermal conductivity tensor is semi-definite positive, b) Check in which scenario the
r
skew part of K ( x ) does not affect the outcome of the heat conduction problem. c) Taking
into account that the material is isotropic, in what format is K ?
Solution: For this problem we know that the stress power is equal to zero, : D = 0 . It then
follows that, the energy equation becomes:
&
u=
r
r
u T
r
r
= : D x q + r = x q + r
{
T t
=0
r
T
r
= x q + r
t
T
r
= x
t
r
[ K (T ) x T ] + r
or
r
r
x [K (T ) x T ] + r = c
T
t
The above equation is called the heat flux equation which is applied to the thermal
conduction problem.

Ciudad Real - Spain
Draft
321
& DT = T .
NOTE: If there is no mass transport it fulfills T
Dt
Then if we take into account the following units: [q] =

r
J
W r
T K
= 2 , x T r = , we
2
x m
m s m
can ensure that the units are consistent if the following is met:
r
[q]
[K ] [ xr T ]
W
W K
J
J
m 2 s = m 2 = s m K = m K m
W
J
=
.
s m K m K
thus, we can draw the conclusion that [K ] =
NOTE: As we will see later, when the stress power is equal to zero, we can decouple the
thermal and mechanical problem. That is, we can study these problems separately.
2) a) We start from the heat conductivity inequality:
r
r
r
r
r
q x T = (K ( x ) x T ) x T 0
r
r
r
xT K ( x) xT 0
or
q i T,i = ( K ij T, j )T,i 0
T,i K ij T, j 0
r
Remember that the arbitrary tensor A is semi-definite positive if it holds that x A x 0

r r
r
for all x 0 thereby demonstrating that K ( x ) is a semi-definite positive tensor. Then, as a
r
result the eigenvalues of K ( x ) are all real values greater than or equal to zero, i.e. K 1 0 ,
r
K 2 0 , K 3 0 . Also remember that since K ( x ) is not symmetric, the principal space of
r
K ( x ) does not define an orthonormal basis. Moreover, it is noteworthy that: the
r
antisymmetric part of K ( x ) does not affect the heat conduction inequality since:
r
r
r
r
x T K ( x ) x T = x T K sym + K skew
r
xT K
sym
r
r
xT + xT K
skew
r
xT
r
xT 0
r
r
r
r
x T K sym x T + K skew : ( x T x T ) 0
r
r
Notice that K skew : ( x T x T ) = 0 , since the double scalar product between an
r
r
antisymmetric tensor ( K skew ) and a symmetric one ( x T x T ) is equal to zero, then:
r
r
r
r
r
0 x T K ( x ) x T = x T K sym x T 0
r
That is, the above inequality is always true whether K ( x ) is symmetric or not.
b) For the proposed problem the only remaining governing equation is the energy
r
r
Du
r
r
&
u = : D x q + r = x q , where u is the specific internal
equation:
Dt
energy, : D is the stress power, and r is the internal heat source per unit volume. Then:
r
r
r
r
&
u = q i ,i = (K ij T, j ) ,i = K ij ,i T, j + K ij T, ji = ( x K T ) ( x T ) + K : x ( x T )
r
r
r
r
= ( x K T ) ( x T ) + K sym + K skew : x ( x T )
r
r
r
r
r
r
= ( x K T ) ( x T ) + K sym : x ( x T ) + K skew : x ( x T )
r
r
r
r
= ( x K T ) ( x T ) + K sym : x ( x T )

Ciudad Real - Spain
Draft
322
r
r
where we have considered the symmetry of [ x ( x T )]ij = T,ij = T, ji . If the material is
homogeneous the implication is that the K field does not depend on ( x ) , so K ij ,i = 0 j . In

this scenario the heat equation reduces to:
r
r
&
u = K sym : x ( x T )
Therefore, when the material is homogeneous, the antisymmetric part of K does not affect
the outcome.
c) The feature of isotropic materials is that their properties (at one material point) do not
change if the coordinate system is changed. It follows then that K must be an isotropic
tensor. An isotropic second-order tensor has the format of a spherical tensor, (see Chapter
1), then the tensor K must be of the type: K = K1 , where K is a scalar:
1 0 0
K ij = K 0 1 0
0 0 1
Problem 4.21
Consider a thermal conduction problem, (see Problem 4.20), in a wall with thickness equal
to h in which the temperature at the outer face ( x1 = 0 ) is equal to 38 C and the
temperature in the interior face ( x1 = h ) is equal to 21 C , (see Figure 4.5). Obtain the heat
flow for case defined by: stationary problem, the temperature field according to x 2 and x3 directions is homogeneous, there is no heat source, and the material is isotropic and
homogeneous.
x2
T ( A) = 38 C
Data:
h = 0.04m
T ( B ) = 21 C
K = 0.19
W
mK
(Interior)
(Exterior)
x1
r
q
h
Figure 4.5
Solution:
As we saw in Problem 4.20 the governing equation for this problem is the equation
T
T
. If we consider the stationary problem we have
= 0 . If
t
t
there is no heat source this implies that r = 0 . With these simplifications the governing
r
r
equation becomes x [K x T ] = 0 , in addition, if the material is homogenous, the tensor
r
r
x [K x T ] + r = c

Ciudad Real - Spain
Draft
323
r
r
r
r
with the thermal properties K do not vary with x , then x [K x T ] = K : x [ x T ] = 0 ,
which in indicial notation is [K ij T , j ],i = K ij ,i T , j + K ij T , ji = K ij T , ji = 0 . By expanding this
123
=0
equation we obtain:
K 11
2T
2T
2T
2T
2T
2T
+ K 12
+ K 13
+ K 21
+ K 22 2 + K 23
+
x 2 x1
x3 x1
x1x 2
x3 x 2
x12
x 2
+ K 31
2T
2T
2T
+ K 32
+ K 33 2 = 0
x1x3
x 2 x3
x3
(4.43)
If the temperature field according to x 2 and x3 -directions is homogenous, this implies that
the temperature gradient components according to these directions are equal to zero, i.e.
T
T
=
= 0 . For an isotropic material, the thermal conductivity tensor components, (see
x 2 x3
Chapter 5 of the textbook), are given by:

K 0 0
K ij = 0 K 0
0 0 K
With these considerations the equation (4.43) becomes:

K 11
2T
=0
x12
By integrating the equation K

K
2T
=0
x12
K11
=K
2T
=0
x12
(4.44)
2T
= 0 we obtain:
x12
integrating

T
+ q1 = 0
x1
q1 = K
dT
dx1
which is the Fouriers law of thermal conduction. Note that for this case q1 is a constant, i.e. it is
independent of x1 . By integrating once more we obtain:
dT =
q1
dx1
K
T ( x1 ) =
q1
x1 + C
K
Applying the boundary condition, x1 = 0 T = T ( A) , we obtain the constant of integration

C = T ( A) . With that we obtain T ( x1 ) =
T ( x1 = h) = T ( B ) =
q1
x1 + T ( A) . In addition, for x1 = h we have
K
q1
h + T ( A)
K
q1 = K
(T ( B ) T ( A) )
h
In this case (unidimensional case), the temperature gradient is the slope of the line defined
by the temperature, which varies linearly in the wall, (see Figure 4.5).
By replacing the problem data (see Figure 4.5), we obtain the heat flux:
q1 = K
W
J
(T ( B ) T ( A) )
W (21 38)( K )
= 0.19
= 80.75 2 = 80.75 2
h
m
m s
mK 0.04(m)
Note that the temperature conversion form degrees Celsius to Kelvin is given by
K = C + 273.15 , then the temperature variation ( T ) either in degrees Celsius or in Kelvin
Ciudad Real - Spain
Draft
324
is the same. Note also that the heat flux flows from the higher temperature to the lower
temperature region.
NOTE: Let us suppose now that we have two walls with different properties as shown in
Figure 4.6.
T ( A)
T (B )
T (C )
K (1)
x1
r
q
K ( 2)
h (2)
h (1)
Figure 4.6
Note that the equation q1 = K (1)
(T ( B ) T ( A) )
is still valid. This also applies to the material
h (1)
(T (C ) T ( B ) )
. To obtain the heat flux we apply the compatibility in
h (2)
temperature on the face B , i.e.:
2 : q1 = K ( 2)
(T ( B _ 1) T ( A) )
h (1)
(T (C ) T ( B _ 2 ) )
q1 = K ( 2)
h ( 2)
q1 = K (1)
q1 h (1)
T ( B _ 1) = T ( A)
T ( B _ 2) = T (C ) +
K (1)
q1 h ( 2)
K ( 2)
T ( B _ 1) = T ( B _ 2 )
T ( A)
q1 h (1)
K (1)
= T (C ) +
q1 h ( 2 )
K ( 2)
thus:
q1 =
(T (C ) T ( A) )
h (1) h ( 2 )
K (1) + K ( 2 )
Problem 4.22
Next, we assume that at a material point there are two types of material that are
represented by a physical quantity per unit volume in such a way that c = c f + c s , and the
r r
r
following holds v = v f + v s , (see Figure 4.7). Considering an isothermal process, an
incompressible medium, and that the property c s does not affect the velocity of the

Ciudad Real - Spain
Draft
325
material f and that the c f -field is homogeneous, and there is no source of the material
f . Show that:
r
c s
r
r
r
Q s x (v f c s ) + x (D x c s ) =
t
Convection-diffusion
equation
(4.45)
r
where the flux of the property s is given by q ( D ) = D x c s .
Control volume
r
vf
cs
r
v
dV
r
vs
Figure 4.7: Heterogeneous medium.

Solution:
Starting from the continuity equation for this physical quantity, we obtain:
Q=
r
+x
t
(v )
Q=
r
r
(c f + c s )
+ r (c f + c s )(v f + v s )
x
t
(4.46)
with Q = Q s + Q f . Thus:
r
r
(c f + c s )
+ r (c f + c s )(v f + v s )
t
x
f
s
r
r
r
(c + c ) f r f
Qs + Q f =
+ r c v + c f v s + csv f + csv s
t
x
f
s
r
r
r
r
c
c
r
Qs + Q f =
+
+ x c f v f + c f v s + csv f + csv s
t
t
f
c
r c s
r
r
r
r
r
r
Qs + Q f =
+ x (c f v f ) +
+ x (c s v f ) + x c f v s + c s v s
t
t
Qs + Q f =
(4.47)
r
c f
r
+ x (c f v f ) = 0 and Q f = 0
t
hold, which is the continuity equation of the physical quantity c f with which the equation
If we assume that there is no ( f )-material source, then

in (4.47) becomes:
Qs =
(4.48)
(4.49)
r
r
r
r
c s
r
r
r
r
+ x (c s v f ) + x ( c s v s ) + x c f v s + c f x v s
t
(4.50)

Ciudad Real - Spain
r
r
r
c s
r
r
r
+ x (c s v f ) + x ( c s v s ) + x (c f v s )
t
Qs =
Qs =
r
r
r
c s
r
r
+ x (c s v f ) + x c f v s + c s v s
t
Draft
326
If the physical quantity c f does not change with x , then the gradient of c f becomes
r
r
x c f = 0 . In addition if we consider the medium ( s ) to be incompressible we obtain
r
r
x v s = 0 . These simplifications indicate that the material ( s ) does not affect the velocity
field of the material ( f ). So, if the amount of the material ( s ) is significant, this approach
is no longer valid. Then, with these approximations we obtain:

r
r
r
r
c s
c s
r
r
r
r
(4.51)
+ x (c s v f ) + x q ( D )
+ x (c s v f ) + x ( c s v s ) =
t
t
r
r
Notice that the term (c s v s ) q ( D ) represents the flux caused by the ( s )-material
r
r
concentration, the diffusive term. The term (c s v f ) q (C ) is related to mass transport, the
r
r
convective term. Considering that q ( D ) = D x c s the equation (4.51) becomes:
Qs =
r
r
c s
r
r
+ x (c s v f ) + x q ( D )
t
r
c s
r
r
r
Qs =
+ x (c s v f ) + x (D x c s )
t
r
c s
r
r
r
Q s x (c s v f ) + x (D x c s ) =
t
Qs =
(4.52)
with that we have demonstrated the equation in (4.45).

Problem 4.23
Consider a water reservoir with
sediment concentration, (see Figure
4.8). The sediment concentration
(concentration density) is given by
( kx3t )
c( x3 , t ) = C t exp
where C and
, per unit volume,

are positive
constants. a) Obtain the total mass of
sediment in the reservoir; b) Obtain
the sediment flux knowing that the
flux is only a function of x3 and time
r r
t , i.e. q = q( x3 , t ) .
x3
x2
b
x1
a
Figure 4.8: Reservoir with sediments.

Ciudad Real - Spain
Draft
327
Solution:
To obtain the total mass we have to solve the integral:
h b a
M = c s dV =
C t exp
( kx3t )
dx1 dx 2 dx3 = ab C t exp
0 0 0
( kx3t )
dx3
( kx3t )
C
= ab
exp
C abC
C
= ab
exp ( kht ) + =
exp ( kht ) 1
k
k
k
To obtain the flux, we can apply the continuity equation of the concentration:
Q=
r
c s
r
+ x q
t
r
c s
r
x q = q i ,i =
t
(4.53)
where we have considered that there is no source of the sediment, i.e. Q = 0 . For this
problem, the flux is not dependent on x 2 and x1 . With this condition we have
q1,1 = q 2, 2 = 0 . Then:
q i ,i = q1,1 + q 2, 2 + q 3,3 =
q1 q 2 q 3
c s
=
+
+
t
x1 x 2 x3
(4.54)
q
c s
3 =
t
x3
where
c s
( kx3t )
( kx3t )
( kx3t )
C t exp
= C exp
C t k x 3 exp
and by replacing into the
=
t
t
equation (4.54) we obtain:
dq 3
c s
( kx3t )
( kx3t )
=
= C exp
+ C t k x3 exp
t
dx3
dq 3 =
q3 =
[ C exp
( kx3t )
+ C t k x3 exp
( kx3t )
]dx
C k x3 t
C
C
( kx3t )
( kx3t )
( kx3t )
exp
exp
exp
+ K3
kt
kt
kt
q 3 = C x3 exp
( kx3t )
(4.55)
+ K3
{
=0
The flux vector in the Cartesian basis is given by q = C x3 exp

Ciudad Real - Spain
Draft
( kx3t )
e3 .
328
4.1.3
Rigid Body Motion
Problem 4.24
Find the linear and angular momentum for a solid subjected to rigid body motion.
x3
r
F(n )
r
F( 2 )
Rigid body
Bt
r
x
x3
x1
G - mass center
r
F(1)
x2
x1
x2
r
v
Figure 4.9
Solution:
According to Problem 2.56 in Chapter 2, we obtained the velocity for rigid body motion
as:
r r r
r r
&
v = c + ( x c)
where is the axial vector (angular velocity) associated with the antisymmetric tensor W
(the spin tensor).
Linear momentum:
r
r r
r
r
r r
r r
r r
&
&
L = v dV = c + ( x c ) dV = c dV + x dV c dV
r
r
r
r r
&
= c dV + x dV c dV
By definition
x dV = mx
is the first moment of inertia, where m is the total mass, and
r
x k is the vector position of the center of mass G . The first moment of inertia is equal to
r r
r
zero if the Cartesian system originates at the center of mass, so, x dV = mx = 0 .
r
r r
r r
&
L = m c + ( x c)
r
= mv
(Linear momentum for rigid body motion)
(4.56)
&
where v = c + ( x c ) is the velocity of the center of mass.
Angular momentum:

Ciudad Real - Spain
Draft
r
r
r
H O = ( x v ) dV =
329
r r
r
r r
&
[x (c + ( x c))] dV
Thus
r
r r
r
r r
r
r r
&
H O = x c dV + x ( x ) dV x ( c ) dV
(4.57)
r
r
r r
r
r r
&
= x dV c + x ( x ) dV x dV ( c )
V
Next, we discuss the second integral of the previous equation.
It was proven in Chapter 1 that given three vectors a , b , c , the relationship

r r
r
r r r
r r r
r r
holds,
thus
when
a=c
it
holds
that
a (b c ) = (a c )b (a b)c
r r
r
r r r
r r r
a (b a) = (a a)b (a b)a , so,
r r r
r r r
x ( x) dV = [( x x) ( x ) x ] dV ,
with
which we obtain:
[x
[
= [x
x k i x p p x i dV = x k x k p pi x p p x i dV = x k x k pi x p x i p dV
x k pi x p x i dV p = I O ip p
r
r r
x ( x ) dV = [( x x ) 1 ( x x )] dV = I O
V
r r
where I O = [( x x ) 1 ( x x )] dV is the inertia tensor with respect to the origin O . As

V
we can observe, I O is a second-order pseudo-tensor, since it depends on the reference
system, and the components I O ij = [x k x k ij xi x j ] dV can be expressed explicitly as:

V
2
2
I O 11 = [( x1 x1 + x 2 x 2 + x 3 x 3 ) 11 x1 x1 ] dV = x 2 + x 3 dV
2
I O 22 = x12 + x 3 dV
2
I O 33 = x12 + x 2 dV
I O 12 = [( x1 x1 + x 2 x 2 + x 3 x 3 ) 12 x1 x 2 ] dV = [x1 x 2 ] dV = I O 12
I O 13 = [x1 x 3 ] dV = I O 13
I O 23 = [x 2 x 3 ] dV = I O 23
where I O 11 , I O 22 , I O 33 , are moments of inertia of the body relative to the reference point O ,
and I O 12 , I O 13 , I O 23 , are the products of inertia of the body relative to the reference point
O . Thus,

Ciudad Real - Spain
Draft
330
2
2
x2 + x3 dV
V
= [x1 x 2 ] dV
V
[x1 x3 ] dV
V
I Oij
[x1 x 2 ] dV
[x1 x3 ] dV
I O 11
V
[x 2 x3 ] dV = I O12
V
I O13
2
x12 + x2 dV
[ + ] dV
[x x ] dV [
x12
2
x3
2 3
I O 12
I O 22
I O 23
I O13
I O 23
I O 33
(4.58)
Returning to the equation in (4.57) we can state that:
r
r
r
r r
r
r r
&
H O = x dV c + x ( x ) dV x dV ( c )
V
r r
r
r r
r
r r
r r
r
& + I m x ( c ) = m x c ( c ) + I
&
=mxc
O
O
r r r
Then by adding and subtracting the term m x x in the above equation we obtain:
r
r r r r
r r r
r r
r
r
r r
r
&
&
H O = m x c c + I O = m x c + ( x c ) m x ( x ) + I O
r r
r r
r r r
r r
r r
r r
r
r
= m x v m (x x) 1 ( x x) + IO = m x v + m ( x x) (x x) 1 + IO
r r
r
= m x v + I
r r r
= m x v + HG
r r
r r
where I = I O + m ( x x ) ( x x ) 1 is the inertia pseudo-tensor, which is related to the
{[
reference system at the center of mass. By means of this equation we can calculate the
inertia tensor in any reference system if we know the inertia tensor at the center of mass:
2
I O ij = Iij m[x i x j ( x12 + x 2 + x 32 ) ij ] . Explicitly, these components can be expressed as:
2
I O 11 = I11 + m( x 2 + x32 ) ; I O 12 = I12 m( x1 x 2 )
I O 22 = I 22 + m( x12 + x32 ) ; I O 23 = I 23 m( x 2 x 3 ) Steiners theorem
2
I O 33 = I33 + m( x12 + x 2 ) ; I O 13 = I13 m( x1 x3 )
(4.59)
Note that, the above equations represent the parallel axis theorem (Steiners theorem) from
Classical Mechanics.
Problem 4.25
Consider a parallelepiped whose dimensions are a b c (see Figure 4.10), in which the
r
mass density filed, ( x ) , is homogeneous. Obtain the inertia tensor with respect to system
in the center of gravity.
Solution:
We use the equation (4.58):
IOij
2
2
x2 + x3 dV
V
= [x1 x2 ] dV
V
[x1 x3 ] dV
V
[x1 x2 ] dV
[ + ]dV
[x x ] dV [
x12
2
x3

Ciudad Real - Spain
[x1 x3 ] dV
I O11
V
[x2 x3 ] dV = I O12
V
I O13
2
x12 + x2 dV
2 3
Draft
I O12
IO 22
I O 23
I O13
I O 23
IO 33
331
Note that, for this problem, the mass density is independent of x (homogeneous material),
and moreover it fulfills that:
m = dV =
V
dV = V = abc
Then, the moment of inertia I O11 becomes:
2
2
I O11 = x2 + x3 dV =
c
2
b
2
a
2
[x
2
2
2
+ x3 dx1dx 2 dx3 =
c b a
2
2
2
Similarly, we can obtain I O 22 =
m 2
(a + c 2 )
12
abc 2
m
(b + c 2 ) = (b 2 + c 2 )
12
12
I O 33 =
m 2
(a + b 2 ) .
12
We leave to the reader show that I O12 = I O 13 = I O 23 = 0 . Recall that the inertia tensor give
us information on how the mass is distributed according to the adopted system, and note
that the mass is equally distributed according to the plane x1 x2 , thus [x1 x 2 ] dV = 0 .
V
Note also that the adopted axes are principal axes of inertia:
I Oij
m 2
2
0
0
12 (b + c )
m 2
=
0
(a + c 2 )
0
12
m 2
2
0
0
(a + b )
12
x3
x2
x1
Figure 4.10: Parallelepiped.

NOTE: A list of inertia tensor for several solids can be found in Wikipedia
http://en.wikipedia.org/wiki/List_of_moments_of_inertia
Problem 4.26
Obtain the principle of conservation of linear momentum and angular momentum for a
solid subjected to rigid body motion.
Solution: We can start from the definition of the principle of conservation of linear
momentum which states that:
Ciudad Real - Spain
Draft
332
r
&
F = Dt v dV = L
V
Then we use the equation of linear momentum obtained in Problem 4.24, L = m v , to

obtain:
r
r
&
r
&
F = Dt v dV = L = m v = m a
V
Then we have:
r
F = m a
Now let us consider the principle of conservation of angular momentum which states:
r
r
r
r
D
D r
&
HO HO
( x v )dV =
Dt V
Dt
By which we obtain:
r
r
&
= HO
or
r
&
= HG
where the equation of angular momentum H O was obtained in Problem 4.24. The set of
equations
equivalent:
r
r
F=ma
and
r
&
= H G inform us that the following systems are
r
&
HG
r
F( 2 )
r
F( n )
G
=
r
F(1)
r
ma
G - center of mass
NOTE: If we are dealing with rigid body motion, the governing equations are:
r
F =ma
and

Ciudad Real - Spain
r
&
= HO
Draft
Governing equations for rigid body

motion
(4.60)
333
Problem 4.27
Consider the beam with the following load and boundary conditions:
P
L
2
HA
L
2
VA
VB
Figure 4.11: Isostatic beam.

Obtain the support reactions VA , VB , H A .
Solution:
Although in the beam there is deformation (small deformation regime) and stress, for
purposes of support reaction calculation of an isostatic beam we can consider as a rigid
body case and the necessary equations (see equations (4.60)) are:
r
r r
F=ma =0
r
r
r
&
MA = HA =0
F
F
F
=0
=0
=0
M
M
M
F
F
=0
=0
= H A + P cos = 0 H A = P cos
= V A + V B P sin = 0 V A = V B + P sin
=0
with which we can obtain V A = V B + P sin =
= V B L P sin
L
P sin
= 0 VB =
2
2
P sin
. Note that we have 3 equations and
2
3 unknowns (isostatic system). If we have a system in which there are more unknown than
equations (hyperstatic system), this procedure is no longer valid since the reactions will
depend on the beam deformation and this depends on the beam stiffness.
Problem 4.28
Find the kinetic energy related to rigid body motion in terms of the inertia tensor, (see
Problem 4.24 and Problem 4.26).
r
&
Solution: The rigid body motion velocity can be expressed as v = c + ( x c ) . Then, the
kinetic energy becomes:
][
r r
1
1
r r
r r r r
r r
&
(v v )dV = c + ( x c ) c + ( x c) dV
&
2V
2V
r
r r r
Using the following vector sum x = x + x , where x is the mass center vector position,
r
and x is the particle vector position with respect to the system that has its origin in the
K (t ) =
center of mass, the energy equation becomes:

Ciudad Real - Spain
Draft
334
{[
][
{[(
]}
)
r r
r r
r r
r r r
r
1
&
&
c + (( x + x ) c) c + (( x + x ) c ) dV
2V
r r
r r
r r
r r
r r
r r
1
&
&
=
c + ( x c ) + ( x ) c + ( x c ) + ( x ) dV
2V
K (t ) =
] [(
]}
&
Note that v = c + ( x c) is the center of mass velocity, thus:
K (t ) =
{[
r
r r
1
v + ( x )
2V
r r
] [vr + ( x )]}dV
or:
r r
r r r
r r r
r r
r r
1
1
1
1
v v dV + v ( x ) dV + ( x ) v dV + + ( x ) ( x ) dV
2V
2V
2V
2V
K (t ) =

K (t ) =
r r
r r r
r r
r r
1
1
v v dV + v ( x ) dV + ( x ) ( x ) dV
2V
2V
V
Next, we discuss separately the terms of the previous equation:

1)
r r
1 r
1
v v dV = v
2
2V
dV = 2 mv
r r
2) v ( x ) dV = v x dV = v ( m {) = 0
x
r
=0
Note that, the system x is located at the center of mass ( G ), hence the center of mass
r
vector position related to the system x is zero.
r
3) ( x ) ( x ) dV
V
[( x ) ( x )] dV =
ijk j x
k
ipq p x q dV = ( jp kq jq kp ) j x k p x q dV
V

= j ( jp kq x k x q jq kp x x ) p dV
k q
V
= j ( jp x x k x p x j ) p dV
k
V
= j ( jp x x x j x p ) dV p
k k
= j I jp p
or in tensorial notation as:

r
[( x ) ( x )] dV = [( x x ) 1 ( x x )] dV
r
r
= I
where I is the inertia pseudo-tensor related to the system located at the center of mass,
(see Problem 4.24).

Ciudad Real - Spain
Draft
335
Then if we bear in mind all the above considerations, the kinetic energy equation for rigid
body motion becomes:
K (t ) =
r r
r r r
r r
r r
1
1
1
v v dV + 2 v ( x ) dV + ( x ) ( x ) dV
2V
2V
2V
144424443
=0
1
2
1r
2
Kinetic energy for rigid

body motion
K(t ) = mv 2 + I
(4.61)
Additionally, if we take into account that:
2
2
x 2 + x3 dV
V

Iij = [x1 x 2 ] dV

[x1 x3 ] dV
V

[x1 x 2 ] dV

[x1 x3 ] dV
I11
V

[x 2 x 3 ] dV = I12
V
I13
x1 2 + x 22 dV
[x + x ] dV
[x x ] dV
[
1
2 3
I12
I 22
I 23
I13
I 23
I 33
we obtain an explicit equation for the kinetic energy as:

1
2
1
2
K(t ) = mv 2 + k Ikj j
1
1
= mv 2 + [1
2
2
=
I11
3 ] I12
I
13
I12
I 22
I 23
I13 1
I 23 2

I 33 3

1
1
2
2
mv 2 + I111 + I 22 2 + I 33 3 2 I12 1 2 2 I13 1 3 2 I 23 2 3
2
2
2
1
2
K(t ) = mv 2 +
1
2
2
I111 + I 22 2 + I 33 3 2 I12 1 2 2 I13 13 2 I 23 2 3
2
2
]
(4.62)
Problem 4.29
Consider the inertia pseudo-tensor, I O , with respect to the system x1 x 2 x3 , (see Figure
4.12). a) Make the physical interpretation of the inertia tensor. b) Given another
* * *
orthonormal system, represented by x1 x 2 x3 . Obtain the inertia tensor components in this
new system. c) Show that the inertia tensor is positive definite tensor. For a solid in
motion, find in which situation the term

Ciudad Real - Spain
DI O &
I O is equal to zero.
Dt
Draft
336
x3
*
x2
*
x3
*
x1
x2
x1
Figure 4.12
Solution:
The inertia pseudo-tensor depends on the adopted coordinate system, and by definition is
given by:
r r
r r
I O = [( x x ) 1 ( x x )] dV
I O ij = x k x k ij xi x j dV
or in components
2
2
x 2 + x3 dV
V
I ij = [x1 x 2 ] dV
V
[x1 x3 ] dV
V
[x1 x 2 ] dV
[x1 x 3 ] dV
V
[x 2 x3 ] dV
2
x12 + x 2 dV
[x + x ]dV
[x x ] dV [
2
1
2
3
2 3
a) The inertia tensor give us the information as the mass into the body is distributed under
the adopted system.
The term [x1 x 2 ] dV indicates as the mass is distributed along the plane x1 x 2 . Then, if
V
the material is homogeneous, i.e. the mass density field is independent of x , and x1 x 2 is
a plane of symmetry, i.e. the mass is distributed equally with respect to plane x1 x 2 , the
term
[x x ] dV
1 2
is equal to zero. With this, we conclude that: if the planes x1 x 2 ,
x1 x 3 , x 2 x3 , are planes of symmetry, the inertia matrix is a diagonal matrix.
Let us consider a student attached to a disc with outstretched arms, each hand holding a
r
weight (see Figure 4.13 initial system). The disk rotates with angular velocity (i ) and the
r
(i
inertia tensor according to the system x is given by I O) . If we consider a system without
energy dissipation, what will happen when the student moves the arms inwardly as shown
in Figure 4.13 final system? As we are dealing with a conservative system, the angular
momentum is conserved too, i.e.:
r(
r(
H Oi ) = H Of )
r
r
(
I ( i ) ( i ) = I Of ) ( f )
O
Since for the final system the mass is more concentrated according to the rotation axis than
r
r
(
i
to the initial system the inequality I Of ) < I (O) holds and as consequence ( f ) > (i ) .

Ciudad Real - Spain
Draft
x3
Initial system
x3
337
Final system
r
(i )
r
( f )
(
I Of )
(i
I O)
Figure 4.13:
b) Let us assume that the given systems (see Figure 4.12) are related by the transformation
law xi* = Aij x j , where Aij is the orthogonal matrix, then it follows that xi = A ji x * . Thus
j
being able to express I O ij as follows:
* *
*
I O ij = x k x k ij xi x j dV = ( x k x k )Aip pq A jq Aip x * A jq x q dV
p
V
* *
*
* *
*
= Aip ( x k x k ) pq x * x q A jq dV = Aip ( x k x k ) pq x * x q dV A jq
p
p
V
{[
]}
= Aip I * ij A jq
O
Abusing a bit of notation, we also use tensorial notation, but bear in mind that we are
working with tensor components, and we are not doing an orthogonal transformation.
r r
r r
I O = [( x x ) 1 ( x x )] dV
r r
r
r
= ( x * x * )A T 1 A (A T x * A T x * ) dV
V
r r
r
r
= ( x * x * )A T 1 A (A T x * x * A ) dV
V
= AT
V
{ [( x * x * )1 ( x * x * )]} A dV
r
r r
r
r
= A T ( x * x * )1 ( x * x * ) dV A
V
= A T I* A
O

Ciudad Real - Spain
Draft
338
I O = A T I* A
O
I O ij = A
*
ip I O ij
Inertia tensor components after a base

change (rotation)
A jq
(4.63)
Then, it is valid I * = A I O A T , which are the new inertia tensor components in the
O
* * *
system x1 x 2 x3 . Note that the equation (4.63) is the same component transformation law
for a second-order tensor, where A is the transformation matrix from the x1 x 2 x3 -system
* * *
to x1 x 2 x3 -system.
c) For a positive definite tensor, by definition, its eigenvalues are greater then zero.
We will start from the kinetic energy obtained in Problem 4.28, i.e.:
1
2
K(t ) = mv 2 +
1
2
2
I111 + I 22 2 + I 33 3 2 I12 1 2 2 I13 13 2 I 23 2 3
2
2
The kinetic energy is a scalar and always a positive number, and only in two situations the
kinetic energy is zero, namely: when there is no mass or when the body is at rest. We adopt
a system such that the origin is at the center of mass and the adopted axes are axes of
symmetry (inertia principal system) and that the body is rotating around the origin (center
of mass). In this situation the kinetic energy becomes:
1
K(t ) = [1
2
0 1
I1 0
0 I
0 2
3 ]
2

0
0 I 3 3
1442443
4
4
Eigenvalues of the
Inertia tensor
1
2
2
I11 + I 2 2 + I 3 3 > 0
2
2
1
2
2
In addition, if we have a motion such that 2 = 3 = 0 , we have K(t ) = I11 , then, the
only way that the kinetic energy is always positive is when I1 > 0 holds. Similarly, we can
conclude that I 2 > 0 and I 3 > 0 . Hence, the inertia tensor is a positive definite tensor.
d) As the inertia pseudo-tensor is dependent on the adopted system, for the following
situations the inertia tensor to a solid in motion does not change with time:
1) If the adopted system is attached to the solid.
2) If the solid is rotating along a axe of symmetry, for example if a cylinder is rotating along
the prismatic axe, then during motion the mass distribution is not changing with respect to
the adopted system, (see Figure 4.14).

Ciudad Real - Spain
Draft
339
reference system fixed in space
Figure 4.14
Problem 4.30
Consider a homogeneous cylinder of radius r and height h = 3r with total mass equal to
2
2
m , (see Figure 4.15). Find the inertia tensor in the system Ox1 x x3 . The system Ox1 x x3 is
2
given by the rotation of the system Ox1x x3 of 45 along the axis x1 . The systems
2
Gx1 x 2 x3 and Ox1x x3 have the same orientation.
Hint: For the reference system Gx1 x 2 x3 we know the inertia tensor components and are
given by:
I G ij
1
2
2
0
12 m(3r + h )
1
0
=
m(3r 2 + h 2 )
12
0
0
2 0 0
mr 2
0 2 0
0 =
0 0 1
1 2
mr
x3
r
x3
x3
x1
x2
r
rG
h = 3r
x2
45
x2

x1, x1
Figure 4.15

Ciudad Real - Spain
Draft
340
Solution:
2
First of all we obtain the inertia tensor in the system Ox1x x3 by means of the Steiner
theorem, (see equation (4.59) in Problem 4.24). After that, we obtain the components due
to a rotation by means of the equation (4.63) in Problem 4.29.
By means of the equations in (4.59) we can obtain:

2
I11 + m( x 2 + x 32 ) = I 11
O
2
2
I 22 + m( x1 + x 3 ) = I 22
O
2
I33 + m( x12 + x 2 ) = I 33
O
; I12 m( x1 x 2 ) = I 12
O
; I 23 m( x 2 x3 ) = I 23
O
; I13 m( x1 x 3 ) = I 13
O
(4.64)
where ( x1 , x 2 , x 3 ) are the coordinates of the center of mass with respect to the system
r
3 3
2
Ox1x x3 . Then, we define the vector rG = x1e1 + x 2 e 2 + x 3 e = 0e1 + re 2 + re . With
2
that we obtain:
3
17
2
I 11 = I11 + m( x 2 + x 32 ) = mr 2 + m r 2 + ( r ) 2 = mr 2
O
2
4
3
13
I 22 = I 22 + m( x12 + x 32 ) = mr 2 + m 0 2 + ( r ) 2 = mr 2
O
2
4
1
3
2
I 33 = I33 + m( x12 + x 2 ) = mr 2 + m 0 2 + r 2 = mr 2
O
2
2
I 12 = I12 m( x1 x 2 ) = 0
O
I 23 = I 23 m( x 2 x3 ) = 0
O
3
3
I 13 = I13 m( x1 x3 ) = m(r )( r ) = mr 2
O
2
2
2
Thus resulting in the following inertia tensor components for the Ox1x x3 -system:
0
34 0
mr 2
0 13 6
I O ij =
4
0 6 6
2
2
Considering the transformation matrix between the systems Ox1x x3 and Ox1 x x3 :
0
0
1
0 cos 45 sin 45
A =
0 sin 45 cos 45
and applying the equation (4.63) we obtain:
I O ij = A I A = Aip I ij A jq
O
O
T

Ciudad Real - Spain
Draft
0
34 0
mr 2
=
0
7 7
8
0 7 31
Problem 4.31
341
Taking into account the angular momentum H O = m x v + I = m x v + H G , find the

rate of change of the angular momentum in such a way that we do not need to calculate at
each instant of time the inertia tensor.
r
- angular velocity of the body
r
HG
r
- angular velocity of the system x *
x3
*
x2
*
x3
r
HO
*
x1
x3
G
r
x
x1
x2
x2
G - center of mass
x1
Figure 4.16
Solution: Applying the material time derivative we obtain:
r r r
r
D r
D
&
m x v + HG
HO HO =
Dt
Dt
r r
D
D r
m x v +
HG
=
Dt
Dt
r
r
r Dv
r
Dx r
&
=m
v +m x
+ HG
Dt
Dt
r r
r r r
&
= m v 2v + m x a + H G
13
r
[ ]
=0
Then, we obtain:
r r r
r
D r
&
&
HO HO = m x a + HG
Dt
(4.65)
r
&
where a is the acceleration of the center of mass. Next, we discuss the term H G . We
2
adopt the mobile system x1 x x3 but with fixed orientation in space which is parallel to the
r
fixed system x1 x 2 x3 , (see Figure 4.16). By expressing the components of I and in the
2
system x1 x x3 , we obtain:
r
r
H G = I
r
r
D r
&
& r
&
H G H G = I + I
Dt
r
Note that, as the solid rotates with respect to the system x the inertia tensor changes,
r
since the mass distribution is changing with respect to the system x . Then, at each time
rate of change

step we have to calculate the inertia tensor. This procedure is very laborious. To solve this

Ciudad Real - Spain
Draft
342
problem, we adopt a new system x * , which has origin at the center of mass, (see Figure
4.16). By means of the component transformation law, the following is true:
(components)
r*
r
r
r*
H G = A H G
HG = A T HG
;
r
r
r*
r*
T
= A ; = A
*
T
*
;
IO = A T I O A
IO = A IO A
where A is the transformation matrix from the x -system to x * -system.

r
The rate of change of H G = A T H G becomes:
r
r*
r*
r*
D r
D
&
&
&
A T HG = A T HG + A T HG
HG HG =
Dt
Dt
(4.66)
By analogy with the rate of change of the orthogonal tensor, (see Chapter of the textbook),
&
&
we can conclude that = A A T A T = A T T , where T is the antisymmetric
r
tensor and represents the rate of change of rotation of the system x * with respect to the
r
system x . Then, we can express (4.66) as follows:
r
r*
r*
&
&
H G = A T T H G + A T H G
(components)
r*
r*
&
= A T T H G + H G
(4.67)
r* r
r*
Resorting to the antisymmetric tensor property such that T H G = H G (see NOTA
r
r r
3), where is the axial vector associated with the antisymmetric tensor T , i.e. = (t )
r
is the angular velocity of the mobile system x * . Proving that (4.67) can still be written as
follows:
r
r*
r*
&
&
H G = A T T H G + H G
(components)
r*
r*
&
T r *
= A HG + HG
(4.68)
where
r*
r*
D * r * DI * r *
&
* D
HG =
I =
+ I
Dt
Dt
Dt
The term
1)
DI *
is equal to zero when one of the two possibilities holds:
Dt
r r
r
DI *
= 0 if the system x * is attached to the solid. In this case, the equation =
Dt
holds, i.e. the mobile system velocity is equal to the angular velocity of the solid.
2)
DI *
= 0 if the solid rotates around a prismatic axis, (see Figure 4.14 in Problem 4.29).
Dt
NOTE 1: The equation in (4.68) can be rewritten as follows:

r
r*
r*
r
&
&
H G = A T * H G + H G
r
r*
r*
r * (components)
&
& r* r *
&
T r*
A H G = A A H G + H G = H G + H G

