Continuum Mechanics For Engineers By Dr.

Djomo Mbong

CHAPTER ONE:

Continuum Theory and Essential Mathematics

1. Continuum theory

1.1. The Continuum Concept

The atomic/molecular composition of matter is well established. On a small enough
scale, for instance, a body of aluminium is really a collection of discrete aluminium atoms
stacked on one another in a particular repetitive lattice. On an even smaller scale, the atoms
consist of a core of protons and neutrons around which electrons orbit. Thus, matter is not
continuous.
At the same time, mathematics teaches us that between any two points in space we can
always find another point, regardless of how close together we choose the original pair.
Therefore, the physical space in which we live is truly a continuum, so the matter also. Indeed,
if we accept the continuum concept of matter, we agree to:
- Ignore the discrete composition of material bodies,
- Assume that the substance of such bodies is distributed uniformly throughout, and
completely fills the space it occupies.
In keeping with this continuum model, we assert that matter may be divided
indefinitely into smaller and smaller portions, each of which retains all of the physical
properties of the parent body. Accordingly, we are able to ascribe field quantities such as
density and velocity to each and every point of the region of space which the body occupies.
The continuum model for material bodies is important to engineers for one major reason.
On the scale by which we consider bodies of steel, aluminium, concrete, etc., the characteristic
dimensions are extremely large compared to molecular distances so that the continuum
model provides a very useful and reliable representation.

1.2. Continuum Mechanics

The analysis of the kinematic and mechanical behaviour of materials modelled on the
continuum assumption is what we know as continuum mechanics. There are two main themes
into which the topics of continuum mechanics are divided:
- In the first, emphasis is on the derivation of fundamental equations which are valid
for all continuous media. These equations are based upon universal laws of physics such
as the conservation of mass, the principles of energy and momentum, etc.
- In the second, the focus of attention is on the development of so-called constitutive
equations characterizing the behaviour of specific idealized materials, the perfectly
elastic solid and the viscous fluid being the best known examples. These equations
provide the focal points around which studies in elasticity, plasticity, viscoelasticity,
and fluid mechanics proceed.
Inasmuch as this is an introduction, we shall make two further assumptions on the
materials we discuss in addition to the principal one of continuity. First, we require the materials

3
 Chapter One: Continuum Theory and Essential Mathematics

to be homogeneous, that is, to have identical properties at all locations. And second, that the
materials be isotropic with respect to certain mechanical properties, meaning that those
properties are the same in all directions at a given point.

2. Essential Mathematics

2.1. Scalars, Vectors, and Cartesian Tensors

The language of continuum mechanics is the algebra and calculus of tensors. Here,
tensors is the generic name for those mathematical entities which are used to represent the
important physical quantities of continuum mechanics.
The tensor equations used to develop the fundamental theory of continuum mechanics
may be written in either of two distinct notations: the symbolic notation or the indicial
notation.
As it happens, a considerable variety of physical and geometrical quantities have
important roles in continuum mechanics, and fortunately, each of these may be represented by
some form of tensor.
- For example, such quantities as density and temperature may be specified completely
by giving their magnitude, i.e., by stating a numerical value. These quantities are
represented mathematically by scalars, which are referred to as zeroth-order tensors.
- Several physical quantities of mechanics such as force and velocity require not only an
assignment of magnitude, but also a specification of direction for their complete
characterization. Quantities possessing such directional properties are represented by
vectors, which are first-order tensors.
- A significant number of physical quantities having important status in continuum
mechanics require mathematical entities of higher order than vectors for their
representation in the hierarchy of tensors. As we shall see, among the best known of
these are the stress tensor and the strain tensors. These particular tensors are second-
order tensors, and are said to have a rank of two. Mathematically, these quantities are
represented by matrices. Third-order and fourth-order tensors are not uncommon in
continuum mechanics, but they are not nearly as plentiful as second-order tensors.

