See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/361108415

Tensor Analysis and its Applications in Physics

Article · June 2022

CITATIONS READS
0 2,213

1 author:

Professor Quddus Khan

Jamia Millia Islamia
33 PUBLICATIONS   151 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Professor Quddus Khan on 06 June 2022.

The user has requested enhancement of the downloaded file.

 ISSN 2347 - 3258
International Journal of Advance Research and Innovation
National Conference on Modern Trends in Analytical & Environmental Physics (Hindu College, Moradabad)

Tensor Analysis and its Applications in Physics

Quddus Khan
Department of Mathematics, Shibli National PG College, Azamgarh (UP)
Abstract
Article Info
Tensor Analysis has their many applications in most all the disciplines of science. This paper, gives
Article history:
the basic history of tensor calculus and in sequence and study about the applications of tensor
Received 02 April 2017
calculus in theory of relativity, elasticity and physics.
Received in revised form
20 May 2017
Accepted 28 May 2017
Available online 15 June 2017
Keywords: Tensor; elasticity,
relativity
1. Introduction
The notion of tensors is as the extension of vectors. The scalar and The laws of physics can not depend on the frame of reference
vector quantities do not cover completely many physical and which the physicist chooses for the purpose of description.
geometrical quantities. For instance, stress in an elastic body and Accordingly it is aesthetically desirable and often convenient to
curl of a vector supposedly misunderstood as vectors are much utilise the tensor calculus as the mathematical background in
more than vectors. Indeed, they are tensors. which such laws can be formula-ted. In particular, Einstein found
The great Mathematician Gregorio Ricci was first introduced and it an excellent tool for the presentation of general relativity theory.
studied about the tensors, so G. Ricci is known as a founder of By use of tensor calculus we can observe that the space is flat or
tensor calculus. The emergence of tensor calculus, otherwise not. For example; if the Riemannian curvature
known as the absolute differential calculus, as a systematic branch h
R ijk or R hijk  gha R ijk
a
,( gha  0) is zero, then the space is flat.
of mathematics is due to Ricci and his pupil Levi - Civita. In The application of tensor calculus are discussed here by taking one
collaboration they published the first memoir on this subject : by one the following topics of application areas:
The tensor formulation was originated by G. Ricci and it become
rather popular when Albert Einstein used it as a natural tool for the
3. Tensor Calculus and Theory of Relativity
description of his general theory of relativity. The concept of a The general theory of Relativity which was developed by Einstein
tensor has its origin in the developments of differential geometry in order to disuss gravitation. He postulated the principle of
by Gauss, Riemann and Christoffel. covariance, which asserts that the laws of physics must be inde-
A tensor is a system of quantities or functions whose components pendent of the space - time coordinates. This swept away the
obey a certain law of transformation of coordinates from one privileged role of the Lorentz transformation. As a result of
system to another. Generally, tensors are three types; one is Minkowski space was replaced by a 4 - dimensional Riemannian
contravariant tensor, second is covariant tensor, and third is mixed space (V4 ) with the general metric
tensor. The rank or order of a tensor is defined as the total number j
ds 2  g dx i dx .
(1)
of indices of the components of a tensor. A tensor of rank zero is a ij
scalar. The examples of scalars are mass, time, distance, speed, Einstein also introduced the principle of equivalence, which in
temperature, energy etc., which are invariant under changes of the essence states that the fundamental tensor g can be chosen to
ij
coordinate system. A tensor of rank one is a vector. The examples
of vectors are displacement, velocity, acceleration, force, electric account for the presence of a gravitational field. That is, g
ij
field etc. The velocity of a fluid at any point is a contra variant
depends on the distribution of matter and energy in physical space.
tensor of rank one. Also, the tangent vector to a curve is a contra
Matter and energy can be specified by the energy - momentum
variant tensor of rank one.
The normal to a surface tensor Tij which in the special theory satisfies the equation
ij j The only forces, namely those due to gravitation, are
f (x i ) = c (Constant) is a covariant tensor of rank one. T F .
,i
A tensor of the second rank is the next in order of complexity after
however already taken into account by the choice of the
scalars and vectors. By a second rank tensor is meant a quantity
uniquely specified by nine real numbers in 3 – dimensional space. fundamental tensor We therefore ignore j and, in accordance
g ij . F
The stress tensor, the moment of inertia tensor, the deformation
with the principle of covariance, the energy - momentum tensor
tensor etc. are the examples of the second rank tensor. In n -
must now satisfy the equation T ij  0. We shall write this equation
dimensional space, it is specified by n2 functions. The Kronecker ,i
delta is a mixed tensor of rank two. in the equivalent form T ij  0 where i is the mixed
,i T, j  g jk Tik
2. Application of Tensor Calculus
In the course of study, we find that the nature of a tensor quantity energy - momentum tensor. The problem now is to determine Ti
does not change by the change of coordinate system. Due to this ,j
property, it is very useful for describing the physical laws as a function of the fundamental tensor g and their derivatives
ij
mathematically and so it is of great use in general Relativity
up to the second order, bearing in mind that Ti
theory, Differential geometry, Riemannian geometry, Mechanics, , j , i  0.
Elasticity, hydrodynamics, electromagnetic theory and many other
The Einstein tensor G i plays a very fundamental role in
disciplines of Science and engineering. ,j
*Corresponding Author, general theory of relativity. The Einstein tensor is defined as
E-mail address: [email protected] 1 (2)
All rights reserved: http://www.ijari.org Gi  g ik R  R i
,j jk j
2

