This action might not be possible to undo. Are you sure you want to continue?

MSC 97U20

PACS 01.30.Pp

R. A. Sharipov. Quick Introduction to Tensor Analysis: lecture notes.

Freely distributed on-line. Is free for individual use and educational purposes.

Any commercial use without written consent from the author is prohibited.

This book was written as lecture notes for classes that I taught to undergraduate

students majoring in physics in February 2004 during my time as a guest instructor

at The University of Akron, which was supported by Dr. Sergei F. Lyuksyutov’s

grant from the National Research Council under the COBASE program. These 4

classes have been taught in the frame of a regular Electromagnetism course as an

introduction to tensorial methods.

I wrote this book in a ”do-it-yourself” style so that I give only a draft of tensor

theory, which includes formulating deﬁnitions and theorems and giving basic ideas

and formulas. All other work such as proving consistence of deﬁnitions, deriving

formulas, proving theorems or completing details to proofs is left to the reader in

the form of numerous exercises. I hope that this style makes learning the subject

really quick and more eﬀective for understanding and memorizing.

I am grateful to Department Chair Prof. Robert R. Mallik for the opportunity

to teach classes and thus to be involved fully in the atmosphere of an American

university. I am also grateful to

Mr. M. Boiwka (mboiwka@hotmail.com)

Mr. A. Calabrese (ajc10@uakron.edu)

Mr. J. Comer (funnybef@lycos.com)

Mr. A. Mozinski (arm5@uakron.edu)

Mr. M. J. Shepard (sheppp2000@yahoo.com)

for attending my classes and reading the manuscript of this book. I would like to

especially acknowledge and thank Mr. Jeﬀ Comer for correcting the grammar and

wording in it.

Contacts to author.

Oﬃce: Mathematics Department, Bashkir State University,

32 Frunze street, 450074 Ufa, Russia

Phone: 7-(3472)-23-67-18

Fax: 7-(3472)-23-67-74

Home: 5 Rabochaya street, 450003 Ufa, Russia

Phone: 7-(917)-75-55-786

E-mails: R Sharipov@ic.bashedu.ru,

r-sharipov@mail.ru,

ra sharipov@hotmail.com,

URL: http://www.geocities.com/r-sharipov

CopyRight c Sharipov R.A., 2004

CONTENTS.

CONTENTS. ............................................................................................... 3.

CHAPTER I. PRELIMINARY INFORMATION. .......................................... 4.

§ 1. Geometrical and physical vectors. ........................................................... 4.

§ 2. Bound vectors and free vectors. .............................................................. 5.

§ 3. Euclidean space. .................................................................................... 8.

§ 4. Bases and Cartesian coordinates. ........................................................... 8.

§ 5. What if we need to change a basis ? ...................................................... 12.

§ 6. What happens to vectors when we change the basis ? ............................ 15.

§ 7. What is the novelty about vectors that we learned knowing

transformation formula for their coordinates ? ....................................... 17.

CHAPTER II. TENSORS IN CARTESIAN COORDINATES. ..................... 18.

§ 8. Covectors. ........................................................................................... 18.

§ 9. Scalar product of vector and covector. .................................................. 19.

§ 10. Linear operators. ............................................................................... 20.

§ 11. Bilinear and quadratic forms. ............................................................. 23.

§ 12. General deﬁnition of tensors. .............................................................. 25.

§ 13. Dot product and metric tensor. .......................................................... 26.

§ 14. Multiplication by numbers and addition. ............................................. 27.

§ 15. Tensor product. ................................................................................. 28.

§ 16. Contraction. ...................................................................................... 28.

§ 17. Raising and lowering indices. .............................................................. 29.

§ 18. Some special tensors and some useful formulas. ................................... 29.

CHAPTER III. TENSOR FIELDS. DIFFERENTIATION OF TENSORS. ... 31.

§ 19. Tensor ﬁelds in Cartesian coordinates. ................................................ 31.

§ 20. Change of Cartesian coordinate system. .............................................. 32.

§ 21. Diﬀerentiation of tensor ﬁelds. ............................................................ 34.

§ 22. Gradient, divergency, and rotor. Laplace and d’Alambert operators. ..... 35.

CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES. ..... 38.

§ 23. General idea of curvilinear coordinates. ............................................... 38.

§ 24. Auxiliary Cartesian coordinate system. ............................................... 38.