2.2. Tensor Algebra in Symbolic Notation - Summation Convention (Einstein

Convention)
The three-dimensional physical space is known as a Euclidean three-space, and its
geometry can be referenced to a system of Cartesian coordinate axes.
Because a scalar has only a single component, it will have the same value in every
system of axes, but the components of vectors and tensors will have different component values,
in general, for each set of axes.
In order to represent vectors and tensors in component form, we introduce in our
physical space a system of rectangular Cartesian axes Ox1 x 2 x 3 , and identify with these axes
the triad of unit base vectors e1 , e 2 , e 3 , shown in Fig.1a. In terms of this basis, an arbitrary
vector v is given in component form by :

4
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

(a) (b)
Figure 1: (a). Unit vectors in the coordinate directions , and . (b) Rectangular components of the
vector
3
v  v1e1  v2e2  v 3e3   v i ei 1
i 1

This vector and its coordinate components are pictured in Fig.1b.

At this juncture of our discussion it is helpful to introduce a notational device called the
summation convention, also called the Einstein Convention, that will greatly simplify the
writing of the equations of continuum mechanics. Stated briefly, we agree that whenever a
subscript appears exactly twice in a given term, that subscript will take on the values 1, 2, 3
successively, and the resulting terms summed. For example, using this scheme, we may now
write Eq.(1) in the simple form:
v  v i ei  2
The symbol  has been deleted. The summed subscripts are called dummy indices since it is
immaterial which particular letter is used. Thus, v j e j is completely equivalent to v i e i , or to
v k ek , when the summation convention is used.

Example 1: Without regard for their meaning as far as mechanics is concerned, expand the
following expressions according to the summation convention:
a ui vi w j e j b Tij vi e j c  Tii v j e j
Solution:
(a) Summing first on i , and then on j , we obtain after rearrangement

ui vi w j e j   u1v1  u2v2  u3v3  w1e1  w2e2  w3e3 

(b) Summing on i then on j , and collecting terms on the unit vectors, we obtain
Tij vi e j  T1 j v1e j  T2 j v 2e j  T3 j v 3e j
  T11v1  T21v2  T31v3  e1   T12v1  T22v2  T32v3  e2   T13v1  T23v2  T33v3  e3
(c) Summing on i , then on j , we obtain
Tii v j e j   T11  T22  T33  v1e1  v2e2  v3e3 
Note the similarity between (a) and (c).

5
 Chapter One: Continuum Theory and Essential Mathematics

With the above background in place we now list, using symbolic notation, several useful definitions
from vector/tensor algebra.

 Addition of vectors:
wuv or wi ei   ui  vi  ei  3
 Multiplication
Of a vector by a scalar:
 v   v i ei  4
Dot (scalar) product of two vectors:
u  v  v  u  uv cos    5
where  is the smaller angle between the two vectors when drawn from a common origin.

The Kronecker Delta Symbol:

From Eq.(5) for the base vectors e i ,  i  1,2,3 
1, if i  j
ei  e j  
 0, if i  j
Therefore, if we introduce the Kronecker delta defined by:
1, if i  j
 ij  
 0, if i  j
we see that
ei  e j   ij ,  i , j  1,2,3  6
Also, note that by the summation convention,
 ii   jj   11   22   33  1  1  1  3
From the definition of  ij , the dot product u  v may be written as

u  v  ui ei  v j e j  ui v j ei  e j  ui v j ij  ui vi  u j v j  7

Cross (vector) product of two vectors:

u  v   v  u   uv sin    e
where 0     , is the angle between the two vectors when drawn from a common origin,
and where e is a unit vector perpendicular to their plane such that a right-handed rotation about
through the angle  carries u into v .

The Permutation Symbol (Levi Civita Symbol):

By introducing the permutation symbol  ijk defined by

1, if the numerical values of ijk appear as 123, 231, 312


 ijk   1, if the numerical values of ijk appear as 132, 321, 213  8
 0, if one of the idices ijk is repeated


6
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

We may express the cross products of the base vectors e i ,  i  1,2,3  by the use of Eq.(8) as
ei  e j   ijk ek ,  i , j, k  1,2,3  9
Therefore, now the vector cross product above becomes:
 
u  v  ui ei  v j e j  ui v j ei  e j   ijk ui v j ek ,  i , j, k  1,2,3 10

Triple scalar product (box product):

u  v  w  u  v  w   u, v , w 
Or
 u, v , w   ui ei   v j e j  wk ek   ui ei    jkqv j wk eq    jkq ui v j wk  ei  eq 
, 11
  jkq ui v j wk iq   ijk ui v j wk

