MMP-I Lecture No.

06

Methods of Mathematical
Physics-I
Lecture No. 02
Course Code: PHY-5502
RECOMMENDED TEXTBOOKS
Mary L. Boas, Mathematical Methods in The Physical Sciences, 3rd Ed., John Wiley & Sons,
Inc. (2006)

REFERENCE BOOKS
(George B. Arfken, Hans J. Weber, Frank E. Harris, Mathematical Physics, Academic Press 7th
Ed. (2006).
E. Butkov, Mathematical Physics, Addison-Wesley 1968

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06

Methods of Mathematical Physics-I

Chapter No. 02
Tensor Analysis:
 Cartesian Tensors,
 Tensor Notation and Operations
 Inertia Tensor, Kronecker Delta and Levi-Civita Symbol
 Pseudo-vectors and Pseudo tensors
 More About Applications
 Curvilinear Coordinates
 Vector Operators in Orthogonal Curvilinear
Coordinates
 Non-Cartesian Tensors.
RECOMMENDED TEXTBOOKS
 Mary L. Boas, Mathematical Methods in The Physical Sciences, 3rd Ed., John Wiley & Sons, Inc. (2006)
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Chapter No. 02
Tensor Analysis
A tensor is a mathematical object that generalizes the concepts of scalars, vectors, and
matrices to higher dimensions. It is a multi-dimensional array of numerical values that
transforms according to specific rules under a change of coordinate system, making it a
fundamental concept in fields like physics, engineering, and mathematics.
Key Features of Tensors:
Order or Rank: The number of dimensions (or indices) a tensor has.
1. A scalar is a rank-0 tensor (single value).
2. A vector is a rank-1 tensor (1D array).
3. A matrix is a rank-2 tensor (2D array).
4. Higher-rank tensors have three or more indices.
In 3-dimensional space
 A scalar has 30 = 1 components
 A vector has 31 = 3 components
 A second-rank tensor has 32 = 9 components
 In general a tensor of rank n has 3n components.
Components and Basis Dependence: A tensor's components depend on the chosen
coordinate system, but the tensor itself is independent of the coordinate system.
RECOMMENDED TEXTBOOKS
 Mary L. Boas, Mathematical Methods in The Physical Sciences, 3rd Ed., John Wiley & Sons, Inc. (2006)
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Chapter No. 02
Tensor Analysis
Transformation Laws: Tensors follow specific transformation rules when the coordinate system
changes:
𝜕𝑥 ′𝑖 𝑗
For example, a vector 𝑉 𝑖 in
one system transforms as = 𝑉 ′𝑖 𝑉
𝜕𝑥 𝑗
Types of Tensors:
Contravariant tensors: Represented with upper indices (e.g., 𝑇 𝑖 ).
Covariant tensors: Represented with lower indices (e.g., 𝑇𝑗 ).
Mixed tensors: Contain both upper and lower indices (e.g.,𝑇𝑗𝑖 ).

Applications:
•Physics: Stress tensors in mechanics, energy-momentum tensors in relativity.
•Engineering: Strain and elasticity tensors.
•Mathematics: Differential geometry, machine learning, and numerical analysis.