§ 25. Coordinate lines and the coordinate grid. ............................................ 39.

§ 26. Moving frame of curvilinear coordinates. ............................................. 41.

§ 27. Dynamics of moving frame. ................................................................ 42.

§ 28. Formula for Christoﬀel symbols. ......................................................... 42.

§ 29. Tensor ﬁelds in curvilinear coordinates. ............................................... 43.

§ 30. Diﬀerentiation of tensor ﬁelds in curvilinear coordinates. ...................... 44.

§ 31. Concordance of metric and connection. ............................................... 46.

REFERENCES. ......................................................................................... 47.

CHAPTER I

PRELIMINARY INFORMATION.

§ 1. Geometrical and physical vectors.

Vector is usually understood as a segment of straight line equipped with an

arrow. Simplest example is displacement vector a. Say its length is 4 cm, i. e.

|a| = 4 cm.

You can draw it on the paper as shown on Fig. 1a. Then it means that point B is

4 cm apart from the point A in the direction pointed to by vector a. However, if

you take velocity vector v for a stream in a brook, you cannot draw it on the paper

immediately. You should ﬁrst adopt a scaling convention, for example, saying that

1 cm on paper represents 1 cm/sec (see Fig. 1b).

Conclusion 1.1. Vectors with physical meaning other than displacement vec-

tors have no unconditional geometric presentation. Their geometric presentation is

conventional; it depends on the scaling convention we choose.

Conclusion 1.2. There are plenty of physical vectors, which are not geomet-

rically visible, but can be measured and then drawn as geometric vectors.

One can consider unit vector m. Its length is equal to unity not 1 cm, not 1 km,

not 1 inch, and not 1 mile, but simply number 1:

|m| = 1.

Like physical vectors, unit vector m cannot be drawn without adopting some

scaling convention. The concept of a unit vector is a very convenient one. By

§ 2. BOUND VECTORS AND FREE VECTORS. 5

multiplying m to various scalar quantities, we can produce vectorial quantities of

various physical nature: velocity, acceleration, force, torque, etc.

Conclusion 1.3. Along with geometrical and physical vectors one can imagine

vectors whose length is a number with no unit of measure.

§ 2. Bound vectors and free vectors.

All displacement vectors are bound ones. They are bound to those points whose

displacement they represent. Free vectors are usually those representing global

physical parameters, e. g. vector of angular velocity ω for Earth rotation about

its axis. This vector produces the Coriolis force aﬀecting water streams in small

rivers and in oceans around the world. Though it is usually drawn attached to the

North pole, we can translate this vector to any point along any path provided we

keep its length and direction unchanged.

6 CHAPTER I. PRELIMINARY INFORMATION.

The next example illustrates the concept of a vector ﬁeld. Consider the water

ﬂow in a river at some ﬁxed instant of time t. For each point P in the water the

velocity of the water jet passing through

this point is deﬁned. Thus we have a

function

v = v(t, P). (2.1)

Its ﬁrst argument is time variable t. The

second argument of function (2.1) is not

numeric. It is geometric object — a

point. Values of a function (2.1) are also

not numeric: they are vectors.

Definition2.1. A vector-valued func-

tion with point argument is called vector

ﬁeld. If it has an additional argument t,

it is called a time-dependent vector ﬁeld.

Let v be the value of function (2.1) at

the point A in a river. Then vector v is a

bound vector. It represents the velocity

of the water jet at the point A. Hence, it is bound to point A. Certainly, one can

translate it to the point B on the bank of the river (see Fig. 3). But there it loses

its original purpose, which is to mark the water velocity at the point A.

Conclusion 2.1. There exist functions with non-numeric arguments and non-

numeric values.

Exercise 2.1. What is a scalar ﬁeld ? Suggest an appropriate deﬁnition by

analogy with deﬁnition 2.1.

Exercise 2.2 (for deep thinking). Let y = f(x) be a function with a non-

numeric argument. Can it be continuous ? Can it be diﬀerentiable ? In general,

answer is negative. However, in some cases one can extend the deﬁnition of conti-

nuity and the deﬁnition of derivatives in a way applicable to some functions with

non-numeric arguments. Suggest your version of such a generalization. If no ver-

sions, remember this problem and return to it later when you gain more experience.