Ciudad Real - Spain
Draft
(4.69)
343
r
r
r
&
&
Note that the term A H G are the components of H G in the system x * , and note also
r
&*
&
that A H G H G , then:
r * r
r
r
&
&
A H = H * + * H * (components)
G
G
G
(4.70)
we can also express the above equation in tensorial notation:

r
DH G
Dt
r
DH G
r r
=
Dt + H G
r
f
(4.71)
r
r
DH G
represents the rate of change of H G with respect to the fixed system,

where
Dt f
r
r
DH G
represents the rate of change of H G with respect to the rotating system with an
Dt
r
angular velocity .
NOTE 2: The equation in (4.71) is valid for any vector (see Figure 4.17), i.e. the rate of
r
r
change of the vector b respect to the fixed system x is equal to the rate of change of the
r
r
vector b respect to the rotating system x * plus the vector product between angular
r
velocity of the system ( which is associated to the antisimetric tensor T ) and the vector
r
b:
r
r
r
r Db
Db
Db
r r
T
=
+ b =
+ b
Dt
Dt
rotating
fixed Dt rotating
(4.72)
r r D
D
D
Note also that
.
=
+ 2 =
1 r3 Dt
r
Dt f Dt r
=0
r
b
x3
*
x2
*
x3
*
x1
x2
x1
Figure 4.17
NOTE 3: Note that the equation (4.72) is the convective rate, (see Chapter on The
C
r
&
Objectivity of Tensors in the textbook), which is defined by a = a + l
l
C
r r
&
= D + W , then a = a + l

Ciudad Real - Spain
a , where
&
a = a + (D + W)T a . Recall from Chapter 2 (Chaves (2013))
Draft
344
1
2
&
&
&
that W = R U U 1 U 1 U R T + R R T holds. And if we are considering rigid solid
C
r r
r
&
&
&
motion we have D = 0 , U = 0 , and W = = R R T , with that we obtain a = a + T a .
NOTE 4: Let us expose a simple example to obtain T . Let us assume that the ei -system
is rotating according to the ei -system (see Figure 4.18), and to obtain T we procedure as
follows. The transformation matrix from ei to ei is given by:

cos
A = sin
d (cos )
dt
d (A )
& = d (sin )
A
dt
dt
d (sin )
dt
d (cos )
dt
0
sin
& A T = cos
&
=A
cos
&
0
&
T = 0
0 0
sin
0
0 0
0 = 3
0 2

3
0
sin
cos
0
0
0
0
&
sin
&
0 = cos

0 0
0 cos
0 sin
0 0
2
1
sin
cos
0
(4.73)
&
cos 0 sin
&
&
sin 0 = cos
cos
sin
0
0
0
&
0
0 1 0 0 0
&
&
0 = 1 0 0 = 0 0
0 0 0 0 0 0
1
0
i = 0

&

r
where is the axial vector associated with the antisymmetric tensor T .
&
3 =
e3
e1
e2
e3
e2
e1
Figure 4.18

Ciudad Real - Spain
Draft
345
NOTE 5: Inertial forces

Let us consider the system OX 1 X 2 X 3 (see Figure 4.19) which is fixed in space. This system
is denoted by inertial reference frame. To this system the Newtons law is applied, and if there
is a falling body it is true that:
r
r
F = mA
Let us consider also that an observer (attached to the system ox1 x 2 x3 ) is moving (for
simplicitys sake we will just consider translation). Since the system ox1 x 2 x3 is moving we
denote it by non-inertial reference frame. By means of vector summation, (see Figure 4.19), we
can obtain:
r r r
X =c+ x
The material time derivative of the above equation becomes:

r r r
& & &
X =c+ x
D
Dt
r r
&& && r
X = c + &&
x
r && r
r
A = c + &&
x
and if we multiply by mass ( m ) we get:

r
r
r
&&
mA = mc + m&&
x
r
r
r
r
&& r
&&
m&& = mA mc = F mc
x
r
r r
&&
ma = F mc
Note that, for the observer it appears the additional force (m&&) to the Newtons law.
c
This additional force is a fictitious force or pseudo force which is denoted by inertial force. In
addition, inertial forces appear if the observers system is rotating, e.g. centripetal force.
r
mA
X2
r
x
r
X
x2
r
c
x1
X1
X3
Figure 4.19

Ciudad Real - Spain
Draft
346
Problem 4.32
Show that the acceleration at a fixed system a f can be expressed as:

r
r
r r
r
r r
a f = ar + 2( v r ) + ( x )
r
(4.74)
where a r and v r are, respectively, the acceleration and the velocity of a particle with
r
respect to an observer that is rotating with the system x * , (see Figure 4.17). Consider also
r r
r
that = is the angular velocity of the system x * , which is constant with time.
Solution:
We use directly the equation in (4.72) to obtain the velocity:
r
r
r r
Dx
Dx
=
+ x
Dt f Dt r
r
r r r
v f = vr + x
We apply the same definition to the above equation in order to obtain the acceleration, i.e.:
r
r r r
r r r
Dv f
r
r r r
D[v r + x ]
D[v r + x ]
=
=
+ [v r + x ]
Dt
Dt
f
r
Dt f
r
r r
r
r r r
r r
Dv D[ x ]
af = r +
+ v r + ( x )
Dt r Dt r
r
r
r
r
r r Dx
r r r
r r
Dv D
x+
af = r +
+ v r + ( x )
Dt r Dt r
Dt r
r
r
r r r r r r r
r r
&
a = a + x + v + v + ( x )
f
r
r
r r
r r
r
r r
&
a f = ar + x + 2( vr ) + ( x )
r r
&
As we are assuming angular velocity constant = 0 , i.e. the angular acceleration is zero,
with that we obtain the equation in (4.74). Then, we can conclude:

r
r
r r
r r
r
r r
&
a f = ar + x + 2( v r ) + ( x )
(4.75)
Note that to obtain the above equation we have not used any principle of conservation.
The above equation is just relating the acceleration in a fixed system in function of
parameters defined in the rotating system.
r
r r r
r r r
NOTE 1: Using the identity a (b c ) = (a c )b (a b )c , (see Problem 1.17), we can

r
r r r
r r r
r r r
conclude that ( x ) = ( x ) ( ) x = ( x ) x . Note that, if = 3e 3 ,

(see Figure 4.20), and also if we adopt the system ( e r , e , e3 ) and taking into account that
2
r r
r
r r
r r r
r 2r
r 2r
r = 0 we obtain the following equation ( r ) = ( r ) r = r , which is
the
centripetal
acceleration,
(see
Problem
2.57).
Earth
rotates
at
rate
r
r r
2 rad
rad
rad
. Note that the term ( x ) is very small
=
0.727 10 4
day 86400 s
s
r r
compared with the term 2( vr ) .
3 = 2

Ciudad Real - Spain
Draft
x3 , z
= 3 e 3
r
r
x3
347
e3
r
x
er
x2 , y
x1 , x
Figure 4.20
r
NOTE 2: The term 2( v r ) , which was established by Gustave-Gaspard Coriolis in

1835, is associated with the fictitious force called Coriolis force. Next, we represent
r r
i
2( v r ) in the system e :
3
e2
Latitude
r
x
3
e
e1

e 3 , e2

e 2 , e1
Equator
3
e1 , e
Figure 4.21

Ciudad Real - Spain
Draft
348
Then, the transformation law from e i to ei is given by:
e1 0
1
0 e1

e
e2 = sin 0 cos 2
e cos 0 sin e
3
3
1
0
0
sin 0 cos
B=
cos 0 sin
3e3
(4.76)
3
= 3 cos( )e2 + 3 sin( )e
e2
3
e
S
Figure 4.22
r
The term 2( v r ) can be obtained as follows:
3
e1
e2
e
r r
2( vr ) = 0 3 cos( ) 3 sin( )
vr1
vr 2
vr 3
3
= 2e1[3 cos( )vr 3 3 sin( )vr 2 ] 2e2 [ 3 sin( )vr1 ] + 2e [ 3 cos( )vr1 ]
3
= 2[3 cos( )vr 3 3 sin( )vr 2 ]e1 + 2[3 sin( )vr1 ]e2 2[3 cos( )vr1 ]e
(4.77)
The term f = 23 sin( ) is known as Coriolis parameter. To small value of vr 3 the above
equation reduce to:
r
r r
Dv r
= 2( v r ) = [23 sin( )vr 2 ]e1 + [ 23 sin( )vr1 ]e2 = [ f vr 2 ]e1 + [ f vr1 ]e2
Dt r
Dv r 1
Dt = f vr 2
Dv r 2 = f v
r1
Dt

Ciudad Real - Spain
Draft
349
Figure 4.23: Coriolis effect (Ref.: Wikipedia Coriolis effect).

NOTE 3: Deflection of vertically falling body
A very simple application of the Coriolis effect is presented next. Let us consider an
observer on the surface of the Earth. Let us consider also that a body of mass m is free
falling from rest with the following initial conditions: at t = 0 . ( x3 = h) , ( x1 = 0) ,

d
x3 = v3 = 0) , (v1 = 0) , (v2 = 0) . As the body is falling we will calculate the deflection of

dt
the body, i.e. we will obtain x1 related to the observer which is attached to a system which
(
is rotating with the Earth. We will adopt the system used in Figure 4.21.
r
The Newtons Second Law ( F = ma f ) (apply to an inertial reference frame), then

r
r
r r
r
r r
F = m[ar + 2( v r ) + ( x )]
r
r r
ar = ge 2( vr )
r
r
r r
r r
3
ma r = F 2m( v r ) = mge 2m( v r )
2[3 cos( )vr 3 3 sin( )vr 2 ] 23 cos( )vr 3

r
2[3 sin( )vr1 ]

(a r ) i =
0
=
g
2[3 cos( )vr1 ] g
where the acceleration a f is given by (4.75), and we are considering that the term
r
r r
r r
( x ) is very small when compared with the term 2( vr ) whose components are
given by (4.77). Then
d 2 x1
2
ar1 dt 23 cos( )vr 3
2
r
d x
(ar )i = ar1 = 22 =
0
a dt
g
r1 d 2 x3
dt 2
(4.78)
Note that
d 2 x3
dx
= g integratin g 3 = gt + C1
vr 3 = gt
2
dt
dt
dx
t2
t2
3 = gt integrating x3 = g + C2
x3 = g + h
dt
2
2

Ciudad Real - Spain
Draft
350
where we have considered the initial conditions, i.e. at t = 0 (vr 3 = 0)
t = 0 ( x3 = h)
C2 = h . Note that x3 = g
C1 = 0 , and
t
gt
.
+h=0h=
2
2
Considering the acceleration vr 3 = gt into the first component of (4.78) we can obtain:
d 2 x'1
dx '
t2
integratin
= 23 cos( )vr 3 = 23 gt cos( ) g 1 = 23 g cos( ) + C1 = vr1
2
dt
dt 2
where the constant of integration is obtained with the initial condition

at (t = 0) {v1 = 0
r
C1 = 0
dx '1
t3
= vr1 = 3 g cos( )t 2 integratin g x'1 = 3 g cos( ) + C2
dt
3
Note also that C2 = 0 , with that we obtain:

1
x'1 = 3 g cos( )t 3
3
As the body is falling from height h we can state that h =
1 2
gt t =
2
2h
, with that the
g
above equation becomes:

3
1
g 2h 2
x'1 = 3 g cos( )t 3 = 3 cos( )
3
3 g


Ciudad Real - Spain
Draft
351
NOTE 4: Acceleration due to sphericity
Local system e1 (east)- e2 (north)- e3 (radially upward)
e2
r
r
Latitude
e3
e1
N North
S South
E East
W West
r
x
3
e
r
r
r = x cos
e3
e2
Equator
e1
e1
S Pole
r d
r
d
= x cos
( e1 )
v r1 = r
dt
dt
e3
vr 2
NP
e2
r
r
r
x
3
e
e2
NP
SP
v r1
r
r
e1
e1
r d
( e2 )
vr 2 = x
dt
Figure 4.24

Ciudad Real - Spain
Draft
352
Previously we have obtained the transformation matrix from ei to ei , (see equation

(4.73)):
cos
A = sin
sin
cos
0
0
0
(4.79)
and the transformation matrix from e i to ei (see equation (4.76)) is given by:
1
0
0
sin 0 cos
B=
cos 0 sin
(4.80)
Then the transformation matrix from ei to ei is given by:

1
0 cos
0
sin 0 cos sin
C = BA =
cos 0 sin 0
sin
cos
0
0 sin
0 = sin cos

1 cos cos

cos
sin sin
cos sin
0
cos
sin
The rate of change of C is given by:
&
&
cos
sin
0
d (C ) & &
&
&
&
&
C = ( cos cos + sin sin ) ( cos sin sin cos ) sin
dt
&
&
&
&
&
( sin cos cos sin ) ( sin sin + cos cos ) cos
&
After the algebraic operation = CC T takes place we obtain:
&
&
0
sin cos
&
v
&
&
= r r1 sin
0
= CC T = sin
&
&
cos
x cos
0

v r1
r
x cos cos
0
vr1 tan vr1
1
0
= r vr1 tan
vr 2
x
v r1
0
vr 2
v r1
sin
r
x cos
0
vr 2
r
x
v
r r1 cos
x cos
vr 2
r
which is an antisymmetric matrix, as expected. Notice that according to Figure 4.24 the
& d = r v r 1
& d = vr 2 hold.
following relationships
and
r
dt
x cos
dt
We apply the definition (see equation (4.71))

r
Dv
Dt
r r
Dv
=
+ vr
f Dt r
r r
r
Note also that T v r = v r holds, so:

Ciudad Real - Spain
Draft
vr1 tan
0
r
1
v r = r vr1 tan
x
v r1
0
vr 2
vr1vr 2 tan + vr1vr 3

r
r
1 2
a f = a r + r vr1 tan + vr 2 vr 3
x
vr21 vr22

v r1 v r1
v = 1 v 2 tan + v v
vr 2 r 2
r r1
r 2 r3
x
vr21 vr22
0 v r 3


r
r
1 2
a r = a f r vr1 tan + vr 2 vr 3
x
vr21 vr22
353
(4.81)
NOTE 5: Coriolis + Curvature acceleration

The acceleration related to the Coriolis terms (see Eq. (4.77) and (4.75)) and curvature is
given by:
r
r
r r
r r
r r
a f = a r + 2( v r ) + T vr + ( x )
(4.82)
vr1vr 2 tan( ) + vr1vr 3

2[3 vr 3 cos( ) 3 vr 2 sin( )]
r r
1 2
T r
2( v r ) + v r =
2[3 vr1 sin( )]
+ r vr1 tan( ) + vr 2 vr 3
2[3 vr1 cos( ) ]

vr21 vr22
(4.83)
where
Problem 4.33
Consider the rigid body in motion in which there are no forces acting on the body and also
consider a torque-free motion. a) Show the Eulers equations of motion:
&
I11 = 2 3 (I 2 I3 )
&
I 2 2 = 13 (I3 I1 )
I = (I I )
1 2 1
2
3&3
Eulers equations of motion
(4.84)
where Ii are the principal moment of inertia related to the system G xyz whose origin is at
r
the center of mass G , i are the components of the body angular velocity ( ), and
&
i
Di
denotes the time derivative of the angular velocity.
Dt
b) Show that the kinetic energy is constant.

Solution:
The governing equations for a rigid body motion (see Problem 4.26) are:
r
r
F = ma
and
r
r
&
MG = HG
If the body is free of forces and torque we have that:

r
F =0
and
r r
&
= 0 = HG
r
&
Next we will evaluate the term H G .
We will consider a mobile system Gxyz attached to the body, (see Figure 4.25), so, in this
r r
situation we have that = .
Ciudad Real - Spain
Draft
354
r
r
HG
r
- angular velocity of the system x
x3
x1 , x2 , x3 - principal axes of inertia
=
x2
x3
x1
G
x1
x2
G - center of mass
Figure 4.25
r
&
In Problem 4.31 we have obtained an efficient equation in order to calculate H G , (see
r r
equation (4.71)), and by considering = we get:
r
DH G
Dt
r
DH G
f Dt
DH G
r r
+ HG =
Dt
r
r
+ H Gxyz
For this problem we have:

I1
(I Gxyz ) ij = 0
0
0
I3
0
I2
0
1
r

( ) i = 2

3
The angular momentum:

r
r
r
r
components
H Gxyz = I Gxyz
( H Gxyz ) i = (I Gxyz )ij ( ) j
r
( H Gxyz )1 I1 0 0 1 I11
r

( H Gxyz ) 2 = 0 I 2 0 2 = I 2 2
r
( H

Gxyz )3 0 0 I 3 3 I 33
The rate of change of the angular momentum:

Note that, since the system G xyz is attached to the body the mass distribution respect to
this system does not change during motion, so, I Gxyz does not change as well, i.e.
&
I Gxyz = 0 . With that we have:
r
DH G
Dt
r
DH Gxyz
=
Dt
r
&
H Gxyz
Gxyz

Ciudad Real - Spain
components
Draft
r
&
(H
) &
&
&
r Gxyz 1 I11 + I11 I11
&
&
&
&
( H Gxyz ) 2 = I 2 2 + I 2 2 = I 2 2
r
I + I I
&
&
&
&
( H Gxyz ) 3 3 3 3 3 3 3
e1
r
r
H Gxyz = 1
e2
2
355
e3
3
I11 I 2 2
I 3 3
= (2I33 3I 2 2 )e1 + (3I11 1I 33 )e 2 + (1I 2 2 2 I11 )e 3
= 2 3 (I 3 I 2 )e1 + 13 (I1 I3 )e 2 + 12 (I 2 I1 )e 3
Components:
2 3 (I3 I 2 )
r
r
H Gxyz i = 13 (I1 I 3 )
(I I )
1 2 2 1
With that we can calculate

r
DH G
Dt
r
r
r
r
&
= H Gxyz + H Gxyz = 0
{ }
} {}
r
r
r
r
&
H Gxyz i + H Gxyz i = 0
&
I11 2 3 (I3 I 2 ) 0

&
I 2 2 + 13 (I1 I3 ) = 0
I (I I ) 0
3&3 1 2 2 1
&
I11 = 2 3 (I 2 I 3 )
&
I 2 2 = 13 (I3 I1 )
I = (I I )
1 2 1
2
3&3
b) The kinetic energy for rigid body motion (see Problem 4.28 equation (4.61)) is given by:
1
2
1r
2
K(t ) = mv 2 + I
Since the origin of the adopted system is at G (mass center) we have v = 0 , with that we
obtain:
1
1
K(t ) = k I kj j = [1 2
2
2
=
1
2
2
I11 + I 2 2 + I33
2
2
I1 0
3 ] 0 I 2
0 0
0 1
0 2

I 3 3

And the rate of change of the kinetic energy becomes:
] [
1 D
1
D
2
2
&
&
&
&
K(t ) = K(t ) =
I11 + I 2 2 + I 33 = 21 I11 + 22 I 2 2 + 23I 33
2
2 Dt
2
Dt
&
&
&
= 1 I11 + 2I 2 2 + 3I 33
If we consider the Eulers equation (4.84) the above equation becomes:

&
&
&
&
K(t ) = 1 I11 + 2I 2 2 + 3I33 = 1 23 (I 2 I3 ) + 213 (I3 I1 ) + 312 (I1 I 2 )
= 1 2 3 (I 2 I3 + I3 I1 + I1 I 2 )
=0
with that we show that the kinetic energy is constant for any problem which is governed by
Eulers equations of motion.
Ciudad Real - Spain
Draft
356
Problem 4.34
Obtain a simplified form of the rigid body governing equations for the particular case:
a) Rigid body rotation around a fixed axis without forces.
Solution:
We will consider the fixed system OX 1 X 2 X 3 and we adopt the rotation axis the X 3 -axis,
(see Figure 4.26) and the mobile system Ox1 x2 x3 attached to the body.
X 3 , x3
X1
x1
= = 3e 3 = 3E3
r
x2
- angular velocity of the system x
X2

system OX 1 X 2 X 3 => orthonormal base (E1 , E 2 , E 3 )

system Ox1 x2 x3 => orthonormal base (e1 , e 2 , e 3 )
Figure 4.26
If the body is free of forces the governing equations becomes:
r
F =0
and
r
&
= HO
r
&
where H O can be calculated by means of
r
r
DH O
&
HO
Dt
r
r
D H Ox
=
r Dt
OX
r
r
r r r D H Ox
+ H Ox =
Dt
r
Ox
r
r
r
+ H Ox
r
Ox
The angular momentum:

r
r
r
r
r
r
r
r
components
H Ox = I Ox
( H Ox )i = (I Ox ) ij ( ) j
r
r
( H Ox )1 IO11
I O12 I O13 0 I O133
r r

IO 22
I O 23 0 = I O 233
( H Ox ) 2 = I O12
r
( H r ) I
IO 33 3 I O 333
I O 23
O13

Ox 3
And its rate of change:

Ciudad Real - Spain
Draft
357
r
&
( H r ) I
&
O13 3
r Ox 1
& r
&
( H Ox ) 2 = I O 233
r
&
( H r ) I
&
Ox 3 O 33 3
r
r
r
And we need to calculate the vector H Ox :
e1
0
e2
0
IO133
I O 233
r
r
r
H Ox =
e3
2
2
3 = I O 233 e1 I O133 e 2
I O 333
thus
Applying
2
2
&
&
I O133 I O 233 I O 233 IO133
r

&
2
2
&
&
( H O ) i = I O 233 + I O133 = I O133 I O 233
I

&
0
IO 333
O 33 & 3

r
r
&
M O = H O we get the following set of equations:
M
M
M
O1
O2
O3
M
M
M
X
Y
Z
2
&
= I O 233 IO133
2
&
= IO133 I O 233
&
= I O 333
&
where = stands for angular acceleration.
NOTE: If the body is prismatic and if we adopt the prismatic axis the same as the rotating
axis the above equations reduce to:
M
M
M
O1
O2
O3
M
M
M
=0
=0
&
= IO 33
since the system Ox1 x2 x3 is principal axes of inertia, e.g.

z
x
IO1
I Gxyz =
IO 2
IO1 = IO 2
IO 3

Ciudad Real - Spain
Draft
358
Problem 4.35
A rigid body consists of two masses m at each extremity of the weightless rod whose
length is 2l . The rod is inclined about respect to the vertical line and rotates with
angular velocity as indicated in Figure 4.27.
a) Find the angular momentum of the body;
r
b) Find the torque ( M ) in order to maintain the rotation.
l
m
Figure 4.27
Solution: We apply the governing equations for a rigid solid motion, (see Problem 4.26). We
will adopt a fixed system OXYZ at the space and a mobile system Oxyz which is attached
to the body, (see Figure 4.28.
Y
= J
l
X
Zz
l
m
x
Figure 4.28
The inertia tensor I (system Oxyz ) is given by:

Ciudad Real - Spain
Draft
IOxyz
0
0
0 2ml 2
=
0
0
359
0
0
2ml 2
The angular velocity (system Oxyz ):

r
= cos() + sin() + 0k
i
j
r
where is the module of .

r
The angular momentum H O :

r
r
HO = I
0
0 cos()
0
H Ox 0
H = 0 2ml 2
sin() = 2ml 2 sin()
Oy
0

H Oz 0

0
2ml 2
0
0
r
2
H O = 0 + 2ml sin() + 0k
i
j
r
The torque M can be evaluated as follows:
r r
r
r r
&
&
M = H O = ( H O ) Oxyz + H O
r
r
r r
&
We can observe that (H O ) Oxyz = 0 and = hold, then:
i
j
k
r r
r r
&
M = H O = H O = cos()
sin()
0
0
2ml 2 sin() 0
r r
&
M = H O = cos()2ml 2 sin()k = 2 ml 2 sin(2)k
= cos() + sin() + 0k
i
j
r
H O = 0 + 2ml 2 sin() + 0k
i
j
r
HO
Zz
l
m
x
Figure 4.29

Ciudad Real - Spain
Draft
360
Solution using the system OXYZ

The transformation matrix from OXYZ to Oxyz is given by:
T
cos 2 sin 2 0
sin () cos() 0
sin cos 0 = cos() sin () 0

A=
0
1
0
0
1
The inertia tensor for the system OXYZ is:

I OXYZ = A T I Oxyz A
I OXYZ
0
sin () cos() 0 0
cos( ) sin ( ) 0 0 2ml 2
=
0
0
1 0
0
0 sin () cos() 0
0 cos() sin () 0
2ml 2 0
0
1
2ml 2 cos 2 ( )
2ml 2 sin ( ) cos()
0
= 2ml 2 sin ( ) cos()

2ml 2 sin 2 ()
0
0
0
2ml 2
The angular momentum becomes:

r
r
H OXYZ = A T H Oxyz
0
H Ox
H = 2ml 2 sin()
Oy
H Oz
2
0
H OX sin () cos() 0
2ml cos() sin()
H OY = cos() sin () 0 2ml 2 sin() = 2ml 2 sin 2 ()
H OZ 0

0
1
0
0
The torque:
M = 2 ml 2 sin(2)K
Rigid Solid Motion Reference

Ferdinand P. BEER; E. Russell JOHNSTON (1987). Vector Mechanics for Engineers: Dynamics.
Seventh Edition. 2 Volumes. McGraw-Hill Science/Engineering/Math; 4 edition.
Ferdinand P. BEER; E. Russell JOHNSTON & William E CLAUSEN (2004). Instructor's and
Solutions Manual to Accompany Vector Mechanics for Engineers - Dynamics. Seventh Edition. 2
Volumes. McGraw Hill Higher Education; Seventh edition (2004)

Ciudad Real - Spain
Draft
5 Introduction to
Constitutive Equations
and IBVP
5.1 Solved Problems
Problem 5.1
Describe the constitutive equation and the free variables for simple thermoelastic materials
when we are considering the Helmholtz free energy .
Solution:
The constitutive equations for a simple material are in function of the following free
variables:
Constitutive equation for energy = ( F , T )
( F , T )
F
( F , T )
Constitutive equation for entropy ( F , T ) =
T
r
r
r
Constitutive equation for heat conduction q0 = q0 ( F , T , X T )
Constitutive equation for stress P( F , T ) = 0
r
The free variables are F -deformation gradient, T -temperature, X T -temperature
gradient. The constitutive equations can also be expressed as follows
= (E ,T )
( E , T )
S = 0
E
;
( E , T )
(E ,T ) =
T
r
r
r
q0 = q0 ( E , T , X T )
= (F , T )
( F , T ) T
F
=
F
( F , T )
(F , T ) =
T
r
1 r
r
q = J q0 ( F , T , X T ) F T
r
r
= J 1F q0 ( F , T , X T )
Problem 5.2
Consider an elastic material in which the energy density (per unit volume) is known and is
given by:
354
1
2
2
( I E , II E ) = ( + 2 )I E 2 II E
where and are material constants. I E = I E (E ) and II E = II E (E ) are, respectively, the

first and second principal invariants of the Green-Lagrange strain tensor. Obtain the
constitutive equations for this problem. Obtain also the explicit expression for the
constitutive equations in terms of , , I E and II E .
Formulary
I E = I E ( E ) = Tr ( E )
II E = II E ( E ) =
1
( TrE ) 2 Tr ( E 2 )
2
I E
=1
E
II E
= Tr ( E )1 E T
E
Solution:
According to the problem, the energy is only a function of the Green-Lagrange strain
tensor. We know that the general expressions for the constitutive equations for a simple
thermoelastic material are:
= (E ,T )
( E , T )
S = 0
E
( E , T )
(E ,T ) =
T
r
r
r
q0 = q0 ( E , T , X T )
Considering the expression of the given energy, we can conclude that the problem is
independent of temperature, since the energy expression is not a function of temperature.
Then, the remaining constitutive equation is the one related to stress, i.e.:
S = 0
( E ) ( I E , II E ) ( I E , II E ) I E ( I E , II E ) II E
=
=
+
E
E
I E
E
II E
E
= ( + 2 )I E (1) + ( 2 ) Tr ( E )1 E T
2
Simplifying the above equation, and taking into account that E T = E , I E = Tr (E ) , we

obtain:
S = I E 1 + 2 E
Problem 5.3
Consider the specific Gibbs free energy G(S, T ) = ( E , T )
S : E as constitutive
equation for energy for thermoelastic materials. Obtain the remaining constitutive
equations for thermoelastic materials, based on the principle that G(S, T ) does not depend
on the temperature gradient.
Solution:
Ciudad Real Spain
Draft
5 INTRODUCTION TO CONSTITUTIVE EQUATIONS AND IBVP
355
We start from the Clausius-Duhem inequality in terms of specific Helmholtz free energy in
the reference configuration:
1r
& &
&
r
S : E 0 + T q0 X T 0
T
(5.1)
Taking into account the specific Gibbs free energy we obtain the rate of change:
1 &
1
&
&
&
G(S, T ) = ( E , T )
S:E
S:E
1 &
1
&
&
&
( E , T ) = G(S, T ) +
S:E +
S:E
and by replacing the above equation into the inequality (5.1) we obtain:
&
1 &
1
&
& & 1r
r
S : E 0 G(S , T ) +
S:E+
S : E + T q 0 X T 0
0
0
T
1r
&
&
&
r
0G(S , T ) S : E 0T q0 X T 0
T
(5.2)
&
&
Note that S : E = E : S holds. The above inequality suggests that for a variation of Gibbs
free energy we must have the following relationships: Strain for variation of stress,
Entropy for a variation of temperature, and heat conduction in terms of temperature
gradient.
&
The term G(S, T ) can also be expressed as follows:
DG(S, T ) &
G(S, T ) & G(S , T ) &
T
G(S, T ) =
:S +
T
Dt
S
and by replacing the above equation into the equation in (5.2) we obtain:
1r
&
&
&
r
0G(S, T ) E : S 0T q0 X T 0
T
1r
G(S, T ) &
G(S, T ) &
&
&
r
:S 0
T E : S 0T q0 X T 0
0
S
T
T
G(S, T )
&
G(S , T )
& 1 r
r
0
+ E : S 0
+ T q0 X T 0
T
S
(5.3)
The above inequality must be satisfied for any admissible thermodynamic process. Let us
r
r
&
now consider the process such that T = 0 (isothermal process), and q 0 = 0 (adiabatic
process), then the above entropy inequality becomes:
G(S, T )
&
0
+ E:S 0
S
(5.4)
Note that the above inequality must also be met for any thermodynamic process. Then if in
the current process the condition in (5.4) is met, we can apply another process such that
&
&
S = S , in which the entropy inequality is violated. Thus, the only way in which the
inequality in (5.4) is satisfied is when:
G(S, T )
+E =0
S
E = 0
G(S, T )
S
Then if we take into account the above equation into the inequality (5.3), we obtain:
University of Castilla- La Mancha

Ciudad Real Spain
Draft
356
& 1 r
G(S, T )
&
G(S, T )
r
0
+ E :S 0
+ T q0 X T 0
S
T
(5.5)
G(S, T )
& 1 r
rT 0
0
+ T q0 X
T
T
r
r
Now let us consider a process where X T = 0 (a uniform temperature field), then the
inequality becomes:
G(S , T )
&
0
+ T 0
T
&
&
Starting from this point, we could apply another process where T = T , in which the
entropy inequality is violated. Thus, the only way in which the above inequality is satisfied
is when:
G(S, T )
+=0
T
G(S, T )
T
Then, the constitutive equations are:

Constitutive equation for energy G(S, T )
G(S , T ) g(S, T )
=
S
S
G(S, T )
Constitutive equation for entropy =
T
r
r
r
Constitutive equation for heat conduction q0 = q0 ( X T )
Constitutive equation for strain E = 0
(5.6)
where g(S, T ) = 0 G(S, T ) . Note that the free variables are (S , T ) .

Problem 5.4
Show that for an isothermal adiabatic process and with no rate of stress the specific Gibbs
free energy cannot increase.
Solution:
We start directly from the equation in (5.3):
1r
&
&
&
r
0G(S, T ) E : S 0T q0 X T 0
T
(5.7)
r
&
Taking into account the isothermal adiabatic process we have T = 0 , q0 = 0 , and with no
&
rate of stress the equation S = 0 holds. With that the inequality in (5.7) becomes:
&
0G(S, T ) 0
(5.8)
&
since 0 > 0 to satisfy the above inequality G(S, T ) 0 must hold.

Ciudad Real Spain
Draft
357
Problem 5.5
Find the governing equations for a continuum solid which has the following features:
Isothermal and adiabatic processes; an infinitesimal strain regime and a linear elastic
relationship between stress and strain.
b) Once the linear elastic, stress-strain relationship has been established, find the equation
in which ( ) is a tensor-valued isotropic tensor function.
Solution:
When we have isothermal and adiabatic processes temperature and entropy play no role.
In an infinitesimal strain regime, the following is satisfied:
r
Strain tensors: E e = sym u

Stress tensors: P S
F 1
r
r
X x
0 . If we take this approach, mass density is no
longer unknown.
Then, taking into account the fundamental equations:
The Fundamental Equations of Continuum Mechanics
(Current configuration)
The Mass Continuity Equation
(The principle of conservation of mass)
The Equations of Motion
(The principle of conservation of linear
momentum)
Cauchy Stress Tensor symmetry
(The principle of conservation of angular
momentum)
The Energy Equation
(The principle of conservation of energy)
The Entropy Inequality
(The principle of irreversibility)
r
D
r
+ ( x v ) = 0
Dt
(5.9)
r
r
&
r
x + b = v
(5.10)
= T
(5.11)
r
r
&
u = : D x q + r
1
T
(5.12)
1
T
&
&
( x , t ) + : D u
1 r r
q xT 0
T2
(5.13)
the remaining equations for the proposed problem are:

1) The equations of motion
2) The energy equation
r
r
&
+ b = v
&
r
&
0 u ( X , t ) = S : E X q0 + 0 r ( X , t )
or in terms of the Helmholtz free energy:
&
&
u = :
Du D
&
[ + T ] = :
=
Dt Dt
& &
&
= e = :
where e is the energy density (also known as strain energy density). Then if we bear in mind
the entropy inequality, we can observe that the proposed problem is characterized by a
process without any energy dissipation (an elastic process), i.e. all stored energy caused by
will recover when = 0 .
Ciudad Real Spain
Draft
358
3) For this problem, the constitutive equations described in Problem 5.1 become:
= ( )
S=
( ) e ( )
=
= ( )
Energy ( ) and stress are only functions of strain. Then, if we calculate the rate of change
( ) &
&
of the Helmholtz free energy, i.e. ( ) =
: , and by substituting it with the equation
& &
&
= e = : , we obtain:
&
( ) & e ( ) &
&
: =
: = :
e ( )
Thus, we can conclude that the energy equation is a redundant one, i.e. if the stress is
known the energy can be evaluated and vice-versa. So, we can summarize the governing
equations for the problem proposed with:
The equations of motion:
r
r
r
&&
&
+ b = v = u (3 equations)
The constitutive equations for stress:

( ) =
Kinematic equations:
(5.14)
e ( )
(6 equations)
r
= symu (6 equations)
The unknowns of the proposed problem are: (6), u (3) and (6), making a total of 15
unknowns and 15 equations, so the problem is well-posed. Then, to achieve the unique
solution of the set of partial differential equations given in (5.14) one must introduce the
initial and boundary conditions, hence defining the Initial Boundary Value Problem (IBVP) for
the linear elasticity problem. The initial and boundary conditions for this problem are:
The displacement boundary condition, on S u :
r r
r r
u( x , t ) =u* ( x , t )
r
r
ui ( x, t ) = u i * ( x, t )
(5.15)
r
jk n k = t j * ( x, t )
(5.16)
r
r
u i ( x , t = 0) = u 0 i ( x )
r
&
u 0 i ( x) = v0 i
(5.17)
The stress boundary condition, on S :

r r
r
( x , t ) n = t * ( x , n, t )
The initial conditions ( t = 0 ):

r r
r
u( x , t = 0) = u0
r r
r r
r r
u0 ( x , t )
&
= u0 ( x , t ) = v 0 ( x )
t
t =0
In the particular case when we have a static or quasi-static problem, the equations of
r r
motion become the equilibrium equations ( + b = 0 ), and the initial conditions
become redundant.

Ciudad Real Spain
Draft
Su
dV
359
r r
t * ( x)
r r
b( x )
Figure 5.1: Solid under external actions.

In subsection 1.6.1 The Tensor Series (Chapter 1-textbook), we saw that we can approach a
tensor-valued tensor function by means of the following series:
1
1 ( 0 )
1
2 ( 0 )
( 0 ) +
: ( 0 ) + ( 0 ) :
: ( 0 ) + L
0!
1!
2!

1
( 0 )
2 ( 0 )
0 +
: ( 0 ) + ( 0 ) :
: ( 0 ) + L
2
( )
Then, by considering the application point 0 = 0 and ( 0 ) = 0 = 0 , and also taking

into account that the relationship - is linear, higher order terms can be discarded, thus:
( ) =
( 0 )
2 e ( 0 )
: =
: = Ce :

where C e =
ij =
ij
kl
kl =
2 e ( 0 )
e
kl = Cijkl kl
ij kl
2 e ( )
is a symmetric fourth-order tensor which is known as the elasticity

tensor, which contains the material mechanical properties.

Note that, the energy equation has to be quadratic with which we can guarantee that the
relationship - is linear, since ( ) =
e ( )
. We can also use series expansion to
represent the strain energy density as follows:
2 e ( 0 )
1 e ( 0 )
1
: ( 0 ) + ( 0 ) :
: ( 0 ) + L

1!
2!
2 e ( 0 )
1
: ( 0 ) + L
= e + 0 : ( 0 ) + ( 0 ) :
0

2
1 2 e ( 0 )
1
: = : Ce :
= :
2

2
1
0!
e ( ) = e ( 0 ) +
where we have also considered that 0 = 0 e = 0, 0 = 0 .

0
NOTE 1: Although the energy equation is a redundant one, at the time of establishing an
analytical or numerical method to solve the problem, we will always start from energy
principles, hence the importance of studying the energy equation in a system.
NOTE 2: Analyzing C e :
e
Note that, according to the equation ij = C ijkl kl and due to the symmetry of ij = ji
e
e
and kl = lk , the tensor C e has minor symmetry, i.e. C ijkl = C ejikl = C ijlk = C ejilk . Note also
that:

Ciudad Real Spain
Draft
360
Ce =
ijkl
2 e ( ) 2 e ( )
=
= C e (major symmetry)
klij
ij kl
kl ij
NOTE 3: To better illustration of the problem established here, let us consider a particular
case (a one-dimensional case) where the stress and strain components are given by:
0 0
ij = 0 0 0
0 0 0
0 0
e
ij = 0 0 0 11 = C111111 = E
0 0 0
In this case, the stress-strain linear relationship becomes = E (Hookes law) and the
1
2
1
2
strain energy density is given by e = = E , y
e ( )
2 e
=
= E , (see Figure 5.2).

Current state
()
1
2
e = E
Stored energy
1
e =
2
E
0 = 0
e =0
0
0 = 0
Figure 5.2: Stress-strain relationship (one-dimensional case).