RECOMMENDED TEXTBOOKS
 Mary L. Boas, Mathematical Methods in The Physical Sciences, 3rd Ed., John Wiley & Sons, Inc. (2006)
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Chapter No. 02
Tensor Analysis
Example 1
Think of a beam carrying a load; there are stresses and strains in the material of the beam. If
we imagine cutting the beam in two by a plane perpendicular to the x direction, we realize
that there is a force per unit area exerted by the material on one side of our imaginary cut on
the other side. This is a vector, so it has three components Pxx, Pxy, Pxz, where the first
subscript x is to emphasize that this is a force across a plane perpendicular to the x direction.
Similarly, if we consider a plane perpendicular to the y direction, there is a force per unit area
across this plane with components Pyx, Pyy, Pyz; and finally across a plane perpendicular to
the z direction there is a force per unit area with components Pzx, Pzy, Pzz. At a point in the
material, then, we have a set of nine quantities which could be displayed as a matrix:.
This is a second-rank tensor known as the stress tensor.
The forces (per unit area) Pxx, Pyy, Pzz are pressures or
tensions; the others are shear forces (per unit area).
For example Pzy is a force per unit area in the y direction
acting across a plane perpendicular to the z direction; this
force tends to shear the beam.
we have simply indicated the number of components that tensors of the various ranks
have. This is not the whole story. To see what else is required, let us talk about first-rank
tensors, that is vectors, which are already familiar to you.
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Chapter No. 02
Tensor Analysis
Example 2
Now let us consider the idea of a vector as a set of three components. In order to talk about
components, we must have a coordinate system. There are infinitely many coordinate
systems—even for rectangular axes (x, y, z) there are infinitely many sets of rotated axes. Thus
we must say that a vector consists of a set of three components in each coordinate system. If
the components of a vector relative to one set of axes are given, we know from elementary
vector analysis that the component of the vector in any direction, or its components relative
to any rotated set of axes, can be found by taking projections. Then the new components are
definite combinations of the old components. This fact allows us to decide whether a physical
quantity is really a vector or not. There is a similar requirement for tensors, for example the
second-rank stress tensor we have described. We could imagine cutting the beam by a plane
oriented in any given direction and ask for the force per unit area acting across this plane.
Each component of this force is a certain combination of the nine components of the stress
tensor. Thus the components of the stress tensor in any other coordinate system are definite
combinations of the nine components of the tensor relative to the (x, y, z) axes. In other
words, tensors of all ranks, like vectors, have a physical meaning which is independent of the
reference coordinate system and there are definite mathematical laws which relate their
components in two systems.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Chapter No. 02
Tensor Analysis
Example 2
You may wonder why we cannot make just any set of components (3 for a vector, 9 for a
second-rank tensor, etc.), given in one coordinate system, a tensor by defining its components
in other systems by the correct transformation laws.
Mathematically, we could! But for a physical entity, we are not free to define its components
in various coordinate systems; they are determined by physical fact. We merely give a
mathematical description of the entity and identify it as a scalar, a vector, a second-rank
tensor, etc. (or perhaps none of these). We can see again now why an arrow associated with a
rotation is not a vector. If we treat the arrow as a vector and take components of it, these
component vectors do not represent rotations which can be combined to give the original
rotation. Thus a vector which looks superficially like the arrow we have defined is not a correct
mathematical representation of the physical entity (a rotation) we are trying to describe.
The ideas of an abstract vector space grew out of the geometry of three-dimensional
displacement vectors. A change of coordinate system (for example, rotation of axes)
corresponds to a change of basis in a vector space.
Here we want the word “vector” to refer to all physical quantities which transform properly.
Thus we shall find the transformation law for a displacement
vector, and then
A tensor define
which a vector properly
transforms as any entity
underwhich obeys the
a rotation same law. (x, y, z) axes is
of rectangular
called a Cartesian tensor
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
A tensor which transforms properly under a rotation of rectangular (x, y, z) axes is called a
Cartesian tensor
Now we want to consider passive rotations (vector fixed,
axes rotated), in order to find how the components of a
displacement vector in one coordinate system are related
to its components in a rotated system.
Let (x, y, z) be a set of rectangular axes and (x, y, z) another
set obtained by rotating the axes in any manner keeping the
origin fixed (Figure ). In the table, we list the cosines of the
nine angles between the (x, y, z) axes and the (x, y, z)
axes.
In the table, l2 means the cosine of the angle between the
x axis and the y axis, etc. A vector r (Figure) has
components x, y, z or x’, y’, z’ relative to the two
coordinate systems; we want to find the relations
between the two sets of components.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
Example 1. Let i, j, k be unit basis vectors along the (x, y, z) axes and I’, j’, k’ be unit basis
vectors along the (x’, y’, z’) axes. Then the vector r can be written in terms of either set of
components and basis vectors as follows:
r = ix + jy + kz = i’x + j’y + k’z.