Triple cross product:

   
u   v  w   ui ei  v j e j  wk ek  ui ei   jkq v j w k eq   jkq ui v j w k ei  eq  ,
12
 
  jkq ui v j wk  iqm em   iqm jkq ui v j w k em   miq jkq ui v j w k em

   Identity:
The product of permutation symbols  iqm jkq in Eq.(12) may be expressed in terms of
Kronecker deltas by the identity:
 miq jkq   mj ik   mk ij , 13

Therefore, Eq.(12) becomes :

 
u   v  w    mj ik   mk ij ui v j wk em   mj ik ui v j wk em   mk ij ui v j wk em
, 14
 ui wi vm em  ui vi wm em   u  w  v   u  v  w

Tensor product of two vectors (dyad):

u v  ui ei v j e j  ui v j ei e j , 15
which in expanded form, summing first on i , then summing on j , yield
ui v j ei e j  u1v1e1 e1  u1v2e1 e2  u1v3e1 e3
 u2v1e2 e1  u2v2e2 e2  u2v3e2 e3 , 16
 u3v1e3 e1  u3v2e3 e2  u3v3e3 e3
This nine-term sum is called the nonion form of the dyad, u v . An alternative notation
frequently used for the dyad product is
u  v  ui ei  v j e j  ui v j ei  e j , 17 
A sum of dyads such as
7
 Chapter One: Continuum Theory and Essential Mathematics

u1  v1  u2  v2  ...  uN  v N , 18
is called a dyadic.
Vector-dyad products:
 
u   v  w   ui ei  v j e j  wk ek  ui vi wk ek , 19
 u  v   w   ui ei  v j e j   wk ek  ui v j w jei ,  20 
 
u   v  w   ui ei  v j e j  wk ek   ijq ui v j wk eq  ek ,  21
 u  v   w  ui ei   v j e j  wk ek    jkq ui v j wk ei  eq ,  22 

Dyad-dyad product :
 u  v    w  s   ui ei   v j e j  wk ek   sqeq  ui v j w j sqei  eq ,  23 

Vector-tensor products :
v  T  vi ei  T jk e j  ek  viT jk ij ek  viTik ek ,  24 
T  v  Tij ei  e j  vk ek  Tij vk jk ei  Tij v j ei ,  25 

Tensor-tensor product :
T  S  Tij ei  e j  S pq e p  eq  Tij S pq jpei  eq  Tij S jqei  eq ,  26 

2.3. Indicial Notation

By assigning special meaning to the subscripts, the indicial notation permits us to carry
out the tensor operations of addition, multiplication, differentiation, etc. We simply agree that
the tensor rank (order) of a term is indicated by the number of “free,” that is, unrepeated,
subscripts appearing in that term. Accordingly,
- A term with no free indices represents a scalar,
- A term with one free index represents a vector,
- A term having two free indices represents a second-order tensor,

Specifically, we have the tensors presented in the table below:

Indicial Name of the Order of the Vector Number of
Notation tensor tensor Notation Components
Zeroth-order
 Scalar  1
tensor
Vector First-order
vi v 3
tensor
Second-order
ui v j Dyad uv 9
tensor

8
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

Second-order
Tij Dyadic T 9
tensor
Third-order
Qijk Triadic Q 27
tensor
Fourth-order
C ijkm Tetradic 81
tensor C

We must emphasize that in the indicial notation exactly two types of subscripts appear:
- “free” indices, which are represented by letters that occur only once in a given term,
- “summed” or “dummy” indices which are represented by letters that appear twice in
a given term.

A tensor is symmetric in any two indices if interchange of those indices leaves the
tensor value unchanged. For example, if S ij  S ji and C ijm  C jim , both of these tensors are
said to be symmetric in the indices i and j .
A tensor is anti-symmetric (or skew-symmetric) in any two indices if interchange of
those indices causes a sign change in the value of the tensor. Thus, if Aij   A ji , it is anti-
symmetric in i and j .