Let A be some ﬁxed point (on the ground, under the ground, in the sky, or in

outer space, wherever). Consider all vectors of some physical nature bound to this

point (say all force vectors). They constitute an inﬁnite set. Let’s denote it V

A

.

We can perform certain algebraic operations over the vectors from V

A

:

(1) we can add any two of them;

(2) we can multiply any one of them by any real number α ∈ R;

These operations are called linear operations and V

A

is called a linear vector space.

Exercise 2.3. Remember the parallelogram method for adding two vectors

(draw picture). Remember how vectors are multiplied by a real number α. Consider

three cases: α > 0, α < 0, and α = 0. Remember what the zero vector is. How it

is represented geometrically ?

§ 2. BOUND VECTORS AND FREE VECTORS. 7

Exercise 2.4. Do you remember the exact mathematical deﬁnition of a linear

vector space ? If yes, write it. If no, visit Web page of Jim Heﬀeron

http://joshua.smcvt.edu/linearalgebra/

and download his book [1]. Keep this book for further references. If you ﬁnd it use-

ful, you can acknowledge the author by sending him e-mail: jim@joshua.smcvt.edu

Conclusion 2.2. Thus, each point A of our geometric space is not so simple,

even if it is a point in a vacuum. It can be equipped with linear vector spaces of

various natures (such as a space of force vectors in the above example). This idea,

where each point of vacuum space is treated as a container for various physical

ﬁelds, is popular in modern physics. Mathematically it is realized in the concept of

bundles: vector bundles, tensor bundles, etc.

Free vectors, taken as they are, do not form a linear vector space. Let’s denote

by V the set of all free vectors. Then V is union of vector spaces V

A

associated

with all points A in space:

V =

¸

A∈E

V

A

. (2.2)

The free vectors forming this set (2.2)

are too numerous: we should work to

make them conﬁne the deﬁnition of a

linear vector space. Indeed, if we have

a vector a and if it is a free vector, we

can replicate it by parallel translations

that produce inﬁnitely many copies of it

(see Fig. 4). All these clones of vector

a form a class, the class of vector a.

Let’s denote it as Cl(a). Vector a is a

representative of its class. However, we

can choose any other vector of this class

as a representative, say it can be vector

˜ a. Then we have

Cl(a) = Cl(˜ a).

Let’s treat Cl(a) as a whole unit, as one indivisible object. Then consider the set

of all such objects. This set is called a factor-set, or quotient set. It is denoted as

V/ ∼. This quotient set V/ ∼ satisﬁes the deﬁnition of linear vector space. For

the sake of simplicity further we shall denote it by the same letter V as original

set (2.2), from which it is produced by the operation of factorization.

Exercise 2.5. Have you heard about binary relations, quotient sets, quotient

groups, quotient rings and so on? If yes, try to remember strict mathematical

deﬁnitions for them. If not, then have a look to the references [2], [3], [4]. Cer-

tainly, you shouldn’t read all of these references, but remember that they are freely

available on demand.

CopyRight c Sharipov R.A., 2004.

8 CHAPTER I. PRELIMINARY INFORMATION.

3. Euclidean space.

What is our geometric space ? Is it a linear vector space ? By no means. It

is formed by points, not by vectors. Properties of our space were ﬁrst system-

atically described by Euclid, the Greek

mathematician of antiquity. Therefore, it

is called Euclidean space and denoted by

E. Euclid suggested 5 axioms (5 postu-

lates) to describe E. However, his state-

ments were not satisfactorily strict from

a modern point of view. Currently E is

described by 20 axioms. In memory of

Euclid they are subdivided into 5 groups:

(1) axioms of incidence;

(2) axioms of order;

(3) axioms of congruence;

(4) axioms of continuity;

(5) axiom of parallels.

20-th axiom, which is also known as 5-th

postulate, is most famous.

Exercise 3.1. Visit the following Non-

Euclidean Geometry web-site and read a

few words about non-Euclidean geometry

and the role of Euclid’s 5-th postulate in

its discovery.

Usually nobody remembers all 20 of these axioms by heart, even me, though I

wrote a textbook on the foundations of Euclidean geometry in 1998. Furthermore,

dealing with the Euclidean space E, we shall rely only on common sense and on

our geometric intuition.

§ 4. Bases and Cartesian coordinates.