NOTE 4: Here it should be pointed out that in the case of elastic processes the
constitutive equation ( ) is only dependent on the current value of , i.e. it is
independent of the deformation history.
b) The tensor-valued tensor function ( ) is isotropic if the following is satisfied:
e (kl ) = e ( kl )
(kl ) = ij (kl )
ij
Then, taking into account that the relationship - is given by ij ( ) = Ce kl (indicial

ijkl
notation), we can conclude that:
(kl ) = ij (kl )
ij
e
Ce kl = Cijkl kl
ijkl
Ce = Ce
ijkl
ijkl
That is, the fourth-order tensor C e is isotropic. An isotropic symmetric fourth-order

e
tensor has the form C ijkl = ij kl + ( ik jl + il jk ) or C e = 1 1 + 2 I , (see Chapter
1), and here the parameters and are known as Lam constants. As we have seen in
Chapter 1, a symmetric isotropic fourth-order tensor is a function of two variables ( , ).
In Chapter 7 of the textbook (Linear Elasticity) we will see that it is possible to express C e
in terms of other parameters, e.g. ( E , ), ( , G ), where E is the Youngs modulus, is the
Poissons ratio, is the bulk modulus, and G = is the shear modulus. The physical meanings
of these parameters are given in the same chapter.

Ciudad Real Spain
Draft
361
NOTE 5: Figure 5.3 shows the stress-strain relationship for an isotropic material. Note
that, for an isotropic linear elastic material in an infinitesimal strain regime the constitutive
equation for stress becomes () = (1 1 + 2 I) : = Tr()1 + 2 :
e ( )
( ) =
144 44
2 3
linear
isotropic

( ) = C e :
( ) = Tr ( )1 + 2
Elastic
It should be emphasized here that due to the fact that the C e -components are independent
of the coordinate system, the tensors and share the same principal space
(eigenvectors), (see Figure 5.3).
22
12
= C e kl
ij
ijkl
11
x1
P
= a ip a jq pq
ij
22
12
= a ip a jq pq
ij
11
22
22
12
12
11
x1
22
Ce
ijkl
11
= C e kl
ij
ijkl
22
Principal space
Isotropic material
Ce
ijkl
11
11
Ce
ijkl
e (kl ) = e ( kl )
x1
Figure 5.3: Stress-strain relationship (isotropic material).

Ciudad Real Spain
Draft
362
NOTE 6: We denote the complementary strain energy density by e ( ) which is a function of

, (see Figure 5.4), and is given by:
2 e ( 0 )
1 e ( 0 )
1
: ( 0 ) + ( 0 ) :
: ( 0 ) + L

1!
2!
2 e ( 0 )
1
: ( 0 ) + L
= e + 0 : ( 0 ) + ( 0 ) :
0

2
1
2 e ( 0 )
1
1
1
: = : De : = : Ce :
= :

2
2
2
1
0!
e ( ) = e ( 0 ) +
Note that if we are dealing with linear elastic material e ( ) = e () holds, and
=
e ( )
.
a) Linear elastic material.
1
Complementary strain energy density - e () = E 1
2
()
e ( ) = e ( )
Stored energy
1
Strain energy density - e () = E
2
E
0 = 0
0 = 0
b) Non-linear elastic material.

Complementary strain energy density - e ()
()
Strain energy density - e ()
() ()
e
0 = 0
0 = 0
Figure 5.4: Complementary strain energy density (one dimensional case).

NOTE 7: Note that e () = e () tensorial e ( ) = g = : e ( ) = 0G( ) ,
where g( ) = 0 G( ) is the Gibbs free energy density (per unit volume) with reversed
sign, (see equations in (5.6) in Problem 5.2).

Ciudad Real Spain
Draft
363
NOTE 8: Taking into account the constitutive equation for stress for an isotropic linear
elastic material ( ) = Tr ( )1 + 2 and considering the additive decomposition of the
tensor into a spherical and deviatoric parts, i.e. = sph + dev =
Tr ( )
1 + dev , we can
3
obtain:
Tr ( )
( ) = Tr ( )1 + 2 = Tr ( )1 + 2 ( sph + dev ) = Tr ( )1 + 2
1 + dev
3
dev
= +
Tr ( )1 + 2
3
= Tr ( )1 + 2 dev
= sph + dev
m
33
23
13
13
11
23
12
12
13
11
23
12
dev
33
23
13
13
22
12
dev
22
12
dev
dev
ij = 2 ij
m
23
23
dev
11
Tr ( ) ij = 3 Tr ( ) ij
33
23
13
13
22
ij = Tr ( ) ij + 2 ij
13
dev
33
23
12
dev
22
12
dev
11
Figure 5.5: Additive decomposition of the constitutive equation.

Recall that, if we are dealing with small deformation regime, the volume ratio (dilatation) is
given by:
v
dV dV0 V
=
= 11 + 22 + 33 = Tr ( ) = I
dV0
dV0
And if we take the trace of ( ) = +
2
dev
Tr ( )1 + 2 we get:
3
Draft
Ciudad Real Spain
364
dev
: 1 = +
Tr ( )1 : 1 + 2 : 1
3
Tr ( ) = 3 +
Tr ( )
3
2
Tr ( )
= m = +
v
3
3
where we have considered that 1 : 1 = 3 and Tr ( dev ) = 0 . If we are dealing with a

compression stress state ( p > 0 ) we have:
0
p 0
0 p 0
ij =
0
0 p
3 m = Tr ( ) = 3 p < 0
p = +
v = v
3
For these reason, the parameter is called bulk modulus (see Figure 5.6), and is given by:
=+
2
3
(5.18)
Just as the spherical part of the tensor ( sph = Tr ( )1 ) is associated with the volume
change, the deviatoric part ( dev = 2 dev ) is associated with the shape change, and the
parameter = G defines the stiffness to the shape change, where G is known as shear
modulus or transversal elastic modulus.
p
p
1
p
Figure 5.6: Bulk modulus.

NOTE 9: In the laboratory the parameters (, ) are not the more appropriated to be
obtained. Next we try to rewrite the constitutive equation in terms of other parameters.
Recall that the reverse form of the constitutive equation ( ) = Tr ( )1 + 2 was
obtained in Problem 1.97 which is:
=
1
1
Tr ( )1 indicial ij =
ij
(11 + 22 + 33 ) ij
2
2 ( 2 + 3 )
2
2 ( 2 + 3 )
11
21
13
12
22
23
13
11
= 1
23
21
2
13
33

Ciudad Real Spain
12
22
23
Draft
13
1 0 0
(11 + 22 + 33 ) 0 1 0
23
2 (2 + 3 )
0 0 1
33
365
Notice also that the normal stress components 11 , 22 , and 33 only produce normal
strain components. Les us consider that we only have the normal stress 11 , 22 = 0 ,
33 = 0 , with that the normal strain components are:
11 =
( + )
1
11
(11 ) 11 =
2 2 (2 + 3 ) 11 = (2 + 3 ) 11
2
2 (2 + 3 )
22 =
(11 ) 22 =
11
2 (2 + 3 )
2 (2 + 3 )
33 =
(11 ) 33 =
11
2 (2 + 3 )
2 (2 + 3 )
As expected, due to the material isotropy, the influence of 11 upon 22 and 33 is the
same. From the equation of 11 we can obtain:
( + )
11 =
(2 + 3 ) 11
where we have denoted by E =
11 =
(3 + 2 )
11
( + )
11 = E11
(3 + 2 )
, which is known as Youngs modulus, or
( + )
longitudinal elastic modulus. We can also obtain:

22 =
(3 + 2 )
11 = 11
11 =
11 =
2 (2 + 3 )
2 (2 + 3 ) ( + )
2( + )
33 =
(3 + 2 )
11 = 11
11 =
11 =
2 (2 + 3 )
2 (2 + 3 ) ( + )
2( + )
where we have denoted by =

Note that =
2( + )
2( + )
, which is known as Poissons ratio.
2
and if we replace it into the equation of E we can
(1 2 )
obtain:
2
3
(1 2 ) + 2
(3 + 2 )
=
E=
=
( + )
2
(1 2 ) +
3
(1 2 ) + 2

=
2
(1 2 ) + 1

+ 2
(1 2 )
+ 1
(1 2 )
6 + 2(1 2 )
(1 2 )
= 2(1 + )
=
2 + (1 2 )
(1 2 )
thus:
E
2(1 + )
and
E
2
=
(1 2 ) (1 + )(1 2 )
We can also express the bulk modulus in function of ( E , ) :

Ciudad Real Spain
Draft
366
=+
E
E + 2 E (1 2 )
2
2
E
E (1 + )
E
=
+
=
=
=
3
(1 + )(1 2 ) 3 [2(1 + )] 3(1 + )(1 2 ) 3(1 + )(1 2 ) 3(1 2 )
So, we can obtain the relationships between these mechanical properties:

G =
f ( E; )
GE
9G 3E
G (E 2G )
3G E
9G
3 + G
E 2G
2G
3 2G
2(3 + G )
G (3 + 2G )
+G
+ G
2G (1 + )
2G (1 + )
3(1 2 )
2( + G )
f (G; )
f (G; )
f (G; )
f (G; E )
E=
2G
1 2
E
3(1 2 )
(9 3E )
9 E
E
(1 + )(1 2 )
9 ( )
3
3(1 2 )
3
1+
3E
9 E
E
2(1 + )
f ( E; )
3( )
2
3(1 2 )
2(1 + )
f ( ; )
f ( ; )
2
3
2G
3
3 E
6
We leave the reader to show:

Tensorial notation
Indicial notation
ij = kk ij + 2 ij
( ) = Tr ( )1 + 2
E
E
Tr ( )1 +
(1 + )(1 2)
(1 + )
=
Tr ( )1 + 2
3
ij =
E
E
kk ij +
ij
(1 + )(1 2)
(1 + )
ij =
kk ij + 2 ij
3
(5.19)
(5.20)
(5.21)
and
Tensorial notation
Indicial notation
1
Tr ( )1
2
2 (3 + 2 )
ij =
1
kk ij
ij
2
2 (3 + 2 )
(5.22)
1+
Tr ( )1 +
E
E
ij =
1+
kk ij +
ij
E
E
(5.23)
2 3
1
ij =
18 kk ij 1 2 ij
(5.24)
2 3
1
=
18 Tr ( )1 2

Ciudad Real Spain
Draft
367
and that the elasticity tensor can be written as follows:

C e = 1 1 + 2 I
E
E
1 1 +
I
(1 + )(1 2 )
(1 + )
Elasticity tensor
1
C e = 1 1 + 2 I 1 1
3
Ce =
(5.25)
and:
Ce
Ce
Ce
(1 + )
1 1 +
I
E
E
1
1 1 +
I
De =
2 (3 + 2 )
2
De =
De =
1
1
1
1 1 +
I 3 1 1
9
2

Ciudad Real Spain
Elastic compliance tensor
Draft
368
Problem 5.6
In tensile testing we evaluated the following points:
Point
1
2
3
4
5
( Pa ) (10 3 )
6.67
13.3
20
24
22
0.667
1.33
2
3
3 .6
Calculate Youngs modulus ( E ) and define the stress-strain curve limit points.
Solution: First, we verify that the first three points maintain the same proportionalities:
E=
(1) ( 2 ) ( 3)
20
= ( 2 ) = ( 3) =
= 10 000 Pa = 10 kPa
(1)
2 10 3
The stress-strain curve can be appreciated in Figure 5.7, in which we define the following
points: e - the proportionality point; Y - the yield point; u - the ultimate strength
point; and r - the rupture strength point.
(Pa ) 30
u
25
20
15
3; 24
3.6; 22
2; 20
1.33; 13.3
10
0.667; 6.67
0; 0
0
0 . 2%
0.5
1.5
2.5
3.5
4
3
(10 )
Figure 5.7: Stress-strain curve.

Problem 5.7
Write in Voigt notation: a.1) the strain energy density and, a.2) the constitutive equations in
stress for an isotropic linear elastic material: a.2.1) in terms of ( , ) and, a.2.2) in terms
of ( E , ) where =
E
E
and =
. b) Expressing the strain tensor in
(1 + )(1 2 )
2(1 + )
[ ] {u } where
Voigt notation such as { } = L(1)
[ ].
{u } is the displacement field, obtain the
(1)
matrix L
c) Write the equations of motion in Voigt notation.

Ciudad Real Spain
Draft
369
Solution:
a.1) The strain energy density can be expressed as follows:
1
2
1
2
1
2
1
2
e ( ) = : C e : = : = : = ij ij
where we have used = C e : . Note that
ij ij = 1 j 1 j + 2 j 2 j + 3 j 3 j
1 3
2
123
1 3
2
1111
21 21
3131
+
+
+
1212
22 22
3232
+
+
+
1313
23 23
3333
thus
1
2
1
2
e () = ij ij = (1111 + 22 22 + 33 33 + 212 12 + 2 23 23 + 21313 )

and
1
2
e ( ) = ij ij =
1
[11 22
2
33
12
23
11

22
33 1
13 ]
T
= { } { }
212 2
2 23
213
Then, the tensors and in Voigt notation are stored as follows:

11

22

{ } = 33 ;
12
23
13
11

22

{ } = 33
212
2 23
213
a.2.1) The constitutive equation for stress in Voigt notation is:
11 + 2

+ 2
22
33
+ 2
= C e : Voigt =
0
0
12 0
23 0
0
0

0
0
13 0

0
0
0
0
0
0
0
0 11
0 22
0 33
{ } = [C ] { }
0 212
0 2 23
213
(5.27)
More detail about this formulation is provided in Problem 1.97 in Chapter 1 where we
have also obtained

Ciudad Real Spain
Draft
370
Tr ( )1
2
2 ( 2 + 3 )
and
( 2 + 3 )
11
2 ( 2 + 3 )
22
33 2 ( 2 + 3 )
=
2 12
2 23
2 13
0
a.2.2) Note that
2 ( 2 + 3 )
2 ( 2 + 3 )
+
2 ( 2 + 3 )
( 2 + 3 )
+
2 ( 2 + 3 )
( 2 + 3 )
0
0
0 11
22
0
33

0 12
23
0 13
0
1
{ } = [C ] 1 { }
(5.28)
E
E
E
+2
=
(1 )
(1 + )(1 2)
2(1 + ) (1 + )(1 2)
E
( )
=
(1 + )(1 2 )
(1 2 )
E
E
=
=
2(1 + ) (1 + )(1 2)
2
+ 2 =
then, the equation (5.27) can be rewritten as follows:
(1 )

11
(1 )
(1 )
22
33
E
0
0
0
=
12 (1 + )(1 2)
0
23
0
0
13
0
0
0
0
0
(1 2)
2
0
0
0 11
0 22

0 33
2 12
0 2 23
(1 2) 2 13
2
0
(1 2 )
2
(5.29)
Note that

Ciudad Real Spain
Draft
+ =
371
E
E
E
+
=
(1 + )(1 2) 2(1 + ) 2(1 + )(1 2)
(2 + 3 ) =
E
E
E
E2
+3
2
=
2(1 + ) 2(1 + )
(1 + )(1 2 ) 2(1 + )(1 2)
+
E
2(1 + )(1 2) 1
=
=
E
(2 + 3 ) 2(1 + )(1 2)
E2
(1 + )(1 2 )
E
=
=
E
2 (2 + 3 ) (1 + )(1 2)
E2
1
2(1 + ) 1
= 2(1 + )
E
E
Then, the equation (5.28) becomes:

11
1

22
33 1
=
2 12 E 0
0
2 23
2 13
0
0
0
1
0
0
11
0
0
0 22
0
0
0 33
2(1 + )
0
0 12
0
2(1 + )
0 23
0
0
2(1 + ) 13
(5.30)
b) According to the definition ij = 1 (u i , j + u j ,i ) we get:

2
11
ij = 21
31
12
22
32
u1
x1
13
1 u
u
23 = 1 + 2
x
2
x1
33 2
u
u
1
1 + 3
x
2 3 x1
u1

x1
x1
u 2
0
11

x 2

22 u 3
0
33
x 3
{ } = = u u =

212 1 + 2
2 23 x 2 x1 x 2
u
u
2 13 2 + 3 0
x3 x 2
u1 u 3
x + x x
3
1
3
x 2
0
x1
x3
0
1 u1 u 2
+
2 x 2 x1
u 2
x 2
1 u 2 u 3
+
2 x3 x 2

u1
x3
u 2
0 u
3

x 2

x1
1 u1 u 3
+
2 x3 x1
1 u 2 u 3
+
2 x3 x 2
u 3
x3
{ } = [L(1) ]{u }
NOTE: If we adopt the engineering notation, i.e. x1 = x , x 2 = y , x3 = z , u1 = u , u 2 = v ,

u 3 = w , 11 = x , 22 = y , 33 = z , 2 12 = xy , 2 23 = yz , 2 13 = xz , the above equation
becomes:

Ciudad Real Spain
Draft
372
u
x x
v
0
11 x
y
22 y w
33 z z 0
{ } = = = u v =
2 12 xy
+
2 23 yz y x y
v w
2 13 xz +
0

z y
u + w
z x z
0
y
u
0
z v { } = L(1) {u }
(5.31)

0 w

x
z y
0
x
r
r
r
&&
&
c) Let us consider the equations of motion + b = v = u (see equation (5.14)) in
&&
indicial notation ij , j + b i = u i and its explicit form:
0
[ ]
&&
ij , j + b i = i1,1 + i 2, 2 + i 3,3 + b i = u i
&&
11,1 + 12, 2 + 13,3 + b1 = u1
21,1 + 22, 2 + 23,3 + b 2 = && 2

u
&&
31,1 + 32, 2 + 33,3 + b 3 = u 3
11 12 13
+
+
+ b1 = &&1
u
x1
x 2
x3
21 22 23
&&
+
+
+ b2 = u2
x1
x 2
x 3
31 32 33
&&
+
+
+ b 3 = u3
x1
x3
x 2
Then, if we consider the stress tensor in Voigt notation, the above set of equations
becomes:

x1
0
x 2
x 3
x 2
x1
0
x3
x 2
&
[ ] { } + { b } = { u&}
L(1)

11
x 3 22 b u
&&
33 1 1

&&
0 + b 2 = u 2

12 b u
&&
3 3
23
x1
13

(5.32)
Problem 5.8
Consider an isotropic homogeneous linear elastic material described in Problem 5.5.
Obtain the governing equation so as to result in a system of three equations and three
unknowns, namely: u1 , u 2 , u 3 , (Displacement Formulation).

Ciudad Real Spain
Draft
373
Solution:
As seen in Problem 5.5, the governing equations, for an isotropic homogeneous linear
elastic material in small deformation regime, are:
Tensorial notation
Indicial notation
r
r
r
&&
&
&&
ij , j + b i = u i (3 equations)

ij = kk ij + 2 ij (6 equations)
( ) = Tr ( )1 + 2 (6 equations)
The kinematic equations:
(5.33)
r
= sym u (6 equations)
ij =
1 u i u j
+
2 x j x i
(6 equations)
which results in a system with 15 equations and 15 unknowns.

The divergence of the Cauchy stress tensor ( ) can be obtained by means of the
constitutive equations for stress, i.e.:
ij = kk ij + 2 ij
ij , j = ( kk ij + 2 ij ) , j
ij , j = , j kk ij + kk , j ij + kk ij , j + 2 , j ij + 2 ij , j
{
{
{
=0 j
= 0i
(5.34)
=0 j
ij , j = kk , j ij + 2 ij , j
ij , j = kk ,i + 2 ij , j
If the mechanical properties and are constants throughout the material, i.e. if they do
r
not vary with x (homogeneous material) we get , j
= 0 j and , j
= 0 j . We
x j
x j
can also express the terms kk ,i and ij, j in terms of displacements. For this, we use the
kinematic equations:
ij =
1 u i u j
+
2 x j
xi
1
u i , j + u j ,i divergence ij , j = 1 u i , jj + u j ,ij

2
2
Note that
2ui
=
x j x j x j
u j ,ij
kk =
2u j
x j xi
u i
x j
r
r
r
u i , jj [ (u)]i 2 u (Laplacian of the vector u )
i
[ ]
2u j
x i x j
1 u k u k
+
2 x k
x k
x i
u j
x j
r
u j , ji [ ( u)]i
u k
gradient
=
x u k ,k kk ,i = u k ,ki = u j , ji
k
With that the equation in (5.34) can be rewritten as:

Ciudad Real Spain
Draft
374
ij , j = kk ,i + 2 ij , j = u j , ji + 2
1
u i , jj + u j , ji = ( + )u j , ji + u i , jj
2
&&
By replacing the above equation into ij , j + b i = u i (equations of motion), we obtain:
&&
ij , j + b i = u i
&&
( + )u j , ji + u i , jj + b i = u i
Thus resulting in 3 equations and 3 unknowns ( u1 , u 2 , u 3 ):

&&
( + )u j , ji + ui , jj + bi = ui
r
r
r
r
&&
( + )[ ( u)] + [ (u)] + b = u
Naviers equations
(5.35)
NOTE 1: The above equations are known as the Naviers equations also known as NavierLam equations. The explicit form of the equation (5.35) is presented as follows:
&&
( + )u j , ji + u i , jj + b i = ( + )(u1,1i + u 2, 2i + u 3,3i ) + (u i ,11 + u i , 22 + u i ,33 ) + b i = u i
&&
( + )(u1,11 + u 2, 21 + u 3,31 ) + (u1,11 + u1, 22 + u1,33 ) + b1 = u1
&&
( + )(u1,12 + u 2, 22 + u 3,32 ) + (u 2,11 + u 2, 22 + u 2,33 ) + b 2 = u 2
&&
( + )(u1,13 + u 2, 23 + u 3,33 ) + (u 3,11 + u 3, 22 + u 3,33 ) + b 3 = u 3
or:
2u
2 u1 2 u1
u1 u 2 u 3
&&
+ b1 = u1
+ 21 +
+
+
+
( + )
2
2
x
x1 x1 x 2 x3
x3
x 2
2
2
2
u2 u2 u2
u1 u 2 u 3
&&
( + )
x + x + x + x 2 + x 2 + x 2 + b 2 = u 2
x 2 1
2
3
2
3
2
2
2
( + ) u1 + u 2 + u 3 + u 3 + u 3 + u 3 + b = u
&& 3
3
2
2
x 2
x 3 x1 x 2 x 3
x 2
x3
NOTE 2: We have proven in Problem 1.105 (Chapter 1) that the following is true:
r
r r
r
r
( a) = ( a) 2 a
indicial
ilq qjk a k , jl = a j , ji a i , jj
Then, we can obtain
r
r r
r
r
r
(u) 2 u = ( u) ( u)
indicial
u i , jj = u j , ji ilq qjk u k , jl
with which the equation (5.35) can also be written as follows:

&&
( + )u j , ji + ui , jj + bi = ui
&&
( + )u j , ji + (u j , ji ilq qjk u k , jl ) + bi = ui
&&
( + 2 )u j , ji ilq qjk uk , jl + bi = ui
and the equivalent in tensorial notation:
r
r
r
r
&&
( + )[ ( u)] + [ (u)] + b = u
r
r r
r
r r
r
&&
( + )[ ( u)] + ( u) ( u) + b = u
r
r
r r
r
r
&&
( + 2 )[ ( u)] ( u) + b = u

Ciudad Real Spain
Draft
r
r
r r
r
r
&&
( + 2 )[ ( u)] ( u) + b = u
&&
( + 2 )u j , ji ilq qjk uk , jl + bi = ui
375
(5.36)
In the Cartesian System we have:

r
u = u i e i = u1 e1 + u 2 e 2 + u 3 e 3
r r
r
r
u
u
u
u
u
u
( u) rot (u) = (rot (u) )i e i = 3 2 e1 + 1 3 e 2 + 2 1 e 3

x
x
x 2
x1
x3
2
3
1
4 244
4 244
4 244
14 r 3
14 r 3
14 r 3
r
r
r
r r (rot (u) )3 (rot (u) )2
( u) =
x 2
x 3
= (rot (u) )1
= (rot (u) )2
r
r
(rot (u) )1 (rot (u) )3

1 +
e
x
x1
3
= (rot (u) )3
r
r
(rot (u) )2 (rot (u) )1

2 +
e
e 3
x1
x 2
r
r
(rot (u) )3 (rot (u) )2 u 2 u1 u1 u 3
x x
x 2
x3
2 1 x 2 x3 x3 x1
r
r
(rot (u) )
r
r r
(rot (u) )3 u 3 u 2 u 2 u1
( u) i =
=
x1
x 3
x 3 x 2 x 3 x1 x1 x 2
r
r
(rot (u) )
(rot (u) )1
2
u1 u 3 u 3 u 2
x1 x3 x1 x 2 x 2 x3
x1
x 2
NOTE 3: If we are dealing with heterogeneous material, the equations in (5.34) become:
ij = kk ij + 2 ij
ij , j = ( kk ij + 2 ij ) , j
ij , j = ( kk ) , j ij + (2 ij ) , j = ( kk ) ,i + ( 2 ij ) , j
Taking into account that 2 ij = u i , j + u j ,i and kk = u k ,k , the above equation becomes:

ij , j = ( kk ) ,i + (2 ij ) , j
ij , j = (u k , k ) ,i + (u i , j + u j ,i ) , j
whereby
ij , j + b i = && i
u
(u k ,k ) ,i + (u i , j + u j ,i ) , j + b i = && i
u
(5.37)
Note that
r
r
u k , k = Tr (u) = ( u)
&
&
&
&
&
&
&
, and its components
u
u
u
u
Du i u i u i
&&
=
+
ui =
v j = i + i v1 + i v 2 + i v3
t
x j
t
x1
x 2
x3
Dt
&
&
&
&
u1 u1
u
u
+
v1 + 1 v 2 + 1 v3
x1
x 2
x3
t
u 2 u 2
&
&
&2
&2
u
u
&&
+
v1 +
v2 +
v3
ui =
x1
x 2
x 3
t
u
& 3 u 3
&
&3
&
u
u
+
v1 +
v 2 + 3 v3
x1
x 2
x3

Ciudad Real Spain
Draft
376
[ (u
i, j
+ u j ,i ) , j =
=
(u i , j + u j ,i )
x j
(u i ,1 + u1,i ) +
(u i , 2 + u 2,i ) +
(u i ,3 + u 3,i )
x1
x 2
x3
2 (u1,1 ) +
(u1, 2 + u 2,1 ) +
(u1,3 + u 3,1 )
x 2
x3
x1
2 (u 2, 2 ) +
(u i , j + u j ,i ) , j =
(u 2,1 + u1, 2 ) +
(u 2,3 + u 3, 2 )
x 2
x3
x1
2 (u 3,3 )
(u 3,1 + u1,3 ) +
(u 3, 2 + u 2,3 ) +
x1
x 2
x3
The three equations in (5.37), ( i = 1,2,3 ), are explicitly given by:
&&
2 (u1,1 ) +
(u1, 2 + u2,1 ) +
(u1,3 + u3,1 ) + b1 = u1
[( u)] +
x1
x2
x3
x1

r
&&
[( u)]+ (u2,1 + u1,2 ) + 2 (u2,2 ) + (u2,3 + u3,2 ) + b 2 = u2
x1
x2
x3
x2

r
&&
[( u)] + (u3,1 + u1,3 ) + (u3,2 + u2,3 ) + 2 (u3,3 ) + b3 = u3
x1
x2
x3
x3
or
r

&&
( u) + 2 (u1,1 ) +
(u1, 2 + u2,1 ) +
(u1,3 + u3,1 ) + b1 = u1
x1
x2
x3
&&
( u) + 2 (u2, 2 ) +
(u2,1 + u1, 2 ) +
(u2,3 + u3, 2 ) + b 2 = u2
x2
x1
x3
( u) + 2 (u3,3 ) +
(u3,1 + u1,3 ) +
(u3, 2 + u2,3 ) + b3 = &&3
u
x3
x1
x2
NOTE 4: Wave equations

If we apply the divergence to the equation (5.36) we obtain:
r
r
r r
r
r
&&
( + 2 ) [ ( u)] ( u) + b = u
144 44
2
3
=0
r
r
r
&&
( + 2 ) [ ( u)] + b = u
r
r
r
&&
( + 2 ) 2 ( u) + b = u
r
r
r
&& ( + 2 ) 2 ( u) + b
u =
r
(5.38)
r
r
( + 2 ) 2
D
( u) =
( u) + b
2
Dt
2
r
D ( + 2 ) 2
=
+ b
Dt 2
r
D 2
= 2 2 + b
Dt 2
r
&
& = 2 2 + b
r r
r
where we considered that = u and ( v ) = 0 (see Problem 1.106). In indicial
notation becomes:
Ciudad Real Spain
Draft
377
&&
( + 2 )u j , jii ilq qjk u k , jli + bi ,i = ui ,i
&&
( + 2 )u j , jii + b i ,i = ui ,i
&&
ui ,i =
D2
2
Dt
( + 2 )
ui
x
i
u j , jii + bi ,i
( + 2 ) 2
=
xi xi
u j
x j
(5.39)
b i
+
xi
D 2
2
b
=2
+ i
2
xi xi xi
Dt
where
=
( + 2 )
P-wave velocity
(5.40)
r
If the body forces do not change in space we have that b = 0 , thus the equation in
(5.38) becomes:
D 2
= 2 2
Dt 2
P- wave equation
(5.41)
P-waves have no rotation.

r
Now if we apply the curl ( ) to the equation (5.36) we obtain:
r
r
r
r r
r r
r &&
r
r
( + 2 ) [ ( u)] ( u) + b = u
r
r
r r
r r
r &&
r
( u) + b = u
r
r
r r
r r
D2 r r
( u) + b = 2 ( u)
Dt
2r
r
r
r r
r
D
( ) + b =
Dt 2
r
r
r r
D 2
(5.42)
( ) =
Dt 2
r
r
r r
D 2
= ( )
Dt 2
r
r r r
D 2
= ( )
Dt 2
2r
r
r r
D
= 2 ( )
2
Dt
r
r r r
r r r
where we considered = u , and that the b -field is conservative thus b = 0 . Note
r
r
r
r
that [ ( u)] = [ ] = 0 (see Problem 1.106), and
[
[
]
]
Shear wave velocity
(5.43)
r
r
r r
r
r r
r r
r
r
r
2 = ( ) ( ) 2 = ( ) , since ( u) = 0
r
r r
r
( ) = ( ( u)) = 0 (see Problem 1.106). With that the equation in (5.42)
Note
that
becomes:

Ciudad Real Spain
Draft
378
r
r
D 2
= 2 2
2
Dt
Shear wave equation

(S-wave equation)
(5.44)
Shear waves have no change in volume.

In the case of = 0 the equation (5.41) becomes the acoustic wave equations:
D 2
= c 2 2
2
Dt
Acoustic wave equation
(5.45)
Speed of propagation
(5.46)
with
c=
Note that the displacement field was split up into: u = + where = 0 . We can
r
r r
r
r
prove this by means of the identity ( a) = ( a) 2a . If we consider the vectors
r
r r
r r
r
r
r
r
u = 2a and = a , and the scalar = a , we obtain u = + , with that we
obtain:
r r
r r r
r
r r
r
r r
r
u = + ( ) = and u = + ( ) = ( )
a)
b)
Figure 5.8: Displacement occurring from a harmonic plane P-wave (a) and S-wave (b). Pwave has no rotation and S-wave no volume change.
If we consider =
E
E
and =
we can obtain:
(1 + )(1 2)
2(1 + )
( + 2 )
E
E
+2
( + 2 )
(2 2 )
(1 + )(1 2)
2(1 + )
=
=
=
E
(1 2)
2(1 + )
With that we conclude that the ratio of P- to S-wave velocities depends only on Poissons
ratio.

Ciudad Real Spain
Draft
379
Problem 5.9
a) Show that:
r
r
( )T = 0
qjk til ij ,kl = 0 qt
(5.47)
where ijk is the permutation symbol.

b) Show also that:
ij , kl + kl ,ij il , jk jk ,il = O ijkl
(5.48)
c) Express the explicit form of the equations (5.47).

Solution:
u
u
1
1
j
The infinitesimal strain tensor is given by ij =
+ i = (u + u i , j ) , and if we
xi x j 2 j ,i
2
take the derivative with respect to ( x ) we get:

ij
x k
ij ,k =
1
(u j ,ik + u i , jk )
2
Note that u i , jk = u i ,kj is symmetric in jk , and if we multiply by the antisymmetric tensor in

jk , i.e. qjk = qkj , we get: u i , jk qjk = 0 iq , thus
1
2
1
2
qjk ij ,k = (u j ,ik + u i , jk ) qjk = u j ,ik qjk

r
once again we take the derivative with respect to ( x ) and we get:

( qjk ij ,k )
xl
1
= qjk ij ,kl = u j ,ikl qjk
2
Note that u j ,ikl = u j ,kil = u j ,kli is symmetric in il and til = tli is antisymmetic in il . With
that, if we multiply both sides of the equation by til we obtain the equation in (5.47), i.e.:
1
2
til qjk ij ,kl = u j ,ikl til qjk = 0 jkt qjk = 0 qt

b) Now, if we multiply both sides of the above equation by tab qmn , we obtain:
tab qmn til qjk ij ,kl = 0 qt tab qmn = O abmn
Remember that the relationships tab til = ai bl al bi and qmn qjk = mj nk mk nj

hold, thus:
tab qmn til qjk ij ,kl = O abmn
( ai bl al bi )( mj nk mk nj ) ij , kl = O abmn
( ai bl mj nk ai bl mk nj al bi mj nk + al bi mk nj ) ij ,kl = O abmn
am ,nb an,mb bm ,na + bn,ma = O abmn

am,bn + bn,am an,mb mb,an = O ambn
Q.E.D.

Ciudad Real Spain
Draft
380
Note that, if we multiply the above equation by bn we get:

am, bn bn + bn, am bn an, mb bn mb, an bn = O ambn bn
am, bb + bb, am ab, bm mb, ba = 0 ambb
r
r
r
r
r
r
r
r
[ x ( x )]am + [ x [ x [Tr ( )]]]am [ x ( x )]am [ x ( x )]ma = 0 ambb
r
r
r
r
r
r
r
r
[ x ( x )]am + [ x [ x [Tr ( )]]]am = [ x ( x )]am + [ x ( x )]ma
[ ]
r
r
r
r
r
r
r
2 am + [ x [ x [Tr ( )]]]am = [ x ( x )]am + [ x ( x )]ma
x
which in tensorial notation becomes:

r
r
r
r
r
r
r
r
x ( x ) + x [ x [Tr ( )]] = x ( x ) + [ x ( x )]
c) Note that in (5.47) we have 6 independent equations since 0 qt is symmetric.

For the case when q = 1, t = 1 we get 1 jk 1il ij ,kl and by expanding the index l we obtain:
1 jk 1il ij ,kl = 1 jk 1i1 ij , k1 + 1 jk 1i 2 ij ,k 2 + 1 jk 1i 3 ij ,k 3 = 1 jk 1i 2 ij ,k 2 + 1 jk 1i 3 ij ,k 3
Expanding the index i the above equation becomes:
1 jk 1il ij ,kl = 1 jk 1i 2 ij ,k 2 + 1 jk 1i 3 ij ,k 3 = 1 jk 132 3 j , k 2 + 1 jk 123 2 j ,k 3 = 1 jk 3 j , k 2 + 1 jk 2 j ,k 3
and by expanding the remaining indices we obtain:
1 jk 1il ij ,kl = 1 jk 3 j ,k 2 + 1 jk 2 j , k 3 = 123 32,32 132 33, 22 + 123 22,33 + 132 23, 23
= 32,32 + 33, 22 + 22,33 23, 23 = 33, 22 + 22,33 2 23, 23 = 0
=
2 33
2
x 2
2 22
2
x3
2 23
=0
x 2 x3
note that 23, 23 = 32,32 .

We leave the reader with the following demonstrations:
when q = 2, t = 2
2 jk 2il ij ,kl = 31,31 + 33,11 + 11,33 13,13 = 33,11 + 11,33 213,13 = 0

=
2 33
x12
2 11
2
x3
2 13
=0
x1x3
when q = 3, t = 3
3 jk 3il ij ,kl = 11, 22 12,12 21, 21 + 22,11 = 11, 22 + 22,11 212,12 = 0

=
2 11
2
x 2
2 22
x12
2 12
=0
x1 x 2
when q = 1, t = 2
1 jk 2il ij , kl = 12,33 + 13, 23 + 32,31 33, 21 = 13, 23 + 23,13 33,12 12,33 = 0

=
23 13 12 2 33
2 12
2 33
2 23
2 13
=0
+
=
+
x2 x3 x1x2
x2 x3 x1x3 x1x2 x3x3 x3 x1
when q = 2, t = 3

Ciudad Real Spain
Draft
381
2 jk 3il ij ,kl = 11,32 + 13,12 + 21,31 23,11 = 13,12 + 12,13 23,11 11, 23 = 0
=
2 23
2 13
2 11
2 12
+
x1 x 2 x1 x3 x1x1 x 2 x 3 x1
13 12 23
x + x x
3
1
2
2 11
x x = 0
2
3
and when q = 1, t = 3
1 jk 3il ij ,kl = 12,32 13, 22 22,31 + 23, 21 = 12, 23 13, 22 22,13 + 23,12 = 0
=
2 23
2 13
2 22
2 12
=
+
x 2 x3 x 2 x 2 x1x 3 x1x 2 x 2
12 13 23
x x + x
2
1
3
2 22
x x = 0
1
3
By regrouping the 6 equations we get:
2 33 2 22
2 23
S11 =
+
2
=0
2
2
x 2 x3
x 2
x3
2
2
2
S 22 = 33 + 11 2 13 = 0
2
x1 x3
x12
x3
2 11 2 22
2 12
S 33 =
+
2
=0
2
x1 x 2
x 2
x12
2
S = 23 + 13 12 33 = 0
12
x3 x1
x 2
x3 x1x 2
23 13 12 2 11
=0
+
+
S 23 =
x1 x1
x 2
x 3 x 2 x3
2
S = 23 13 + 12 22 = 0
13
x 2 x1
x 2
x3 x1 x3
Compatibility equations for 3D
(5.49)
The above equations in Voigt notation become:
2
S11 x 2
3
S
2
22
2

S 33 x 2
=

S12 0
S 23

2
S13

x 2 x 3
2
2
x3
0
2
x12
2
2
x 2
2
x12
0
2
x1x 2
2
x1x3
2
x 2 x3
2
x1x 2
2
1 2
2
x3
2
1
2
x1x3
2
1
2
x 2 x3
0
2
x1x3
2
1 2
2
x1
2
1
2
x1x 2
1
2
{S } = [L( 2) ] { } = {0}
x1x3 11 0

22 0
0
0
33 =

2
212 0
2
x 2 x 3 2 23 0

2 2 13 0
1
2
x1 x 2
2
1 2
2
x 2
0
(5.50)
NOTE 1: The equations in (5.49) are known as the compatibility equations. The
compatibility equations guarantee that the displacement field is unique and continuous (see
Figure 5.9). In other words, the 6 components of the strain tensor are not independent and
cannot be arbitrary.