Example 2: By direct expansion of the expression vi   ijkW jk , determine the components

of the vector v i in terms of the components of the tensor W jk .
Solution:
By summing first on j and then on k and then omitting the zero terms, we find that:
vi   i 1kW1k   i 2 kW2 k   i 3 kW3 k
  i 12W12   i 13W13   i 21W21   i 23W23   i 31W31   i 32W32
Therefore,
v1   123W23   132W32  W23  W32
v2   213W13   231W31  W31  W13
v3   312W12   321W21  W12  W21
Note that if the tensor W jk were symmetric, the vector v i would be a null (zero) vector.

2.4. Matrices and Determinants

For computational purposes it is often expedient to use the matrix representation of
vectors and tensors.

2.4.1. Definition
A matrix is an ordered rectangular array of elements enclosed by square brackets and
subjected to certain operational rules. The typical element Aij of the matrix is located in the i th

(horizontal) row, and in the j th (vertical) column of the array. A matrix having elements Aij ,

9
 Chapter One: Continuum Theory and Essential Mathematics

which may be numbers, variables, functions, or any of several mathematical entities, is

designated by  Aij  , or symbolically by the kernel letter A . An M by N matrix (written
M  N ) has M rows and N columns, and may be displayed as:
 A11 A12 A1 N 
A A22 A2 N 
A   Aij    21
    27 

 
 AM 1 AM 2 AMN 
- If M  N , the matrix is a square matrix.
- A 1  N matrix  A1N  is an row matrix, and an M  1 matrix  AM 1  is a column
matrix. Row and column matrices represent vectors, whereas a 3  3 square matrix
represents a second-order tensor. A scalar is represented by a 1  1 matrix (a single
element).
- A zero, or null matrix has all elements equal to zero.
- A diagonal matrix is a square matrix whose elements not on the principal diagonal,
which extends from A11 to ANN , are all zeros. Thus for a diagonal matrix, Aij  0 for
i  j.
- The unit or identity matrix I , which, incidentally, is the matrix representation of the
Kronecker delta, is a diagonal matrix whose diagonal elements all have the value one.

2.4.2. Transpose matrix

The N  M matrix formed by interchanging the rows and columns of the M  N
T
matrix A is called the transpose of A , and is written as A T , or  Aij  . By definition,
the elements of a matrix A and its transpose are related by the equation AijT  Aji .

- A square matrix for which A  A T , or in element form, Aji  AijT , is called a

symmetric matrix; one for which A  A , or Aji   AijT is called an anti-
T

symmetric, or skew-symmetric matrix.

- The elements of the principal diagonal of a skew-symmetric matrix are all zeros.
- Two matrices are equal if they are identical element by element.
- A square matrix A can be expressed as the sum of a symmetric and a skew-
symmetric matrix by the decomposition
A  AT A  AT
A  B C    28 
2 2
Aij  AijT Aij  A ji ATji  A ji
Where Bij     B ji  BijT is the symmetric part of the A
2 2 2
Aij  AijT Aij  A ji A ji  ATji
and C ij      C ji   C ijT is the skew-symmetric part
2 2 2
of the A .

10
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

2.4.3. Determinant of a matrix

The determinant of the matrix A is designated by either det A , or by Aij , and for
a 3  3 matrix A ,
A11 A12 A13
det A  Aij  A21 A22 A23  29 
A31 A32 A33
Which can be expanded using the method called expansion by cofactors as:
A22 A23 A21 A23 A21 A22
det A  A11  A12  A13  30 
A32 A33 A31 A33 A31 A32
which upon complete expansion gives
 A11  A22 A33  A23 A32   A12  A21 A33  A23 A31 
det A  31
 A13  A21 A32  A22 A31 

2.4.4. Matrix form of vector/tensor equations

As mentioned near the beginning of this section, a vector may be represented by a row
or column matrix, a second-order tensor by a square 3  3 matrix. For computational purposes,
it is frequently advantageous to transcribe vector/tensor equations into their matrix form.
As a very simple example, the vector-tensor product u  T  v ,(symbolic notation)
or ui  Tij v j (indicial notation) appears in matrix form as:

 u1  T11 T12 T13   v1 

    
 ui 1   Tij   v j 1  or
 u2   T21 T22 T23   v2   32 
 u3  T31 T32 T33   v3 

In much the same way the product w  v  T or wi  v jT ji appears as

T
T11 T12 T13 
 w1i    v1 j  T ji     33 
or  w1 w2 w3   v1 v2 v3  T21 T22 T23 
T31 T32 T33 