Taking the dot product of this equation with i’, we get

Similarly, dotting r into j’ and k’, and using above table can be written as

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
Example 2. In the same way, dotting r with i, j, k in turn, we get equations for x, y, z
in terms of x’, y’, z’:

r = ix + jy + kz = i’x + j’y + k’z.

These transformation equations may be written more concisely in matrix notation.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
Example 2. In the same way, dotting r with i, j, k in turn, we get equations for x, y, z
in terms of x’, y’, z’: r = ix + jy + kz = i’x + j’y + k’z.

These transformation equations may be written more concisely in matrix notation.

where AT is the
transpose of A.
A rotation matrix is an orthogonal matrix, and for an orthogonal matrix, AT = A-1.
Above equations tell us how displacement vectors in a rectangular coordinate system transform
under a rotation of axes. We now use this result to define Cartesian vectors, that is, vectors
which transform in the same way that displacement vectors do under rotations of rectangular
(Cartesian) axes. We will then generalize this to define Cartesian tensors of other ranks.
Dr. Nasar Ahmed:
Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
A Cartesian vector V consists of a set of three numbers (components) in every rectangular
coordinate system; if Vx, Vy, Vz are the components in one system and Vx’, Vy’, Vz’ are the
components in a rotated system, these two sets of components are related by an equation
namely,

where A is
the rotation
matrix

We can simplify our notation by making the following changes.

In this notation and become

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
A Cartesian vector V consists of a set of three numbers (components) in every rectangular
coordinate system; if Vx, Vy, Vz are the components in one system and Vx’, Vy’, Vz’ are the
components in a rotated system, these two sets of components are related by an equation
namely,

Become

Become

Now element i, j of AT is the same as element j, i of A, so the coefficients in (above) are

aji instead of aij as they were in (1). It is now straightforward to define tensors.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
A tensor of rank zero has one component which is unchanged by a rotation of axes; it is called
an invariant or a scalar. Simple examples are the length of a vector, or the dot product of two
vectors. A first rank tensor is just a vector. A tensor of second rank has nine components (in
three dimensions) in every rectangular coordinate system. If we call the components in one
system Tij the components T’kl in a rotated coordinate system are given by (below), where the
a’s are the direction cosines in the rotation matrix A.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
Example 3. Let U and V be vectors; we form the following array (in each coordinate system)
from the components U1, U2, U3 and V1, V2, V3 of U and V (in that coordinate system):

We can show that these nine quantities are the components of a second-rank tensor
which we shall denote by UV. Note that this is not a dot product or a cross product; it is
called the direct product of U and V (or outer product or tensor product). Since U and V
are vectors, their components in a rotated coordinate system are, by

Hence the components of the second-rank tensor

UV are

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
MMP-I Lecture No. 06
Tensor Analysis
CARTESIAN TENSORS
Example 3. Let U and V be vectors; we form the following array (in each coordinate system)
from the components U1, U2, U3 and V1, V2, V3 of U and V (in that coordinate system):

which is just
generalizes immediately. For example, a 4th-rank Cartesian tensor
is defined as a set of 34 or 81 components Tijkl, in every rectangular
coordinate system, which transform to a rotated coordinate system
by the equations

where i, j, k, l take the values 1, 2, 3. Note that a 4th-rank tensor has 4 indices and
requires four a’s in its definition. Similarly, an nth-rank tensor has n indices and requires n
a’s in its definition.

Dr. Nasar Ahmed:

Associate Professor. Department of Physics, University of Azad Jammu and Kashmir, Muzaffarabad
Solve all the problems of SECTION 2