Ciudad Real Spain
Draft
382
6
7
The compatibility equations are

not satisfied
1
4
7
The compatibility
equations are satisfied
3
6
9
2
5
8
Figure 5.9
NOTE 2: When using numerical method for obtaining the solution, e.g. finite element
method, the way to ensure the compatibility equations is by means of the continuity of the
displacement field. With regards the finite element method, when we assembly the
elements (tie nodes) we are ensuring that the compatibility equations are satisfied.
r
NOTE 3: When the displacement field is independent of one direction, i.e. u = u( x1 , x 2 ) ,

the compatibility equations reduce to:
S 33 =
2 11
2
x 2
2 22
x12
2 12
=0
x1x 2
Compatibility
equations for 2D
(5.51)
since i 3 = 3i = 0 .
NOTE 4: To understand the compatibility condition let us consider an example in two
dimensional case (2D), where we have the scalar field = ( x1 , x2 ) and we know the
following derivatives:
= x1 + 3x2 and
= x12 , we see clearly that this scalar field is
x1
x2
incompatible since
= x1 + 3 x2 = F1
x1
= x12 = F2
x2
x2

2
( x1 + 3 x2 )
=
=3
x x x =
x2
2 1
1
2
( x12 )

=
=
= 2 x1
x2
x1 x2 x1x2
The scalar field = ( x1 , x2 ) will be compatible if and only if:
= F1 ( x1 , x2 )
x1
compatible iff F1 F2
x2 x1
= F2 ( x1 , x2 )
x2
(5.52)
If we consider the Greens theorem (see Chapter 1 in the textbook) which states:
r r
r
r
F
F
components
F d = (
r
x
F) e 3 dS F1dx1 + F2 dx2 = 2 1 dS3
x1 x2

Ciudad Real Spain
Draft
383
r
and also considering the equation in (5.52), we conclude that: if F = x , is compatible
r r
r
r r
r
r
r
r
if and only if F d = ( x F) e 3dS = 0 x F = 0 .
dS = dSe 3
x2
x3
e3
x1
Figure 5.10: Greens theorem.

r
r
r
NOTE 5: Let us consider that F = ( x ) a = a ( x )T , where is a second-order
r
r
tensor field and a is an arbitrary vector independent of x (constant). Note also that the
following relations are true:
r
r r
r
r r
r
r
r
r r
r
r
r
(a) F d = ( x ) a d = a ( x )T d = a ( x )T d
(b)
and
r
r
r
r
( x F) dS =
{ [(
r
x
r
x
]}
r
r
) a dS =
r
r r
r
r
= a x ( x )T
r
{ [a (
r
x
r
r
} dS = a {
r
r
x
r
x
)T
r
]} dS
r
r
( x )T
} dS
T
In indicial notation becomes

r
r
r
r
r
r
r
r
r
r
(a) Fi (d )i = ( x )ij a j (d )i = a j ( x )ij (d )i = a j ( x )ij (d )i
(b)
r
r
x
[ [
r
r
r
r
r
F) i (dS ) i = ijk Fk , j (dS )i = ijk a p ( x )T
]]
kp , j
r
(dS ) i
r
r
r
r
r
r
r
r
= ijk a p , j ( x )T kp + a p ( x )T kp, j (dS ) i = ijk a p ( x )T kp , j (dS ) i
{
=0
r
r
r
r
T
r
= a p ijk ( x ) kp, j (dS ) i = a p ijk psq qk ,s , j (dS ) i = a p ijk psq qk ,sj (dS )i
= ap
r
= a
r
[
r
x
r
r
( x )T
r
x
r
r
( x )T
} dS
ip
T
r
(dS )i
r
It would be worth reviewing the Problem 1.108, where we have shown that the following
r
r
r
r
relationship ( x ) = ksq qp , s e k e p holds, thus ( x )T = psq qk , s e k e p . Also in

Ciudad Real Spain
Draft
384
r
r
Problem 1.108 we shown that x ( x )T = ipq tsj qj , ps e t e i , which is equivalent to
r
r
r
r
x ( x )T = psq ijk qk ,sj e i e p .
And by consider the Stokes Theorem (see Chapter 1 in the textbook) we conclude:
r r
r
r
r
F d = (
r
x
F ) dS
14444 4444
2
3
r r
r
r
r
r
r
r
r
a ( x )T d = a x ( x )T
r
r
r
r
x ( x ) T
r
r
r
( x )T d =
} dS
} dS
Then, for a compatible field it must fulfill:
{
r
r
x
r
r
( x )T
=0
r
r
r
r
x ( x )T = 0
r
x
X
Now let us consider that A = F a where F is the gradient deformation, F = r , and by

apply the Stokes theorem we get:
r r
r
r
r
r
A d = ( x A ) dS
r
r
r
r
r
r
( F a) d = ( x ( F a)) dS
144444444444444 44444444444444
2
3
r r
r
a F T d = a
r
r
x F
r
dS
F T d =
r
{
r
x
} dS
T
Then, for a compatible field it must fulfill:
{
r
r
x
=0
r
r
x F = 0
More detail about these algebraic manipulations is provided in Problem 1.108.

Problem 5.10
a) Show that the fundamental equations for an isotropic linear elastic material (see
equations in (5.33)), can be replaced by six equations and six unknowns ( ij ), (Stress
Formulation), i.e.:
Indicial notation
ij , kk +
2( + )
&&
kk ,ij
ll , kk ij = 2 ( ui ), j
(2 + 3 )
(2 + 3 )
Tensorial notation
r
2 +
x
sym
2 ( bi ), j
sym
(5.53)
r sym
r
2( + ) r r
&& sym
r
r
r
x [ x [Tr ( )]]
2 [Tr ( )]1 = 2 x ( u) 2 x ( b)
x
(2 + 3)
(2 + 3)
r
r
r
r
r
r
where 2 = x ( x ) and 2 [Tr ( )] = x [ x [Tr ( )]] .
x
x

Ciudad Real Spain
Draft
385
b) or by:
Indicial notation
ij , kk +
2( + )
&&
kk ,ij =
u
( b k ), k ( &&k ), k ij + 2 ( ui ), j
(2 + 3 )
(2 + )
Tensorial notation
r
2 +
x
2 ( bi ) , j
sym
] [
sym
r
r sym
r
r
2( + ) r r
&&
&& sym
r
r
r
x [ x [Tr ( )]] =
x ( b) ( u) 1 + 2 x ( u) 2 x ( b)
( 2 + 3)
(2 + )
(5.54)
E
E
and =
, express the equations (5.53) and
(1 + )(1 2 )
2(1 + )
(5.54) in function of ( E , ) .
r
Hint: The kinematic equations = sym u can be replaced by:
c) Considering that =
ij , kl + kl ,ij il , jk jk ,il = O ijkl
(5.55)
(see Problem 5.9).

Solution:
a) We obtain the inverse of the constitutive equation in stress ( = C e : ):
Ce
: = Ce
: C e : = I sym : = sym =
Remember that the tensor C e

=
= Ce
for isotropic material is given by:
1
1
Tr ( )1 indicial ij =
ij
ss ij .
2
2 (2 + 3 )
2
2 (2 + 3 )
r
If we consider a homogeneous material, the mechanical properties do not vary with x , i.e.
,i
= 0 i and ,i
= 0 i , then:
x i
xi
2 ij
1
ij , kl =
2 ij 2 (2 + 3 ) ss ij = 2 ij ,kl 2 (2 + 3 ) ss , kl ij
x k xl
,kl
(5.56)
Moreover, if we multiply the equation in (5.55) by jk we get:

ij , kl jk + kl ,ij jk il , jk jk jk ,il jk = O ijkl jk
ik ,kl + kl ,ik il , kk kk ,il = 0 il
(5.57)
Note that, according to equation (5.56) the following is true:

ik ,kl =
1
1
ik ,kl
ss ,kl ik =
ik ,kl
ss ,il
2
2 (2 + 3 )
2
2 (2 + 3 )
kl ,ik =
1
1
kl ,ik
ss ,ik kl =
lk ,ki
ss ,il
2
2 (2 + 3 )
2
2 (2 + 3 )
il ,kk =
1
il ,kk
ss ,kk il
2
2 (2 + 3 )

Ciudad Real Spain
Draft
386
kk ,il =
=
1
1
3
kk ,il
ss ,il kk =
{ 2 kk ,il 2 (2 + 3 ) ss ,il
2
2 (2 + 3 )
=3
2
1
3
3
ss ,il
ss ,il =
2 2 (2 + 3 ) ss ,il = 2 (2 + 3 ) ss ,il
2
2 (2 + 3 )
2 ij
x k x l
ij ,kl =
1
ij ,kl
ss ,kl ij
2
2 (2 + 3 )
With that the equation in (5.57) becomes:

ik ,kl + kl ,ik il ,kk kk ,il = 0 il
1
2
2
2
ik ,kl
ss ,il + lk ,ki il , kk +
ss ,kk il
ss ,il = 0 il
(2 + 3 )
(2 + 3 )
(2 + 3 )
2
2
ik ,kl
(2 + 3 ) + (2 + 3 ) ss ,il + lk , ki il ,kk + (2 + 3 ) ss , kk il = 0 il
2( + )
ik ,kl
ss ,il + lk , ki il ,kk +
ss , kk il = 0 il
(2 + 3 )
(2 + 3 )
2( + )
ss ,il il ,kk +
ss ,kk il = ik , kl lk ,ki
(2 + 3 )
(2 + 3 )
(5.58)
&&
From the equations of motion ij , j + b i = u i we can obtain:
&&
ij , jk + ( b i ) , k = ( u i ) ,k
with that the following is true:

&&
ik ,kl + ( b i ) ,l = ( u i ) ,l
&&
ik ,kl = ( b i ) ,l ( u i ) ,l
&&
lk , ki + ( b l ) ,i = ( u l ) ,i
&&
lk ,ki = ( b l ) ,i ( u l ) ,i .
&&
&&
&&
And note that ik ,kl lk , ki = ( b i ) ,l ( u i ) ,l + ( b l ) ,i ( u l ) ,i = 2[( b i ) ,l ]sym 2[( u i ) ,l ]sym
By replacing the above equation into the equation (5.58) we get:
2( + )
ss ,il il ,kk +
ss ,kk il = 2 ( b i ) ,l
(2 + 3 )
(2 + 3 )
2 ( b i ) , j
sym
&&
2 ( u i ) ,l
sym
Restructuring the above and considering that ( l = j ) we obtain:

ij ,kk +
2( + )
&&
kk ,ij
ll ,kk ij = 2 ( u i ) , j
(2 + 3 )
(2 + 3 )
sym
sym
Q.E.D.
which is the same as (5.53).

b) Starting from the above equation we get:
ij ,kk +
2( + )
&&
kk ,ij =
ll ,kk ij + 2 ( u i ) , j
(2 + 3 )
(2 + 3 )
sym
2 ( b i ) , j
sym
(5.59)
Our goal now is to obtain an expression for ll, kk . If we multiply equation (5.55) by jk li
we obtain:

Ciudad Real Spain
Draft
387
ij , kl jk li + kl ,ij jk li il , jk jk li jk ,il jk li = O ijkl jk li

ij , ji + ji ,ij ii , jj jj ,ii = 2 ij ,ij 2 ii , jj = 0
(5.60)
ij ,ij ii , jj = 0
If we use the inverse of the constitutive equation (see equation (5.56)), we can obtain:
ij ,ij =
ii , kk
1
1
ij ,ij
ss ,ij ij =
ij ,ij
ss ,ii
2
2 (2 + 3 )
2
2 (2 + 3 )
(5.61)
2
1
=
ii ,kk
ss ,kk ii =
2 (2 + 3 ) ii , kk
2
2 (2 + 3 )

ij ,ij ii , jj = 0
1
ij ,ij
ss ,ii
2 (2 + 3 ) ii ,kk = 0
2
2 (2 + 3 )
ij ,ij
(2 + 3 ) + (2 + 3 ) ii ,kk = 0
(5.62)
2 +
ij ,ij =
(2 + 3 ) ii , kk
&&
Now, by means of the equations of motion ij , j + b i = u i we can obtain:
&&
ij , ji + ( b i ) ,i = ( u i ) ,i
&&
ij , ji = ( u i ) ,i ( b i ) ,i

2 +
ij ,ij =
(2 + 3 ) ii ,kk
2 +
&&
( u i ) ,i ( b i ) ,i =
(2 + 3 ) ii ,kk
ii ,kk = ll ,kk =
(5.63)
(2 + 3 )
(2 + 3 )
&&
&&
( u k ) , k ( b k ) , k =
( b k ) , k ( u k ) , k
2 +
2 +
Replacing equation (5.63) into (5.59), we obtain:

ij ,kk +
2( + )
&&
kk ,ij =
ll ,kk ij + 2 ( ui ) , j
(2 + 3)
(2 + 3)
sym
2 ( b i ) , j
sym
ij ,kk +
2( + )
( 2 + 3)
&&
&&
kk ,ij =
( b k ) ,k ( uk ) ,k ij + 2 ( ui ), j
( 2 + 3)
(2 + 3) 2 +
ij ,kk +
2( + )
&&
&&
kk ,ij =
( b k ) ,k ( u k ) ,k ij + 2 ( ui ) , j
( 2 + 3)
(2 + )
sym
2 ( b i ) , j
Q.E.D.
2 ( b i ) , j
sym
sym
(5.64)
Thus obtaining the equation in (5.54)

c) After some algebraic manipulations we can obtain:
(1 2)
1
,
=
(2 + 3 )
E

Ciudad Real Spain
Draft
sym
388
(2 + 3 )
(2 + 3 )
(1 2 )
E
,
=
E
(1 + )(1 2) (1 + )
(1 2)
(1 2)
E
,
=
E
2(1 + ) 2(1 + )
2( + )
(1 2)
1
,
=2
+2
=
(2 + 3 )
(1 + )
2(1 + ) (1 + )
(2 + ) = 2
E (1 )
E
E
,
+
=
2(1 + ) (1 + )(1 2) (1 + )(1 2)
(1 + )(1 2)
E
,
=
=
(2 + ) (1 + )(1 2)
E (1 )
(1 )
whereby the equation (5.53) becomes:

ij , kk +
&&
kk ,ij
ll , kk ij = 2 ( ui ), j
(1 + )
(1 + )
Tensorial notation
r
2 +
x
sym
2 ( b i ) , j
sym
r sym
r
1
&& sym
r
r
r
r
r
x [ x [Tr ( )]]
2 [Tr ( )]1 = 2 x ( u) 2 x ( b)
x
(1 + )
(1 + )
(5.65)
and the equation (5.54) becomes:
ij , kk +
Tensorial notation
r
2 +
x
&&
&&
kk ,ij =
( bk ), k ( uk ), k ij + 2 ( ui ), j
(1 + )
(1 )
] [
sym
2 ( b i ), j
sym
r
r sym
r
r
1
&&
&& sym
r
r
r
r
r
x [ x [Tr ( )]] =
x ( b) ( u) 1 + 2 x ( u) 2 x ( b)
(1 + )
(1 )
(5.66)
NOTE 1: For a static problem the above equation becomes:
ij ,kk +
r
2
x
kk ,ij =
( b k ) , k ij 2 ( b i ) , j
(1 + )
(1 )
] [
sym
r
r sym
1
r
r
r
r
+
x [ x [Tr ( )]] =
x ( b ) 1 2 x ( b )
(1 + )
(1 )
Michells equations
(5.67)
which are known as Michells equations.

r
If the body forces do not vary with x , the Michells equations reduce to:
ij ,kk +
r
2
x
1
kk ,ij = 0 ij
(1 + )
1
r
r
+
x [ x [Tr ( )]] = 0
(1 + )
Beltramis equations
(5.68)
which are known as Beltramis equations.

Ciudad Real Spain
Draft
389
&&
NOTE 2: For a static problem (u k = 0 k ) , the equation in (5.63) becomes:
ll ,kk =
r
(1 + )
( b )
(1 )
r
(1 + )
2 [Tr ( )] =
( b)
(1 )
(2 + 3 )
(1 + )
( b k ) , k =
( b k ) , k
2 +
(1 )
{ [Tr ( )]} =
(5.69)
The above equation can also be obtained by means of equation (5.67) with ( i = j ), i.e.:
ii ,kk +
1
kk ,ii =
( b k ) , k ii 2 ( b i ) ,i
{
(1 + )
(1 )
=3
1
ii ,kk =
1 +
(1 ) 2 ( b k ) , k
(1 + )
( 2 + )
(2 + )
(1 + ) ii ,kk = (1 ) ( b k ) ,k
ii ,kk =
(1 + )
( b k ) , k
(1 )
]
(5.70)
Note that ii ,kk = kk ,ii and ( b k ) ,k = ( b i ) ,i .

Problem 5.11
a) Prove that:
Indicial notation
, iijj = 0
Tensorial notation
{ [ ( )]} = 0
(i, j = 1,2)
2 2 = 0
4
4 4
4 +2 2 2 + 4 =0
x1
x1 x2 x2
(5.71)
4 = 0
with
11 =
2
2
x 2
22 =
2
x12
12 = 21 =
2
x1 x 2
(5.72)
Consider a linear elastic material, a static problem, and no body forces. Consider also that
the Cauchy stress tensor is only dependent of x1 and x 2 , i.e. = ( x1 , x 2 ) .
b) Show whether the equilibrium are satisfied.
Solution:
a) In Problem 5.10 (see equation (5.70)) we have shown that:
ii ,kk =
(1 + )
( b k ) , k = 0
(1 )
where we have considered that ( b k ) ,k = 0 . For the proposed problem we have i, k = 1,2 ,
with which:

Ciudad Real Spain
Draft
390
ii ,kk = 0
ii ,11 + ii , 22 = 0
11,11 + 22,11 + 11, 22 + 22, 22 = 0
2 11
x12
2 22
x12
2 11
2
x 2
2 22
2
x 2
=0
Using the definition given by (5.72), we can conclude that:

2 11
x12
2 22
x12
2 11
2
x 2
2 22
2
x 2
=0
2 2 2 2 2 2 2 2
+ 2
+ 2
+ 2
=0
2
2
x12 x 2
x1 x12 x 2 x 2
x 2 x12
4
4
4
+2 2 2 + 4 =0
x14
x1 x 2
x 2
Q.E.D.
NOTE: In the literature, is known as the Airy stress function.

b) For the bidimensional case (2D), the equilibrium equations (without body forces) reduce
to:
ij , j = 0 i
i , j ,2)
(=1
i1,1 + i 2, 2 = 0 i
11 12
x + x = 0
1
2
21 + 22 = 0
x1
x 2
Using the definition (5.72), we can obtain:

11 12
x + x = 0
1
2
21 22
+
=0
x1
x 2
2
2
3
3
=0
2
2
2
x 2 x1 x 2 x1 x 2 x1 x 2
x1 x 2
2
2
3
3
+ = + = 0
x x x
x 2 x12
x12 x 2 x 2 x12
1
1
2
With this, we show that the expressions of stresses given by (5.72) satisfy the equilibrium
equations.

Ciudad Real Spain
Draft
391
Problem 5.12
Consider the governing equation for the elastic, linear problem. Obtain an equivalent
r
formulation such as the unknowns are displacement u and stress (Mixed Formulation).
Use Voigt notation.
Solution:
Taking into account the governing equations for the elastic, linear problem:
Tensorial notation
r
r
r
&&
&
Voigt notation
&
[L ] { } + { b } = { u&} (3 equations)
( ) = C e : (6 equations)
{ } = [C ]{ } (6 equations)
(1) T
(5.73)
{ } = [L(1) ] {u } (6 equations)
r
= sym u (6 equations)
where the equations in Voigt notation were obtained in Problem 5.7, where
[L ]
(1) T
x1
= 0
x 2
x 2
x1
0
0
x3
x3
x 2
x 3
0

x1
To eliminate the strain from the equations, we replace the kinematic equation into the
constitutive equations for stress, thus:
{ } = [C ]{ }
{ } = [C ] [L(1) ]{u }
1
1
[C ] { } = [C ] 4 ] [L(1) ]{ }
1 2[3
u
4 C
=[1 ]
[ ]{u } = {0 }
[C ] { } L
(1)
whereby the system (5.73) becomes:
[ ]
L(1) T { } + { b } = { u&}
&
1
(1)
u
[C ] { } L { } = {0 }
[ ]
It is also possible to write as follows:

[0 ]
(1)
L
[ ]
&
[L ] {u } = { b } + { u&}

(1) T
[C ]1 { }
{0 }
NOTE: The above formulation is known as Mixed Formulation. It is interesting to note that
in the formulations either in displacement or in stress, (see Problem 5.8 and Problem
5.10), we have second derivative of the unknowns, meanwhile in the mixed formulation we
deal only with the first derivative of the unknowns, and moreover this formulation does
not deal with the derivative of the mechanical properties.

Ciudad Real Spain
Draft
392
Problem 5.13
Let us consider two systems made up by the same linear elastic material but with different
load conditions as shown below:
r
System I
Surface force - t *
Specific body force - b

S
Su
r
u*
dV
r r
t * ( x)
r r
b( x )
Stress field -
Strain field -
System II
Su
r
u*
Displacement field - u
Surface force - t *
S
B
dV
r r
t * ( x)
r
b
Specific body force - b

Stress field -
Strain field -
Displacement field - u
Figure 5.11: Two systems under external actions.

Show the Bettis theorem also known as Bettis reciprocal theorem:
: dV = : dV
Bettis theorem
(5.74)
Solution:
Taking into account the constitutive equation for stress, = C e : , in indicial notation:
e
ij = C ijkl kl
And by multiplying both sides of the equation by the field we get:

ij ij = ij C e kl
ijkl
Major Simmetry
of C
e
ij ij = ij C ijkl kl = kl C e ij
klij
e
where we have applied the major symmetry of the elasticity tensor ( C ijkl = C e ). Since the
klij
both systems are made up by the same material the relationship = C e : holds. With that
the above equation becomes:
ij ij = ij C e kl = kl C e ij = kl kl
ijkl
klij
Tensorial notation
: = :
If now we integrate over the whole volume we get the Bettis theorem:
: dV = : dV

Ciudad Real Spain
Draft
(5.75)
393
e
NOTE 1: The above equation is only valid if C ijkl = C e holds, i.e. if C e has major
klij
e
symmetry. In other words, the condition C ijkl = C e enforces the existence of the storedklij
energy function, such as:
Ce =
ijkl
2 e
2 e
=
= Ce
klij
ij kl kl ij
NOTE 2: The Bettis theorem is the start point to obtain the formulation of the Boundary
Element Method.
NOTE 3: The Bettis theorem can also be expressed in another form which we show
below.
2 x j
1 u
Recall that ij = i +
ij =
u j 1
= (u i , j + u j ,i ) , which is also valid for the system II, i.e.
xi 2
1
( ui , j + u j ,i ) . Then:
2
ij ij dV
= ij ij dV
1
1
ij (u i , j + u j ,i )dV =
ij ( ui , j + u j ,i )dV
2V
2V
u
ij
i , j dV
(5.76)
= ij ui , j dV
where ij u i , j = ij u j ,i and ij ui , j = ij u j ,i hold due to the symmetry of and ,

respectively. Also note that:
( ij u i ), j = ij , j u i + ij u i , j
ij u i , j = ( ij u i ), j ij , j u i
( ij ui ), j = ij , j ui + ij ui , j
ij ui , j = ( ij ui ), j ij , j ui
u
ij
= ij ui , j dV
i , j dV
( u ),
( u ),
ij
ij , j u i dV = ( ij ui ), j ij , j ui dV
ij
(5.77)
dV ij , j u i dV = ( ij ui ), j dV ij , j ui dV
Applying the divergence theorem to the first one integral on both sides of the equation, we
get:
u n dS
ij
ij , j u i dV
= ij ui n j dS ij , j ui dV
S
(5.78)
t i u i dS ij , j u i dV = t i ui dS ij , j ui dV
S
where we have applied the definition n = t and n = t . The above equation in

tensorial notation becomes:
r
t

Ciudad Real Spain
udS ( ) udV = t u dS ( ) u dV
S
Draft
(5.79)
394
If we resort to the equations of motion is satisfied that:

r
r
&&
+ b = u
r
r
&&
+ b = u
r r
&&
= (b u )
r r
&&
= (b u)
with that the equation in (5.79) becomes:
r r r
r r
r r r
r r
&&
&&
t udS + (b u ) udV = t u dS + (b u) u dV
Bettis theorem
(5.80)
Note that, if we consider S = S u + S we have:
r r
t udS =
r r
t u dS =
r r
r r
t * udS + t u* dS
r r
r r
t * u dS + t u * dS
Su
(5.81)
Su
For the particular case when the system is in equilibrium and in the absence of body force,
the equation (5.80) becomes:
r r
r r
t udS = t u dS
(5.82)
In addition, if we have concentrated forces instead of surface force, we obtain:

r
r
r
r
F loc u loc = F loc u loc
Fi loc u loc = Filoc uiloc

i
(5.83)
Problem 5.14
Let us consider two systems as described in Figure 5.11. Show the Principal of Virtual
Work:
r r r
r r
&&
t * u dS + (b u) u dV = : dV
1444442444443
Total external virtual work
Principle of Virtual Work
V 4 4
1 2 3
(5.84)
Total internal
virtual work
with u = u* on S u .
Solution:
We can prove the Principle of Virtual Work by starting directly from the relationship:
ij ij dV
1
ij ( ui , j + u j ,i )dV = ij ui , j dV
2V
V
Note that ( ij ui ), j = ij , j ui + ij ui , j
ij ij dV
ij ui , j = ( ij ui ), j ij , j ui , thus:
= ij ui , j dV = ( ij ui ), j ij , j ui dV
V
ij ij dV = ( ij ui ), j dV ij , j ui dV
V

Ciudad Real Spain
(5.85)
(5.86)
Draft
395
by applying the divergence theorem to the first volume integral on the right side of the
equation, we get:
ij ij dV
= ( ij ui ), j dV ij , j ui dV = ij ui n j dS ij , j ui dV
(5.87)
= t * ui dS ij , j ui dV
i
S
where we have applied the definition n = t * . The above equation in tensorial notation
becomes:
r
r r
: dV = t * u dS ( ) u dV
&&
If we use the equations of motion we get + b = u
that the equation in (5.88) becomes:
(5.88)
r r
&&
= (b u) , with
r r r
r r
&&
t * u dS + (b u) u dV = : dV
14444 244444
4
3
V 4 4
1 2 3
Total internal
virtual work
which is known as the Principle of Virtual Work. Note that, for the demonstration, we
have not used the major symmetry of C e .
For the particular case when the system is in equilibrium and in the absence of body force,
the above equation becomes:
r r r
t * u ( x )dS = : dV
(5.89)
In addition, if we have concentrated forces instead of surface force, we obtain:

Tensorial notation
Voigt notation
r
r
F loc u loc = : dV
{F } {u }= { } { } dV
loc T
loc
(5.90)
where the direction of uiloc -component is the same as the Filoc -component direction,
where {F loc }= {F1 , F2 ,..., Fn }T , {u loc }= {U1 , U2 ,..., Un }T .
F2
F1
F3
u
u
,
REAL

Ciudad Real Spain
VIRTUAL
Draft
396
NOTE 1: The Principle of Virtual Work states: A structure is in equilibrium, under a

system of external forces, if and only if the total external virtual work equals the total internal
r
virtual work for every virtual displacement field ( u ).
NOTE 2: The Principle of Virtual Work is used to discretization techniques of the
problem such as the Finite Element Technique, in which the fundamental unknown is the
displacement.
NOTE 3: It is easy to show that the equation in (5.84) is also valid for rate of change of
r
& &
virtual fields u , , i.e.:
r r r
r r
&
&& &
&
t * u dS + (b u) u dV = : dV
S
V
1444442444443
V 4 4
1 2 3
(5.91)
Total internal
virtual work
Also it is valid for a variation of the virtual field u , i.e.:
r r
r
r
r
&&
t * u dS + (b u) u dV = : dV
144444 2444444
4
3
V 4 4
1 2 3
(5.92)
Total internal
virtual work
NOTE 4: We can also define the Principle of complementary virtual work as follows:
r r
r
&& r
r
t u* dSu + (b u ) udV =
r
Su
14444 244444
4
3
Total external complementary virtual work
: dV
Principle of Complementary
Virtual Work
V 4 4
1 2 3
Total internal
complementary virtual work
(5.93)
with n = t * on S .
NOTE 5: Note that, if we are using the Principle of Virtual Work the fundamental
unknowns are displacements (strains), if we are using the Principles of Complementary
Virtual Work the fundamental unknowns are forces (stresses), and if we are using the
Bettis reciprocal theorem the fundamental unknowns are displacements and forces
simultaneously (see equation (5.79)).
Problem 5.15
Consider a sub-domain ( ) made up by a homogeneous, isotropic linear elastic material.
Consider also that at some points of the sub-domain boundary there are concentrated
r
r
forces {F (e ) }, and that the displacement field is approximated by {u( x )} = [N ( x )]{ ( e ) }
u
where { (e ) } are the displacements at the points where concentrated forces are applied.
u
Prove that the fundamental equations for a linear elastic problem can be replaced by:
{F }= [K ]{u }
(e)
(e )
(e)
with
[K ] = [B] [C ][B] dV
T
(e)
(5.94)
where [C ] is the elasticity tensor in Voigt notation, and obtain an expression for [B( x )] .
Hint: Use the Principle of Virtual Work.
Solution:
Ciudad Real Spain
Draft
397
We can start directly from the equation in (5.90), which is equivalent to:
r
r r
F u = : dV = : ( sym u ) dV
(5.95)
r r
r
u F = ( sym u ) : dV
The above equation in Voigt notation becomes:

r r
r
u F = ( sym u ) : dV Voigt u (e )
{ } {F }= { } { } dV
(e)
(5.96)
Note that, the above equation is already considering the equilibrium equation (see equations
(5.88)-(5.90)). The constitutive equation in stress, in Voigt notation is given by
r
r
{ ( x )} = [C ]{ ( x )} , where the strain tensor field is given by the kinematic equations
r
r
( x ) = sym u . In Problem 5.7 we have obtained the symmetric part of the displacement
field gradient, ij = 1 (u i , j + u j ,i ) , in Voigt notation, i.e.:
2

u1

x1
x1
u 2
0
11

x 2

22 u 3
0
33
r
x3
{ ( x )} = = u u =

2 12 1 + 2
2 23 x 2 x1 x 2
u
u
2 13 2 + 3 0
x3 x 2
u1 u 3
x + x x
3
1
3
x 2
0
x1
x 3
0

u1
x3
u 2
0 u
3

x 2

x1
r
r
{ ( x )} = [L(1) ]{u( x )}
then
r
r
r
r
{ ( x )} = [L(1) ]{u( x )} = [L(1) ][N ( x )]{u (e) }= [B( x )] {u (e) }
where
r
r
[B( x )] = [L(1) ] [N ( x )]
(5.97)
Note that { (e ) } is independent of x . The stress field can be obtained as follows:

u
r
r
r
r
{ ( x )} = [C ]{ ( x )} = [C ] [B ( x )] {u (e ) }
We can adopt the same displacement field approach to approximate the virtual
displacement field, with which we can obtain:
r
r
{u ( x )} = [N ( x )]{u (e ) }
r
r
{ ( x )} = [B( x )] {u (e) }
Then, the equation in (5.96) becomes:

r
r
{u } {F }= { } {} dV = {[B( x)]{u }} [C ][B( x)]{u }dV
(e) T
(e)
(e)
(e)
or:

Ciudad Real Spain
Draft
398
r
r
{u } {F }= {u } [B( x)] [C ][B( x )]{u }dV
(e) T
(e ) T
(e)
(e)
(5.98)
Note that neither { (e ) } nor {u (e ) } are dependent of x , then:

u
r
r
r
{u } {F }= {u } [B( x)] [C ][B( x)] dV {u }
(e) T
(e) T
(e)
(e)
r T
r
F ( e ) = [B ( x )] [C ][B ( x )] dV u ( e )
{ }
u
{F } = [K ]{ }
(e)
(e)
(5.99)
{ }
(e)
[ ]
NOTE: K (e ) is known as the stiffness matrix of the sub-domain (finite element), and the
r
r
r
matrix [N ( x )] from the relationship {u( x )} = [N ( x )]{ ( e ) } is known as the matrix shape
u
function. The shape functions are functions defined into the domain that allows us to
r
obtain the value of an unknown function {u( x )} at any point of the domain through the
nodal values of the function { (e ) }. For example, let us suppose that the sub-domain ( )
u
r
is characterized by a quadrilateral then we can obtain {u( x )} by means of its nodal values
{u (e) } as follows:
U(e ) _ 2
U
(e ) _ 3
r
{u( x )}
(e ) _ 1
r
r
{u( x )} = [N ( x )]{u(e) }
U(e ) _ 4
U(e ) _ 1
(e ) _ 2
r U
= [N ( x )] (e ) _ 3
U
(e ) _ 4
Problem 5.16
For a equilibrium system let us consider the total potential energy :
r r
r r
r
1
(u) =
: dV t * udS ( b) udV
2
V
S
V
The total potential energy
(5.100)
where
U int =
2 : dV =
( ) dV
(5.101)
r r
r r
U ext = t * udS + ( b) udV
(5.102)
and
Also let us consider that the first variation of , denoted by , equals cero for a
r
stationary value of . Show that, if = 0 is equivalent to a stationary value , so (u)
assume a minimum value.

Ciudad Real Spain
Draft
399
r r
NOTE 1: Consider that during the deformation process, the external actions ( t * , b ) do
not vary, and also consider a linear elastic material.
Solution:
The first variation ( ) can be obtained as follows:
1
r r
r r
=
: dV t * udS ( b) udV
2
S
V
V
r r
r* r
1
= : dV t udS ( b) udV
2
V
S
V
(5.103)
r
r
r
r
1
=
( : ) dV t * udS ( b) udV
2
V
S
V
Note that:
1
1
1
e ( ) = ( : ) = ( : + : ) = (C e : ) : + :
2
2
2
1
1
1
= (C e : ) : + : = : C e : + : = [ : + : ]
2
2
2
= :
] [
(5.104)
e
:
where we have considered =
e
(see Problem 5.5). For small deformation regime we
can also write the above equation as follows:

e ( ) =
r
r
r
e
: = : = : ( symu) = : ( sym u) = : (u)
(5.105)
where we have used the property A sym : B = A sym : (B sym + B skew ) = A sym : B sym . Then, the
equation in (5.103) becomes:
r
r
r
r
1
( : ) dV t * udS ( b) udV
2
V
S
V
r
r* r
r
= : dV t udS ( b) udV
r
r
r
r
= e dV t * udS ( b) udV
The expression (u + u) can be obtained as follows (see equation (5.100)):

r r
r r
r
r
r
r
(u + u) = e ( + ) dV t * (u + u)dS ( b) (u + u)dV
(5.106)
By using the Taylor series to approach e ( + ) we get:
e ( + ) = e () +

Ciudad Real Spain
1
e ( )
2 e ( )
: + :
: + ...
2
Draft
(5.107)
400
Note that
2 e ( )
e ( )
: = e (see equation (5.104)) and C e =
(see Problem 5.5),

with which the equation in (5.107) becomes:
1
e ( )
2 e ( )
: + :
: + ...
2

1
e ( ) + e + : C e :
2
e ( + ) = e () +
and by replace the above equation into the equation (5.106) we obtain:
r r
r
r
r
r
r
r
(u + u) = e ( + ) dV t * (u + u)dS ( b) (u + u)dV
r r
r
r
r
r
1
: C e : dV t * (u + u)dS ( b) (u + u)dV
2
V
V
V
S
V
r r
r* r
= e ( ) dV t udS ( b) udV +
= e ( ) dV + e dV +
r
r
r
r
1
+ e dV t * udS ( b) udV +
: C e : dV
2
V
S
V
V
Note that:
r r
r r
r
(u) = e ( ) dV t * udS ( b) udV
and
r
r
r
r
= e dV t * udS ( b) udV = 0
Thus:
r
r
r
1
(u + u) = (u) + +
: C e : dV
2
V
r
r
r
1
(u + u) (u) = +
: C e : dV
2
V
r
r
r
1
(u + u) (u) =
: C e : dV
2
V
where we have considered = 0 . Note that the term : C e : > 0 is always positive for
any valor of since C e is a positive definite tensor (see Chapter 1). Then, we guarantee
that:
r
r
r
r
r
r
1
= (u + u) (u) =
: C e : dV > 0
(u + u) > (u)
2
V
r
So,
= 0
(u) is a minimum
NOTE 2: For a system made up by a linear elastic material, the equilibrium point
corresponds to minimum value of . This is known as the principle of minimum potential
energy.