IJARI 47
 ISSN 2347 - 3258
International Journal of Advance Research and Innovation
National Conference on Modern Trends in Analytical & Environmental Physics (Hindu College, Moradabad)
satisfies the equation Gi The equations of motion require on change of the coordinates by the positive orthogonal
, j , i  0. transformation y  a y . We see from y  a y that both
i ij j i ij j
T,i j , i  0, but very remarkably G,i j , i  0 is an identity in
y j and their differentials dy i are Cartesian vectors. Also, the
Riemannian geometry. This led Einstein to propose the relation
Koneker delta is a Cartesian tensor of the second order because
i i (3)
 T, j  G, j  0. ij  air a js rs  air a jr  ij
In effect these equations form the link between the physical in virtue of Similarly, a permutation tensor e is a
a ji aki   jk . ijk
energy - momentum tensor Ti and the geometrical tensor Gi of
,j ,j Cartesian tensor of the third order.
the M4 of general relativity. In order that Newton’s theory of The fundamental tensor of the Euclidean space is the Koneker
gravitation can be deduced as a first approximation from Einstein delta  . Hence all the Christoffel symbols are zero and so the
ij
theory, it was found necessary to choose   8  k / c 4 where k is comma notation for covariant derivatives now denotes the familiar
the usual gravitational constant 6.664  10 8 cm.3 gm.1 sec .2. partial derivatives which are Cartesian tensors.
The study of elasticity in terms of tensors falls into three parts: a
The value of c is 2.99796  1010 cm. sec .1, and
description of the strain or deformation of the elastic substance, a
 is 2.0 73  10 48 cm.1 gm.1 sec .2 in c. g. s. units. description of the force or stress which produces the deformation,
The motion of a particle which moves under the action of some and a generalization of Hook’s law in terms of tensor form.
force system can be represented in Minkowski space of curve, Deformation of a body may be described by giving the change in
called the world- line of the particle. In special theory, the world - the relative position of the parts of the body when the body is
lines of free particles and of light rays are respectively the subjected to some external force. As the result of this force the
geodesics and the null - geodesics of Minkowski space. The configuration of the body changes and we say that the body is in a
principle of equivalence demands that all particles be regarded as strained state.
free particles when gravitation is the only force under Let us consider a point P of the body, at the position
consideration. Then it follows from the principle of covariance r  (x1, y 1 , z1 ) relative to some fixed origin O. Let there is another
that the world - line of a particle under the action of gravita-tional point Q, at the position r  r  (x1  x1, y 1  y 1 , z1  z1 ) or ,
forces is a geodesic of the V4 with the metric (1). If no forces act
( x i  x i ). In the unstrained state the coordinates of Q relative to P
on the particle, then the world - line of a free particle is a geodesic
of the Minkowski space. Similarly, the world - line of a light ray is are x . Let each point of the body is subjected to an external
i
a null - geodesic. force which causes displacement u (r ) varying from point to
The velocity of a light ray is the constant c, and so we see from the
relation d 2  (dx 1 )2 (dx 2 )2  (dx 3 )2  c 2 (dx 4 )2 that for such point. Thus the point P is shifted to P1 with the displacement
a ray d   0. accordingly, the world - line of a light ray is a null u (r ) and point Q is shifted to Q 1 with the displacement
geodesic of the Minkowski space. v  u (r  r ). The coordinates of relative to P1 are
Q1
Also, the study of planetary motion, Einstein’s field equations,
poisson’s equations as an approximation of Einstein’s field y i  x i  ui (see in figure 1). Using Taylor’s expansion-
equations, Einstein’s universe and De Sitter’s universe etc.
become so simple and interesting by use of tensor calculus. Thus,
tensor calculus provide the tool for description of the general
theory of Relativity.
4. Tensor Calculus and Elasticity
The forces acting on a body are either external or internal. The
external forces may consist either of body forces such as gravity
which act on every particle of it, or of surface forces which act on
the external surface of the body, for example the pressure between
two bodies in contact. When an elastic body is subjected to an
external force or stress, it becomes deformed or strained. In the
formulation of the equations governing the equilibrium of an
elastic solid the Cartesian tensor of second rank are used. A
Cartesian tensor is defined as-
A Cartesian tensor of the n th order in a 3 - dimensional Euclidean
space is defined as a set of 3 n quantities which transform
Fig.1: The point P and Q and its shifting point P 1 and Q1 Related
according to equation (2.8) when the coordinates undergo a each other
positive orthogonal transformation. This is a less straingent
u
condition than that imposed on a tensor. So we see that all tensors u (r  r )  u i  x i  x i   u i  x i   i x j  ... (4.1)
are Cartesian tensor but a Cartesian tensor is not necessarily a x i
tensor in the usual sence. A tensor A is a Cartesian
k1 k2 ...kn Neglecting the second and higher order differentials, the above
th equation reduces to
tensor of the n order if the transformed com-ponents satisfy u
A i i ...i  Ak k ...k ai k ai k ... ai k . (4) u (r  r )  u i  x i   i x j  u (r )  grad u .r (4.2)
1 2 n 1 2 n 1 1 2 2 n n xi
which shows that the displacement of the point Q consists of two
parts, one of which, that is, u ( r ) is same for all the points of the