Thus, E is composed by points. Let’s choose one of them, denote it by O and

consider the vector space V

O

composed by displacement vectors. Then each point

B ∈ E can be uniquely identiﬁed with the displacement vector r

B

=

−→

OB. It is

called the radius-vector of the point B, while O is called origin. Passing from

points to their radius-vectors we identify E with the linear vector space V

O

. Then,

passing from the vectors to their classes, we can identify V with the space of free

vectors. This identiﬁcation is a convenient tool in studying E without referring to

Euclidean axioms. However, we should remember that such identiﬁcation is not

unique: it depends on our choice of the point O for the origin.

Definition 4.1. We say that three vectors e

1

, e

2

, e

3

form a non-coplanar

triple of vectors if they cannot be laid onto the plane by parallel translations.

These three vectors can be bound to some point O common to all of them, or

they can be bound to diﬀerent points in the space; it makes no diﬀerence. They

also can be treated as free vectors without any deﬁnite binding point.

§ 4. BASES AND CARTESIAN COORDINATES. 9

Definition 4.2. Any non-coplanar ordered triple of vectors e

1

, e

2

, e

3

is called

a basis in our geometric space E.

Exercise 4.1. Formulate the deﬁnitions of bases on a plane and on a straight

line by analogy with deﬁnition 4.2.

Below we distinguish three types of bases: orthonormal basis (ONB), orthogonal

basis (OB), and skew-angular basis (SAB). Orthonormal basis is formed by three

mutually perpendicular unit vectors:

e

1

⊥ e

2

,

e

2

⊥ e

3

,

e

3

⊥ e

1

,

(4.1)

|e

1

| = 1,

|e

2

| = 1,

|e

3

| = 1.

(4.2)

For orthogonal basis, the three con-

ditions (4.1) are fulﬁlled, but lengths of

basis vectors are not speciﬁed.

And skew-angular basis is the most

general case. For this basis neither angles

nor lengths are speciﬁed. As we shall see

below, due to its asymmetry SAB can

reveal a lot of features that are hidden in

symmetric ONB.

Let’s choose some basis e

1

, e

2

, e

3

in E. In the general case this is a skew-

angular basis. Assume that vectors e

1

, e

2

, e

3

are bound to a common point O as

shown on Fig. 6 below. Otherwise they can be brought to this position by means

of parallel translations. Let a be some arbitrary vector. This vector also can be

translated to the point O. As a result we have four vectors e

1

, e

2

, e

3

, and a

beginning at the same point O. Drawing additional lines and vectors as shown on

Fig. 6, we get

a =

−→

OD =

−→

OA +

−→

OB +

−→

OC. (4.3)

Then from the following obvious relationships

e

1

=

−−→

OE

1

, e

2

=

−−→

OE

2

, e

3

=

−−→

OE

3

,

−−→

OE

1

−→

OA,

−−→

OE

2

−→

OB,

−−→

OE

3

−→

OC

we derive

−→

OA = αe

1

,

−→

OB = β e

2

,

−→

OC = γ e

3

, (4.4)

where α, β, γ are scalars. Now from (4.3) and (4.4) we obtain

a = αe

1

+ β e

2

+ γ e

3

. (4.5)

10 CHAPTER I. PRELIMINARY INFORMATION.

Exercise 4.2. Explain how, for what reasons, and in what order additional

lines on Fig. 6 are drawn.

Formula (4.5) is known as the expansion of vector a in the basis e

1

, e

2

, e

3

,

while α, β, γ are coordinates of vector a in this basis.

Exercise 4.3. Explain why α, β, and γ are uniquely determined by vector a.

Hint: remember what linear dependence and linear independence are. Give exact

mathematical statements for these concepts. Apply them to exercise 4.3.

Further we shall write formula (4.5) as follows

a = a

1

e

1

+ a

2

e

2

+ a

3

e

3

=

3

¸

i=1

a

i

e

i

, (4.6)

denoting α = a

1

, β = a

2

, and γ = a

3

. Don’t confuse upper indices in (4.6)

with power exponentials, a

1

here is not a, a

2

is not a squared, and a

3

is not a

cubed. Usage of upper indices and the implicit summation rule were suggested by

Einstein. They are known as Einstein’s tensorial notations.

Once we have chosen the basis e

1

, e

2

, e

3

(no matter ONB, OB, or SAB), we

can associate vectors with columns of numbers:

a ←→

a

1

a

2

a

3

, b ←→

b

1

b

2

b

3

. (4.7)

§ 4. BASES AND CARTESIAN COORDINATES. 11

We can then produce algebraic operations with vectors, reducing them to arith-

metic operations with numbers:

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

,

αa ←→α

a

1

a

2

a

3

=

αa

1

αa

2

αa

3

.