Ciudad Real Spain
Draft
401
NOTE 3: When the external action is characterized by concentrated forces and in the
absence of body forces, the equation (5.100) becomes:
{ } {u }
r
1
(u) = U int + U ext =
: dV F loc
2
V
loc
The total potential energy
(5.108)
(u )
F
(u ( 2 ) )
(u )
=0
u
(u ( 3 ) )
Deformed corresponding to the

equilibrium
u (2)
u ( 3)
NOTE 4: By means of equation (5.108), it is easy to show the Castiglianos Theorem Part I:
r
(u)
U int
U ext
U int
{ }{ }
{ }
T
F loc uloc
=0

loc
u
{ } {u } {u } {u }
U
{F } =
{ }
u
where {F }= {F , F ,..., F } , {u }= {U , U ,..., U } . Note that the term U
function of { } .
u
uloc
loc
loc
loc
int
loc
loc
loc
loc
int
must be in
loc
NOTE 5: For better illustration of the proposed problem, we will consider a rod of length
L and cross-sectional area A . Consider also that the stress and strain fields are
homogeneous and given by:
0 0
ij = 0 0 0
0 0 0
0 0
e
ij = 0 0 0 11 = C111111 = E
0 0 0
Consider also that the displacement field is approached by a linear function

( u ( x) = a1 + a2 x ), and that on the extremities of the bar, we have:

Ciudad Real Spain
Draft
402
(3D )
V = AL (volume)
F = dA = A
(1)
F ,U
(1D )
(1)
( 2)
( 2)
,U
The goal now is to express the total potential energy in terms of U(1) ,U( 2) . Note that, due to
the concentrated forces we have:
{ } {u }= {F
U ext = F loc
loc
U(1)
F ( 2 ) ( 2 ) = F (1)U (1) + F ( 2 )U ( 2 )
U
(1)
(5.109)
For this case, (see NOTE 3 of Problem 5.5), the linear stress-strain relationship is given
1
2
1
2
by = E , and the energy density by e = = E . Then the total internal energy is

given by:
U int =
2 : dV
1D
U int =
note that 11 =
1
1
1
dV =
E dV =
E 2 dV
2V
2V
2V
u1 u ( x)
=
= , thus
x1
x
U int =
1
1
u ( x)
E 2 dV =
E
dV
2V
2 V x
(5.110)
Our goal now is to express the displacement field in terms of their nodal values U(1) ,U( 2) .
To do this we use the approach adopted u ( x) = a1 + a2 x , where:
= a1 + a2 L
u ( x = 0) = U(1) = a1
u ( x = L ) = U( 2 )
U(1) 1 0 a1
( 2) =

U 1 L a2
Next we evaluate the coefficients a1 and a 2 . To do this, we obtain the reverse form of the
above relationship:
U(1) 1 0 a1
(2) =

U 1 L a2
a1 1 L 0 U(1)
=
a2 L 1 1 U( 2)
reverse
a1 = U(1)
1 (2)
(1)
a 2 = U U
L
with which we can obtain:

u ( x) = a1 + a2 x = U(1) +
1 (2)
x
x
U U(1) x = 1 U(1) + U( 2 )
L
L
L
(1)
r
x x U
u ( x) = 1 ( 2 ) = [N ( x )] u(e )
L L U
{ }

Ciudad Real Spain
Draft
(5.111)
403
and the equation in (5.110) becomes:

U int =
2
2
1
E 1 (2)
E
u ( x)
(1)
E
( 2) 2U(1)U( 2 ) + U(1) dV
U U dV = 2 U
dV =
2 V x
2VL
2L V
Note that U(1) and U( 2) are not dependent of x , then:

U int =
2
2
2
E (2) 2
E
2U(1)U( 2 ) + U(1) dV = 2 U( 2 ) 2U(1)U( 2 ) + U(1) V
2L
2L
V
(5.112)
2
2
EAL ( 2) 2
EA ( 2 ) 2
=
2U(1)U( 2 ) + U(1) =
2U(1)U( 2 ) + U(1)
U
2L
2L
Then, the total potential energy (see equation (5.108)) is given by equations (5.109) and
(5.112), i.e.:
r
2
EA ( 2 ) 2
(u) = U int U ext =
2U(1)U( 2 ) + U(1) F (1)U(1) + F ( 2 )U( 2 ) = (U(1) , U( 2 ) )
U
2L
As we look for a stationary state, the following must be true:

(U(1) , U( 2 ) )
U(1)
(1)
( 2)
(U , U )
U( 2 )
2
EA ( 2 ) 2
U 2U(1)U( 2 ) + U(1) F (1)U(1) + F ( 2 )U( 2) = 0
(1)
u 2 L
EA
=
2U( 2) + 2U(1) F (1) = 0
2L
EA (1)
=
U U( 2 ) F (1) = 0
L
2
EA ( 2 ) 2
= ( 2)
U 2U(1)U( 2 ) + U(1) F (1)U(1) + F ( 2 )U( 2 ) = 0
U 2 L
EA
2U( 2 ) 2U(1) F ( 2 ) = 0
2L
EA ( 2 )
U U(1) F ( 2 ) = 0
=
L
Rearranging the above equations in matrix form we get:

(1)
(1)
EA 1 1 U F
= ( 2)
L 1 1 U( 2 ) F
[ ]{ } { }
] has no inverse, since det[K ] = 0 . To solve the problem we have to
K ( e ) u( e ) = F ( e )
Note that K (e )
introduce the boundary conditions.
(e)
[ ]
Note that the matrix K (e ) of the above equation could have been obtained by means of
the equation (5.99) (see Problem 5.15), and for this particular case we have [C ] = E , then,
the equation (5.111) becomes:
r
r
[B( x )] = [L(1) ][N ( x )] =
x x 1 1
1 L L = L L

thus

Ciudad Real Spain
Draft
404
1

r T
r
1 1 1
L 1 1
= [B ( x )] [C ][B ( x )] dV =
E dV = E 2
dV
1 L L
L 1 1
V
V
V

L
[K ]
(e )
1 1
EA 1 1
E 1 1
1 1 dV = 2 1 1 V =
L 1 1
L
r
NOTE 6: Analyzing [N ( x )]
=
E
L2
r
r
r
x x
Note that the shape functions are [N ( x )] = [N1 ( x ) N 2 ( x )]T = 1 . If we draw
L L
these functions inside the domain we get:

N1
N1 ( x = 0) = 1
x
L
N1 ( x = L ) = 0
N 2 ( x = 0) = 0
x
N 2 ( x) =
L
N 2 ( x = L) = 1
N2
N1 ( x ) = 1
N1 ( x ) + N 2 ( x ) = 1
The adopted approximation for [N ( x )] will depend on the problem. For the previous
problem we have that the strain is constant into the domain, so, it is sufficient to adopt a
linear approximation for displacement since by definition =
r
u ( x)
. As consequence we
x
just need two points on the boundary to define [N ( x )] . If a problem requires a cubic
r
displacement approximation, so, we will need three points to define [N ( x )] , and so on.
NOTE 7:
Principle of the Stationarity of Potential Energy

In this problem we have establish the principle of the stationarity of Potential Energy, (see
equation in (5.100)):
r r
r r
r
(u) = e ( ) dV t * udS ( b) udV
(5.113)
1
2
where we have considered e () = : . The functional is stationary if and only if

r
r
u (u) = 0 .
Hellinger-Reissners Variational Principle

In Problem 5.5 (see NOTE 7) have established that

Ciudad Real Spain
Draft
e () = e () tensorial e ( ) = : e ( ) = 0G( ) = g( )
e
( ) = : g( )
405
(5.114)
where g ( ) is the Gibbs free energy density with reversed sign.
g() - Complementary strain energy
e () - Strain energy
Figure 5.12: Strain energies.
By replacing e ( ) = : g( ) into the functional (5.113) we can obtain:

r r
r r
r
(u) = e ( ) dV t * udS ( b) udV
r r
r r
r
HR (u, ) = : g( ) dV t * udS ( b) udV
(5.115)
Note that : = : ( sym u) = : (u) . Then, we obtain:

r r
r r
r
r
HR (u, ) = : (u) g( ) dV t * udS ( b) udV
(5.116)
r
The functional (5.115) is stationary for variation of u vanishing on S u if and only if
satisfies the equilibrium equations, and is stationary for variation of if and only if they
satisfy the constitutive equation (strain-stress).
r
r
r
r
r
r
r
u HR (u, ) = : (u) dV t * udS ( b) udV = 0
r
= ij (u) i , j dV t * (u) i dS ( b) i (u) i dV = 0
i
(5.117)
r
= ij (u) i n j dS ij , j (u) i dV t * (u) i dS ( b) i (u) i dV = 0

i
r
= ij , j + ( b) i (u) i dV +
V
t * (u) i dS = 0
i
ij n j
In the volume we obtain the equilibrium equations: ij , j + ( b) i = 0 i .

Ciudad Real Spain
Draft
406
On surface S we obtain the boundary condition in stress: ij n j t * = 0 i

i
r
r
HR (u, ) = : ( sym u) g( ) dV = 0
r
g( )
: dV = 0
= ( sym u) :
(5.118)
r
g( )
: dV = 0
= ( sym u)

In the volume we obtain the constitutive equation for strain: ( sym u)
g( )
=0.
Hu-Washizus Variational Principle

The Hu-Washizus principle is a generalization of the Hellinger-Reissners principle, in
r
which the functional, in addition of the independent fields (u, ) , is also dependent of field:
r r
r r
r
r
r r
HW (u, , ) = e ( ) : ( symu) ( b) u dV ( n) (u* u)dS t * udS
r
Su
(5.119)
and is stationary if and only if:
r
u HW (u, , ) = 0
r
r r
r
On Su
u = 0
Equilibrium equations
HW (u, , ) = 0
Kinematic Equations
r
Boundary condition on S u
HW (u, , ) = 0
Constitutive equations for stress
That is:
r
r
u HW (u, , ) = 0
r
r
r
r
[ : ( u) ( b) u] dV tr udS
r
r
r
r
r
r
(u, , ) = [( ) (u) ( b) u] dV t udS
r
r
r
r
r
(u, , ) = [( ) ( b)] (u) dV t udS
r
r
u HW (u, , ) =
sym
r
u HW
r
u HW
Ciudad Real Spain
Draft
407
r
HW (u, , ) = 0
r
r
r
r
HW (u, , ) = : ( sym u) dV ( n) (u* u)dS = 0
r
Su
r
r
HW (u, , ) = ( sym u) : dV
V
[n (u
r*
r
u) : dS = 0
r
Su
r
HW (u, , ) = 0
r
HW (u, , ) = e ( ) : ( ) dV = 0
r
e ( )
e ( )
HW (u, , ) =
: : ( ) dV =
: dV = 0
V
V
NOTE 8: Discretization of the Fields

The variation of the Hu-Washizus principle can be written as follows
r r
r r
r
HW = e ( ) : ( symu) ( b) u dV t * udS = 0
V
r
r
r
r
r
= e ( ) : dV : ( sym u) dV ( b) u dV t * udS = 0
r
r
r
r r
= : e ( ) dV + : ( sym u ) dV u ( b)dV u t * dS = 0
r
r
r
r r
= : dV + : ( symu ) dV u ( b) dV u t * dS = 0
In the implementation of finite element methods we often use Voigt notation when we are
dealing with symmetric matrix. Using Voigt notation the above equation becomes:
HW =
T
T
T
T
sym
*
{} {} dV + {} { u }dV {u} {b}dV {u} {t }dS = 0
{} {} dV + {} {
T
r
symu dV =
{ }
r
r
{u}T b dV +
{}
r r
{u}T t * dS
(5.120)
Consider as approximation for displacement, strain, and stress fields, respectively, (see
Jirsek (1998)), as follows:
r
{u} [N ]{d} + [N c ]{d c }
{} [B ]{d} + [G]{e}
{} [S]{s}
and
r
{u} [N ]{d} + [N c ]{d c }
{} [B ]{d} + [G]{e}
{} [S]{s}
(5.121)
where the matrices [N ] and [B] contain the displacement interpolation functions and their
derivatives (strain interpolation matrix), respectively. [N c ] and [G] are matrices containing
some enrichment terms for displacement and strain respectively. [S] is a stress
interpolation matrix. {d} , {d c } , {e} and {s} collect the degrees of freedom corresponding

Ciudad Real Spain
Draft
408
to nodal displacement, enhanced displacement modes, enhanced strain modes, and stress
parameters, respectively. If we consider the variation of the Hu-Washizus principle:
T
T
T
T
sym
*
{} {} dV + {} { u }dV = {u} {b}dV + {u} {t }dS
V 4 244
1 4
3
V
144424443
(5.122)
144444 2S 44444
4 4
3
3
we can obtain:
(1)
{} {} dV = {[B]{d} + [G]{e}} {} dV = {[B]{d}} {} dV + {[G]{e}} {} dV

T
= {d}
[B] {} dV + {e} [G] {} dV

T
r
T
(2) {} symu dV =
{} {
T
sym r
u dV +
V 44424443
1
{} {
T
sym r
u dV =
sym
r
u dV
2.2
{[S]{s}} { ([N ]{d} + [N ]{d }) ([B]{d} + [G]{e})}dV

T
sym
{} {
V 444 24444
1
4
3
2.1
(2.1)
{[S]{s}} { ([N ]{d} + [N ]{d })}dV {[S]{s}} {([B]{d} + [G]{e})}dV

T
sym
= {s}
[S] { ([N ]{d} + [N ]{d })}dV {s} [S] {([B]{d} + [G]{e})}dV

T
sym
= {s}
[S] { ([N ]{d})}+ [S] { ([N ]{d })}dV {s} [S] {([B]{d} + [G]{e})}dV
T
sym
sym
= {s}
[S] [B]{d} + [S] [B ]{d }dV {s} [S] [B]{d} + [S] [G]{e}dV
T
[S] [B ]{d }dV {s} [S] [G]{e}dV

T
= {s}
= {s}
[S] { [B ]{d } [G]{e} }dV

T
where we have considered { sym ([N ]{d})} = [B]{d} and { sym ([N c ]{d c })} = [B c ]{d c } .

Ciudad Real Spain
Draft
{} {
T
( 2 .2 )
sym
r
u dV =
sym
409
} {}dV
r
u
{ ([N ]{d} + [N ]{d }) ([B]{d} + [G]{e}) } {[S]{s}}dV

T
sym
{ ([N ]{d}) + ([N ]{d }) ([B]{d} + [G]{e}) } {[S]{s}}dV

sym
sym
{ [B]{d} + [B ]{d } ([B]{d} + [G]{e}) } {[S]{s}}dV

T
{ [B ]{d } [G]{e} } {[S]{s}}dV

T
{ {d } [B ] {e} [G] }{[S]{s}}dV

T
= {d c }
[B ] {[S]{s}}dV {e} [G] {[S]{s}}dV

T
(3)
T
T
*
{u} {b}dV + {u} {t }dS
= {d}
{ f ext } + {d c }T { f c }
Taking into account the previous terms, the equation in (5.122) becomes:
{d}T [B]T {} dV + {e}T [G]T {{} {[S]{s}}} dV + {s}T [S]T { [B c ]{d c } [G]{e} }dV +
V
+ {d c }
[B ] {[S]{s}}dV = {d} { f } + {d } { f }
T
ext
(5.123)
r
Since {u} , {} and {} are variables of the independent fields, so, we can say that:
T
T
T
{d} [B] {} dV = {d} { f ext }
T
T
{e} [G] {{} {[S]{s}}} dV = {0}
T
T
{s} [S] [B c ]{d c } [G]{e} dV = {0}
V
{d }T [B ]T {[S]{s}}dV = {d }T { f } = {0}
c
c
c
c
V
(5.124)
If we consider { f c } = {0}, the above equation can be rewritten:

Ciudad Real Spain
Draft
410
T
[B] {} dV = { f ext }
V
T
[G] {{} {[S]{s}}} dV = 0
V
T
[S] [B c ]{d c } [G]{e} dV = 0
V
[B ]T {[S]{s}}dV = 0
c
(5.125)
Taking into account that the stress-strain relationship is given by the following expression:
{} = [C ]{} = [C ]{[B]{d} + [G]{e}}
(5.126)
and by substituting into the equation in (5.125) we can obtain:
T
T
[B] [C ][B] dV {d} + [B] [C ][G] dV {e} = { f ext }
V
V
T
T
T
[G] [C ][B ] dV {d} + [G] [C ][G] dV {e} [G] [S] dV {s} = {0}
V
V
V
T
T
[S] [B c ] dV {d}c [S] [G] dV {e} = {0}
V
V
B T [S] dV {s} = {0}

c
[ ]
(a)
(b)
(5.127)
(c)
(d)
Rewriting the above equation in matrix form we get:

[B]T [C ][B ]
T
[G] [C ][B]
[B]T [C ][G]
[G]T [C ][G]
T
[S] [G]
0
T
[G] [S]
0
[B c ] [S]
T
dV
[S]T [B c ]
0
0
{d} { f ext }
{e} {0}
{s} {0}
{d c } {0}
(5.128)
Let us suppose that we do not introduce any displacement enhancement terms, thus
{d c } = {0} [B c ] = [0] , with that the equation in (5.127)(c) becomes:
[S] [G] dV {e} = {0}

T
(5.129)
Thus, piecewise constant stress functions {} will require [S] = [1] (unit matrix). The
compatibility conditions (5.129) now read:
[G] dV = [0]
(5.130)

Ciudad Real Spain
Draft
411
Discontinuity on displacement and strain fields Applying the Principle of Virtual

Work
As we have seen before, virtual work is the work done by real force acting through virtual
displacements. A virtual displacement is any displacement consistent with the constraints
of the structure, i.e. which satisfies the boundary conditions.
The principle states that the virtual work of the internal forces must be equal to the virtual
work of the external forces:
r r r
r r
&&
: dV = t * u dS + (b u) u dV
V 4 4
1 2 3
Total internal
virtual work
(5.131)
144444 44444
2
3
Wext
Wint
for all the admissible virtual displacements u .

Let us consider a discretized system where we can say that all the forces are applied in the
nodes of the finite element (CST-Constant Strain Triangle) (see Figure 5.13).
(a) nodal forces
(b) nodal displacements
Figure 5.13: Discontinuous CST finite element.

The nodal forces and nodal displacements can be expressed as follows:
{F}
{F}
Fx1
Fy1
Fx2
Fy2
Fx3
Fy3
0
0
{a e }
{ e }
u1
x
1
uy
u 2
x
2
u y
= 3
ux
u 3
y
4
x
4
x
(5.132)
Hence, the external virtual work becomes:
{ } {F} + { } {F} = {{a } { } }{F}

{F}
Wext = a *
e

Ciudad Real Spain
* T
e
* T
e
Draft
* T
e
(5.133)
412
We consider the strain field and the virtual strain field are compound by two parts:
~
{} = {} + {}
{ } = { }+ {~ }
(5.134)
thus the internal virtual work becomes:

Wint =
{} {} d V = {} {{ }+ { }}d V = {{ }+ { }} {} d V
T
~*
~*
(5.135)
Symmetric formulation
The discretization for the first approximation is:
{a
{ }= [123}+ [124 ] = {[B] [G ]}{ }
a
B]{
G ][
4 3
}
{a }
{} = [B2a} + [Ge23 = {[B] [Ge ]} e ;
] { 4 ]{ e }
13 1 4
{ e }
{ }
{~}
*
e
e
~*
{ }
{ }
*
e
*
e
*
e
(5.136)
Notice that we have used the same approximation function [B] , [Ge ] for virtual and real
strains. Then, the stress field can be written as follows:
{a }
{} = [C ]{} = [C ]{[B] [Ge ]} e
{ e }
(5.137)
By replace the approximations (5.136) and (5.137) into the equation (5.135), the internal
virtual work becomes:
Wint =
{{ } { }}
~
+ *
*
{} d V = {[B]
{ } [C ]{[B] [G ]}{a } d V
{ }
{ }
*
[Ge ]} a e*
e
{a }
{ }
{ } {[B] [G ]} [C ]{[B] [G ]}{ } d V
a*
= e
*
V e
(5.138)
{a
{ } { } [B] [C ]{[B] [G ]} d V { }
= a*
e
* T
e
[Ge ]
By apply Wext = Wint (see Eq. (5.133) and (5.138)) we can obtain
[B ]T [C ][B]
{F}
=

T
{F} Ve [Ge ] [C ][B]
[B]T [C ][Ge ] dV {a e }
[Ge ]T [C ][Ge ] { e }
(5.139)
and considering the traction vector continuity, i.e. {F} = {0} , we obtain:
[B]T [C ][B ] [B]T [C ][Ge ]
{a e }
{F}
=
dV

T
T
{0} Ve [Ge ] [C ][B ] [Ge ] [C ][Ge ]
{ e }
144444 2444444
4
3
(5.140)
[K e ]

Ciudad Real Spain
Draft
413
Anti-symmetric formulation
Now consider the real and virtual strain approximation by:

{a }
{} = [B]{a3 + [Ge23 = [[B] [Ge ]] e
{e
e } 1 ] 4}
12
4
{ e }
{ }
{~}
(5.141)
{
{ }= [B]{a }+ [G ]{ }= [[B] [ ]] { }
}
*
*
e
*
e
a*
G* e
e
*
e
*
e
where we are considering different approximation functions for virtual and real strains i.e.
[G ] [G ].
*
e
Using equation (5.135), and discretization (5.141) we can obtain:
{{
Wint = a *
e
{a
} { } } [B] [C ]{[B] [G ]} dV { }
[G ]
}
* T
e
Ve
* T
e
(5.142)
Considering Wext = Wint and considering the traction vector continuity, we can obtain:
{ }
{ }
[B]T [C ][B] [B]T [C ][G e ]

ae
{F}
dV
= * T
* T
{0} Be G e [C ][B] G e [C ][Ge ]
3
44444 2444444 e
1
4
[ ]
[ ]
(5.143)
[K e ]
According to Jirsek(1998) there are three major classes of these models:
SOS (Statically Optimal Symmetric) formulation cannot properly reflect the

kinematics of a completely open crack but it gives a natural stress continuity
condition;
KOS (Kinematically Optimal Symmetric) formulation describes the kinematic

aspects satisfactorily but leads to an awkward relationship between the stress in the
continuous part of the element and the tractions across the discontinuity line.
These findings motivate the development of the nonsymmetric;
SKON (Statically and Kinematically Optimal Nonsymmetric) formulation, which

combines the strong points of each of the symmetric formulations.
Reference
JIRSEK, M. (1998). Finite elements with embedded cracks. LSC Internal Report 98/01,
April.

Ciudad Real Spain
Draft
414
Problem 5.17
Show that:
r r
r
r
r
r* r
r
&
&
&
&& r
r
( X , t ) u dS 0 + 0 b( X , t ) u( X , t ) u dV0 = P : X u dV0
V0
(5.144)
V0
where u is the virtual displacement field, and P is the first Piola-Kirchhoff stress tensor.
Reference
configuration
Current
configuration
S 0
V0
B0
r r
r
t * ( X , t) = t *
0
dV0
r r
u( X , t )
B
r r
u( x , t )
r r
r
0 b( X , t ) = 0 b 0
Solution:
r
r r
t * ( x, t )
dV
r r
b( x , t )
r r
Although the variables t * ( X , t ) and b( X , t ) are not intrinsic variables of the reference
r r
r
configuration like the variables 0 , S 0 , V0 , for simplicity, we denote t * ( X , t ) = t * and
0
r r
r
b( X , t ) = b 0 .
Remember also (see Chapter 2 of the textbook) that:
r
r
r
r
&
D
xi ( X , t )
xi ( X , t ) u i ( X , t ) &
&
Fij Fij =
=
=
= ui, J ( X , t )
Dt
t X j
X j 1 24
X j
4t 3
&
xi
r r
&
r r
&
& = l F = r u( X , t ) = u( X , t )
or F
r
X
X
r r
&
r &
and l = F F 1 = X u( X , t ) F 1
r r
r r
&
r &
r&
F 1 = F 1 l = F 1 X u( X , t ) F 1 = F 1 x u( x , t )
Taking into account the above relations, it is also valid that:

r r
r r
r r
&
&
&
&
&
r
r
r
F = X u ( X , t ) y F 1 = F 1 X u ( X , t ) F 1 = F 1 x u ( x , t )
With that we can get:

&
P : F dV = P
V0
&
&
&
(PiJ ui ) , J = PiJ , J ui + PiJ ui , J
iJ
r
&
&
FiJ dV0 = PiJ ui , J ( X , t ) dV0
V0
V0
&
&
&
PiJ ui , J = (PiJ ui ) , J PiJ , J ui
thus:

Ciudad Real Spain
Draft
415
r
r
r
&
&
&
&
P : F dV0 = PiJ ui , J ( X , t ) dV0 = (PiJ ui ( X , t )) , J PiJ , J ui ( X , t ) dV0
V0
V0
&
V0
P : F dV = (P
iJ
V0
V0
&
P : F dV = P
iJ
V0
r
r
&
&
ui ( X , t )) , J dV0 PiJ , J ui ( X , t ) dV0
V0
r
r
&
&
ui ( X , t )n J dS 0 PiJ , J ui ( X , t ) dV0
S0
V0
where we have applied the divergence theorem. The above in tensorial notation becomes:
r
&
P : F dV = (P n) u ( X , t ) dS (
&
V0
S0
r
X
&
P) ui ( X , t ) dV0
V0
Remember that the equations of motion in the reference configuration are given by:
r
r
&& r
r
X P + 0 b 0 = 0 u( X , t )
r
r
&& r
r
X P = 0 b 0 u( X , t )
r r
&
&
r
and taking into account that F = X u ( X , t ) and t * = P n we get:

0
r
&
P : F dV = (P n) u ( X , t ) dS (
&
V0
S0
r
X
V0
&
P) ui ( X , t ) dV0
r
r r
r && r
r
r
r
&
&& r
&
r
P : X u ( X , t ) dV0 = t * u( X , t ) dS 0 + 0 b 0 u( X , t ) ui ( X , t ) dV0
0
V0
S0
V0
Reminder: Recall from Chapter 5 of the textbook that the stress power can be expressed
in different ways, namely:
1
1
&
&
&
&
&
S : C dV0 =
P : F dV
w int (t ) = P : F dV0 = S : E dV0 = P : F dV0 =
2V
J
V
V
V
V
&
P : F dV = : D dV = { : D dV0 = : D dV0
J
V
V0
V0
&
&
NOTE 1: Remember that neither P nor F are in any configuration, but the scalar P : F
is in the reference configuration.
NOTE 2: Taking into account the above. The total external virtual work can also be
expressed as follows:
sym
r r
r r
r r
&
&
&
&
r
r
r
: D dV = : x u ( x , t ) dV = : x u ( x , t ) dV = P : F dV0 = P : X u ( X , t ) dV0
V
V
V
V
V
r r sym
r&
= x u( x , t)
(see Problem 2.37). Note that, due to
sym
r r
r r
&
&
r
r
the symmetry of the relationship : x u ( x , t ) = : x u ( x , t ) holds.
where we have used that D = l
sym
NOTE 3: From a Variational Principle point of view, (see Holzapfel (2000)), the equation
in (5.144) is also valid for a variation of the virtual field:
r r
r
r
r
r r
r
&& r
r
t * ( X , t ) u dS 0 + 0 b( X , t ) u( X , t ) u dV0 = P : X u dV0
V0

Ciudad Real Spain
V0
Draft
(5.145)
416
Problem 5.18
a) Show that: given a symmetric second-order tensor A = A sym we can split up into

A = A P + A S where A P = P P : A , A S = P S : A , with P P = (b b) (b b) and

P S = I sym (b b) (b b) , where b is a unit vector according to certain direction, and
I sym is the symmetric part of the fourth-order unit tensor. b) Show that the constitutive
equation for stress = C e : can be written as follows:
P C PP
S = SP
C
P
PP
ij Cijkl

S = SP
ij Cijkl

C PS
:
C SS
PS
Cijkl kl
SS
Cijkl kl
(5.146)
where
C PP = P P : C e : P P
PP
P
P
Cijkl = Pijpq Cepqst Pstkl
C PS = P P : C e : P S
PS
P
S
C SP = P S : C e : P P
SP
S
P
C SS = P S : C e : P S
(5.147)
SS
S
S
Solution:
a) We start using the Cartesian system where it fulfills:
A = A ij (e i e j ) = A i1 (e i e1 ) + A i 2 (e i e 2 ) + A i 3 (e i e 3 )
= A 11 (e1 e1 ) + A 21 (e 2 e1 ) + A 31 (e 3 e1 ) + A 12 (e1 e 2 ) + A 22 (e 2 e 2 ) + A 32 (e 3 e 2 )
+ A 13 (e1 e 3 ) + A 23 (e 2 e 3 ) + A 33 (e 3 e 3 )
Components:
A 11
A ij = A 21
A 31
A12
A 22
A 32
A13 0
A 23 = A 21

A 33 A 31

A 12
A 22
A13 A11
A 23 + 0

A 33 0

A 32
0 0
0 0 = A ij + A ij
0 0
Note also that the normal component A 11 = A (Ne1 ) (according to e1 -direction) can also be
obtained by A11 = A : (e1 e1 ) = (e1 e1 ) : A , so the tensor A = A11 (e1 e1 ) can be

written as follows:
A = (e1 e1 ) A 11 (e1 e1 ) A 11 = (e1 e1 ) (e1 e1 ) : A
thus
A = A A = A (e1 e1 ) (e1 e1 ) : A = I sym : A (e1 e1 ) (e1 e1 ) : A
= I sym (e1 e1 ) (e1 e1 ) : A
Although we showed for the unit vector e1 the above is valid for any unit vector, i.e.:

A P = (b b) (b b) : A = P P : A

A S = I sym (b b) (b b) : A = P S : A
Note that A (N ) = (b b) : A = b A b is the normal component according to b -direction,
i.e. parallel to b . It is interesting to review the Problem 1.114.

Ciudad Real Spain
Draft
417
b) We apply the above definition to obtain:

= P + S = PP : + PS :
= P + S = PP : + PS :
with that, and considering = C e : , we can get:

P = P P : = P P : C e : = P P : C e : ( P + S ) = P P : C e : P + P P : C e : S
= P P : Ce : P P : + P P : Ce : P S :
S = P S : = P S : C e : = P S : C e : ( P + S ) = P S : C e : P + P S : C e : S
= P S : Ce : P P : + P S : Ce : P S :
thus
P P P : C e
S= S
e
P : C
P P : C e P
:
P S : C e S
or
P P P : C e : P P
S= S
e
P
P : C : P

Ciudad Real Spain
Draft
P P : C e : P S
:
P S : C e : P S
418

Ciudad Real Spain
Draft
7 Linear Elasticity
7.1 Solved Problems
Problem 7.1
The cylinder shown in Figure 7.1 is made up of a linear isotropic elastic material, and is
subjected to a strain state (in cylindrical coordinates) as follows:
err = e = a sin
a cos
2
e zz = ez = erz = 0
e r =
(7.1)
where eij are the Almansi strain tensor components.

r
Calculate the traction vector t on the boundary, in cylindrical coordinates.

Assumptions: 1) , are the Lam constants; 2) Small deformation regime.
x3
ez
er
x2
x1
Figure 7.1
Solution:
416
Small deformation regime: e E

rr
( r , , z ) = r
rz
r

z
a sin
rz
a cos
z =
2
zz 0
a cos
2
a sin
0
(7.2)
= Tr ( )1 + 2
(7.3)
Tr ( ) = 2a sin
(7.4)
thus,
a sin
1 0 0
a cos
= 2a sin 0 1 0 + 2
2
0 0 1
a cos
2
a sin
0
0
a cos
2a sin + 2 a sin
0
a cos
2a sin + 2 a sin
(r,r, =
0
0
2a sin
(7.5)
(7.6)
The traction vector t :

r
t (n ) = n
(7.7)
(
t 1n) 2a sin + 2 a sin
(n)
a cos
t 2 =
t (n)
0
3
(7.8)
n = (1,0,0)
Problem 7.2
The parallelepiped of Figure 7.2 is deformed as indicate by the dashed lines. The
displacement components are given as follows:
u = C1 xyz
v = C 2 xyz
= C 3 xyz
(7.9)
a) Obtain the strain state at the point E , which the current reference of the point is
represented by E whose coordinates are E (1.503; 1.001; 1.997) ;
b) Obtain the normal strain at the point E in the direction of the line EA ;
c) Calculate the angular distortion at the point E that undergoes the angle formed by
the lines EA and EF .
d) Find the volume variation and the average volumetric deformation.

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
417
z
F = F
G = G
D = D
2m
E
C = C
O = O
y
1 .5 m
A = A
B = B
1m
Figure 7.2
Solution:
a) The strain state in function of the displacements is given by:
ij =
1 u i u j
+
2 x j x i
(7.10)
which in engineering notation is:
x
1
ij = xy
2
1
2 xz
1
xy
2
y
1
yz
2
u
1

xz
x
2
1 v u
1
yz = +
2
2 x y
z 1 u

+
2 x z
1 v u
+
2 x y
v
y
1 v
+
2 y z
1 u
+
2 x z
1 v
+
2 y z
(7.11)
To determine the strain state at any point we need a priori to determine the displacement
field.
Calculation of the constants:
By substituting the values given for the point E (1.5; 1.0; 2.0) , we obtain:
u ( E ) = 1.503 1.5 = C1 (1.5)(1.0)(2.0) C1 = 0.001
0.001
3
= 1.997 2.0 = C 3 (1.5)(1.0)(2.0) C 3 = 0.001
v ( E ) = 1.001 1.0 = C 2 (1.5)(1.0)(2.0) C 2 =

w(E)
where
u
(E)
we
have
used
the
displacement
= (1.503 1.5; 1.001 1.0; 1.997 2.0) .
of
(7.12)
the
point
Then, the displacement field becomes:

Ciudad Real - Spain
Draft
418
Engineering notation
Scientific notation
u = 0.001 xyz
0.001
v=
xyz
3
w = 0.001 xyz
u1 = 0.001 X 1 X 2 X 3
0.001
X1X 2 X 3
3
u 3 = 0.001 X 1 X 2 X 3
u2 =
(7.13)
u
= 0.001 yz = 0.002 = 11
x
v 0.001
y =
=
xz = 0.001 = 22
3
y
w
z =
= 0.001xy = 0.0015 = 33
z
v u 0.001
0.011
xy =
x + y = 3 yz + 0.001xz = 3 = 2 12
x =
w u
xz =
+
= 0.001 yz + 0.001xy = 0.0005 = 2 13
x z
w v
0.001
yz =
y + z = 0.001xz + 3 xy = 0.0025 = 2 23
The strain field becomes:
yz
1 yz
ij = 0.001
2 3 + xz
1
( xy yz )
2
1 yz
+ xz
2 3
xz
3
1 xy
xz
2 3
1
(xy yz )
1 xy
xz
2 3
xy
The strain state at the point E ( x = 1.5; y = 1.0; z = 2.0) is:
ij
x
1
= xy
2
1
2 xz
1
xy
2
y
1
yz
2
0.011
1

0.00025
xz 0.002
6
2
0.011
1
0.00125
yz =
0.001
6
2
0.00025 0.00125 0.0015

z
(7.14)
b) The normal strain component is obtained as follows:
M = M M
components

M = ij M i M j
(7.15)
By expanding the above equation and by considering the symmetry of the strain tensor we
obtain:
2
2

M = 11 M 12 + 22 M 2 + 33 M 3 + 2 12 M 1 M 2 + 2 13 M 1 M 3 + 2 23 M 2 M 3
(7.16)
in engineering notation:
2
2

M = x M 12 + y M 2 + z M 3 + xy M 1 M 2 + xz M 1 M 3 + yz M 2 M 3
(7.17)
The normal vector is given by the direction cosines of the direction of the line EA :

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
M2 =
;
5
M1 = 0;
419
M3 =
5
(7.18)
By substituting the corresponding values into the equation (7.17), we obtain:

2
2

M = y M 2 + z M 3 + yz M 2 M 3
1
4
2
+ (0.0015) + (0.0025)
5
5
5
3
= 2 10
M = 0.001
(7.19)
c) For small deformation, the distortion of the angle at the point E formed by the lines
EA and EF , with = 90 ), becomes:
MN =

1
1 2M N
MN =

2
2
sin
= M N

MN = ij M i N j (7.20)

components

More details of the above equation is provided in the textbook in Chapter 2 Continuum
kinematics (small deformation regime). Expanding the above expression and considering
the symmetry of the strain tensor we obtain:

MN = 11 M 1 N 1 + 22 M 2 N 2 + 33 M 3 N 3 + 12 M 1 N 2 + M 2 N 1 +

+ 13 M 1 N 3 + M 3 N 1 + 23 M 2 N 3 + M 3 N 2
(7.21)
or in engineering notation:
MN

2

= x M 1 N1 + y M 2 N 2 + z M 3 N 3 +
xy
2
(M

N 2 + M 2 N1 +
yz

+ xz M 1 N 3 + M 3 N 1 +
M 2 N3 + M 3 N2
2
2
(7.22)
where:
M i = 0
1
5
N i = [ 1 0 0]
(7.23)
thus:
MN

2
MN

2
1
0.011

+ ( 0.00025)( 1)
= 12 M 2 N 1 + 13 M 3 N 1 =
(1)
6
5
5
(7.24)
= 5.96284793998 10 4
2
MN = 1.1925696 10 3

Alternative Solution
We can construct an orthogonal basis associated with the unit vectors M and N by means
of the cross product P = M N . Then, we obtain the components of the unit vector P :
e1
P=MN = 0
1
e2
1
e3
2
5
0
5
0

Ciudad Real - Spain
2
1
e2
e3
5
5
Draft

Pi = 0
2
5
(7.25)
420
Then, the transformation matrix from the X 1 X 2 X 3 -system to the base constituted by the

unit vectors M , N and P are given by:
M 1

A = a ij = N1
P
M2
N2
P
2
M3 0

N 3 = 1
P3 0

1
5
0
2
5
5
0
1
5
(7.26)
Applying the component transformation law for a second-order tensor components, i.e.
ij = a ik a jl kl or in matrix form = A A T :
0.011
0.00025
2 0.002
0
6
5 0.011
0
0.001
0.00125 1
1
0.00025 0.00125 0.0015 0
= 1
0
5
0
2
5
1
5
0
2
5
5
0
1
5
(7.27)
Thus:
MN =

2 10 3
= 5.96284794 10 4
ij
4
2.5 10
MN

2
5.96284794 10 4
2 10 3
1.75158658 10 3
3
1.75158658 10
1.5 10 3
2.5 10 4
(7.28)
NOTE: Note that this example is not a case of homogeneous deformation, i.e. a straight
edge in the reference configuration is no longer straight in the current configuration. To
obtain the deformed unit vector we must apply the linear transformation m = F M and
n = F N , where F is the deformation gradient.

d) The volume strain by definition is V =
( dV )
where dV is the differential volume.
dV
For small deformation we have:

(dV )
= x + y + z
dV
(dV ) = x + y + z dV
V =
(7.29)
by integrating we obtain the volume variation:

V =
2.0
+ y + z dV = 0.001
1.5
yz +
z = 0 y =0 x = 0
xz
xy dxdydz
3
(7.30)
thus:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
421
V = 1.125 10 3 m 3
(7.31)
Then:
V =
(dV )
1.125 10 3
= 0.375 10 3
=
dV
1 .5 1 .0 2 .0
(7.32)
Problem 7.3
The stress state at one point of a structure, which is made up of an isotropic linear elastic
material, is given by:
6 2 0
ij = 2 3 0 MPa
0 0 0
a) Obtain the engineering strain tensor components. Consider the Youngs modulus
( E = 207GPa ) and the shear modulus ( G = 80GPa ).
b) If a cube of side 5cm is subjected to this stress state. Obtain the volume variation.
Solution:
The strain components can be obtained by means of the equations:
)]
1
x y + z = 3.333 10 5
E
1
y =
y ( x + z ) = 2.318 10 5
E
1
z =
z x + y = 4.348 10 6
E
1
xy = xy = 2.5 10 5
G
1
xy = xy = 0
G
1
xy = xy = 0
G
x =
)]
(7.33)
where the Poissons ratio can be obtained by:

G=
E
2(1 + )
E
207
1 =
1 0.29375
160
2G
Thus:
0
33.24 12.5
12.5 23.01
0 10 6
ij =
0
0
4.257
1
In the textbook (Chaves(2013)) we have demonstrated that C e =
(1 + )
I 1 1 , with
E
E
that we obtain

Ciudad Real - Spain
Draft
422
(1 + )
(1 + )
(1 + )
= Ce : =
I 1 1 : =
I : 1 1: =
Tr ( )1
E
E
E
E
E
E
In components:
11
(1 + )
ij =
12
E
13
13
1 0 0
Tr ( ) 0 1 0
23
E
0 0 1
23 33
0
6 2 0
1 0 0 33.24 12.5
4.2609 10 6 0 1 0 = 12.5 23.01
6
= 6.251 10 2 3 0
0 10 6
0 0 0
0 0 1 0
0
4.257
where we have used
12
22
(1 + )
1
= 6.25 10 6
Tr ( ) = 4.25725 10 6 .
,
E
MPa E
In the small deformation regime the volumetric deformation (linear) is equal to the trace of
the strain tensor:
L
DV V = I = (33.24 23.01 4.257 ) 10 6 = 5.973 10 6
Then, the volume variation is:

V = V V0 = 5.973 10 6 (5 5 5) = 7.466 10 4 cm 3
Problem 7.4
A parallelepiped of dimensions a = 3cm , b = 3cm , c = 4cm , is made up of an isotropic
homogeneous linear elastic material, which is accommodated in a cavity of the same shape
and dimensions, whose walls are made up of a very rigid material (undeformable).
Via a rigid plate (dimensions a b ) of negligible weight and negligible friction we apply a
perpendicular compression force equal to F = 200 N which compresses the elastic block.
If = 0.3 is the Poissons ratio and E = 2 10 4 N / cm 2 is the Youngs modulus, calculate:
a) The lateral force exerted by the wall of the cavity on the parallelepiped;
b) The height variation experienced by it, i.e. find c .
F
z
a
y
c
x
Figure 7.3

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
423
Solution:
At any point of the elastic body the stress state is characterized only by normal components
x , y and z . The stress z is given by:
z =
200
200
200 N
=
=
ab
3 3
9 cm 2
(7.34)
Note that, because of the problem symmetry the stresses x and y are equal, then:
)]
1
x y + z = 0
E
x ( x + z ) = 0
x = y =
1
[ x ( x + z )] = 0
E
(7.35)
z
x =
(1 )
thus:
x =
z
0.3 200
200 N
=
=
(1 ) (1 0.3)
9
21 cm 2
(7.36)
The force exerted by the wall on the elastic body is given by:
200
3 4 = 114.28 N
21
200
3 4 = 114.28 N
Fx = x b c =
21
Fy = y a c =
(7.37)
The strain z can be obtained as follows:
)]
1
z x + y
E
1
= [ z 2 x ]
E
1 200
200
=
+ 2 0 .3
= 8.25 10 4
4
21
2 10 9
z =
(7.38)
Then, the height variation is given by:

c = z c = 8.25 10 4 4 = 0.0033cm
(7.39)
Problem 7.5
Figure 7.4 (a) shows a support device for a machine. Said support apparatus is made up of
a neoprene block of dimensions ( 50 20cm ), and is characterized by the element ABCD of
Figure 7.4(b).