48
IJARI
 ISSN 2347 - 3258
International Journal of Advance Research and Innovation
National Conference on Modern Trends in Analytical & Environmental Physics (Hindu College, Moradabad)
body and therefore, corresponds to a translation of the body as a A1  dx 2 dx 3 and so on. The tangential stresses
whole. Relative displacement Q 1 and P1 (that is, deformation) P12 , P21 ,P13 ,P31 ,P23 ,P32 are the shearing stresses. Thus the total
is stress may be completely specified by the array matrix
u  u  r  r   u  r    r  grad  u .
  P11 P12 P13 
 (5)
   
u i  P    P  (8)
or, u i  u i x i  x i  u i x i 
x j
x j . 
 ij   21 22
P P23 ,
 P 
u i  31 P32 P33 
,
x j where the diagonal elements are normal stresses and the non-
The quantity being the gradient of a vector is a dyadic, that
diagonal elements are shearing stresses.
is, a tensor of second rank. Now,
Let us assume that the stresses are homogeneous. Then the forces
u 
1 u u 
j  
1 u u 
j  on the opposite faces will be reversed in sign as shown in figure 3.
u i  i x j   i  x   i  x
x j 2  x j x i  j 2  x j x i  j P = P ,P = P and P = P or, in general P = P . (9)
    12 21 13 31 32 23 ij ji
and hence the array of stresses, given equation (9) is symmetric.
 ij x j  ij x j , (6) The total force along x i is
 u j   u u  (10)
1 u
where ij   i   and   1  i  j
ij

(7) 
m
 aim a jk Pmk da  Pij da
2  x j x i  2  x j x i  k
    for static equilibrium. Thus in summation convention, we have
are symmetric and ant - symmetric respectively. Pij  aim a jk Pmk (11)
The term  x of the equation (6) is due to the rigid body
ij j which is the law of transformation of a tensor of second rank as
rotation of the element. Therefore the quantity  is known as defined by equation Tik   3
Therefore, the
ij  aij a Ti . k
rotation tensor. The remaining symmetric part ij is taken as a
j , 1
array (9), with the condition (10) represents a symmetric tensor of
pure strain known as strain tensor. second rank. This tensor is defined as stress tensor.
The stress is defined as the internal force per unit area acting on a
deformed body. The force in the x
i  direction acting on the face
dA whose normal in the x
j  direction is Pij dA , where

Pij's are actually pressures in the sence of force / area.

