1

Vector and tensor analysis

Vectors and scalars
Vector methods have become standard tools for the physicists. In this chapter we
discuss the properties of the vectors and vector elds that occur in classical
physics. We will do so in a way, and in a notation, that leads to the formation
of abstract linear vector spaces in Chapter 5.
A physical quantity that is completely specied, in appropriate units, by a single
number (called its magnitude) such as volume, mass, and temperature is called a
scalar. Scalar quantities are treated as ordinary real numbers. They obey all the
regular rules of algebraic addition, subtraction, multiplication, division, and so
on.
There are also physical quantities which require a magnitude and a direction for
their complete specication. These are called vectors if their combination with
each other is commutative (that is the order of addition may be changed without
aecting the result). Thus not all quantities possessing magnitude and direction
are vectors. Angular displacement, for example, may be characterised by magni-
tude and direction but is not a vector, for the addition of two or more angular
displacements is not, in general, commutative (Fig. 1.1).
In print, we shall denote vectors by boldface letters (such as A) and use ordin-
ary italic letters (such as A) for their magnitudes; in writing, vectors are usually
represented by a letter with an arrow above it such as

A. A given vector A (or

A)
can be written as
A = A
^
A. (1.1)
where A is the magnitude of vector A and so it has unit and dimension, and
^
A is a
dimensionless unit vector with a unity magnitude having the direction of A. Thus
^
A = A,A.
1
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
A vector quantity may be represented graphically by an arrow-tipped line seg-
ment. The length of the arrow represents the magnitude of the vector, and the
direction of the arrow is that of the vector, as shown in Fig. 1.2. Alternatively, a
vector can be specied by its components (projections along the coordinate axes)
and the unit vectors along the coordinate axes (Fig. 1.3):
A = A
1
^ e
1
A
2
^ e
2
A^ e
3
=

3
i=1
A
i
^ e
i
. (1.2)
where ^ e
i
(i = 1. 2. 3) are unit vectors along the rectangular axes x
i
(x
1
= x. x
2
= y.
x
3
= z); they are normally written as
^
i.
^
j.
^
k in general physics textbooks. The
component triplet (A
1
. A
2
. A
3
) is also often used as an alternate designation for
vector A:
A = (A
1
. A
2
. A
3
). (1.2a)
This algebraic notation of a vector can be extended (or generalized) to spaces of
dimension greater than three, where an ordered n-tuple of real numbers,
(A
1
. A
2
. . . . . A
n
), represents a vector. Even though we cannot construct physical
vectors for n 3, we can retain the geometrical language for these n-dimensional
generalizations. Such abstract ``vectors'' will be the subject of Chapter 5.
2
VECTOR AND TENSOR ANALYSIS
Figure 1.1. Rotation of a parallelpiped about coordinate axes.
Figure 1.2. Graphical representation of vector A.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
Direction angles and direction cosines
We can express the unit vector
^
A in terms of the unit coordinate vectors ^ e
i
. From
Eq. (1.2), A = A
1
^ e
1
A
2
^ e
2
A^ e
3
, we have
A = A
A
1
A
^ e
1