Ciudad Real - Spain
Draft
424
a)
b)
1 .1
D
A
1 .2
A
1
20
1 .1
50
Dimensions in centimeters - cm
Figure 7.4
Under the action of vertical and horizontal loads the neoprene deforms as shown in Figure
7.4 (b) (ABCD) in which the displacement field ( u, v) is represented as follows:
u = a1 x + b1 y + c1
v = a2 x + b2 y + c2
where a1 , b1 , c1 , a 2 , b2 , c 2 are constants to be determined.

a) Calculate the strain tensor components and the volumetric deformation at any
point;
b) Calculate the stresses at any point;
c) The maximum normal stress;
d) Obtain the unit extension according to the direction of the diagonal AC .
Hypothesis:
1 Isotropic linear elastic material with Youngs modulus equals to 1000 N / cm 2 and the
shear modulus equals to
1
N / cm 2 .
0.0028
2 It is assumed a state of plane strain.

Solution:
u = a1 x + b1 y + c1
v = a 2 x + b2 y + c 2
(7.40)
According to Figure 7.4 we obtain:

u (0;0) = 1 = c1
u (50;0) = 1.1 = 50a1 + 1 a1 = 0.002
(7.41)
u (0;20) = 1.1 = 20b1 + 1 b1 = 0.005
thus
u = 0.002 x + 0.005 y + 1

Ciudad Real Spain
Draft
(7.42)
7 LINEAR ELASTICITY
425
For the vertical displacement:

v(0;0) = 0 = c 2
u (50;0) = 0 = 50a 2 a 2 = 0
(7.43)
u (0;20) = 1 = 20b2 b2 = 0.05

v = 0.05 y
(7.44)
u = 0.002 x + 0.005 y + 1
v = 0.05 y
(7.45)
Then:
a)
Strains
x =
u
= 0.002
x
y =
v
= 0.05
y
xy =
u v
= 0.005
+
y x
(7.46)
The linear volumetric deformation (small deformation regime):

L
DV = V = x + y + z = I = 0.048
b)
G=
(7.47)
Stresses
E
E
=
1 = 0.4
2(1 + )
2G
E
(1 ) x + y
(1 + )(1 2v)
= 3571.4286 [(0.6) 0.002 0.4 0.05] = 67.1428
x =
E
(1 ) y + x
(1 + )(1 2v)
= 3571.4286 [(0.6) (0.05) + 0.4 0.002] = 104.2857
y =
(7.48)
xy = G xy
=
1
0.005 = 1.785714
0.0028
An alternative solution is to use ij =

x
ij = 1 xy
2
1 xz
2
1
2
xy
y
1
2
yz
1
2
1
2
ETr ( )
E
ij +
ij , where:
(1 + )(1 2 )
(1 + )
xz 0.002

yz = 1 (0.005)
2
z
0

1
2
(0.005) 0
0.05 0
0
0
ETr ( )
N
E
N
,
= 714.285714
= 68.571429
2
(1 + )(1 2)
(1 + )
cm
cm 2

Ciudad Real - Spain
Draft
426
0.002
1 0 0
0 1 0 + 714.285714 1 (0.005)
ij = 68.571429
2
0 0 1
0
1
2
(0.005) 0
0.05 0
0
0
0
67.1428 1.785714
N
1.785714 104.2857
0
cm 2
0
0
68.571
c)
The principal stresses

(1,2) =
x + y
2
x y

+ 2
xy
(7.49)
2
67.1428 104.2857
67.1428 + 104.2857
2

+ 5.35714 =
2
2
= 171.4285 19.328675
(1,2) =
1 = 152.099824 N cm 2
2
2 = 190.757175 N cm
d)
(7.50)
(7.51)
The unit extension
The diagonal ( AC ) in the reference configuration measures:

L0 = AC = 50 2 + 20 2 = 53.852cm
(7.52)
and the deformed diagonal

AC = 50.2 2 + 19 2 = 53.675cm
(7.53)
L = AC AC = 0.177cm
(7.54)
L 0.177
=
= 0.0033
L0
53.852
(7.55)
The unit extension is:

=
Problem 7.6
Consider a bar of length L , which has a squared cross section of side a . The elastic
constants of the material is assumed to be known ( E and = 0.25 ).
a) In the case of Figure 7.5(a), calculate the stored energy (strain energy density) in the
bar during the deformation, and also obtain the total strain energy;
b) Determine the stored energy corresponds to the change of volume and to the
change of shape;
c) The same question as the paragraph (a), but considering the case of Figure 7.5(b).

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
427
A -Area
P
P
a
a
M
a)
cross section
b)
Figure 7.5
Solution:
Considering a one dimensional case:
x = E x x =
x
E
x =
with
P
A
(7.56)
We know that the strain energy per unit volume is given by:
1
2
1
2
1
2
e = : unidimensional e = x x = x

x 1 P2
=
E 2 EA 2
(7.57)
Then, the total energy U is given by:
e (volume) = L A
x
P2
2 EA 2
(7.58)
P2L
U =
2 EA
The strain energy (per unit volume) can also be expressed as follows:
e =
1
1
2
I
II dev
6(3 + 2 )
2
14 244 1 24
4
3
4 3
vol
(7.59)
shape
Considering:
x
ij = 0
0 0
P
0 0 I = x =
A
0 0
(7.60)
Calculation of II dev :
1
I2
2
II dev = (3 II I ) = = x
3
3
3
(7.61)
Then, the strain energy associated with the change of volume is:
e vol =
(1 2) 2 (1 2) 2
1
2
I =
I =
x
6(3 + 2 )
6E
6E

Ciudad Real - Spain
Draft
(7.62)
428
(1 2 ) P 2
(per unit volume)
6E A 2
e vol =
(7.63)
The strain energy associated with the change of shape is:
e shape =
1
1 2(1 + )
II dev =
II dev
E
2
2
(1 + ) x
=
E
3
(7.64)
shape
(1 + ) x
(1 + ) P 2
(per unit volume)
=
=
E
3
3E A 2
(7.65)
Checking:
vol
shape
=
=
=
=
(1 2) P 2 (1 + ) P 2
+
6E A2
3E A 2
P2
[(1 2) + 2(1 + )]
6 EA 2
P2
[1 2 + 2 + 2]
6 EA 2
P2
= e
2
2 EA
For the case of bending, we consider the following relationships:

y =
M y
I
I=
y =
a4
12
12 M y
a4
y = E y y =
y
E
The strain energy becomes:

1 12 M y y
E
e = y y = 4
2
2 a
1 12 M y 12 M y
2 a4
Ea 4
72 M 2 y 2
Ea 8
e =

Ciudad Real Spain
72 M 2 y 2
Ea 8
Draft
(7.66)
(7.67)
7 LINEAR ELASTICITY
429
Problem 7.7
Under the approximation of small deformation theory, the displacement field is given by:
r
u = ( x1 x3 ) 2 10 3 e1 + ( x 2 + x3 ) 2 10 3 e 2 x1 x 2 10 3 e 3
a) Obtain the infinitesimal strain tensor, the infinitesimal spin tensor at the point
P (0,2,1) ;
Solution:
u1
x1
u i u 2
=
x j x1
u 3
x1
u1
x 2
u 2
x 2
u 3
x 2
u1
x3 2( x x )
0
2( x1 x3 )
1
3
u 2
0
2( x 2 + x 3 ) 2( x 2 + x3 ) 10 3
=
x3
x2
0
x1
u 3
x3
and at the point P(0,2 1) we have

u i
x j
2( x1 x 3 ) 2 0 2
0
2( x1 x 3 )
=
0
2( x 2 + x3 ) 2( x 2 + x 3 ) = 0 2 2 10 3

x2
2 0 0
x1
0
u i
= ij + ij
x j

Infinitesimal strain tensor
u j
1 u
ij = i +
2 x j xi
Infinitesimal spin tensor
2 0 2

u j
= 0 2 1 10 3 ij = 1 u i

2 x j xi
2 1 0
0 0 0

(7.68)
= 0 0 1 10 3

0 1 0
Problem 7.8
Under the restriction of small deformation theory, the displacement field is given by:
r
2
u = a ( x12 5 x 2 ) e1 + (2 a x1 x 2 )e 2 (0) e 3
a) Obtain the linear strain tensor and the linear spin tensor;
b) Obtain the principal strains and principal stresses;
c) Given the shear modulus G , obtain the Youngs modulus E to guarantee the balance at
any point.
NOTE: The body forces are negligible.
Solution:
2
a) Considering that u1 = a ( x12 5 x 2 ) , u 2 = 2 a x1 x 2 , u 3 = 0 , the displacement gradient
2 x1a
r
ui
r
( x u)ij =
= 2ax2
x j
0

Ciudad Real - Spain
Draft
10ax2
2ax1
0
0
0
430
Decomposing additively the displacement gradient in a symmetric part (the linear strain
tensor - ij ) and an antisymmetric part (the infinitesimal spin tensor- ij ) we obtain:
u i
= ij + ij
x j
where
u j
1 u
ij = i +
x j xi
2
2 x1 a
= 4ax 2
2x a
1 1
= 2ax 2
2
4ax 2
2ax1
0
10ax 2
2ax1
0
0 2 x1 a
0 + 10ax 2

0 0

2ax 2
2ax1
0
0
0
and
u j
1 u
ij = i
x j xi
2
2x a
1 1
= 2ax 2
2
10ax 2
2ax1
0
0 2 x1 a
0 10ax 2

0 0

2ax 2
2ax1
0
0 0
0 = 6ax 2
0 0

6ax 2
0
0
0
0
b) The principal strains.

2 x1 a
4ax 2
4ax 2
=0
2ax1
(
) (4ax )
(2 x a ) = (4ax )
2 x1 a
=0
1 = 2 x1 a 4ax 2
2 x1 a = 4ax 2
2 = 2 x1 a + 4ax 2
2 x1a + 4ax2
0
0
=
0
2 x1a 4ax2 0 (principal strains)
ij
0
0
0
Since the strain and stress share the same principal space we can use the equation
ij = 4 x1a ij + 2 ij to obtain the principal stresses:
0
1 0 0
2 x1a + 4ax2
0 1 0 + 2
= 4 x1a ij + 2 = 4 x1a
0
2 x1a 4ax2
ij
ij
0 0 1
0
0
0
0
4 x1a + 2 (2 x1a + 4ax2 )
0
0
=
4 x1a + 2 ( 2 x1a 4ax2 )
0
0
4 x1a
0
0
c) Starting from the equilibrium equations:
r r
+ b = 0 Indicial ij , j = 0 i
{
r
=0
and by expanding the above, we obtain:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
11,1 + 12, 2 + 13,3 = 0
21,1 + 22, 2 + 23,3 = 0
31,1 + 32, 2 + 33,3 = 0
431
11 12 13
+
+
=0
x1
x 2
x 3
21 22 23
+
+
=0
x1
x 2
x3
31 32 33
+
+
=0
x1
x 2
x3
and considering that kk = 4 x1 a , the stress tensor components ij = kk ij + 2 ij

becomes ij = 4 x1 a ij + 2 ij , thus
11 = 4 x1 a 11 + 2 11 = 4 x1 a + 2 (2 x1 a ) = 4 x1 a ( + )
12 = 4 x1 a 12 + 2 12 = 2 ( 4ax 2 ) = 8 ax 2
13 = 0
or we can use:
ij = 4 x1a ij
1
= 4 x1a 0
1 0 0
11 12 13
0 1 0 + 2
+ 2 ij = 4 x1a
12 22 23
0 0 1
13 23 33
0 0
0
8 ax2
2 x1a 4ax2 0 4 x1a ( + )
+ 2 4ax
= 8 ax
1 0
2ax1 0
4 x1a ( + )
0
2
2
0
0 1
0
0
0
0
4 x1a
Then, the first equilibrium equation becomes:

11 12 13
+
+
=0
x1
x 2
x3
4a( + ) 8 a = 0
+ = 2
= = G
G (3 + 2G )
, which was obtained by means of the relationships
+G
E
E
and = G =
. Then, we conclude that:
=
(1 + )(1 2)
2(1 + )
In addition, note that E =
E=
G (3 + 2G ) G (3G + 2G )
= 2.5G
=
+G
G+G
Problem 7.9
Consider a soil made up of a linear elastic material. At a point in the soil the volumetric
deformation is V = 2 10 3 , the shear deformation is 12 = 3 10 3 and the normal
strain is 11 = 0 . The soil is subjected to a state of plane strain according to the plane
x1 x 2 .
a) Obtain the Cartesian components of the infinitesimal strain tensor. Obtain the
principal strains, and the directions where they occur.

Ciudad Real - Spain
Draft
432
1
4
b) Assuming that the elastic constants are E = 50MPa and = . Obtain stress tensor
components and the principal stresses. Obtain the maximum normal and shear
stresses.
c) Obtain the strain energy density (per unit volume).
NOTE: Remember the expression for the stress tensor:
= Tr ( )1 + 2
with =
E
E
, =
(1 + )(1 2)
2(1 + )
Solution:
a) The infinitesimal strain tensor components are given by:
ij = 3 10 3
0
0
3 10 3
22
0
L
The volumetric deformation DV V = I = 11 + 22 + 33 = 2 10 3 22 = 2 10 3 .
ij = 3
0
3 0
0
plane strain
2 0 10 3 ij =
3
0
0
3
3
10
2
The principal strains

0 3
=0
3 2
2 + 2 3 = 0
1 = 1
2 = 3
thus
1 = 1 10 3
2 = 3 10 3
b)
1 10 3
=
ij
0
0
3
3 10
1
y
xy
yy

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
433
Mohrs circle in strain

S =
(10 3 )
2
( N = 0; S = 3 )
2
I =1
III = 3
N 10 3
( N = 0; S = 2)
( N = 2; S = 3 )
Note that the radius is R = (1 (3)) / 2 = 2 . Then:

tg(2) =
3
1
2 = arctg( 3 )
b) Applying ij = Tr ( ) ij + 2 ij , where =
= 30
E
= 20MPa ,
(1 + )(1 2)
E
= 20 MPa , Tr ( ) = 2 10 3 . Then:
2(1 + )
3 0
2 0 10 3
0
0
40
0
0
0
3 0
0
+ 40 3 2 0 10 3 MPa
0
=
40
1 24
4 3
0
=103 Pa
0
0
0
0
40
0
1 0 0
0 1 0 + 2 3
ij = Tr ( )
0
0 0 1
Thus:
40
0
40 3
0 kPa
ij = 40 3 120
0
0
40
As the material is isotropic, the stress and strain share the principal space. In addition, the
eigenvalues of and can be related to each other as follows.
By replacing the value of = Tr ( )1 + 2 into the definition of the eigenvalueeigenvector, we obtain:

Ciudad Real - Spain
Draft
434
n = n
(Tr ( )1 + 2 ) n = n
Tr ( )1 n + 2 n = n
Tr ( )n + 2 n = n
2 n = n Tr ( )n
2 n = ( Tr ( ) )n
Tr ( )
n
n =
n = n
Then:
=
Tr ( )
2
= 2 + Tr ( )
And the eigenvalues of can be obtained as follows:

(1
(
) I = 2 1) + Tr ( ) = (40 10 6 ) (1 10 3 ) + (20 10 6 ) (2 10 3 ) = 0
(
(
2 ) II = 2 2 ) + Tr ( ) = (40 10 6 ) (0) + (20 10 6 ) (2 10 3 ) = 40 10 3 Pa
(
(
3) III = 2 3) + Tr ( ) = (40 10 6 ) (3 10 3 ) + (20 10 6 ) (2 10 3 ) = 160 10 3 Pa
We can also use the equation = Tr ( )1 + 2 in the principal space:

40
0
0
0
0
0
1 0 0
+ 40 0 3 0 10 3 MPa = 0 160
= 0
40
0
0 kPa
ij
4 3
1 24
0
=103 Pa
0
0 0 0
40
40
0
0
Mohrs circle in stress:
160
40
N (kPa)
1
2
c) The strain energy density is e = : . We can use the principal space in order to
obtain the strain energy, i.e.:
1
0
0
2
0
0
0
0
= 0 160
0 10 3 Pa

3 0
40
0

0
0

Ciudad Real Spain
Draft
1
0
0
2
0
0 1 0 0
0 = 0 3 0 10 3

3 0 0 0

m
m

7 LINEAR ELASTICITY
435
1
1
2
2
m
N m
J
1
= (0)(1) + (160 10 3 )(3 10 3 ) + ( 40 10 3 )(0) = 240 Pa = 240 2
= 240 3
m
2
m m
m
e = ij ij = (11 + 2 2 + 3 3 )
Problem 7.10
A solid is subjected to state of plane strain, and at one point the infinitesimal strain tensor
3 0
2
3 10 0 10 3
ij =
0
0
0
Consider that the material has an isotropic linear elastic behavior defined by the Youngs
modulus E = 10MPa and Poissons ratio = 0.25 .
a) Obtain the volumetric deformation and the deviatoric part of the strain tensor;
b) Obtain the principal strains and the principal directions;
c) Obtain the Cauchy stress tensor components;
d) Obtain the maximum and minimum normal stress;
e) It is known that the material fails when the tangential stress exceeds the value 40 kPa .
Check if the material fails.
Solution:
a) Volumetric deformation ( V ):
V = I = Tr ( ) = ( 2 10) 10 3 = 12 10 3
Additive decomposition of the strain tensor = sph + dev , where the spherical part is
given by:
sph
ij
0
4 0
Tr ( )
0 4 0 10 3
=
ij =
3
0
0 4
and the deviatoric part is:

dev
ij
= ij
sph
ij
2
0
3 0 4 0
2 3 0
0 4 0 10 3 = 3 6 0 10 3
= 3 10 0
0
0 0 4
0
0 4
0
0
b) The principal strains are obtained by means of the characteristic determinant:

2
3
=0
3
10
2 + 12 + 11 = 0
Solving the quadratic equation:

(1, 2 ) =
(12) (12) 2 4(1)(11)

2(1)
12 10
2
(1) = 1.0
( 2 ) = 11
Then, the principal strains are:

Ciudad Real - Spain
Draft
436
1 = 1.0 10 3
2 = 11.0 10 3
The principal directions can be obtained by solving ( ij ij )n (j ) = 0 i
(i, j = 1,2)
The principal direction associated with the eigenvalue (1) = 1.0 :

(1)
(1)
(1)
(1)
(
3
2 (1)
n11) 0 n1 + 3n 2 = 0 n1 = 3n 2
(
=
10 (1) n (1) 0 3n11) 9n (1) = 0

3
2

2
2
(
restriction n11) + n (21) = 1 , with that we obtain (3n (21) ) 2 + n (21) = 1 n (21) =
(
n11) =
1
10
, and
3
10
The principal direction associated with the eigenvalue (1) = 11.0 :

( 2)
(
( 2)
3
2 (11)
n12) 0 9n1 + 3n 2 = 0
( 2) = ( 2 )
(
10 (11) n 2 0 3n1 + n ( 2 ) = 0 n ( 2) = 3n12 )
3
2
2
(
(
with the restriction n12) + n (22) = 1 , we obtain n12) =
10
, and n (22) =
3
10
We summarize as follows:
1 = 1 10 3
2 = 11 10 3
1 = 0
n i(1) =
10
principal
direction
principal
direction
principal direction
n i(3)
1
i
n ( 2) =
10
= [0 0 1]
1
10
3
10
c) The Cauchy stress tensor components are given by:

ij = Tr ( ) ij + 2 ij
where =
E
E
= 4MPa , = G =
= 4 MPa , Tr ( ) = 12 10 3 :
(1 + )(1 2 )
2(1 + )
3
0
1 0 0
2
0 1 0 + 2 (4) 3 10 0 10 3 MPa
ij = 4 (12)
0 0 1
0
0
0
24
0
64
24 128
=
0 kPa
0
48
0
As the material is isotropic, the principal directions of the stress and strain coincide. The
principal stresses can be obtained by working on the principal space
= Tr ( ) ij + 2 :
ij
ij

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
437
1 0 0
1 0 0
0 1 0 + 2 (4) 0 11 0 10 3 MPa
= 4 (12)
ij
0 0 1
0
0 0
0
0
56
0
=
136
0 kPa
0
48
0
d) By considering that I = 48kPa , II = 56kPa , III = 136kPa , the Mohrs circle in

stress can be obtained as follows:
S (kPa )
S max = 44
II = 56
III = 136
I = 48
N (kPa)
The maximum shear stress can be obtained as follows:

S max =
I III
(48) (136)
=
= 44kPa
2
2
Then, the material fails.

Problem 7.11
The stress state at a point of the continuum is represented by the Cauchy stress tensor
components:
26 6 0
ij = 6
9 0 kPa
0
0 29
Consider an isotropic linear elastic material.

a) Obtain the principal invariants of ;
b) Obtain the spherical and deviatoric parts of ;
c) Obtain the eigenvalues and eigenvectors of ;
d) Draw the Mohrs circle in stress. Obtain the maximum normal and tangential stress.
e) Considering a small deformation regime and taking into account that the elastic
mechanical properties are = 20000kPa and = 20000kPa ( , are the Lam constants).
Obtain the infinitesimal strain tensor;
f) Obtain the eigenvalues and eigenvectors of .
Ciudad Real - Spain
Draft
438
Solution:
a) The principal invariants
I = 12 10 3
II =
0 29
10 6 +
26
29
26 6
10 6 +
10 6 = 763 10 6
III = det ( ) = 7830 10 9

dev
sph
The spherical and deviatoric parts ij = ij + ij :
1
3
The mean stress: m = ii =
( 29 26 + 9)
= 4 10 3 Pa
3
hyd
ij
dev
ij
sph
ij
4 0 0
= 0 4 0 kPa
0 0 4
6
0 30 6 0
26 4
6
5 0 kPa
=
94
0 = 6

0
0 25
0
29 4 0
By solving the characteristic equation we obtain the eigenvalues:

I = 29kPa , II = 10kPa , III = 27 kPa :
The eigenvectors:
I = 29kPa
II = 10kPa
n i(1) = [0 0 1]
principal direction

principal direction

III = 27kPa
n i( 2 ) = [0.1644 0.98639 0]
principal direction

i
n (3) = [0.98639 0.1644 0]
S (kPa)
S max =
29 (27)
= 28
2
N (kPa)
III = 27
I = N max = 29
II = 10
ij = Tr ( ) ij + 2 ij

Ciudad Real Spain
inverse
Draft
ij =
1
Tr ( ) ij +
ij
2 (3 + 2 )
2
7 LINEAR ELASTICITY
where
439
= 5 10 9 ( Pa ) 1 , Tr ( ) = 1.2 10 4 ( Pa)
2 (3 + 2 )
26 6 0
1 0 0
0 1 0 + 2.5 10 8 6
9 0 10 3
ij = (5 10 )(1.2 10 )
0
0 0 1
0 29
26 6 0
1 0 0
+ 2.5 10 5 6
5
9 0
= 6 10 0 1 0
0
0 0 1
0 29
0
7.1 1.5
1.5 1.65
0 10 4
=
0
0 6.65
As the material is isotropic, the stress and strain have the same principal directions. We
work on the principal space to obtain:
=
ij
Tr ( ) ij +
ij
2 (3 + 2 )
2
0 66.5 0
0
1 0 0
29 0
+ 2.5 10 5 0 10
= 0 19
= 6 10 0 1 0
0
0 10 5
0 0 1
0 0 27 0
0 73.5
Problem 7.12
Show that the constitutive equations in stress, for an isotropic linear elastic material, can be
represented by the set of equations:
dev = 2 dev
Tr ( ) = 3Tr ( )
where = G is the shear modulus, and is the bulk modulus.

Solution:
= Ce :
= [1 1 + 2 I] :
= Tr ( )1 + 2
= dev + sph = Tr ( )1 + 2 ( dev + sph )

Tr ( )
1 = Tr ( )1 + 2 ( dev + sph )
3
Tr ( ) Tr ( )
1
= Tr ( )1 + 2 dev + 2
3
3
2
Tr ( )
1
= +
Tr ( )1 + 2 dev
3
3
dev +
dev
dev
The trace of the stress tensor:

Tr ( ) = : 1 = [Tr ( )1 + 2 ] : 1 = Tr ( )3 + 2 Tr ( ) = (3 + 2 )Tr ( )

Ciudad Real - Spain
Draft
440

2
Tr ( )
dev
1
dev = +
Tr ( )1 + 2
3
3
(3 + 2 )Tr() 1
2
dev
dev = +
Tr ( )1 + 2
3
3
(3 + 2 )Tr() 1 + 2 dev
2
dev = +
Tr ( )1
3
44444424443 443
1
4
4
4
=0
To the equations dev = 2 dev we must add the constraint:
(3 + 2 )Tr()
2
1=0
+
Tr ( )1
3
3
2
Tr ( )
+
1=0
Tr ( )1
3
3
Tr ( )1 = 3 +
Tr ( )1
3
Tr ( )1 = 3 Tr ( )1
or
Tr ( ) = 3 Tr ( )
33
23
13
13
23
12
12
ij = Tr ( ) ij + 2 ij
13
11
+
m
22
12

Ciudad Real Spain
12
dev
22
12
dev
dev
ij = 2 ij
dev
33
23
12
m
23
23
dev
11
Tr ( ) ij = 3 Tr ( ) ij
33
23
13
13
22
11
13
dev
33
Draft
23
13
13
23
12
dev
22
12
dev
11
7 LINEAR ELASTICITY
441
Starting from the constitutive equation in stress for an isotropic linear elastic material
= ( ) = Tr ( )1 + 2 , and by considering the linear regime the relationship
= ( ) = ( sph + dev ) = ( sph ) + ( dev ) holds, where:
( sph ) = Tr ( sph )1 + 2 sph
2
Tr ( )
Tr ( )
Tr ( )
sph = Tr
11 + 2
1 = Tr ( )1 + 2
1 = ( +
) Tr ( )1 = Tr ( )1
3
3
3
3
Tr ( )
1 = Tr ( )1
3
Tr ( )1 = 3 Tr ( )1
( dev ) = Tr ( dev )1 + 2 dev = 2 dev
1 24
4 3
=0
[ ]
Note that Tr ( sph ) = Tr sph
= Tr[ ] holds.
NOTE: Note that for an isotropic material if we have a purely spherical state of
compression:
p
p>0
p
0
p 0
0 p 0
ij =
0
0 p
Tr ( ) = 3 p < 0
E
, we can conclude
3(1 2)
that: if > 0.5 this implies that < 0 and as consequence Tr ( ) > 0 , i.e. an expansion,
we have that Tr ( ) = 3 Tr ( ) < 0 , and considering that =
which has no physical meaning for a compression state in isotropic materials. With that we
conclude that < 0.5 .
Problem 7.13
A parallelepiped of dimensions a = 0.10m , b = 0.20m , c = 0.30m , (see Figure 7.6), which is
made up of an elastic material with Poissons ration = 0.3 and Youngs modulus
E = 2 10 6 N / m 2 . Said parallelepiped is introduced into a cavity of width b whose walls are
very rigid, so that two opposite faces of the parallelepiped are in contact with the cavity
walls.
Once the parallelepiped is this position the temperature is raised in T = 30 C .
1) Calculate the values of the principal stresses at the point of the parallelepiped.
2) Find the strain components.
Consider that the thermal expansion coefficient of the material is 1.25 10 5 C 1 .

Ciudad Real - Spain
Draft
442
Solution:
As the solid can deform freely according to the directions x and z , hence the normal
stresses are x = z = 0 . The solid is restricted to move according to the y -direction,
hence y = 0 :
y =
[ ]
1
1
y ( x + z ) + T = y + T = 0
E
E
y = ET
thus:
y = ET = 2 10 6 1.25 10 5 (30) = 750
N
m2
The Cauchy stress tensor components:

0
0
0
0 750 0 Pa
ij =
0
0
0
x = z =
b)
y
E
+ T = 4.875 10 4
The strain tensor components:

0
4.875 0
0
ij =
0
0 10 4
0
0 4.875
Data:
a = 0.10m
b = 0.20m
c = 0.30m
E = 2 10 6 N / m 2
= 0 .3
T = 30 C
= 1.25 10 5 C 1
c
y
b
x
Figure 7.6
Problem 7.14
Consider a container with very rigid walls, and a squared cross section with dimensions
0.10 0.10m . In the interior of said container is placed a synthetic rubber block whose
Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
443
dimensions are 0.10 0.10 0.5m , as shown in Figure 7.7(a). The rubber block fits
perfectly into the rigid container.
The elastic characteristics of the rubber are E = 2.94 10 6 N / m 2 and = 0.1 .
Above the rubber is poured 0.004m 3 of mercury, whose mass density is 13580kg / m 3 .
a) Obtain the height H that reach the mercury, (see Figure 7.7(b));
b) The stress state at a generic point of the rubber block.
Hypothesis: 1) the weight of the rubber is negligible. 2) Consider the acceleration of gravity
equal to g = 10m / s 2 . 3) Between the rubber block faces and the container walls there is no
friction.
z
Rigid walls
Rigid walls
Mercury
Rubber
L 0 = 0 .5 m
y
a)
b)
Figure 7.7
Solution:
First we calculate the total force due to the weight of mercury on the rubber:
kg
kgm
m
F = V g = 0.004(m 3 ) 13580 3 10 2 = 543.20 2 N
s
m
s
The normal stress according to the z -direction is given by:

z =
543.20
F
N
=
= 54.320 10 3 2
A
(0.1 0.1)
m
According to the directions x and y the rubber does not deform, hence x = y = 0

Ciudad Real - Spain
Draft
444
x = ( y + z )
y = ( x + z )
1
x ( y + z ) = 0
E
1
y = y ( x + z ) = 0
E
x =
y = ( x + z )
{[
y = ( y + z ) + z
y =
( 2 + )
z =
z = 6035.55 Pa = x
2
(1 )
(1 )
The normal strain according to the z -direction is given by:
1
z ( x + y )
E
1
=
{ 54320 0.1[2(6035.55)]} = 0.0180656
2.94 10 6
z =
b) The length variation of the rubber block is:

L = L0 z = 0.5 ( 0.018656 ) = 0.00903m
The height H reached by the mercury is given by:

H = hmer + ( L0 L)
where hmer is given by:

Vmer = b 2 hmer = 0.004
hmer =
0.004
= 0.4m
0.1 0.1
thus:
H = hmer + ( L0 L)
= 0.4 + (0.5 0.00903)
= 0.891m
Problem 7.15
By means of a material test in the laboratory, it was obtained the following relationships:
1
21
31
x = x +
E
E y + E z
1
2
3
32
12
1
y =
E x + E y + E z
2
1
3
1
13
23
z =
E x + E y + E z
1
2
3
(7.69)
where 12 = 0.2 , 13 = 0.3 , 23 = 0.25 , E1 = 1000 MPa , E 2 = 2000MPa , E3 = 1500MPa .

Knowing that the analyzed material is orthotropic, obtain 21 , 31 , 32 .
Solution:
The constitutive matrix for orthotropic material has the following format:
Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
C11
C
12
C
[C ] = 13
0
0
C12
C22
C23
0
0
0
0
0
0
0
0
C44
0
0
C23
C33
0
0
0
0
C66
0
0
C13
0
C55
0
445
Orthotropic symmetry
9 independent constants
(7.70)
By restructuring (7.69) we obtain

1

E
xx 1
12
yy E
zz 1
= 13
2 xy
E1
2 yz
0
2 xz
0
0
21
2
1
2
23
2
0
0
0
31
E 0
3
32
E 0
3
1
E
3
0
C44
0
0
0
0
0
0
0
0
C55
0
0
xx
0 yy

zz
0 xy

0 yz

0 xz
C66
(7.71)
Then, for orthotropic material it fulfills that:

21 12
E = E

2 1
31 13
E = E

3 1
32
E
3
23
=
E
2
with that we obtain

21 12
=
E2
E1
21 =
E 2 12
2000 0.2
=
= 0.4
1000
E1
31 13
=
E3
E1
31 =
E 3 13 1500 0.3
= 0.45
=
1000
E1
32
= 23
E3
E2
32 =
E 3 23 1500 0.25
=
= 0.1875
2000
E2
Problem 7.16
Given an isotropic linear elastic material whose elastic properties are E = 71 GPa ,
G = 26.6 GPa , find the strain tensor components and the strain energy density at the point
in which the stress state, in Cartesian basis, is represented by:
20 4 5
ij = 4 0 10 MPa
5 10 15
Solution: Poissons ratio can be obtained by means of the equation: G =

Ciudad Real - Spain
Draft
E
2(1 + )
446
=
11 =
E
1 = 0.335
2G
1
[11 ( 22 + 33 )] = 1 9 [20 0.335 (0 + 15)]10 6 = 211 10 6
E
71 10
1
[ 22 (11 + 33 )] = 1 9 [0 0.335 (20 + 15)]10 6 = 165 10 6
E
71 10
1
1
= [ 33 (11 + 22 )] =
[15 0.335 (20 + 0 )]10 6 = 117 10 6
E
71 10 9
22 =
33
12 =
1+
1 + 0.335
12 =
( 4 10 6 ) = 75 10 6
9
E
71 10
1+
1 + 0.335
13 =
(5 10 6 ) = 94 10 6
E
71 10 9
1+
1 + 0.335
=
23 =
(10 10 6 ) = 188 10 6
9
E
71 10
13 =
23
thus:
211 75 94
ij = 75 165 188 10 6
94
188 117
Then, the strain energy density for an elastic material is obtained by the equation:
1
2
1
2
e = : Ce : = :
indicial
1
2
e = ij ij
Next, by considering the symmetry of the tensors and , the strain energy density can
be calculated as follows:
1
[1111 + 22 22 + 33 33 + 212 12 + 2 23 23 + 21313 ]
2
1
= [( 211)( 20) + ( 165)(0) + (117 )(15) + 2( 75)( 4) + 2(188)(10) + 2(94)(5) ] = 5637 .5 J / m 3
2
e=
We can also obtain the strain energy density by using the equation:
e =
1
1
1
1
2
2
I II dev =
I + J2
6(3 + 2 )
2
6(3 + 2 )
2
and if we consider that I = 3.5 10 7 ; II = 2.4933 1014 ; 5.3804 10 10 Pa ; = G , we

can obtain e 5638 .03 J / m 3 . Note that any discrepancies in the numerical results of e
are due to numerical approximations.
Problem 7.17
Find the strain energy density in terms of the principal invariants of .
Solution:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
1
2
1
2
e = : = : [Tr ( )1 + 2 ] =
=
=
[Tr ( )]
[Tr ( )]2
2
+ : =
[Tr ( )]
+ Tr ( ) =
447
Tr ( )
2
2
1
1:3 + :
Tr ( )
+ Tr ( T )
[Tr ( )]2
2
+ Tr ( 2 )
We can add and subtract the term [Tr ( )]2 without altering the above outcome:
e=
[Tr ( )]2
2
1
( + 2 )[Tr ( )]2 [Tr ( )]2 Tr ( 2 )
2
+ [Tr ( )] + Tr ( 2 ) [Tr ( )] =
2
Finally, if we consider that the principal invariants of the strain tensor are I = Tr ( ) ,
II =
1
[Tr ()]2 Tr ( 2 ) , we can obtain:
2
1
2
2
e = ( + 2 )I 2 II = e ( I , II )
Problem 7.18
The responses of a liner thermoelastic solid due to two actions are known, namely:
r
r
r (I )
r ( II )
r (I )
r ( II )
r
r
I (b ( I ) , t * on S ; u* ; on S u ; T ( I ) ) and II (b ( II ) , t * on S ; u* ; on S u ; T ( II ) ) .
Obtain the response of the system formed by I + II and justify.
Solution:
As we have a linear regime the following is satisfied:
r r
r
b = b ( I ) + b ( II )
T = T ( I ) + T ( II )
r
r ( I ) r ( II )
t* = t* + t*
r
r ( I ) r ( II )
u* = u* + u*
The same is true for the fields:

r r
r
u = u ( I ) + u ( II )
= ( I ) + ( II )
= ( I ) + ( II )
Starting from the governing equations of linear thermoelastic equilibrium we have:

] [
r
r
r
r
r
r
r
r
r
r
x + b = x ( ( I ) + ( II ) ) + (b ( I ) + b ( II ) ) = x ( I ) + b ( I ) + x ( II ) + b ( II ) = 0

Ciudad Real - Spain
Draft
448
] [
r
r
1 r r (I )
1 rr
r
r
x u + ( x u ( I ) ) T + x u ( II ) + ( x u ( II ) ) T
2
2
r ( II )
r (I )
r ( II ) T
r
r
r
+ xu
+ xu + xu
= ( I ) + ( II ) =
{[
{ [
]}
] [
r
1
r
x u( I )
2
r
r
r
rT
T
1 r r
1 rr
r
r
= x u ( I ) + u ( II ) + x (u ( I ) + u ( II ) ) = x u + [ x u] =
2
2
=
]} {
] [
The constitutive equations in stress:

= C e : + MT
where M is the thermal stress tensor

= C e : + MT = C e : ( ( I ) + ( II ) ) + M (T ( I ) + T ( II ) )
= (C e : ( I ) + MT ( I ) ) + (C e : ( II ) + MT ( II ) )
= ( I ) + ( II )
Then, we conclude that all the conditions are met. Then, we can apply the principle of
superposition to linear thermoelastic problem, as expected, since we are in the linear
regime.
Problem 7.19
Let us consider a length rod equal to L = 7.5m , whose diameter is equal to 0.1m , which is
made up of a material whose properties are: E = 2.0 10 11 Pa and = 20 10 6
Initially the rod has a temperature equal to 15 C which later rises to 50 C .
1
.
C
1) Considering that the rod can expand freely, calculate the total elongation of the rod, L ;
2) Now assume that the rod can not expand freely because concrete blocks have been
placed at its ends, (see Figure 7.8(b)). Find the stress in the rod.
Hint: Consider the problem in one dimension.
x
L = L(1) + L( 2 )
L(1)
T
L
L
L( 2)
b)
a)
Figure 7.8: Rod under thermal effect.
Solution: 1) To obtain the elongation, we pre-calculate the thermal strain according to the
rod axis direction ij = (T ) ij . Since this is a one-dimensional case, we need only
consider the normal strain component according to the x -direction, 11 = x , then:
11 = x = 20 10 6 (50 15) = 7 10 4
Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
449
Then, the total elongation, L = L(1) + L( 2) , is obtained by solving the integral:

L
L = x dx = x L = 7 10 4 7.5 = 5.25 10 3 m
0
Note that as the rod can expand freely, it is stress-free.