Columns of numbers framed by matrix delimiters like those in (4.7) are called

vector-columns. They form linear vector spaces.

Exercise 4.4. Remember the exact mathematical deﬁnition for the real arith-

metic vector space R

n

, where n is a positive integer.

Definition 4.1. The Cartesian coordinate system is a basis complemented

with some ﬁxed point that is called the origin.

Indeed, if we have an origin O, then we can associate each point A of our space

with its radius-vector r

A

=

−→

OA. Then, having expanded this vector in a basis, we

get three numbers that are called the Cartesian coordinates of A. Coordinates of

a point are also speciﬁed by upper indices since they are coordinates of the radius-

vector for that point. However, unlike coordinates of vectors, they are usually

not written in a column. The reason will be clear when we consider curvilinear

coordinates. So, writing A(a

1

, a

2

, a

3

) is quite an acceptable notation for the point

A with coordinates a

1

, a

2

, and a

3

.

The idea of cpeciﬁcation of geometric objects by means of coordinates was

ﬁrst raised by French mathematician and philosopher Ren´e Descartes (1596-1650).

Cartesian coordinates are named in memory of him.

12 CHAPTER I. PRELIMINARY INFORMATION.

§ 5. What if we need to change a basis ?

Why could we need to change a basis ? There may be various reasons: we may

dislike initial basis because it is too symmetric like ONB, or too asymmetric like

SAB. Maybe we are completely satisﬁed; but the wisdom is that looking on how

something changes we can learn more about this thing than if we observe it in a

static position. Suppose we have a basis e

1

, e

2

, e

3

, let’s call it the old basis, and

suppose we want to change it to a new one ˜ e

1

, ˜ e

2

, ˜ e

3

. Let’s take the ﬁrst vector

of the new basis e

1

. Being isolated from the other two vectors ˜ e

2

and ˜ e

3

, it is

nothing, but quite an ordinary vector of space. In this capacity, vector ˜ e

1

can be

expanded in the old basis e

1

, e

2

, e

3

:

˜ e

1

= S

1

e

1

+ S

2

e

2

+ S

3

e

3

=

3

¸

j=1

S

j

e

j

. (5.1)

Compare (5.1) and (4.6). Then we can take another vector ˜ e

2

and also expand it

in the old basis. But what letter should we choose for denoting the coeﬃcients of

this expansion? We can choose another letter; say the letter “R”:

˜ e

2

= R

1

e

1

+ R

2

e

2

+ R

3

e

3

=

3

¸

j=1

R

j

e

j

. (5.2)

However, this is not the best decision. Indeed, vectors ˜ e

1

and ˜ e

2

diﬀer only in

number, while for their coordinates we use diﬀerent letters. A better way is to add

an extra index to S in (5.1). This is the lower index coinciding with the number

of the vector:

˜ e

1

= S

1

1

e

1

+ S

2

1

e

2

+ S

3

1

e

3

=

3

¸

j=1

S

j

1

e

j

(5.3)

Color is of no importance; it is used for highlighting only. Instead of (5.2), for the

second vector e

2

we write a formula similar to (5.3):

˜ e

2

= S

1

2

e

1

+ S

2

2

e

2

+ S

3

2

e

3

=

3

¸

j=1

S

j

2

e

j

. (5.4)

And for third vector as well:

˜ e

3

= S

1

3

e

1

+ S

2

3

e

2

+ S

3

3

e

3

=

3

¸

j=1

S

j

3

e

j

. (5.5)

When considered jointly, formulas (5.3), (5.4), and (5.5) are called transition

formulas. We use a left curly bracket to denote their union:

˜ e

1

= S

1

1

e

1

+ S

2

1

e

2

+ S

3

1

e

3

,

˜ e

2

= S

1

2

e

1

+ S

2

2

e

2

+ S

3

2

e

3

,

˜ e

3

= S

1

3

e

1

+ S

2

3

e

2

+ S

3

3

e

3

.

(5.6)

§ 5. WHAT IF WE NEED TO CHANGE A BASIS ? 13

We also can write transition formulas (5.6) in a more symbolic form

˜ e

i

=

3

¸

j=1

S

j

i

e

j

. (5.7)

Here index i runs over the range of integers from 1 to 3.