Whenever the term force is used, it is understood that the Pij are

to be multiplied by the appropriate area. There are the forces

acting on the small parallelopiped see in figure 2. The stress may
be directed normally or tangentially to the surfaces on which they Fig.3 Forces on the opposite faces will be reversed in sign
act. In case deforming forces act normally to a given area of an In the elementary theory of elasticity, Hooke’s law states that the
elastic medium, they produce pure alongation and the stresses are tension of an elastic string is proportional to the extension. In
called the normal stresses. If the deforming forces act tangentially other words, stress is proportional to strain. The corresponding
to the surface they produce shearing stresses. assumption in the general theory of elasticity is that the stress
tensor is a linear homogeneous function of the strain tensor. That
is
Pij  c e .
ijk k
It follows from the quotient that c is a Cartesian tensor of the
ijk
fourth order, and it is called the elasticity tensor. Further from the
symmetry of P and e , we find that c is symmetric not
ij k ijk
only with respect to the indices i and j but also with respect to
k and .
A body is said to be homogeneous if the elastic properties of the
body are independent of the point under consideration. This means
Fig.2 Forces acting on parallelepiped that the components of the elasticity tensor are all constants for a
Then obviously P   F1 , P   F2 , and P   F3
homogeneous body. We call a body isotropic if the elastic
33 properties at a point are the same in all directions at that point.
A1 A2 A3
11 22
This means that the elasticity tensor c transforms to c
are the normal stresses for ijk ijk
x 2  x 3 , x 1  x 3 and x 1  x 2 faces respectively, where the itself under any rotation of axes. A Cartesian tensor which
transforms into itself under a rotation of axes is called an isotropic
component F1 of the total force F act normal to the face area tensor.

49
IJARI
 ISSN 2347 - 3258
International Journal of Advance Research and Innovation
National Conference on Modern Trends in Analytical & Environmental Physics (Hindu College, Moradabad)
5. Tensor Calculus and Physics  4  (14)
 ,  J ,
Tensor analyses have developed as the natural generalization of c
vector analysis, and therefore, it may be expected that tensor whilst the remaining two of the relation (13) are accounted for by
analysis has relatively vider field of applications. As pointed out the equations of the set
earlier, the general theory of relativity was developed by using  ,    ,    ,   0 (15)
metric tensor, Christoffel symbols, curvature tensor, Ricci tensor,
which do not vanish identically.
and Einstein tensor. By use of electromagnetic field tensor, inertia
Maxwell’s equations in tensor form in Minkowski space. Thus
tensor, stress tensor, fundamental tensors etc., the readers feel
they are invariant under the Lorentz group of transformations.
interest in the study of physics and its relative field like dynamics
Position of a moving particle P be determine by a vector r . If the
and electrodynamics. The classical theory of electrodynamics,
according to Lorentz, is specified by the electric potential φ curvilinear coordi-nates of the terminal point of r are denoted by
x i (t ), then the equations of the path C of the particle can be
which is a scalar and the magnetic potential Ai which is a
written in the form
i i
vector. The electric field strength vector Ei and the magnetic C : x  x (t ), (16)
field strength vector H are derived from these potentials by the and we call the curve C, the trajectory of the particle.
i Velocity of a particle P is a vector dr whose components are
equations v  ,
dt
1 Ai dxi
Ei   grad Φ  and H i  curl A i . vi  (17)
c t dt
Using electrostatic units, Maxwell’s equations are as the velocity of the particle.
In transformed coordinates the components of the velocity are
div Ei  4 ,  dx i dx i dx
j
 dx i using (17)
vi   v j j ,
div Hi  0,  dt j
dx dt dx
 (12)
1 Hi
which follows the tensor law of transformation and hence the
curl Ei   0,  velocity of a particle at any point is a contravariant tensor (or, a
c t  contravariant vector) of rank one.