2) If the ends can not move, there will be a homogeneous stress field equal to:
x = E (T ) ij = E " x " = 2.0 1011 7 10 4 = 1.4 10 8 Pa
Note that in the case 2) there is no strain, since L = 0 . Moreover, it is the same as when
the initial length is equal to L + L in which we apply compression stress in order to
obtain a final length equal to L .
Problem 7.20
Consider an isotropic linear elastic material with the following thermo mechanical
properties E = 10 6 Pa (Youngs modulus), = 0.25 (Poissons ratio), = 20 10 6 C 1
(Thermal expansion coefficient).
Consider that that at one point of the solid the stress tensor components are given by:
12 0 4
ij = 0 0 0 Pa
4 0 6
a) Obtain the principal stresses and directions of the stress tensor; Obtain the maximum
shear stress.
b) Obtain the strain related to paragraph (a). And find the principal strains and directions.
c) Obtain the strain energy density.
d) If the solid undergoes a change in temperature T = 50 C , what is the final strain state
at this point?
e) We can say that we are dealing with a state of plane stress?
Solution:
We obtain the eigenvalues by solving the characteristic determinant. Note that we already
know an eigenvalue 2 = 0 which is associated with the direction n i( 2) = [0 1 0] . Then,

to obtain the remaining eigenvalues, it is sufficient to solve:
12
=0
2 18 + 56 = 0
Solving the quadratic equation we obtain:

(1,3) =
18 324 224
2
1 = 14
3 = 4
14 0 0
' ij = 0 0 0 Pa
0 0 4
And the eigenvectors (unit vectors) are given by:

Ciudad Real - Spain
Draft
450
eigenvector

1 = 14
2 = 0
eigenvector

3 = 4
eigenvector

n i(1) =
0
5
n i( 2 ) = [0 1 0 ]
1
n i(3) =
5
1
= [0.8944 0 0.4472]
5
2
= [0.4472 0 0.8944]
5
Making the change of nomenclature such that I > II > III , we have I = 14 , II = 4 ,
III = 0 .
S ( Pa )
S max = 7
N ( Pa)
II = 4
III = 0
I = 14
We can obtain the maximum normal stress as follows:

S max =
I III
(14) (0)
=
= 7 Pa
2
2
The Cauchy stress tensor components are given by:

ij = Tr ( ) ij + 2 ij
inverse
ij =
Tr ( ) ij +
ij
2
2 (3 + 2 )
1
Remember that = C e : , and the reciprocal form = C e : .

where
E
= 4 10 5 Pa ,
(1 + )(1 2 )
=G =
E
= 4 10 5 Pa ,
2(1 + )
Tr ( ) = 18 ,
= 2.5 10 7 Pa
2 (3 + 2 )
0
4
1 0 0
12
+ 1.25 10 6 0 10 0
6
ij = 4.5 10 0 1 0
0 0 1
4
0
6
0
5
10.5
0
=
4.5 0 10 6
5
0
3

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
451
For an isotropic linear material the principal directions of the stress and strain match. The
principal strains can be obtain by means of the equation ij =
Tr ( ) +
ij
2 (3 + 2 )
2
in the principal space, i.e.:
1 0 0
14 0 0
+ 1.25 10 6 0 0 0 =
6
= 4.5 10 0 1 0
ij
0 0 1
0 0 4
1
2
0
0
13
0 4.5 0 10 6
0
0
0 .5
1
2
The strain energy density is given by e = : = ij ij . We can use the principal space
to obtain the energy, i.e.:
0
0
13
0 4.5 0 10 6
=
ij
0
0
0 .5
14 0 0
= 0 0 0
ij
0 0 4
With that we can obtain:
e =
J
1
1

= [ 11 11 + ] = 92 10 6 3
ij ij
33 33
2
2
m
Using the principle of superposition:

ij = ij ( ) + ij (T )
= ij ( ) + T ij
we can obtain:
0
5
0
5
10.5
1 0 0 1010.5
0
10 6 + 20 10 6 (50) 0 1 0 = 0
=
4 .5 0
995.5
0 10 6
ij
5
0 0 1 5
0
3
0
1003
The principal directions of the infinitesimal strain tensor are the same as the stress tensor
principal directions.
e) We can not say that we are dealing with a state of plane stress, since we do not know
information about how stresses vary in the continuum, i.e. the stress field. Remember that
the state of plane stress is considered when the stress tensor field is independent of one
direction.

Ciudad Real - Spain
Draft
452
Problem 7.21
Let us consider a bar to which at one end we apply a force equal to 6000 N as shown in
Figure 7.9. Find x , y , z , and the length change of the bar. Let us consider that the bar is
made up of a material whose properties are: Youngs modulus: E = 10 7 Pa ; Poissons ratio:
= 0 .3 .
1m
100 m
1m
y, v
y =
x, u
6000
11
F = 6000 N
z, w
Figure 7.9
Solution: Using the normal strain expressions we can obtain:
)]
(0.3)(6000 )
1
x y + z = y =
= 0.00018
E
E
10 7
y 6000
1
y = y ( x + z ) =
=
= 0.0006
E
E
10 7
x =
z =
)]
z x + y = y = 0.00018
E
E
The total change in cross-sectional dimensions is u = w = 0.00018 1 = 1.8 10 4 m , and

the total change in length is v = 0.0006 100 = 6.0 10 2 m .
Problem 7.22
A strain gauge (or strain gage) is a device used to obtain the strain in only one direction.
Consider a strain rosette that contains three strain gauges where there are 45 internal angles,
(see Figure 7.10). At one point we have calculated the following strain values:
x = 0.33 10 3
x = 0.22 10 3
y = 0.05 10 3
Find the maximum shear stress at the point in question.

Then consider an isotropic linear elastic material with the following mechanical properties:
E = 29000 Pa (Youngs modulus); = 0.3 (Poissons ratio).

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
453
a)
Obtain the eigenvalues (principal strains) and eigenvectors (principal directions) of
the strain tensor;
b)
Obtain the eigenvalues (principal stresses) and eigenvectors (principal directions) of
the stress tensor.
Hint: Consider the state of plane strain.
y
x
strain gauge
45
45
Figure 7.10: Strain rosette.

Solution:
Firstly, we have to obtain the strain tensor components in the system x, y, z and to do so
we will use the coordinate transformation law in order to obtain the component xy = 212 .
Remember that in two-dimensional cases, the normal component in a new system is given
by (see Problem 1.98 in Chapter 1):
11 =
11 + 22 11 22
+
cos( 2) + 12 sin( 2)
2
2
The above equation was obtained by means of the transformation law, (see Chapter 1 of
the textbook), which in engineering notation becomes:
x =
x + y
2
x y
2
cos( 2) +
xy
2
sin( 2)
Then, xy can be obtained as follows:

xy =
( + y ) ( x y )
2
x x
cos( 2) = 0.16 10 3
sin( 2)
2
2
thus
0.33 0.08 0
ij = 0.08 0.05 0 10 3
0
0
0
Then, the stress components can be evaluated as follows:

Ciudad Real - Spain
Draft
454
E
(1 2 ) x + y = 12 .0462 Pa
(1 + )(1 2 )
E
(1 2 ) y + x = 3.5692 Pa
y =
(1 + )(1 2 )
E
E
xy =
xy = 1.7846 Pa ; z =
x + y = 4.684 Pa
2(1 + )
(1 + )(1 2 )
x =
Additionally, the maximum shear stress is given by:

max
x + y
=
+ 2 = 4.5988 Pa
xy
a) The characteristic equation for the strain tensor (2D) is:

2 0.28 2.29 10 2 = 0
( 10 3 )
Then, by solving the above equation we can find the eigenvalues (principal strains) given
by:
1 = 0.346155 10 3
2 = 0.06615528 10 3
Then, the eigenvectors of the strain tensor are:

1 0.9802 0.1979 0
2 0.1979 0.9802 0
3
0
0
1
b) Given the stress tensor components, we have:

12.0462 1.7846
ij = 1.7846 3.5692
0
0
0
0 Pa
4.684
We now obtain the characteristic determinant and in turn the eigenvalues (principal
stresses) 1 = 12.40654 , 2 = 3.208843 . Additionally, the eigenvectors of the stress tensor
are:
1 0.9802
2 0.1979
3
0
0.1979
0.9802
0
0
0
As expected, the eigenvectors of stress and strain are the same; since we are working with
isotropic linear elastic material.
b) Alternative solution for the stress tensor components:
Knowing the strain tensor components:
0.33 0.08 0
ij = 0.08 0.05 0 10 3
0
0
0
We apply the constitutive equation: ij = Tr ( ) ij + 2 ij , where the Lam constants are

given by:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
455
E
= 16.7307692 10 3 Pa
(1 + )(1 2)
E
=
= 11.15384615 10 3 Pa
2(1 + )
and Tr ( ) = 0.27999972 10 3 0.28 10 3 , with that ij = Tr ( ) ij + 2 ij becomes:

1 0 0
11 12 13
1 0 0
0.33 0.08 0
0 1 0 + 2
= Tr ( ) 0 1 0 + 2 0.08 0.05 0 10 3
ij = Tr ( )
12 22 23
0 0 1
13 23 33
0 0 1
0
0
0
0
12.0461 1.784615
1.784615 3.5692
0 Pa
=
0
0
4.6846
As the material is isotropic, the tensors and share the same principal directions, then
we can use the same equation ij = Tr ( ) ij + 2 ij on the principal space, i.e.:
1 0 0
1 0 0
0 1 0 + 2 0
0
= Tr ( )
2
ij
0 0 3
0 0 1
1 0 0
0.346155
0
0
0 1 0 + 2
0
0.0662 0 10 3
= Tr ( )
0 0 1
0
0
0
0
0
12.40752
0
3.20783
0 Pa
=
0
0
4.6846

Ciudad Real - Spain
Draft
456
Problem 7.23
A strain gauge (or strain gage) is a device used to obtain the strain in only one direction.
Consider a strain rosette that contains three strain gauges arranged according to a equilateral
triangle, (see Figure 7.11), and records the strain values according to the directions x1 , x1
and x1 .
x2
x1
x1
30
60
30
60
60
x1
Figure 7.11
The strain calculated according to the directions x1 , x1 and x1 are respectively:
11 = 4 10 4
11 = 1 10 4
11 = 4 10 4
Obtain 22 = y , 212 = xy , 22 y . Show that 11 + 22 = 11 + 22 .
Hypothesis: Consider a state of plane strain.

Solution:
Using the component transformation law for the second-order tensor and considering the
2D-plane state, we obtain that:
11 =
11 + 22 11 22
+
cos( 21 ) + 12 sin( 21 )
2
2
(7.72)
11 =
11 + 22 11 22
+
cos( 2 2 ) + 12 sin( 2 2 )
2
2
(7.73)
where 1 = 60 and 2 = 120 . Then, by combining the above equations, it is possible to

eliminate 12 , i.e.:
22 =
11 + 11 11 = 4.66667 10 4
3
2
Once the value 22 = 4.66667 10 4 is obtained, we can replace it into the equation (7.72)
and we obtain:
xy = 2 12 =
1
3
(411 11 3 22 ) = 3.46410 10 4

Ciudad Real Spain
Draft
12 = 1.73205 10 4
7 LINEAR ELASTICITY
457
To obtain 22 , first we obtain the angle of rotation with respect to x1 , which is

3 = 60 +90 = 150 , thus:
22 =
11 + 22 11 22
+
cos( 2 3 ) + 12 sin( 2 3 ) = 0.33333 10 4
2
2
Checking that:
11 + 22 = 11 + 22 = 0.66667 10 4
As expected, since the trace is an invariant.

Problem 7.24
Consider a dam section in which the displacement is known and given by:
y, v
u ( x, y ) = 4 x 2 y 2 + 2 xy + 2
v ( x, y ) = 4 y 2 x 2 + 2 xy + 5
x, u
This structure consists of a material with following elastic properties: E = 100 MPa ,
G = 35 .7 MPa , = 0.4 . Assuming that the structure is under a small deformation regimen:
a) Find the stress field, b) For the given displacement field, show whether the equilibrium
equations are satisfied or not.
Solution:
a) We can calculate the strain tensor components as:
x =
u
= 8 x + 2 y
x
y =
v
= 8 y + 2 x
y
xy =
u v
=0
+
y x
which in matrix form is:

8 x + 2 y
ij =
0
0
0
8 y + 2 x 0
0
0
b) For a dam, as we have seen, we can adopt the approximation of plane strain condition:
0 x
0 .6 0 .4 0 8 x + 2 y
1
E

1
0 y = 357 .1428 0.4 0.6 0 8 y + 2 x MPa
y = (1 + )(1 2 )
1
xy
0
xy
0 0.3
0
0
0
x
4 x 2 y

y = 357.1428 2 x 4 y MPa
xy

z =
E
x + y = 357 .1428 [( 8 x + 2 y ) + ( 8 y + 2 x )]
(1 + )(1 2 )

Ciudad Real - Spain
Draft
458
Then, the equilibrium equations become:

x xy xz
+
+
+ b x = 0
y
z
x
xy y yz
+
+
+ b y = 0
y
z
x

yz z
xz +
+ b z = 0
+
z
y
x
4 + 0 + 0 + 0 0 (fails)
(fails)
0 4 + 0 + 0 0
z
0 + 0 +
+0=0
z
So, the given displacement field does not satisfy the equilibrium equations.
Problem 7.25
Let us consider a regular quadrangular prism whose material is characterized by
E = 27.44 10 5 N / cm 2 (Youngs modulus) and = 0.1 (Poissons ratio). The side length of
the square section is a = 20cm . In both bases of the prism are placed two plates perfectly
smooth and rigid, they are joined together by four cables of area section A1 = 1cm 2 and
Youngs modulus equals to E1 = 19.6 10 6 N / cm 2 . Initially the length of the prism is equal
to l = 1m , (see Figure 7.12).
On two opposite sides of the prism we apply a compressive pressure p = 7350 N / cm 2 .
1. Obtain the stress on the cable C ;
2. Obtain the principal stresses in the prism;
3. Obtain the volume variation experienced by the prism.
a) Reference configuration
b) current configuration
z
z
l = 1m
a
x
Figure 7.12

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
459
Solution:
Verify that the cable and the prism deform in the same way according to the z -direction,
thus:
P = C
z
z
On the cable it fulfills that:

C = E1 C
z
C =
z
C
E1
The stress field in the prism is given by:
0 0
0 p
P
0
ij =
0 0 4 C A1
a2
The strain in the prism according to the direction z :

P =
z
)]
1
1
z x + y =
E
E
4 c A1
+
a2
Applying that P = C :
z
z
P = C
z
z
1
E
4 c A1
+
a2

p = C
E1
After some algebraic manipulations we obtain the stress on the cable:

c =
E1 pa 2
( Ea 2 + 4 E1 A1 )
0.1 19.6 10 6 7350 20 2

N
= 4900 2
5
2
6
(27.44 10 20 + 4 19.6 10 1)
cm
The normal stress in the prism according to the z -direction becomes:

P =
z
4 C A1
P
ij
4 4900 1
N
= 49 2
2
20
cm
0
0
0
N
0 7350
=
0
cm 2
0
49
0
The volume variation experienced by the prism is obtained as follows:

V = V V0
where V = I is the linear volumetric deformation (small deformation regime):

V = I = x + y + z =
4
x + y + z
E
(1 2 ) = 2.12857 10 3
and V0 = 4 10 cm is the initial prism volume, thus:

V = V V0 = (2.12857 10 3 )(4 10 4 ) = 85.1428cm 3

Ciudad Real - Spain
Draft
460
Problem 7.26
Two parallelepiped of the same material and the same shape a b c are placed on either
side of a rigid flat plate attached thereto by their sides a c . Both parallelepipeds, together
with the plate, are introduced into a cavity such as indicated in Figure 7.13. The walls of the
cavity are flat, rigid and perfectly smooth.
We apply the pressures (force per unit surface area) p1 and p 2 on the upper faces of the
prisms as indicated in Figure 7.13.
Consider the Youngs modulus E and the Poissons ratio .
a) Obtain the principal stresses in both prisms;
b) Obtain the block edge length variations.
z
plate
a
p1
p2
cavity
c
y
Figure 7.13
Solution:
Prism 1
(x1) = 0
(y1)
(z1) = p1
(x2) = 0
(y2)
(z2) = p 2
Prism 2
For compatibility of stress:
(y1) = (y2) = y

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
461
(y1) + (y2 ) = 0
(
[
] [
[ + p ] + [
)]
)]
1 (1)
1
y (x1) + (z1) + (y2 ) (x2 ) + (z2 ) = 0
E
E
(1)
(2)
y z + y z = 0
+ p2 = 0
thus
y =
( p1 + p 2 )
2
Prism 1:
(x1) = 0
(y1) =
( p1 + p 2 )
2
(z1) = p1
(x2 ) = 0
(y2) =
( p1 + p 2 )
2
(z2 ) = p 2
Prism 2:
The strains in each prism are given by:

Prism 1:
(x1) =
)]
(y1)
)]
)]
(x2 ) =
)]
(y2 )
)]
)]
(z1)
1 (1)
[( p1 + p 2 ) + 2 p1 ]
x (y1) + (z1) =
E
2E
1
= (y1) (x1) + (z1) =

( p1 p 2 )
E
2E
1
1 2
= (z1) (x1) + (y1) =
( p1 + p 2 ) 2 p1
E
2E
Prism 2:
(z2 )
1 ( 2)
[( p1 + p 2 ) + 2 p 2 ]
x (y2 ) + (z2 ) =
E
2E
1
= (y2) (x2 ) + (z2 ) =

( p 2 p1 )
E
2E
1
1 2
= (z2) (x2 ) + (y2 ) =
( p1 + p 2 ) 2 p 2
E
2E
The edge variations:

Prism 1
a
[( p1 + p 2 ) + 2 p1 ]
2E
b
= (y1) b =
( p1 p 2 )
2E
c 2
= (z1) c =
( p1 + p 2 ) 2 p1
2E
Prism 2
a
[( p1 + p 2 ) + 2 p 2 ]
2E
b
= (y2 ) b =
( p 2 p1 )
2E
c 2
= (z2 ) c =
( p1 + p 2 ) 2 p 2
2E
a (1) = (x1) a =
a ( 2 ) = (x2) a =
b (1)
b ( 2 )
c (1)

Ciudad Real - Spain
c ( 2 )
Draft
(7.74)
462
Problem 7.27
A gravity dam of triangular cross section is made up of concrete with specific weight equal
to
5
, where is the specific weight of water. The shape and dimensions of the cross
2
section are indicated in Figure 7.14, and the stress field in the dam (state of plane strain) is
given by:
11 = x 2
( x1 3 x 2 )
2
= x1
22 =
12
1
4
Consider: Poissons ratio: = ; Youngs modulus E .

a) Obtain the graphical representation of the surface force (traction vector) acting on the
face AB due to the ground reaction;
b) Obtain the principal stresses at the points A and B . Starting from the Mohrs circle in
stress, obtain the extreme values of the stresses at the respective points.
c) Obtain the strain field in the dam.
x1
O
= g
45
g -acceleration of gravity
- mass density
[ ] =
kg m
N
= 3
3
2
m s
m
x2
Figure 7.14
NOTE: Although in the literature is known as the specific weight, also known as the
unit weight, in reality is the module of the body force per unit volume, i.e.
[]
r
r
r
r
N
m
= p = b = g , where b is the body force per unit mass b =
= 2 . Remember that
kg s
in the International System of Units (SI) the term specific is related to per unit mass,
which is not the case of , the correct term would be the weight density, since the term
density is related to per unit volume.
Solution:
a) The stress and strain fields in the dam:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
x 2
ij = x1
x1
0
( x1 3 x 2 ) 0
2
0
33
463
11
ij = 12
12
22
0
0
0
We obtain the surface force by means on the traction vector t (n) = n . For the side AB
whose normal unit vector is n i = [0,1,0] , we obtain:
t 1 ( AB ) x 2
( AB )
t 2
= x1
t ( AB )
3
0
0 0 x1
x1
( x1 3 x 2 ) 0 1 = ( x1 3 x 2 )
2
0
33 0
0
The surface force on the base of the dam:

h
( AB )
3h
2
t1
t2
(according to x1 -direction)
( AB )
(according to x 2 -direction)
x1
O
45
2
B
(n )
x2
b) Note that 33 is already a principal stress. Starting from = Tr ( )1 + 2 we can

obtain 33 , i.e.:
ij = Tr ( ) ij + 2 ij 33 = Tr ( ) 33 + 2 33 33 = Tr ( )
The term Tr ( ) can be obtained by means of the double scalar product between
= Tr ( )1 + 2 and the second-order unit tensor, thus:

Ciudad Real - Spain
Draft
464
: 1 = Tr ( )1 : 1 + 2 : 1
Tr ( ) = 3Tr ( ) + 2 Tr ( ) = [3 + 2 ]Tr ( )
Tr ( ) =
+ 22 + 33
Tr ( )
= 11
3 + 2
3 + 2
Then, the component 33 is defined as follows:

33 = Tr ( ) =
33
3 + 2
3 + 2
33 =
(11 + 22 + 33 )
(11 + 22 )
3 + 2
33 1
3 + 2 = 3 + 2 (11 + 22 )
33 =
2( + )
(11 + 22 )
By replacing the values of 11 and 22 , we obtain:

33 =
(11 + 22 ) = x 2 + ( x1 3x 2 ) = [x1 5 x 2 ] = [x1 5 x 2 ]
2
2( + )
2
8
where we have considered =
.
2( + )
The stress state at the point A( x1 = 0; x 2 = h) is given by:
(
ijA)
x 2
= x1
x1
( x1 3 x 2 )
2
0

h

0
= 0

[x1 5 x 2 ] 0

8

0
0
3h
2
0

0 1 0

3
0 =0
2

5h
0
0
8

0 h
5
8
Note that this space is already the principal space.

Mohrs circle in stress at the point A :
S ( h)
S max = 0.4375
1.5
0.625
N (h)
The stress state at the point B( x1 = h; x 2 = h) is given by:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
(
ijB )
x 2
= x1
x1
( x1 3 x 2 )
2
0

h

0
= h

[x1 5 x 2 ] 0

8

0
465
(h 3h)
2
0
1 1 0

0
= 1 1 0 h
1

0
0
[h 5h]
2
8
The principal stresses at the point B( x1 = h; x 2 = h) are given by:

1
1
=0
1
1
(1 ) 2 1 = 0
(1 ) 2 = 1
( 1 ) = 1
1 = 2
2 = 0
S ( h)
S max = 1
0.5
N (h)
c) We can obtain the expression of the strain field by starting from the equation:
= Tr ( )1 + 2 :
= Tr ( )1 + 2
2 = Tr ( )1
=
Tr ( )1
2
2
Remember that we have obtained that Tr ( ) =

=
Tr ( )
, then:
3 + 2
1
1
Tr ( )1 =
Tr ( )1
2
2
2
2 (3 + 2 )
We can also express the above equation in terms of E and :
=G =
E=
(1 + )
E
1
=
2(1 + )
2
E
(3 + 2 )
1
1
=
+
(3 + 2 ) E ( + )
=
=
2 (3 + 2 ) 2 E ( + ) E
Then:

Ciudad Real - Spain
Draft
466
Tr ( )1
2
2 (3 + 2 )
(1 + )
Tr ( )1
E
E
The trace of is given by:

Tr ( ) = 11 + 22 + 33

5
= ( x 2 ) + ( x1 3 x 2 ) + ( x1 5 x 2 ) = ( x1 5 x 2 )
2
8
8
1
4
With that we obtain the strain tensor components, with = , as follows:

ij =
x 2
5
ij =
x1
4E
0
5
5
ij
( x1 5 x 2 ) ij
4E
32 E
x1
( x1 3 x 2 )
2
0
1 0 0
5
( x1 5 x 2 ) 0 1 0
0
32 E
0 0 1
[x1 5 x 2 ]
1
x1
0
8 ( x1 3 x 2 )
5
1
=
x1
(3 x1 + 7 x 2 ) 0
4E
8
0
0
0
Problem 7.28
A metallic cube with sides a = 0.20m is immersed in the sea at a depth z = 400m .
Knowing the Youngs modulus of the metal E = 21 1010 Pa , and the Poissons ratio
= 0.3 , calculate the volume variation that is experienced by the cube. Consider the
acceleration of gravity equals to g = 10m / s 2 .
Hypothesis: Although the mass density varies with temperature, salinity, and pressure
(depth), consider that the mass density of seawater equal to = 1027 kg / m 3 .
Solution:
Because of the depth and cube dimensions we can take as a good approximation that the
whole cube is subjected to the same pressure, (see Figure 7.15).

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
h = 400m
467
Figure 7.15
F
, where A is the area and F can be obtained by
A
means of the Newtons second law F = ma = V g (weight of water column). Then:
The pressure can be obtained by p =
p=
Ah g
kg
kg m
m
F V g
=
=
= gh = 1027 3 10 2 400m = 4.108 10 6 2 2 = 4.108 10 6 Pa
A
A
A
m
s
m s
The stress tensor components in the cube are given by:

0 4.108
0
0
p 0
0 p 0 = 0
4.108
ij =
0 MPa

0
4.108
0 p 0
0
As we have only normal stress components and the material is isotropic, only normal
strains appear:
z = y = x =
)]
1
1
[ 4.108 0.3 ( 4.108 4.108 )] 10 6
x y + z =
10
E
21 10
Thus
z = y = x = 7.82 10 6
In small deformation regime the linear volumetric deformation is equal to the trace of the
infinitesimal strain tensor:
V
L
= DV V = Tr ( )
V0
V = V0 Tr ( ) = 0.2 3 (2.346 10 5 ) = 1.8768 10 7 m 3
where we have considered that Tr ( ) = 2.346 10 5 .

Problem 7.29
A solid cylinder of radius 0.05m and height 0.25m is made up of a material whose
mechanical properties are: E = 3 10 4 MPa (Youngs modulus) and = 0.2 (Poissons
ratio). Said cylinder is placed between two pistons, which can be considered infinitely rigid,
and all of this is enclosed in a hermetically sealed container as shown in Figure 7.16.

Ciudad Real - Spain
Draft
468
The container is filled with oil, and by suitable mechanism, the fluid pressure is raised to
the value 15MPa . By operating the mechanical press, we apply a total axial force of
F = 2.35619 10 5 N (piston force+pressure) on the bases of the cylinder.
At a generic point of the body:
a) Obtain the stress tensor components;
b) Obtain the strain tensor components;
c) Obtain the displacement field components ( u , v , w ).
F
y
0.25m
Corte AA
F
0.1m
Figure 7.16: Triaxial compression test.

Solution:
a) The stress tensor components
z =
F
2.35619 10 5
=
= 30MPa
A
(0.05) 2
x = y = 15MPa
Thus:
0
15 0
0
ij =
15
0 MPa
0
0
30
b) For a isotropic linear elastic material, the normal stress only produce normal strain, then:

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
469
)]
x = E x y + z
y = y ( x + z )
E
z = E z x + y
)]
By substituting the values of the variables we obtain the following strain tensor
components:
0
2 0
0 2 0 10 4
ij =
0
0 8
c) The displacement field

As we are in small deformation regime, the following is fulfilled:
x =
u
x
y =
v
y
z =
w
z
Integrating and obtaining the constants of integration we finally obtain the displacement
field:
u = 2 10 4 x
v = 2 10 4 y
w = 8 10 4 z
Problem 7.30
A hexahedron with sides 0.1m is made up of a material whose mechanical properties are
represented by the Lam constants: = 8333.33MPa , = 12500 MPa .
A deformation is imposed to the material as shown in Figure 7.17, in which all faces
remains planar, the faces AEFB and DHGC become parallelogram and the remaining
faces continue squares:
a)
b)
c)
d)
e)
f)
g)
Obtain the displacement field;

Obtain the strain field;
Obtain the stress field;
Obtain the stress in the center of the hexahedron;
Obtain the principal strains in the center of the hexahedron;
Obtain the principal stresses in the center of the hexahedron;
Obtain the actions performed by the testing machine on the faces ABFE and
BCGF .

Ciudad Real - Spain
Draft
470
z
H
G
tg( ) = 0.001
C = C
D = D
A = A
B = B
Figure 7.17: The deformed hexahedron.

Solution:
a) According to Figure 7.17 we can verify that there are only shear strain components.
Moreover we can also verify that there is no displacement in the directions x and z , then
u = 0 , w = 0 . By means a triangle analogy we can obtain the displacement v :
tg( ) = 0.001 =
v
z
v( z ) = 0.001z
The displacement field:

u = 0
v( z ) = 0.001z
w = 0
b) By considering the strain tensor components:
x
1
ij = xy
2
1
2 xz
1
xy
2
y
1
yz
2
u
1

xz
x
2
1 v u
1
yz = +
2
2 x y
z 1 u

+
2 x z
1 v u
+
2 x y
v
y
1 v
+
2 y z
1 u
+
2 x z
1 v
+
2 y z
we can conclude that x = y = z = xy = xz = 0 and the component yz is given by:

yz =
x
1
ij = xy
2
1
2 xz

Ciudad Real Spain
1
xy
2
y
1
yz
2
v w
= 0.001
+
z y
1
xz
2
0
0
0
1
0
yz =
0
0.0005
0 0.0005
0
Draft
7 LINEAR ELASTICITY
471
E
v(z )
c) The stress field = Tr ( )1 + 2

Considering Tr ( ) = 0 , = 8333.33MPa , = 12500MPa , we obtain:
0
0
0
0
0 0
0
MPa = 0 0 12.5 MPa
ij = 2 (12500)
0
0.0005
0 0.0005
0 12.5 0
0
e, f) The principal strains:
0.0005
=0
0.0005
2 = 0.0005 2
2 = +0.0005
= 0.0005
3 = 0.0005
Remember that in the small deformation regime, the stress and strain share the same
principal directions, then we work in the principal space to obtain the principal stresses by
using = Tr ( )1 + 2 , i.e.:
0
0
0
0
0 0
0 0.0005
MPa = 0 12.5
0
0 MPa
= 2 (12500)
ij
0
0 0
0
0.0005
12.5
g) To obtain the total force acting on one surface, we multiply the surface force by the area
of the corresponding face. The surface force is obtained by means of the traction vector
t (n) = n . For the face ABFE the unit vector is given by n i = [1,0,0] , thus:
t 1 ( ABFE ) 0 0
0 1 0
( ABFE )

= 0 0 12.5 0 = 0
t 2
t ( ABFE ) 0 12.5 0 0 0

3
For the face BCGF , the unit vector is given by n i = [0,1,0] , thus
t 1 ( BCGF ) 0 0
0 0 0
( BCGF )

= 0 0 12.5 1 = 0 MPa
t 2
t ( BCGF ) 0 12.5 0 0 12.5


Ciudad Real - Spain
Draft
472
If we do the same procedure for the other faces we obtain the representation of the surface
forces on the faces as indicated in Figure 7.18:
z
C = C
D = D
A = A
B = B
Figure 7.18: The surface forces in the hexahedron.

Problem 7.31
Consider the prism as indicated in Figure 7.19, we apply the forces F1 = 10 N and F2 = 2 N
as indicated in said figure. The prism edge lengths are: AB = 4cm , AD =
Consider the following material properties: E = 2.5 10 6
= 0.25 (Poissons ratio), and = 5 10 8
N
cm 2
10
cm , AA = 2cm .
3
(Youngs modulus),
1
(thermal expansion coefficient).
C
a) Obtain the principal stresses; b) Obtain the traction vector on the plane . Is it on that
plane where the maximum shear acts? Justify your answer. c) Obtain the values of the
forces F1 and F2 to be applied to guarantee that in the solid there is no displacement
according to the directions x1 and x2 , when the prism is subjected to a temperature
variation of T = 20 C .

Ciudad Real Spain
Draft
7 LINEAR ELASTICITY
x2
473
A2
A1
F2
F1
F1
x1
60
F2
x3
Figure 7.19
a) The stress field
A1 = 8.0 , A2 = 4
10
3
F1
A
1
ij = 0
F2
A2
0
0
0
0
1.25
N
= 0
0
0.15 0
cm 2
0
0
0 0
whose values are the principal stresses, since there is no shear stresses.
b)
x2
r
t (n)
D
x1
60
x3

Ciudad Real - Spain
Draft
474
r
3
The unit vector components are: ni = ;

; 0 . Then, the traction vector t (n) is
given by:
t (n) = n
t i(n)
t (n) = ij n j
i
0
0
1.25
= 0
0.15 0
0
0
0
2 1.0825
1
= 0.075
2
0
0
The normal stress component is:

r
N = t (n) n = t i(n) n i
= [1.0825 0.075 0]
2
1
= 0 .9
2
0
The tangential stress component is obtained as follows:

r
t (n)
= 2 + 2
N
S
S =
r
t (n)
2
N
where
r
t (n)
1.0825
r (n) r (n)
(n ) ( n )
= t t = t i t i = [1.0825 0.075 0] 0.075 = 1.1775
Thus:
S =
r
t (n)
2 = 1.1775 0.9 2 = 0.60621778

N
The Mohrs circle in stress is drawn as follows
III = 0.15

Ciudad Real Spain
I = 1.25
Draft
N ( N / cm 2 )
7 LINEAR ELASTICITY
475
We verify that for any point in the solid, the maximum tangential stress is on the plane
2
defined by the unit vector n i = ;

2
max =
2
; 0 and the maximum tangential stress is:
2
I III
= 0.7 > S
2
c) We consider the following strain field:
1+
Tr ( )1 + T 1
E
E
1+
ij =
ij Tr ( ) ij + T ij
E
E
=
For the particular case Tr ( ) = 11 + 22 we have:

0 0 0
11
0 0 0 = 1 + 0
E
0 0 33
0
0
22
0
0
1 0 0
+ T Tr ( ) 0 1 0
0
0 0 1
0
Then, we set the following system:
1+
1+
11 = 0 = E 11 + T E Tr ( ) = E 11 + T E (11 + 22 )
1+
1+
= 0 =
22 + T (11 + 22 )
22 + T Tr ( ) =
22
E
E
E
E
By solving the above set of equation we obtain:

11 = 22 =
ET
N
= 3.33333 2
(1 )
cm
Then, the forces are given by:

F1 = 11 A1 = 26.66666 N
F2 = 22 A2 = 44.44444 N

Ciudad Real - Spain
Draft
476

Ciudad Real Spain
Draft
11 Introduction to Fluids
11.1 Solved Problems
Problem 11.1
Demonstrate whether the following statements are true or false:
a) If the velocity field is steady, then the acceleration field is also;
b) If the velocity field is homogeneous, the acceleration field is always equal to zero;
c) If the velocity field is steady and the medium is incompressible, the acceleration is always
zero.
Solution:
r r
v ( x , t) r
a) In a steady velocity field we have
= 0 whereby the acceleration field becomes:
t
r
v ( x , t )
&
a i = vi = i
+ vi ,k v k = v i ,k v k
1 24
4t 3
=0i
r r
r r v ( x )
r r r r
r r r r
&
r
r
a=v =
+ x v ( x) v ( x) = x v ( x) v ( x)
14 244
4
3
t
Independent of time
Then, assumption (a) is TRUE.
r r
b) A homogeneous velocity field implies that v ( x , t ) = v (t ) , whereby:

r r
r r
r r
r r
r r v ( x , t )
v ( x , t )
&=
r v ( x , t ) v ( x, t ) =
a=v
+x
1 24
4 3
t
t
=0
Then, assumption (b) is FALSE.
r r
r r
c) A steady velocity field implies that v ( x , t ) = v ( x ) and an incompressible medium means

r r
r
that x v ( x , t ) = 0 , so, we can conclude that:
r r
r r r r
r r r r
r r v ( x )
&=
r
r
a=v
+ x v ( x) v ( x) = x v ( x) v ( x)
t
Then, assumption (c) is FALSE.
478
Problem 11.2
Show the Navier-Stokes-Duhem equations of motion:
&
vi = b i p,i + ( * + * )v j , ji + *vi , jj
Navier-Stokes-Duhem
r
r
r
r equations of motion
&
r
r
r
r
v = b x p + ( * + * ) x ( x v ) + * x 2 v
(11.1)
Solution:
The Navier-Stokes-Duhem equations of motion are a combination of the equations of
r
r
&
r
&
motion x + b = v ( ij , j + b i = vi ) and the constitutive equations:
= p1 + * Tr (D)1 + 2 *D
ij = p ij + * ijD kk + 2 *Dij
(11.2)
r
Then, the Cauchy stress tensor divergence ( x ) can be evaluated as follows:
ij , j = ( p ij + * ijD kk + 2 *Dij ) j = p, j ij + * ijD kk
,j
+ 2 *Dij , j
= p,i + * ijD kk , j + 2 *Dij , j = p,i + *D kk ,i + 2 *D ij , j
(11.3)
Note that, in this formulation, we are considering that the material is homogeneous, i.e.
*, j = *, j = 0 .
In addition, by considering 2D ij = vi , j + v j ,i and 2D kk = v k ,k + v k , k = 2v k ,k , we obtain:
2D ij , j = vi , jj + v j ,ij = vi , jj + v j , ji
D kk , j = vk ,kj
(11.4)
whereby the equation in (11.2) becomes:
ij , j = p,i + *D jj ,i + 2 *Dij , j = p, i + *v j , ji + * vi , jj + v j , ji
= p,i + + v j , ji + vi , jj
*
(11.5)
&
Then, by substituting the equation in (11.5) into the equations of motion ( ij , j + b i = vi ),
we obtain the Navier-Stokes-Duhem equations of motion for homogeneous materials.
NOTE 1: The explicit form of the equation (11.1) is presented as follows:

&
( * + * )v j , ji + vi , jj + b i p,i = vi
&
( * + * )(v1,1i + v2, 2i + v3,3i ) + * (vi ,11 + vi , 22 + vi ,33 ) + b i p,i = vi
( * + * )(v1,11 + v2, 21 + v3,31 ) + * (v1,11 + v1, 22 + v1,33 ) p,1 + b1 = v1

&
*
*
*
&
( + )(v1,12 + v2, 22 + v3,32 ) + (v2,11 + u 2, 22 + v2,33 ) p, 2 + b 2 = v2
*
*
*
&
( + )(v1,13 + v2, 23 + v3,33 ) + (v3,11 + v3, 22 + v3,33 ) p,3 + b 3 = v3
or:

Ciudad Real Spain
Draft
11 INTRODUCTION TO FLUIDS
479
*
2v 2v 2v
v1 v2 v3
*
+ * 21 + 21 + 21 + b1 p,1 = v1
&
+
+
( + )
x
x1 x1 x2 x3
x3
x2
2v
2 v2 2 v2
v1 v2 v3
*
*
+ b 2 p, 2 = v2
+ * 22 +
&
+
+
+
( + )
2
2
x
x2 x1 x2 x3
x2
x3
2
2
2
( * + * ) v1 + v2 + v3 + * v3 + v3 + v3 + b p = v
2
&3
3
,3
2
2
x
x3 x1 x2 x3
x3
x2
NOTE 2: We have proven in Problem 1.105 (Chapter 1) that the following is true:
r
r r
r
r
( a) = ( a) 2 a
indicial
ilq qjk a k , jl = a j , ji a i , jj
Then, we can obtain
r
r
r rr
r
2r
r
r
r
r
r
r
x ( x v ) x v = x ( x v ) x ( x v )
indicial
vi , jj = v j , ji ilq qjk vk , jl
with which the equation (11.1) can also be written as follows:

&
(* + * )v j , ji + *vi , jj + bi p,i = vi
&
(* + * )v j , ji + * (v j , ji ilq qjk vk , jl ) + bi p, i = vi
&
(* + 2 * )v j , ji *ilq qjk vk , jl + bi p, i = vi
and the equivalent in tensorial notation:
r
r
r
r
&
r
r
r
r
r
(* + * )[ x ( x v )] + * [ x ( x v )] + b x p = v
r
r
r
r r
r
r
&
r
r
r
r
r
r
r
(* + * )[ x ( x v )] + * x ( x v ) x ( x v ) + b x p = v
r
r
r
r
r
r
&
r
r
r
r
r
(* + 2 * )[ x ( x v )] * x ( x v ) + b x p = v
In the Cartesian System we have:

r
v = vi e i = v1e1 + v2 e 2 + v3e 3
r r
v
r
r v
v
v
v v
( v ) rot (v ) = (rot (v ) )i e i = 3 2 e1 + 1 3 e 2 + 2 1 e 3
x
x
x2
x1
x3
2
3
1
4243
4243
4243
1 4 r 4
1 4 r 4
1 4 r 4
=(rot (u ) )1
r
r
r
r r (rot (v ) )3 (rot(v ) )2
( v ) =
x2
x3
=(rot (u) )2
=(rot (u) )3
r
r
r
r
(rot (v ) )1 (rot (v ) )3
(rot (v ) )2 (rot (v ) )1
e1 +
e 2 +
e 3
x1
x1
x2
3
r
r
(rot (v ) )3 (rot (v ) )2 v2 v1 v1 v3
x2 x1 x2 x3 x3 x1
x2
x3
r
r
r
r r
(rot (v ) )1 (rot (v ) )3 v3 v2 v2 v1
( v ) i =
=
x1
x3
x3 x2 x3 x1 x1 x2
r
r
(rot (v ) )
(rot (v ) )1 v
2
v v3 v2