Look at index i in formula (5.7). It is a free index, it can freely take any

numeric value from its range: 1, 2, or 3. Note that i is the lower index in both

sides of formula (5.7). This is a general rule.

Rule 5.1. In correctly written tensorial formulas free indices are written on the

same level (upper or lower) in both sides of the equality. Each free index has only

one entry in each side of the equality.

Now look at index j. It is summation index. It is present only in right hand

side of formula (5.7), and it has exactly two entries (apart from that j = 1 under

the sum symbol): one in the upper level and one in the lower level. This is also

general rule for tensorial formulas.

Rule 5.2. In correctly written tensorial formulas each summation index should

have exactly two entries: one upper entry and one lower entry.

Proposing this rule 5.2, Einstein also suggested not to write the summation

symbols at all. Formula (5.7) then would look like ˜ e

i

= S

j

i

e

j

with implicit

summation with respect to the double index j. Many physicists (especially those

in astrophysics) prefer writing tensorial formulas in this way. However, I don’t like

omitting sums. It breaks the integrity of notations in science. Newcomers from

other branches of science would have diﬃculties in understanding formulas with

implicit summation.

Exercise 5.1. What happens if ˜ e

1

= e

1

? What are the numeric values of

coeﬃcients S

1

1

, S

2

1

, and S

3

1

in formula (5.3) for this case ?

Returning to transition formulas (5.6) and (5.7) note that coeﬃcients in them

are parameterized by two indices running independently over the range of integer

numbers from 1 to 3. In other words, they form a two-dimensional array that

usually is represented as a table or as a matrix:

S =

S

1

1

S

1

2

S

1

3

S

2

1

S

2

2

S

2

3

S

3

1

S

3

2

S

3

3

(5.8)

Matrix S is called a transition matrix or direct transition matrix since we

use it in passing from old basis to new one. In writing such matrices like S the

following rule applies.

Rule 5.3. For any double indexed array with indices on the same level (both

upper or both lower) the ﬁrst index is a row number, while the second index is a

column number. If indices are on diﬀerent levels (one upper and one lower), then

the upper index is a row number, while lower one is a column number.

Note that according to this rule 5.3, coeﬃcients of formula (5.3), which are

written in line, constitute ﬁrst column in matrix (5.8). So lines of formula (5.6)

turn into columns in matrix (5.8). It would be worthwhile to remember this fact.

14 CHAPTER I. PRELIMINARY INFORMATION.

If we represent each vector of the new basis ˜ e

1

, ˜ e

2

, ˜ e

3

as a column of its

coordinates in the old basis just like it was done for a and b in formula (4.7) above

e

1

←→

S

1

1

S

2

1

S

3

1

, e

2

←→

S

1

2

S

2

2

S

3

2

, e

3

←→

S

1

3

S

2

3

S

3

3

, (5.9)

then these columns (5.9) are exactly the ﬁrst, the second, and the third columns

in matrix (5.8). This is the easiest way to remember the structure of matrix S.

Exercise 5.2. What happens if ˜ e

1

= e

1

, ˜ e

2

= e

2

, and ˜ e

3

= e

3

? Find the

transition matrix for this case. Consider also the following two cases and write the

transition matrices for each of them:

(1) ˜ e

1

= e

1

, ˜ e

2

= e

3

, ˜ e

3

= e

2

;

(2) ˜ e

1

= e

3

, ˜ e

2

= e

1

, ˜ e

3

= e

2

.

Explain why the next case is impossible:

(3) ˜ e

1

= e

1

−e

2

, ˜ e

2

= e

2

−e

3

, ˜ e

3

= e

3

−e

1

.

Now let’s swap bases. This means that we are going to consider ˜ e

1

, ˜ e

2

, ˜ e

3

as

the old basis, e

1

, e

2

, e

3

as the new basis, and study the inverse transition. All

of the above stuﬀ applies to this situation. However, in writing the transition

formulas (5.6), let’s use another letter for the coeﬃcients. By tradition here the

letter “T” is used:

e

1

= T

1

1

˜ e

1

+ T

2

1

˜ e

2

+ T

3

1

˜ e

3

,

e

2

= T

1

2

˜ e

1

+ T

2

2

˜ e

2

+ T

3

2

˜ e

3

,

e

3

= T

1

3

˜ e

1

+ T

2

3

˜ e

2

+ T

3

3

˜ e

3

.