2.5. Transformations of Cartesian Tensors

Although vectors and tensors have an identity independent of any particular reference
or coordinate system, the relative values of their respective components do depend upon the
specific axes to which they are referred.
The relationships among these various components when given with respect to two
separate sets of coordinate axes are known as the transformation equations. In developing
these transformation equations for Cartesian tensors, we consider two sets of rectangular
Cartesian axes, Ox1 x 2 x 3 and Ox1 x 2 x 3 , sharing a common origin, and oriented relative to

11
 Chapter One: Continuum Theory and Essential Mathematics

(a) (b)
Figure 2: (a). Rectangular coordinate system relative to . (b). Transformation
table between and axes.

one another so that the direction cosines between the primed and unprimed axes are given by a
 
aij  cos xi , x j as shown in Fig.2(a).
The square array of the nine direction cosines displayed in Fig.2(b) is useful in relating
the unit base vectors ei and ei to one another, as well as relating the primed and unprimed
coordinates xi and x i of a point. Thus, the primed base vectors are given in terms of the
unprimed vectors by the equations,
 e1  a11e1  a12e2  a13e3  a1 j e j

 e2  a21e1  a22e2  a23e3  a2 j e j  34 
 
 e3  a31e1  a32e2  a33e3  a3 j e j
Or in compact indicial form
ei  aij e j  35 
By defining the matrix A whose elements are the direction cosines aij , Eq.(35) can be written
in matrix form as
 e1   a11 a12 a13   e1 
ei1    aij   e j 1  or  e2    a21 a22
 
a23   e2   36 

 e3   a31 a32 a33   e3 
Where the elements of the column matrices are unit vectors. The matrix A is called the
transformation matrix.
Reciprocally, we can also move from the Ox1 x 2 x 3 base to the Ox1 x 2 x 3 base, by
writing:
ei  a ji ej  37 
which in matrix form is
T
 a11 a12 a13 
 
e1i    e1 j   a ji  or e1 e2 e3   e1 e2 e3  a21 a22 a23   38 
 a31 a32 a33 

12
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

Consider next an arbitrary vector v having components v i in the unprimed system, and
v i in the primed system. Then using Eq.(37),
v  vj ej  vi ei  vi a ji ej
From which by matching coefficients on e j we have (in both the indicial and matrix forms),

vj  a ji vi or v  Av  vAT  39 
which is the transformation law expressing the primed components of an arbitrary vector in
terms of its unprimed components. Reciprocally, the inverse law expressing the unprimed
components of an arbitrary vector in terms of its primed components, is given by:
vk  a jk v j or v  vA  AT v  40 
By a direct application of Eq.(40) to the dyad u  v we have
ui v j  aqi uq amj vm  aqi amj uq vm  41
But a dyad is, after all, one form of a second-order tensor, and so by an obvious
adaptation of Eq.(41), we obtain the transformation law for a second-order tensor, T , as

Tij  aqi amjTqm or T  AT T A  42 
Reciprocally, we have the inverse law as
Tij  aiq a jmTqm or T   ATAT  43 

Example 3: Let the primed axes Ox1 x 2 x 3

be given with respect to the unprimed axes
by a 45° counterclockwise rotation about
the x2 axis as shown. (a)-Determine the
primed components of the vector given by
v  e1  e2  e3 . (b)-Determine also the
primed components of the tensor
 2 6 4
Tij    0 8 0  .
 
 4 2 0 

Solution:
Here the transformation matrix is
 2 2 0  2 2
 
 aij    0 1 0 
 
 2 2 0 2 2
(a)-From Eq.(39) in matrix form

13
 Chapter One: Continuum Theory and Essential Mathematics

 v1   2 2 0  2 2  1  0 
     
 v2    0 1 0  1   1 
 v3   2 2 0   
2 2  1  2 

(b)-From Eq.(43) in matrix form
 2 2 0  2 2  2 6 4  2 2 0 2 2   3 4 2 1
     
Tij    0 1 0   0 8 0   0 1 0 0 8 0
   
 2 2 0 2 2   4 2 0    2 2 0 2 2   1 8 2 5 

2.6. Principal Values and Principal Directions of Symmetric Second-Order

Tensors
First, let us note that in view of the form of the product of a second-order tensor T with
the arbitrary vector u ,