1 3
x x
x1
x2
1 3 x1 x2 x2 x3
NOTE 3: If we are dealing with heterogeneous material, we have:

ij = p ij + * ijD kk + 2 *D ij
ij , j = ( p ij + * ijD kk + 2 *D ij ) , j = p, j ij + (*D kk ), j ij + 2( *D ij ), j
ij , j = p,i + (*D kk ) ,i + 2( *Dij ), j
Taking into account that 2D ij = vi , j + v j ,i and D kk = vk ,k , the above equation becomes:
ij , j = p,i + (*D kk ),i + 2( *D ij ) , j = p,i + (*vk ,k ),i + * (vi , j + v j ,i ) , j
whereby
Ciudad Real Spain
Draft
480
&
ij , j + bi = vi
&
p, i + (*vk , k ), i + * (vi , j + v j ,i ) , j + bi = vi
(11.6)
Note that
r
r
vk ,k = Tr (v ) = ( v )
&
v
v
v v
Dvi vi vi
&
v j = i + i v1 + i v2 + i v3 , and its explicit components are:
vi =
+
=
x3
x2
t x1
Dt
t x j
v1 v1
v
v
+
v1 + 1 v2 + 1 v3
x2
x3
t x1
v
v
v
v
&
ai = vi = 2 + 2 v1 + 2 v2 + 2 v3
x2
x3
t x1
v3 v3
v
v
+
v1 + 3 v2 + 3 v3
x2
x3
t x1
* (vi , j + v j ,i ) , j =
* (vi , j + v j ,i )
x j
* (vi ,1 + v1,i ) +
* (vi , 2 + v2,i ) +
* (vi ,3 + v3,i )
x1
x2
x3
* (v1, 2 + v2,1 ) +
* (v1,3 + v3,1 )
2 * (v1,1 ) +
x3
x2
x1
*
*
*
*
(vi , j + v j ,i ) , j =
(v2,1 + v1, 2 ) +
(v2,3 + v3, 2 )
2 ( v2 , 2 ) +
x3
x2
x1
* (v3,1 + v1,3 ) +
* (v3, 2 + v2,3 ) +
2 * (v3,3 )
x3
x2
x1
The three equations in (11.6), ( i = 1,2,3 ), are explicitly given by:
*
r
&
2 * (v1,1 ) +
( v ) +
* (v1, 2 + v2,1 ) +
* (v1,3 + v3,1 ) + b1 p,1 = v1
x1
x1
x2
x3
*
r
*
&
2 * ( v2 , 2 ) +
( v ) +
(v2,1 + v1, 2 ) +
* (v2,3 + v3, 2 ) + b 2 p, 2 = v2
x2
x1
x2
x3
*
r
*
&
( v ) +
(v3,1 + v1,3 ) +
* (v3, 2 + v2,3 ) +
2 * (v3,3 ) + b3 p,3 = v3
x3
x1
x2
x3
(11.7)
or
*
r
&
( v ) + 2 * (v1,1 ) +
* (v1, 2 + v2,1 ) +
* (v1,3 + v3,1 ) + b1 p,1 = v1
x1
x2
x3
*
r
&
( v ) + 2 * (v2, 2 ) +
* (v2,1 + v1, 2 ) +
* (v2,3 + v3, 2 ) + b 2 p, 2 = v2
x2
x1
x3
*
r
&
( v ) + 2 * (v3,3 ) +
* (v3,1 + v1,3 ) +
* (v3, 2 + v2,3 ) + b3 p,3 = v3
x3
x1
x2
(11.8)

Ciudad Real Spain
Draft
481
NOTE 4: To obtain the dynamic equations of motion in a rotating fluid on the sphere, we
have to consider the rotation of the Earth and the curvature, and the acceleration for a
fixed system was obtained in Problem 4.32:
r
r
r r
r r
r r
a f = a r + 2( v r ) + T v r + ( x )
r
r
r r
r r
r r
a r = a f 2( v r ) T v r ( x )
r
(11.9)
where (2( v r )) is the Coriolis term, ( T v r ) is the curvature term, ( x ) is the

centrifugal term, and the components of these terms are.
2[3 vr 3 cos( ) 3 vr 2 sin( ) ]
r r
2( v r ) i =
2[3 vr1 sin( )]
2[3 vr1 cos( ) ]
vr1vr 2 tan( ) + vr1vr 3
1 2
T r
( v r ) i = r vr1 tan( ) + vr 2 vr 3
x
vr21 vr22
r
r r
r 2r
( ( r )) i = r
(11.10)
Then, if we want to consider these terms we replace a r into the Navier-Stokes-Duhem

equations.
Problem 11.3
Consider
&
vi = b i p,i + ( * + * )v j , ji + * vi , jj
Navier-Stokes-Duhem
r
r
r
r equations of motion
&
r
r
r
r
v = b x p + ( * + * ) x ( x v ) + * x 2 v
(11.11)
Show the equation of vorticity:

r
r r
* r2 r r
r
+ 2 x ( v ) skew
= 0 The equation of vorticity
x
t
r
r
r
r
r
r
where is vorticity vector and is given by rot (v ) x v .
(11.12)
Solution:
Taking into account the material time derivative of the Eulerian velocity we obtain:
&
vi
r r
r r
r r
r r v ( x , t )
&
r
+ x v ( x, t ) v ( x, t )
a=v =
t
Dvi vi v i
v
=
+
v j = i + vi , j v j
Dt
t x j
t
(11.13)
The resulting components of the algebraic operation vi , j v j are the components of

r r
r
( x v ) v , (see Chapter 1 in Problem 1.106), and it was shown that:

Ciudad Real Spain
Draft
482
r
r r 1
r r
r r
r
r
r
( x v ) v = ( x v ) v + x (v v )
2
r
r r 1
r
r
= ( x v ) v + x (v 2 )
2
r r 1
r
= v + x (v 2 )
2
(11.14)
Then, the Eulerian acceleration can also be represented by:

r r
r r
r v ( x , t ) r r 1 r 2
&=
+ v + x (v )
a ( x, t ) = v
2
t
(11.15)
Taking into account (11.13) and (11.14), the equation in (11.11) becomes:
&
vi = b i p,i + ( * + * )v j , ji + * vi , jj
r
&
r
r
r
r
v = b x p + ( * + * ) x ( x v ) + * x 2 v
r
r
r
v r r 1 r 2
2r
r
r
r
r

+ v + x (v ) = b x p + ( * + * ) x ( x v ) + * x v
2
t
r
r 1
( * + * ) r r r * r 2 r r
v r r 1 r 2
r
+ v + x (v ) b + x p
x ( x v )
v =0
x
t
2
(11.16)
Then we take the curl of the above equation:

r
r 1
r
v r r 1
( * + * ) r r r * r 2 r r
r
r
r
x + v + x (v 2 ) b + x p
x ( x v )
v = 0
2
x
t
(11.17)
Note that the following relationships hold:
[
r
[(
r
r r
r r
r
r
r
r
r
r
r
r
r
r
x x (v 2 ) = 0 , x [ x p ] = 0 , x [ x ( x v )] = 0 ;
r
r
x
r
r r
r r
r
r r
r r
r
r
r
r
r
r
r
v ) v = ( x v )( x v ) + x ( x v ) v x v ( x v )
r r
r r
r r
r r
r
r
r
r
x [ v ] = ( x v ) + [ x ] v x v
r
x
[
]
r
r
r
r
r
r
r
2r
2
2 r
r
r
r
r
r
r
r
r
x x v = x x ( x v ) = x x v = x ;
r
r
r
v r r r
r
x =
x v =
;
t
t t
r
Considering that the field b is conservative, and considering that the curl of any
r r
r
r
conservative vector field is always zero we obtain x b = 0 .
Considering the previous expression, the equation (11.17) becomes:

r
r
r r
r r
r r *
2 r
r
r
r
r
+ ( x v ) + ( x ) v x v
x =0
(11.18)
Note that the following relationships hold:

(v i j ), i = vi , i j + vi j , i
(v i j ), j = v i , j j + vi j , j
v i , i j = (vi j ), i vi j , i
vi , j j = (v i j ), j v i j , j = (vi j ), j
(11.19)
which in tensorial notation become:

Ciudad Real Spain
Draft
r r
r r
r r
r
r
r
( x v ) = x [ v ] ( x ) v
r r
r r
r r
r
r
r
r
( x v ) = x [v ] ( x )v = x
483
[v ]
r
(11.20)
where we have applied the definition that the divergence of the curl of a vector is zero, i.e.
r
r
r
r
r
r
x = x ( x v ) = 0 . Taking into account (11.20), the equation (11.18) becomes:
r
r r
r r
r r * r2 r r
r
r
r
=0
+ ( x v ) + ( x ) v x v
x
t
r
r r
r r
r r
r r * r2 r r
r
r
r
r
=0
+ x [ v ] ( x ) v + ( x ) v x [v ]
x
t
r
r r
r r * r2 r r
r
r
+ x [ v ] x [v ]
=0
x
t
r
r
r r r r *
2 r
r
r
+ x [ v v ]
x =0
t
r
r r
* r2 r r
+ 2 x ( v ) skew
=0
x
t
(11.21)
With that we prove the equation of vorticity given by the equation in (11.12).

Ciudad Real Spain
Draft
484
Problem 11.4
Let us consider a body immersed in a Newtonian fluid. Find the total traction force E
acting on the closed surface S which delimits the volume V . Consider that the bulk
viscosity coefficient to be zero.
r
t (n)
Figure 11.1
Solution: We know that the following holds:
dE i = t i(n) dS
The total traction force is given by the following integral:
E i = t i(n ) dS = ij n j dS = ij , j dV
S
where we have used the relationship ij n j = t i(n) .
2
3
Then, if the bulk viscosity coefficient is zero, we have: * = 0 * = * (Stokes

condition).
Next, by considering the stress constitutive equation for Newtonian fluids, we obtain:
ij = p ij + * ij D kk + 2 *D ij = p ij
D
2 *
ij D kk + 2 *D ij = p ij + 2 * D ij kk ij
3
3
4 244
1 4
3
dev
Dij
dev
ij = p ij + 2 *D ij
Then
dev
E i = ( p ij + 2 *D ij )n j dS
S
and by applying the Gauss theorem, we obtain:

Ei =
( p
ij
dev
+ 2 *D ij

Ciudad Real Spain
,j
dV =
( p
,j
Draft
dev
dev
ij + 2 *D ij , j )dV = ( p ,i + 2 *D ij , j )dV
V
485
where we have considered that *j = 0 j , i.e. * is a homogenous scalar field (homogenous

,
material). Then, the above equation in tensorial notation becomes:
r
E=
r
x
r
p + 2 * x (D dev ) dV
(11.22)
Problem 11.5
Let us consider a fluid at rest which has the mass density f . Prove Archimedes
Principle: Any body immersed in a fluid at rest experiences an upward buoyant force equal to the
weight of the volume fluid displaced by the body.
If mass density in the body is equal to s and the body force (per unit mass) is given by
b i = g i 3 , obtain the resultant force and acceleration acting on the body.
Solution:
r
r
r
In Problem 11.4 we showed that E = x p + 2 * x (D dev ) dV . If the fluid is at rest
D dev = 0 holds, and the thermodynamic pressure is equal to the hydrostatic pressure, i.e.
p = p 0 whereby we have:
r
E=
r
x
p 0 ]dV
(11.23)
r
E
p0
V -volume
r
Ws
x3
x2
x1
Figure 11.2
The weight of the fluid volume displaced by the body is given by:

Ciudad Real Spain
Draft
486
r
W
r
= f bdV
(11.24)
Then, by applying the equilibrium equations we have:

r r
r
x + f b = 0
r
r
x = f b
ij , j + f b i = 0 i
ij , j = f b i
r
r
x ( p 0 1) = f b
r
r
x p0 = f b
(11.25)
( p 0 ij ), j = f b i
p 0,i = f b i
Next, by considering both (11.23) and (11.24), we can conclude that:

r
W
r
r
r
= f bdV = x p 0 dV = E
(11.26)
which thereby proves Archimedes principle.

Now, the body weight, with mass density s , can be obtained as follows:
r
r
W s = s bdV
and the resultant force acting on the body is given by:

r
r
r
r r r
R = E + W s = f bdV + s bdV = ( s f )bdV
Ri = ( s f )b i dV = g ( s f ) i 3 dV =
0
V
V
g ( f s )dV
V
thereby verifying that: if the body has a mass density lower than fluid mass density, e.g. if
r r
the body is a gas, the body rises, i.e. f > s R > 0 , and if not the body falls. Moreover,
r
if we consider that R = m s a , where m s is the total mass of the submerged body, we can
obtain the acceleration of the body as:
a3 =
R3
ms
g (
s )dV
ms
g (
s)
ms
s
dV
s
g (
=
s)
ms
dV
=
g (
s)
NOTE: It is interesting to note that if the medium ( f ) is such that f = 0 we have

a 3 = g , i.e. the acceleration is independent of the mass. Here we have clearly seen, as did
Galileo, by means of a simple experiment, that a freely falling body was independent of the
mass. For example, on the moon where we can consider that the mass density of air is
equal to zero, two bodies with different masses in free fall, e.g. a feather and a hammer, will
have the same acceleration and will reach the moon surface at the same time.

Ciudad Real Spain
Draft
487
Problem 11.6
Prove that the Cauchy deviatoric stress tensor dev is equal to dev , where
ij = p ij + ij .
Solution
If we consider that kk = 3 p + kk we can obtain:
dev
ij = ij
kk
( 3 p + kk )
dev
ij = p ij + ij
ij = ij kk ij = ij
3
3
3
Problem 11.7
Obtain the one-dimensional mass continuity equation for a non-viscous incompressible
fluid flow through a pipeline. Then, consider the volume V between the two arbitrary
cross sections A and B .
B
nB
r
vB
A
nA
r
vA
Figure 11.3
Solution:
D
&
=0.
Dt
r
r
&
Moreover, here we can consider the mass continuity equation + v k ,k = ( x v ) = 0 ,
r
r
where x v = 0 or v k , k = 0 holds. Then, by considering the entire volume we obtain:
In an incompressible medium, the mass density is independent of time
r
x
v dV
=0
k , k dV
=0
(11.27)
and by applying the divergence theorem (Gauss theorem) we obtain:

r
v n dS = 0
k n k dS
=0
(11.28)
Thus:
r
n A dS +
SA
Ciudad Real Spain
n B dS = 0
SB
Next, the velocities at the cross sections
SA
and
Draft
SB
can be expressed as follows:
488
v A = v A n A
v B = vBn B
and by substituting the velocity into the integral we can obtain:
v A n A n A dS + v B n B n B dS = 0
SA
SB
v AS A = vB S B
Problem 11.8
The velocity field of a gas in motion through a pipeline, whose prismatic axis is x 2 , is
defined by its components as follows:
v1 = 0 ;
v 2 = 0.02 x 2 + 0.05 ;
At x 2 = 0 the mass density is equal to 1.5
v3 = 0
kg
. Find at x 2 = 5m .
m3
Solution:
r r
Note that the velocity field is stationary, i.e. v = v ( x ) . From the mass continuity equation
we can obtain:
r
r
r
r
+ x ( v ) = 0 x ( v ) = 0
t
r
Then, we can conclude that v is constant along x 2 -direction, so:
( v 2 )
x2 = 0
= ( v 2 )
x2 =5
v 2 ( x 2 = 0) = 0.02 0 + 0.05 = 0.05 and v 2 ( x 2 = 5) = 0.02 5 + 0.05 = 0.15 , thus:
( v 2 )
x2 = 0
1.5 0.05 = 0.15
= ( v 2 )
x2 =5
( x 2 = 5) = 0.5
kg
m3
r
r
x ( v ) = 0
indicial
( v i ) ,i = ,i v i + v i ,i = 0
v1 v 2 v3
v
vi + i =
x v1 + x v 2 + x v3 + x + x + x
x i
xi 1
2
3
2
3
1
=0
Thus:

x (0.02 x 2 + 0.05) + (0.02 ) = 0
0.02
=
x
(0.02 x 2 + 0.05) 2
By integrating the above equation, we obtain:

Ciudad Real Spain
Draft
489
Ln = Ln(0.02 x 2 + 0.05) + LnC =
C
Ln = Ln
(0.02 x 2 + 0.05)
C
=
(0.02 x 2 + 0.05)
The constant of integration can be obtained by applying the boundary condition, i.e. at
x 2 = 0 = 1.5 , with that we obtain C = 0.075 :
0.075
(0.02 x 2 + 0.05)
x2 =
5
kg
0.075
= 0.5 3
(0.02 5 + 0.05)
m
Problem 11.9
The Cauchy stress tensor components at one point of a Newtonian fluid, in which the bulk
viscosity coefficient is zero, are given by:
6 2 1
ij = 2 9 4 Pa
1 4 3
Obtain the viscous stress tensor components.

Solution:
In the case in which the bulk viscosity coefficient is zero (Stokes condition) we have
p = p = p 0 , and, in addition, we can obtain:
ij = p ij + ij
* = * +
2 *
=0
3
ii = 3 p
p=
ii
( 6 9 3) = 6
=
3
3
Then:
ij = ij + p ij
2 1
6 2 1 6 0 0 0
2 9 4 + 0 6 0 = 2 3 4 Pa
=

1 4 3 0 0 6 1 4
3
Problem 11.10
Determine the conditions needed for mean normal pressure p =
to thermodynamic pressure p for a Newtonian fluid.
kk
= m to be equal
3
Solution:
It was deduced that:

Ciudad Real Spain
Draft
490
dev
dev
ij = 2 *D ij
1
kk = p + k *D ii
3
kk
= p = p + *D kk
3
{
p
Thus, p = p holds when the following is satisfied:

* = 0
or
D ii = 0
Tr (D) = 0
or
2
3
* = *
Problem 11.11
A baratropic perfect fluid has as equation of state = 0 +
p
, where k is constant. Obtain
k
the pressure field for a quasi-static regime (zero acceleration), under the action of the
gravitational field b i = [0 0 g ]T .
Solution:
The constitutive equation in stress of a perfect fluid:
= p1
The equations of motion become:

r
r
&
r
x + b = v
indicial
&
ij , j + b i = vi = 0 i
{
= 0i
( p ij ), j + b i = 0 i
r r
r
x p + b = 0
p, j ij + b i = 0 i
tensorial

(11.29)
p, i + b i = 0 i
Considering the body force vector b i = [0 0 g ] we obtain:
p
p
+ b1 = 0
= 0 p = p ( x1 , x 2 , x3 )
(i = 1)
x1
x1
p
p
p, i + b i = 0 i (i = 2)
+ b2 = 0
= 0 p = p ( x1 , x 2 , x3 )
x 2
x 2
dp ( x3 )
p
(i = 3) x + b 3 = 0 dx + g = 0
3
3
(11.30)
With that we can conclude that the pressure field is only a function of the coordinate x 3 ,
i.e. p = p( x3 ) .
By the fact we are dealing with a barotropic fluid, this implies that the mass density is only
a function of pressure = ( p ) . This relationship is precisely the equation of state of the
problem statement:
= ( p)
=0 +
p
k
Then:

Ciudad Real Spain
Draft
491
dp( x3 )
+ g = 0
dx3
dp( x3 )
p
+ 0 + g = 0
dx3
k
dp( x3 ) g
+ p = 0 g
dx3
k
(11.31)
The solution of the above differential equation is the sum of a particular solution and a
homogeneous solution:
g
( x3 )
dp( x 3 ) g
Homogeneous solution:
+ p = 0 p = Cexp k
dx3
k
Particular solution: p = k 0
Thus:
p = Cexp
g
x )
k 3
k 0
Problem 11.12
A perfect gas is an ideal and incompressible fluid in which in the absence of heat sources
the pressure is proportional to (barotropic motion), where is a constant and > 1 .
Show that when r = 0 (no internal heat source), the specific internal energy for a ideal gas
is given by:
u=
1 p
+ constant
( 1)
Solution:
For the proposed problem, the energy equation becomes:
r
r
&
u = :D x q + r = :D
&
u :D = 0
For a perfect gas the stress tensor is a spherical tensor and is given by:
( p ) = p1
where p is the thermodynamic pressure. Then, the energy equation becomes:

&
u :D = 0
&
u + p1 : D = 0
&
u + pTr (D) = 0
r
r
&
u + p x v = 0
For a barotropic motion, the specific internal energy is a function of the mass density,
u = u ( ) , thus:
r
r
&
u + p x v = 0

Ciudad Real Spain
r
u
r
&
+ p x v = 0
Draft
492
Considering the mass continuity equation
r
r
D
r
r
&
+ x v = 0 = x v , the energy
Dt
equation becomes:
r
u
r
&
+ p x v = 0
r
r
u
r
r
x v + p x v = 0

r
u
r
2
+ p x v = 0
1 3
with that the following holds:

2
u
+ p=0
p
u
= 2

As pressure is proportional to , we can state that p = p ( ) = k , where k is a

proportionality constant, then:
2
u
+ p=0
u k
=
= k ( 2 )
2
By integrating the above equation we obtain:

u=
=
Problem 11.13
k
+ constant
( 1)
1 p
+ constant
( 1)
A fluid moves with velocity v around a sphere of radius R , where the velocity
components in spherical coordinates ( r , , ) are given by:
R 3 3R
v r = c 3
+ 1 cos()
2r
2r
3
R
3R
v = c 3 +
1 sin()
4r
4r
v = 0
(11.32)
where c is a positive constant.

Check whether we are dealing with an isochoric motion or not.
r
Note: Given a vector u , the divergence of this vector in spherical coordinates is:
u cot()
r
r u
1 u
1
2
r
div u x u = r +
+
+
u + u r
r r sin()
r
r
r
Solution:
r
To demonstrate that a motion is isochoric, we must prove that x v = 0
We obtain the following derivatives:

Ciudad Real Spain
Draft
493
3R 3 3R
v r
R 3 3R
= c 3
+ 1 cos() = c
2r 4 + 2r 2 + 1 cos()
2r
2r
r r

R 3 3R
v
R 3 3R
c 3 +
1 cos()
=
1 sin() = c 3 +
4r
4r
4r
4r
With that it is possible to obtain the divergence of the velocity field:

v cot()
r v
1 v
1
2
r
+
v + v r
x v = r +
+
r
r
r r r sin() {
=0
3R
3R
1 R 3 3R
= c
+ 2 + 1 cos() + c 3 +
1 cos() +
2r 4
4r
r 4r
2r
2 R 3 3R
cos() 1 R 3 3R
1 sin() + c 3
+ 1 cos()
c 3 +
4r
2r
sin() r
4r
2r
r
By simplifying the above we obtain:
r
r
x v = 0
So, we are dealing with an isochoric motion.

Problem 11.14
A barotropic fluid flows through a pipeline as shown in Figure 11.4, and said fluid has as
equation of state:

p = Ln
( and 0
are constants )
where p is pressure, and is the mass density.

Calculate, in steady state, the output pressure p ( 2) as a function of the other variables of
Figure 11.4.
p (1)
p( 2)
v (1)
v( 2)
S (1)
S (2)
Figure 11.4
p - pressure, v - velocity, S - cross section area.
Solution:
According to the principle of conservation of mass we have:
Ciudad Real Spain
Draft
494
D
dV = 0
Dt V
and given a property ( x , t ) , it fulfills that:

r
D
DdV
D ( x, t )
dV + ( x , t )
( x , t )dV =
Dt V
Dt
Dt
V
r r
D ( x, t )
r
dV + ( x , t ) x v ( x , t )dV
=
Dt
r r
D ( x, t )
r
=
+ ( x , t ) x v ( x , t ) dV
Dt
r r
r r
( x , t )
r
r
=
+ x ( x , t ) v ( x , t ) + ( x , t ) x v ( x , t ) dV
t
( x , t )
r
=
+x
t
V
(( x, t )v ( x, t )) dV
=
=
( x , t )
r
dV + x
t
V
( x , t )
dV +
t
r r
r r
[( x, t )v ( x, t )]dV
r r
[( x, t )v ( x, t )] ndS = 0
S
by denoting = ( x , t ) we obtain:
r
r r
( x , t )
D
( x , t )dV =
dV + [ ( x , t )v ( x , t )] ndS = 0
t
Dt V
V
S
By applying the steady state condition, i.e.

r r
[ ( x)v ( x)] ndS = 0
(1) v (1) dS
S(1)
S(1)
( x , t )
= 0 , we obtain:
t
r r
r r
[ ( x )v ( x )] ndS + [ ( x )v ( x )] ndS = 0
( 2 ) v ( 2 ) dS
S( 2 )
=0
S( 2 )
(1) v (1) S (1) + ( 2 ) v ( 2) S ( 2) = 0
Thus:
(1) v (1) S (1) = ( 2) v ( 2) S ( 2)
(11.33)
Remember that q = v is the mass flux, and the SI unit is [q] =

r
kg
.
m2s
Introducing the equation of state, we obtain the following equation:

p = Ln

p
= Ln

exp =
( x ) = 0 exp
Then:

Ciudad Real Spain
Draft
495
(1) v(1) S (1) = ( 2) v( 2) S ( 2)

0 exp
p ( 1)
exp
v(1) S (1) = 0 exp
p ( 2 ) p (1 )
p( 2 ) p(1)
p( 2 ) p(1)
p( 2 )
v( 2 ) S ( 2 )
v(1) S (1)
v( 2 ) S ( 2 )
v(1) S (1)
= Ln
v( 2 ) S ( 2 )
v(1) S (1)
= Ln
v( 2 ) S ( 2 )
(11.34)
v(1) S (1)
p( 2 ) = p(1) + Ln
v( 2 ) S ( 2 )
NOTE: The volumetric flow rate, (also known as volume flow rate), often represented by
Q , is the specific total flow, i.e.:
Q=
r r
q dS
r r
v dS
=
= v dS
S
S
Volumetric flow rate
m3

s
(11.35)
r r
q dS kg m 3 2 m 3
We check the SI unit [Q ] =
. In this example, we have
m =
= 2
s
S m s kg
obtained (1) v (1) S (1) = ( 2) v ( 2) S ( 2) , which can be rewritten as:
(1) v (1) S (1) = ( 2) v ( 2) S ( 2)
(1) Q(1) = ( 2) Q( 2)
For the particular case of an incompressible medium we have (1) = ( 2) , then:

v (1) S (1) = v ( 2) S ( 2)
Q(1) = Q( 2 ) (see Problem 11.7)
Problem 11.15
Starting from the Navier-Stokes-Duhem equations of motion, obtain the Bernoullis
equation:
gh +
v2
= constant
2
Bernoullis equation
(11.36)
Hypothesis: incompressible and non-viscous fluid. Consider that the velocity field is steady
and irrotational.
Solution:
r
r
Considering an incompressible medium ( x v ) = 0 , and a non-viscous fluid ( * = * = 0) ,
the Navier-Stokes-Duhem equations of motion become:
&
vi = b i p,i + ( * + * )v j , ji + * v i , jj
r
&
r
r
r
r
v = b x p + ( * + * ) x ( x v ) + * x 2 v
r
r
&
r
v = b x p
Ciudad Real Spain
Draft
(11.37)
496
&
r
Note that the v = b x p are the Euler equations of motion. The material time derivative
of the velocity, (see equation (11.15)), becomes:
r
r v r r 1 r 2
1 r
&=
+ v + x ( v ) = x (v 2 )
v
2
2
t
r
v r
= 0 , and irrotational
where we have considered that the steady velocity field
t
r
r
r
r r
r
x v = rot v = = 0 . With that the equation (11.37) can be rewritten as follows:
r
r
r
x (v 2 ) = b x p
r 1
r
1 r 2
r
x (v ) b + x p = 0
(11.38)
r
r
Considering that the body force (conservative field) can be represented by b = x ,
where is a potential, and also by considering that the mass density field is homogeneous
p 1
r
= x p holds. Then, the equation in (11.38) becomes:

r
the relationship x
p v2
r
=0
x + +
2 i
v2
= constant
2
(11.39)
Considering that the potential can be represented by = gh , where g is the acceleration

of gravity and h is the piezometric height, we obtain the Bernoullis equation:
gh +
v2
= constant
2
v 2 p N m 3 Nm J m 2
, which is the unit of
=
=
=
= = 2
kg kg s 2
2 m kg
We check the SI unit: [gh] =
specific energy, i.e. unit of energy per unit mass.

Note that the Bernoullis equation is the application of the conservation of energy, i.e. in
the system there is no energy loss:
constant energy
energy at A
energy at B
v2
2
v2
2
gh A
gh B
Figure 11.5

Ciudad Real Spain
Draft
497
Problem 11.16
Let us consider a perfect and incompressible fluid in steady regime that is flowing through
the channel as shown in Figure 11.6. Obtain the value of H .
Hypothesis: No energy loss is considered.
v (1) = 1m / s
h(1)
v( 2 ) = 2 m / s
h( 2)
Figure 11.6
Solution:
The mass continuity equation:
v (1) h(1) = v ( 2 ) h( 2)
h( 2 ) =
v (1)
v( 2)
h(1) =
1
h(1)
2
The Bernoullis equation:

( H + h(1) ) + 0 +
h( 2 ) + 0 +
v (22)
2g
v (21)
2g
H = h( 2 ) h(1) +
v (22 ) v (21)
2g
H=
h(1)
2
3
2g
Problem 11.17
A large diameter circular tank is filled with water. The water pours through a small orifice
located at a height H below the water level of the reservoir, (see Figure 11.7). If the
volumetric flow rate is Q , obtain the orifice diameter D .
Hypothesis: Consider that H does not vary with time (steady state). Consider that in the
section BB , the flow pressure is equal to atmospheric pressure:

Ciudad Real Spain
Draft
498
A
H
( p patm )
patm
Pressure
Pressure
Cross section BB
Cross section CC
Figure 11.7
Solution:
The water can be considered as an incompressible perfect fluid. Applying the Bernoullis
equation:
z+
p
v2
+
= const.
g 2g
where it fulfills that:

Point A
Point B
p atm
+0
g
2
v( B )
p
0 + atm +
g
2g
H+
v ( B ) = 2 gH
Considering that the volumetric flow rate is given by Q = v ( B ) S ( B ) , we can conclude that:
Q = v ( B ) S ( B ) = 2 gH

Ciudad Real Spain
D 2
4
Draft
D=
4Q
2 gH
499
Problem 11.18
Consider a pipeline which has be introduced a pitot tube as shown in Figure 11.8. Obtain
the velocity at the point 1 in terms of h(1) and h( 2) . Consider that there is no energy loss in
the system.
Pitot tube
Pipeline
h( 2)
h(1)
v(1)
1
p(1)
v( 2) = 0
2
p( 2 )
Figure 11.8: Pitot tube.

Solution:
By applying the Bernoullis equation between the point 1 and 2 , we obtain:
gh +
p (1)
p (1)
v (21)
v (21)
2
= gh +
p( 2)
v (22 )
2
p ( 2)
2( p ( 2 ) p (1) )
v (1) =
The pressure values at the points 1 and 2 are given, respectively, by:
p (1) = gh(1)
p ( 2 ) = gh( 2 )
with that the velocity v (1) is obtained as follows:

v (1) =
2( p ( 2) p (1) )

Ciudad Real Spain
2( gh( 2 ) gh(1) )
Draft
= 2 g (h( 2 ) h(1) )
500
Problem 11.19
Consider an incompressible non-viscous fluid, which has a steady velocity field and
irrotational. Consider also that the velocity field is independent of x3 -direction. Obtain the
governing equations of the proposed problem in terms of the velocity potential and
streamlines .
Solution:
Velocity potential: In this example we can represent the velocity field by means of a potential
r
r
, i.e. v = x . With that we are considering that the velocity field is conservative, hence
r
r
the curl of the velocity field is zero, i.e. x v = rot v = = 0 . Remember that a field
whose curl is zero it does not necessarily imply that the field is conservative, but for a
conservative field the curl is always equals zero.
r
Note that the velocity has the same direction as x , i.e. it is normal to the isosurfaces
= const .
Streamline: Given a spatial velocity field at time t , we can define a streamline ( ) to the
curve in which the tangent at each point has the same direction as the velocity. In general,
the streamline and trajectory do not coincide, but in steady state motion they do. Two
streamlines can not intersect.
(5) = const.
Control volume
( 4) = const.
r r
v ( x)
Streamlines
( 2) = const.
r
x
(1) = const.
r
x
Figure 11.9
Based on the definition of differential d (the total derivative), and the definition of
r
r
r
gradient x we obtain the relationship d = x dx .

Ciudad Real Spain
Draft
501
r
r
v = x
+ d
r
x
= const. d = 0
in the streamline
Figure 11.10
r
r
r
Note that it holds that x x = 0 . The differential dx in the streamline, at a point, has
the same direction as the velocity at this point. Hence, the following is fulfilled:
r r r
dx v = 0
In the Cartesian system:
e1
r r
dx v = dx1
e2
dx 2
e3
r
dx3 = 0
v1
v2
v3
= (v3 dx 2 v 2 dx3 )e1 + (v 3 dx1 v1 dx3 )e 2 + (v 2 dx1 v1 dx 2 )e 3 = 0
Components:
(v3 dx 2 v 2 dx3 ) 0
r r
(dx v ) i = (v 3 dx1 v1 dx3 ) = 0

(v 2 dx1 v1 dx 2 ) 0
For this example the velocity field is independent of x3 , i.e. the problem is state on the
plane x1 x 2 (2D-case). With that we can conclude that:
0
0
r r
= 0
( dx v ) i =
0

(v 2 dx1 v1 dx 2 ) 0

v 2 dx1 v1 dx 2 = 0
(11.40)
Given that in a streamline it holds that = const. d = 0 and also by applying the
r
r
definition d = x dx , we obtain:
r
r
d = x dx
indicial
d = ,i dxi = 0
d = ,1 dx1 + , 2 dx 2 + ,3 dx3 = 0
d =
dx1 +
dx 2 +
dx3 = 0
x 2
x3
x1
For the 2D-case (bidimensional) we have:
dx1 +
dx 2 = 0
x1
x 2
(11.41)
If we compare the equations (11.40) and (11.41) we conclude that:

Ciudad Real Spain
Draft
502
v1 =
x 2
v2 =
x1
(11.42)
r
1) Starting from an incompressible fluid: ( x v ) = 0 we obtain:
v i ,i =
v1 v 2 v3
+
+
=0
x1 x 2 x3
2D
v1 v 2
+
=0
x1 x 2
r
Considering that v = x , we obtain:
2 2
r
+ 2 =0
2 = 0
x
2
x1
x 2
r
r
r r r
r
2) Based on the fact that the fluid is irrotational x v rot v = = 0 we obtain:
v1 v 2
+
=0
x1 x 2
e1 e 2
e3
r
r
r
r
r
= ijk v k , j e i = 0
rot (v ) = x v =
x1 x 2 x3
v1
v2
v3
v
v
v
v
v
v
= 3 2 e 1 + 1 3 e 2 + 2 1
x
x
1 x 2
2 x3
3 x1
(11.43)
(11.44)
r
e 3 = 0
Then:
v 3 v 2
x x
3
2
0
v1 v3
= 0

x3 x1 0
v
v
2 1
x1 x 2
As we are dealing with a 2D-case, the above equation becomes:

v 2 v1
=0
x1 x 2
Taking into account the equations in (11.42) we conclude that:

v 2 v1
=0
x1 x 2
2 2
=0
+
x 2
x1
r
2 = 0
x
With which the problem is stated by the equations:

r
2 = 0
x

Ciudad Real Spain
Draft
r
2 = 0
x
(11.45)
Bibliography
ASARO, R.J. & LUBARDA, V.A. (2006). Mechanics of solids and materials. Cambridge University
Press, New York, USA.
BATRA, R. C. (2006). Elements of Continuum Mechanics. John Wiley & Sons Ltd., United
Kingdom.
CASANOVA, J.C. (1993). Ejercicios de elasticidad. Editorial UPV.
CHADWICK, P. (1976). Continuum mechanics concise theory and problems. George Allen & Unwin
Ltd.Great Britain.
CHAVES, E.W.V. (2013). Notes on Continuum Mechanics. Series: Lecture notes on numerical
methods in engineering and sciences, vol. 4. Springer/CIMNE. ISBN:978-94-0075985-5(HB), ISBN:978-94-007-5986-2(e-book).
GOICOLEA, J.M. Mecnica del medio continuo web: http://w3.mecanica.upm.es/mmc-ig/
HOLZAPFEL, G.A. (2000). Nonlinear solid mechanics. John Wiley & Sons Ltd. England.
MASE, G.E. (1977). Teora y problemas de mecnica del medio continuo. McGraw-Hill, USA.
OLIVER, J. & AGELET DE SARACBAR, C. (2000). Mecnica de medios continuos para ingenieros.
Ediciones UPC, Barcelona, Espaa.
OLIVER, J. & AGELET DE SARACBAR, C. (2000). Cuestiones y problemas de mecnica de medios
continuos. Ediciones UPC, Barcelona, Espaa.
ORTIZ BERROCAL, L. (1985). Elasticidad. E.T.S. de Ingenieros Industriales. Litoprint. U.P.
Madrid.
PARKER, D.F. (2003). Fields, Flows and Waves: An introduction to continuum models. SpringerVerlag London, UK.

MMC Problems PDF

Uploaded by

Copyright:

Available Formats

MMC Problems PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MMC Problems PDF

Uploaded by

Copyright:

Available Formats

[

2 = 21x1 a + 4axz= 0 2 X1 = 0 2 + x =l2 1 x

2,5 Xx a l = 4ax 1 2 1+ b ax0

d+2 x yt+ 0 + x1 a b x 0ax 2 0

1 2 x1x 10 ax 2y 0 xy5 = z 1 xy + y 10 6yz + b = 2,318

(dx2xy=4ax1 t3,33311=XXx x1p=t2r 2dt= x1 = X + b = 0

x1Xrx a =n223 = (4axx2 )2xln2 x2 =

Solving Problems by means of

MECNICA DEL MEDIO CONTINUO

Rigid Body Motion

Fundamental equations of C.M.

MECNICA DEL MEDIO CONTINUO

Initial Boundary Value Problem

Operators and Symbols

Operators and Symbols

energy, work, heat

Acceleration (reference configuration)

Continuum medium in the reference configuration at

Continuum medium in the current configuration at

Body force (per unit mass)

Left deformation Cauchy-Green tensor, Finger

Entropy created inside

local entropy per unit mass per unit time

Right deformation Cauchy-Green tensor

Green-Lagrange strain tensor, or Lagrangian finite

Almansi strain tensor, or Eulerian finite strain tensor

Youngs modulus, or elastic modulus

Deformation gradient (pseudo-tensor)

material displacement gradient tensor

spatial displacement gradient tensor

thermal conductivity tensor

Spatial velocity gradient

body force (per unit volume)

First Piola-Kirchhoff stress tensor

Cauchy heat flux (non-convective vector)

MECNICA DEL MEDIO CONTINUO

Orthogonal tensor of polar decomposition

Biot stress tensor

rate of change of the internal energy

Specific internal energy

Displacement vector (Eulerian)

Velocity (reference configuration)

Velocity (current configuration)

Spin tensor, rate-of-rotation tensor, or vorticity

Vector position (material coordinate)

coefficient of thermal expansion

Permutation symbol, or Levi-Civita tensor

Infinitesimal strain tensor

Solute mass density

fluid mass density

Mass density (reference configuration)

Cauchy stress tensor, or true stress tensor

Helmholtz free energy, specific (per unit mass)

Helmholtz free energy (per unit volume)

Strain energy density

MECNICA DEL MEDIO CONTINUO

The indicial notation was introduced by Einstein (1916, sec.

1.1 Solved Problems

Vectors, Indicial Notation