1 Ei 4
2
 The acceleration of a particle P is a vector a  dv  d r , and in
curl Hi   j ,
c t c i  dt dt 2
the general case, velocity is the function of of time t. Let us write
where ji is the current density vector and  is the charge its intrinsic derivative with respect to t according to the equation
A dA
 

   dx
density.
Skew - symmetrical tensor  defined by   A . Then
 t dt dt

   
j
v i dv i j dx
k
d 2x i k
i dx dx , (18)
Φ Φ ai    ijk v  
  Φ ,   Φ ,    t dt dt dt 2 jk dt dt
x  x 
 
i
Where, v is the intrinsic derivative and i are the Christoffel
and we immediately calculate that its non - vanishing components t jk
in the given coordinate system are
symbols calculated from metric tensor g . In the cartesian
23   η32   H1 ; η31   η13   H 2 ; η12   η21  H3 ; ij
η14   η41   c E1 ; η24   η42   c E2 ; η34   η43   c E3 . 2 i
rectangular coordinates the equation (17) reduces to a i  d x ,

The non - vanishing contra variant components 
2
may now dt
which is obviously the acceleration of the particle. Therefore, the
be obtained and which are
η23   η32   H1 ; η31   η13   H2 ; η12   η21   H3 ; quantity a i given by the equation (18) is called acceleration
vector in curvilinear coordinates system. If m is the mass of the
1 1 1
η14   η41   E1 ; η24   η42   E2 ; η34   η43   E3 . particle P, then the force vector (contravariant) in curvilinear
c c c coordinate is defined as
We now write Maxwell’s equations (12) in terms of  and J and
d 2x i
 
j k
Fi  ma i  m  m i dx dx . (19)
the results are readily verified to be respectively jk dt dt
dt 2
14 24 34 4 
   The covariant vector associated with F  may be readily be
4
J , 
x 1 x 2 x 3 c  written from equation (2.24), that is,
23 31 12  (13)
   0,  Fi  g i  F  (20)
x 1 x 2 x 3 
ij  j 4   and Fi  g i  F .

(21)
  4ji  0, 
x 4 x i x  The quantity W  F x , is scalar accordingly to quotient rule.
 i i
1i 2i 3i  4i
4 i  This quantity is defined as work done when the point of
    J . 
x 1 x 2 x 3 x 4 c  application of covariant force is moved through a small
The first and last equations of the relation (13) combine together displacement x i . In cartesian coordinate system there is no
into the form
distinction between covariant and contravariant components and
hence W  F x  F i x i .
i i

50
IJARI
 ISSN 2347 - 3258
International Journal of Advance Research and Innovation
National Conference on Modern Trends in Analytical & Environmental Physics (Hindu College, Moradabad)
From the study it clear that the tensor calculus are used so much in
physics. Thus the relations in between tensor calculus and physics
are much wide and by use tensor calculus, the physics becomes so
simple and interesting for readers.
References
[1.] L Brand, Vector and tensor calculus, 1947.
[2.] A Einstein. The principle of theory of relativity, 1923
[3.] GE Hay. Vector and tensor analysis, Dover Publications, New
York, 1953
[4.] Methodes de calcul differential absoluetleurs applications,
Mathematische Annalen, 54, 1901.
[5.] AD Michal. Matrix and tensor calculus, 1947.
[6.] M Ray. Theory of Relativity (Special and general), S. Chand
and Co., 1970.
[7.] JA Schouten. Tensor Calculus for Physicists, 1951.
[8.] JA Schouten. Ricci - Calculus (2nd Edition), An Introduction to
tensor analysis and its geometrical applications, Springer -
Verlag, Ferly, Berlin, 1954.
[9.] B Spain B. Tensor calculus, Oliver and Boyd, Edinburgh, 3rd
ed. 1960.
[10] JH Taylor. Vector analysis with an introduction to tensor
analysis, Prentice - Hall,
[11] NJ Englewood Clifffs.U S A , 1939
[12] CE Weatherburn. Riemannian geometry and the tensor calculus,
1938.
[13] CE Weatherburn. An Introduction to tensor Calculus and
Riemannian geometry, Cambridge University Press, London, 1942.
[14] AP Wills. Vector analysis with an introduction to tensor
analysis, Prentice - Hall, 1931 and Dover, New York, (1958).

51
IJARI
View publication stats