Tij u j  vi or T uv  44 
any second-order tensor may be thought of as a linear transformation which transforms the
antecedent vector u into the image vector v in a Euclidean three-space. In particular, for

every symmetric tensor T having real components Tij , and defined at some point in physical
space, there is associated with each direction at that point (identified by the unit vector n i ), an
image vector v i given by:

Tij n j  vi or T nv  45 
If the vector v i determined by Eq.(45) happens to be a scalar multiple of n i , that is, if

Tij n j   ni or T  n  n  46 
The direction defined by n i is called a principal direction, or eigenvector, of T , and the scalar

 is called a principal value, or eigenvalue of T . Using the substitution property of the

Kronecker delta, Eq.(46) may be rewritten as

T ij 
  ij n j  0 or T   I   n  0  47 
or in expanded form
 T11    n1  T12 n2  T13 n3  0



T21n1   T22    n2  T23 n3  0  48 


T31n1  T32 n2   T33    n3  0


14
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

This system of homogeneous equations for the unknown direction n i and the unknown  will
have non-trivial solutions only if the determinant of coefficients vanishes. Thus,
Tij   ij  0  49 
Which upon expansion leads to the cubic in  (called the characteristic equation)
 3  I 1 2  I 2   I 3  0  50 
where the coefficients here are expressed in terms of the known components of T by:
 I  T  Tr T
 1

ii 


  

 1  2

1
 
 I 2  TiiT jj  TijT ji   Tr T   Tr T 
2  
2
  51
 2 


 I   T T T  det T


3 ijk 1 i 2 j 3 k 
and are known as the first, second, and third invariants, respectively, of the tensor T .
q
It is important to note that the eigen vector ni associated to the eigenvalue  q 

( q  1, 2,3 and where  q  is one of the solution of Eq.(50)), have to be normalize, that is we
have to make sure that:
ni  ni   1 ,  q  1,2,3   52 
q q

The set  n  , n  , n   of the three principal directions, that can be written in the
i
1
i
2
i
3

vector form  n   , n   , n    or  e , e , e  , will define a new base called the principal base
1 2 3 * * *
1 2 3

(also noted Ox1* x2* x3* ). The transformation matrix A to move from the base Ox1 x 2 x 3 to the
principal base Ox1* x2* x3* is defined as:
 
a13   n1 n2  n3  
1 1 1
 a11 a12
 Aij    a21   2 
 53 
a23    n1  n2
2
n3  
2
 a22
 a31  3 
a32 a33   n3 3 n3
3
n3  

And the second-order tensor T * in the principal base Ox1* x2* x3* is a diagonal matrix whose
elements are the principal values  q  , given by:

 0 0 
  1 
T  AT  A   0
* T
 2 0   54 
 
 0 0  3 

15
 Chapter One: Continuum Theory and Essential Mathematics

Example 4: Determine the principal values and principal directions of the second-order
5 2 0
tensor T whose matrix representation is Tij    2 2 0  .
   
 0 0 3 
Solution:
Let us evaluate the characteristic equation using Eq.(49). We have
5 2 0
2 2 0 0   3      2  7  6   0
0 0 3

Hence, 1  3 ,  2   6 ,  3  1 are the principal values of T . Let evaluate now the

principal directions ni  of T .

q

- For  1  3 , Eq.(48) yields the equations

 n1  n2  0

 2n1  n2  0
which are satisfied only if n1  n2  0 . Since, ni  ni  1 , we have
n1  n1  n2  n2  n3  n3  1  n32  1  n3  1 .
- For  2   6 , Eq.(48) yields the equations

  n1  2n2  0
  n1  2n2
 2n1  4n2  0  
 3 n  0  n3  0
 3

Since, ni  ni  1 , we have

 2n2 
2
n1  n1  n2  n2  n3  n3  1   n22  1
1 2 .
 n2    n1  
5 5
- For  3  1 , Eq.(48) yields the equations

 4n1  2n2  0
  n2  2n1
 2n1  n2  0  
 2n  0  n3  0
 3
Since, ni  ni  1 , we have

n12   2n1   1
2
n1  n1  n2  n2  n3  n3  1 
1 2 .
 n1    n2 
5 5
From these results the transformation matrix A is given by:

16
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

 0 0 1
 
 aij    2 5 1 5 0
 1 5 0 
 2 5
Using the upper set of the  signs, we define the three principal directions (or eigenvector)
bellow:
2 1 1 2
n    e3 ; n    e1  e2 ; n    e1 
1 2 3
e2
5 5 5 5
Using Eq.(54), we have
 0 0 1 5 2 0 0 2 5 1 5   3 0 0
   
2 5 1 5 0   2 2 0   0 1 5 2 5    0 6 0 
1 5  0   0 0 1 
 2 5 0   0 0 3   1 0


Since the obtained matrix is a diagonal matrix which elements are the principal values of T
1

located on the diagonal, it prove that the base n   , n   , n  
2 3
 is the principal base.
2.7. Tensor Fields, Tensor Calculus

A tensor field assigns to every location x , at every instant of time t , a tensor

Tij ... k  x , t  , for which x ranges over a finite region of space, and t varies over some interval
of time.
- The field is continuous and hence differentiable if the components Tij ... k  x , t  are
continuous functions of x and t .
- Tensor fields may be of any order. For example, we may denote typical scalar, vector,
and tensor fields by the notations   x , t  , vi  x, t  and Tij  x , t  ,respectively.

Partial differentiation of a tensor field with respect to the variable t is symbolized by

the operator  t and follows the usual rules of calculus. On the other hand, partial
differentiation with respect to the coordinate xq will be indicated by the operator  xq , which

may be abbreviated as simply  q . Likewise, the second partial  2 xq xm may be written
 qm , and so on. As an additional measure in notational compactness it is customary in
continuum mechanics to introduce the subscript comma to denote partial differentiation with
respect to the coordinate variables. For example, we write :
 vi Tij  2 ui
,i  ; vi , j  ; Tij ,k  ; ui , jk   55 
x i x j x k x j xk

A consequence occurs:

17
 Chapter One: Continuum Theory and Essential Mathematics

xi
 xi , j   ij  56 
x j
In the notation adopted here the operator  (del) of vector calculus, which in symbolic
notation appears as:
   
 e1  e2  e3  e    i  ei  57 
x1 x 2 x 3 x i i
Therefore, we may write:
- The scalar gradient   grad  as
  grad     i  ei  ,i ei  58 
- The vector gradient   v as
 
  v    i  ei  v j e j   i v j ei  e j  v j , i ei  e j  59 

- The divergence of v , div v    v as

 
div v    v    i  ei  v j e j   i v j ei  e j   ij v j ,i  vi ,i  60 

- The curl of v , curl v  rot v    v as

 
rot v    v    i  ei  v j e j   i v j ei  e j   ijk v j ,i ek  61

2.8. Integral Theorems of Green-Ostrogradski and Stokes

2.8.1. Green-Ostrogradski Formula

Let V be a volume in the space

with a closed surface S bounding the
volume, and let the outward normal to this
bounding surface be n as shown in Fig.3,
so that the element of surface is given by

dS  n  dS . Let  , v and T being a

scalar, vector and second-order tensor
fields respectively, continuously
differentiable and defined on that region of
Figure 3: Volume with infinitesimal
physical space. We define the different
surface element having a unit normal
forms of the divergence theorem (Green-
Ostrogradski Formulas): vector .

- Scalar field:

18
 Continuum Mechanics For Engineers By Dr. Djomo Mbong

   n dS   grad  dV or   n dS    dV i ,i  62 
S V S V

- Vector field:

 v  n dS   div v dV or  v n dS   v
i i i ,i dV  63 
S V S V

- Tensor field:

 T  n dS   div T dV or  T n dS   T
ij j ij , j dV  64 
S V S V

2.8.2. Stokes Formula

Let C be the bounding space curve

to the surface S , and let dxi be the
differential tangent vector to C as
shown in Fig.4. If n i is the outward
normal to the surface S , and v is any
vector field defined on S and C , Stokes’
theorem asserts that:

 v  dx   rot v
C S
 ndS
Figure 4: Bounding space curve with
or tangential vector and surface element

 v dx   
i i ijk v j ,i nk dS  65  for partial volume.
C S

19