A
2
A
^ e
2

A
3
A
^ e
3
_ _
= A
^
A.
Now A
1
,A = cos c. A
2
,A = cos u, and A
3
,A = cos are the direction cosines of
the vector A, and c, u, and are the direction angles (Fig. 1.4). Thus we can write
A = A(cos c^ e
1
cos u^ e
2
cos ^ e
3
) = A
^
A;
it follows that
^
A = (cos c^ e
1
cos u^ e
2
cos ^ e
3
) = (cos c. cos u. cos ). (1.3)
3
DIRECTION ANGLES AND DIRECTION COSINES
Figure 1.3. A vector A in Cartesian coordinates.
Figure 1.4. Direction angles of vector A.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
Vector algebra
Equality of vectors
Two vectors, say A and B, are equal if, and only if, their respective components
are equal:
A = B or (A
1
. A
2
. A
3
) = (B
1
. B
2
. B
3
)
is equivalent to the three equations
A
1
= B
1
. A
2
= B
2
. A
3
= B
3
.
Geometrically, equal vectors are parallel and have the same length, but do not
necessarily have the same position.
Vector addition
The addition of two vectors is dened by the equation
A B = (A
1
. A
2
. A
3
) (B
1
. B
2
. B
3
) = (A
1
B
1
. A
2
B
2
. A
3
B
3
).
That is, the sum of two vectors is a vector whose components are sums of the
components of the two given vectors.
We can add two non-parallel vectors by graphical method as shown in Fig. 1.5.
To add vector B to vector A, shift B parallel to itself until its tail is at the head of
A. The vector sum A B is a vector C drawn from the tail of A to the head of B.
The order in which the vectors are added does not aect the result.
Multiplication by a scalar
If c is scalar then
cA = (cA
1
. cA
2
. cA
3
).
Geometrically, the vector cA is parallel to A and is c times the length of A. When
c = 1, the vector A is one whose direction is the reverse of that of A, but both
4
VECTOR AND TENSOR ANALYSIS
Figure 1.5. Addition of two vectors.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
have the same length. Thus, subtraction of vector B from vector A is equivalent to
adding B to A:
A B = A (B).
We see that vector addition has the following properties:
(a) A B = B A (commutativity);
(b) (A B) C = A (B C) (associativity);
(c) A 0 = 0 A = A;
(d) A (A) = 0.
We now turn to vector multiplication. Note that division by a vector is not
dened: expressions such as k,A or B,A are meaningless.
There are several ways of multiplying two vectors, each of which has a special
meaning; two types are dened.
The scalar product
The scalar (dot or inner) product of two vectors A and B is a real number dened
(in geometrical language) as the product of their magnitude and the cosine of the
(smaller) angle between them (Figure 1.6):
A B = ABcos 0 (0 _ 0 _ ). (1.4)
It is clear from the denition (1.4) that the scalar product is commutative:
A B = B A. (1.5)
and the product of a vector with itself gives the square of the dot product of the
vector:
A A = A
2
. (1.6)
If A B = 0 and neither A nor B is a null (zero) vector, then A is perpendicular to B.
5
THE SCALAR PRODUCT
Figure 1.6. The scalar product of two vectors.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
We can get a simple geometric interpretation of the dot product from an
inspection of Fig. 1.6:
(Bcos 0)A = projection of B onto A multiplied by the magnitude of A.
(Acos 0)B = projection of A onto B multiplied by the magnitude of B.
If only the components of A and B are known, then it would not be practical to
calculate A B from denition (1.4). But, in this case, we can calculate A B in
terms of the components:
A B = (A
1
^ e
1
A
2
^ e
2
A
3
^ e
3
) (B
1
^ e
1
B
2
^ e
2
B
3
^ e
3
); (1.7)
the right hand side has nine terms, all involving the product ^ e
i
^ e
j
. Fortunately,
the angle between each pair of unit vectors is 908, and from (1.4) and (1.6) we nd
that
^ e
i
^ e
j
= c
ij
. i. j = 1. 2. 3. (1.8)
where c
ij
is the Kronecker delta symbol
c
ij
=
0. if i ,= j.
1. if i = j.
_
(1.9)
After we use (1.8) to simplify the resulting nine terms on the right-side of (7), we
obtain
A B = A
1
B
1
A
2
B
2
A
3
B
3
=

3
i=1
A
i
B
i
. (1.10)
The law of cosines for plane triangles can be easily proved with the application
of the scalar product: refer to Fig. 1.7, where C is the resultant vector of A and B.
Taking the dot product of C with itself, we obtain
C
2
= C C = (A B) (A B)
= A
2
B
2
2A B = A
2
B
2
2ABcos 0.
which is the law of cosines.
6
VECTOR AND TENSOR ANALYSIS
Figure 1.7. Law of cosines.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
A simple application of the scalar product in physics is the work W done by a
constant force F: W = F r, where r is the displacement vector of the object
moved by F.
The vector (cross or outer) product
The vector product of two vectors A and B is a vector and is written as
C = A B. (1.11)
As shown in Fig. 1.8, the two vectors A and B form two sides of a parallelogram.
We dene C to be perpendicular to the plane of this parallelogram with its
magnitude equal to the area of the parallelogram. And we choose the direction
of C along the thumb of the right hand when the ngers rotate from A to B (angle
of rotation less than 1808).
C = A B = ABsin 0^ e
C
(0 _ 0 _ ). (1.12)
From the denition of the vector product and following the right hand rule, we
can see immediately that
A B = B A. (1.13)
Hence the vector product is not commutative. If A and B are parallel, then it
follows from Eq. (1.12) that
A B = 0. (1.14)
In particular
A A = 0. (1.14a)
In vector components, we have
A B = (A
1
^ e
1
A
2
^ e
2
A
3
^ e
3
) (B
1
^ e
1
B
2
^ e
2
B
3
^ e
3
). (1.15)
7
THE VECTOR ( CROSS OR OUTER) PRODUCT
Figure 1.8. The right hand rule for vector product.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
Using the following relations
^ e
i
^ e
i
= 0. i = 1. 2. 3.
^ e
1
^ e
2
= ^ e
3
. ^ e
2
^ e
3
= ^ e
1
. ^ e
3
^ e
1
= ^ e
2
.
(1.16)
Eq. (1.15) becomes
A B = (A
2
B
3
A
3
B
2
)^ e
1
(A
3
B
1
A
1
B
3
)^ e
2
(A
1
B
2
A
2
B
1
)^ e
3
. (1.15a)
This can be written as an easily remembered determinant of third order:
A B =
^ e
1
^ e
2
^ e
3
A
1
A
2
A
3
B
1
B
2
B
3

. (1.17)
The expansion of a determinant of third order can be obtained by diagonal multi-
plication by repeating on the right the rst two columns of the determinant and
adding the signed products of the elements on the various diagonals in the result-
ing array:
The non-commutativity of the vector product of two vectors now appears as a
consequence of the fact that interchanging two rows of a determinant changes its
sign, and the vanishing of the vector product of two vectors in the same direction
appears as a consequence of the fact that a determinant vanishes if one of its rows
is a multiple of another.
The determinant is a basic tool used in physics and engineering. The reader is
assumed to be familiar with this subject. Those who are in need of review should
read Appendix II.
The vector resulting from the vector product of two vectors is called an axial
vector, while ordinary vectors are sometimes called polar vectors. Thus, in Eq.
(1.11), C is a pseudovector, while A and B are axial vectors. On an inversion of
coordinates, polar vectors change sign but an axial vector does not change sign.
A simple application of the vector product in physics is the torque s of a force F
about a point O: s = F r, where r is the vector from O to the initial point of the
force F (Fig. 1.9).
We can write the nine equations implied by Eq. (1.16) in terms of permutation
symbols
ijk
:
^ e
i
^ e
j
=
ijk
^ e
k
. (1.16a)
8
VECTOR AND TENSOR ANALYSIS
a
1
a
2
a
3
b
1
b
2
b
3
c
1
c
2
c
c
_
_
_
_
a
1
a
2
b
1
b
2
c
1
c
2




-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-

Cambridge University Press www.cambridge.org

Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
where
ijk
is dened by

ijk
=
1
1
0
if (i. j. k) is an even permutation of (1. 2. 3).
if (i. j. k) is an odd permutation of (1. 2. 3).
otherwise (for example. if 2 or more indices are equal).
_
_
_
(1.18)
It follows immediately that

ijk
=
kij
=
jki
=
jik
=
kji
=
ikj
.
There is a very useful identity relating the
ijk
and the Kronecker delta symbol:

3
k=1

mnk

ijk
= c
mi
c
nj
c
mj
c
ni
. (1.19)

j.k

mjk

njk
= 2c
mn
.

i.j.k

2
ijk
= 6. (1.19a)
Using permutation symbols, we can now write the vector product A B as
A B =

3
i=1
A
i
^ e
i
_ _

3
j=1
B
j
^ e
j
_ _
=

3
i.j
A
i
B
j
^ e
i
^ e
j
_ _
=

3
i.j.k
A
i
B
j

ijk
_ _
^ e
k
.
Thus the kth component of A B is
(A B)
k
=

i.j
A
i
B
j

ijk
=

i.j

kij
A
i
B
j
.
If k = 1, we obtain the usual geometrical result:
(A B)
1
=

i.j

1ij
A
i
B
j
=
123
A
2
B
3

132
A
3
B
2
= A
2
B
3
A
3
B
2
.
9
THE VECTOR ( CROSS OR OUTER) PRODUCT
Figure 1.9. The torque of a force about a point O.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information
The triple scalar product AE (BC)
We now briey discuss the scalar A (B C). This scalar represents the volume of
the parallelepiped formed by the coterminous sides A, B, C, since
A (B C) = ABC sin 0 cos c = hS = volume.
S being the area of the parallelogram with sides B and C, and h the height of the
parallelogram (Fig. 1.10).
Now
A (B C) = A
1
^ e
1
A
2
^ e
2
A
3
^ e
3
( )
^ e
1
^ e
2
^ e
3
B
1
B
2
B
3
C
1
C
2
C
3

= A
1
(B
2
C
3
B
3
C
2
) A
2
(B
3
C
1
B
1
C
3
) A
3
(B
1
C
2
B
2
C
1
)
so that
A (B C) =
A
1
A
2
A
3
B
1
B
2
B
3
C
1
C
2
C
3

. (1.20)
The exchange of two rows (or two columns) changes the sign of the determinant
but does not change its absolute value. Using this property, we nd
A (B C) =
A
1
A
2
A
3
B
1
B
2
B
3
C
1
C
2
C
3

=
C
1
C
2
C
3
B
1
B
2
B
3
A
1
A
2
A
3

= C (A B).
that is, the dot and the cross may be interchanged in the triple scalar product.
A (B C) = (A B) C (1.21)
10
VECTOR AND TENSOR ANALYSIS
Figure 1.10. The triple scalar product of three vectors A, B, C.
Cambridge University Press www.cambridge.org
Cambridge University Press
0521652278 - Mathematical Methods for Physicists: A Concise Introduction - Tai L. Chow
Excerpt
More information