(5.10)

Here is the short symbolic version of transition formulas (5.10):

e

i

=

3

¸

j=1

T

j

i

˜ e

j

. (5.11)

Denote by T the transition matrix constructed on the base of (5.10) and (5.11). It

is called the inverse transition matrix when compared to the direct transition

matrix S:

(e

1

, e

2

, e

3

)

S

−−−−→

←−−−−

T

(˜ e

1

, ˜ e

2

, ˜ e

3

). (5.12)

Theorem 5.1. The inverse transition matrix T in (5.12) is the inverse matrix

for the direct transition matrix S, i. e. T = S

−1

.

Exercise 5.3. What is the inverse matrix ? Remember the deﬁnition. How

is the inverse matrix A

−1

calculated if A is known? (Don’t say that you use a

computer package like Maple, MathCad, or any other; remember the algorithm for

calculating A

−1

).

CopyRight c Sharipov R.A., 2004.

§ 6. WHAT HAPPENS TO VECTORS WHEN WE CHANGE THE BASIS ? 15

Exercise 5.4. Remember what is the determinant of a matrix. How is it usually

calculated? Can you calculate det(A

−1

) if det A is already known?

Exercise 5.5. What is matrix multiplication ? Remember how it is deﬁned.

Suppose you have a rectangular 5 ×3 matrix A and another rectangular matrix B

which is 4 ×5. Which of these two products AB or BA you can calculate ?

Exercise 5.6. Suppose that A and B are two rectangular matrices, and suppose

that C = AB. Remember the formula for the components in matrix C if the

components of A and B are known (they are denoted by A

ij

and B

pq

). Rewrite

this formula for the case when the components of B are denoted by B

pq

. Which

indices (upper, or lower, or mixed) you would use for components of C in the last

case (see rules 5.1 and 5.2 of Einstein’s tensorial notation).

Exercise 5.7. Give some examples of matrix multiplication that are consis-

tent with Einstein’s tensorial notation and those that are not (please, do not use

examples that are already considered in exercise 5.6).

Let’s consider three bases: basis one e

1

, e

2

, e

3

, basis two ˜ e

1

, ˜ e

2

, ˜ e

3

, and basis

three

˜

˜ e

1

,

˜

˜ e

2

,

˜

˜ e

3

. And let’s consider the transition matrices relating them:

(e

1

, e

2

, e

3

)

S

−−−−→

←−−−−

T

(˜ e

1

, ˜ e

2

, ˜ e

3

)

˜

S

−−−−→

←−−−−

˜

T

(

˜

˜ e

1

,

˜

˜ e

2

,

˜

˜ e

3

). (5.13)

Denote by

˜

˜

S and

˜

˜

T transition matrices relating basis one with basis three in

(5.13):

(e

1

, e

2

, e

3

)

˜

˜

S

−−−−→

←−−−−

˜

˜

T

(

˜

˜ e

1

,

˜

˜ e

2

,

˜

˜ e

3

). (5.14)

Exercise 5.8. For matrices

˜

˜

S and

˜

˜

T in (5.14) prove that

˜

˜

S = S

˜

S and

˜

˜

T =

˜

T T.

Apply this result for proving theorem 5.1.

§ 6. What happens to vectors when we change the basis ?

The answer to this question is very simple. Really nothing ! Vectors do not

need a basis for their being. But their coordinates, they depend on our choice

of basis. And they change if we change the basis. Let’s study how they change.

Suppose we have some vector x expanded in the basis e

1

, e

2

, e

3

:

x = x

1

e

1

+ x

2

e

2

+ x

3

e

3

=

3

¸

i=1

x

i

e

i

. (6.1)

Then we keep vector x and change the basis e

1

, e

2

, e

3

to another basis ˜ e

1

, ˜ e

2

, ˜ e

3

.

As we already learned, this process is described by transition formula (5.11):

e

i

=

3

¸

j=1

T

j

i

˜ e

j

.

16 CHAPTER I. PRELIMINARY INFORMATION.

Let’s substitute this formula into (6.1) for e

i

:

x =

3

¸

i=1

x

i

3

¸

j=1

T

j

i

e

j

=

3

¸

i=1

3

¸

j=1

x

i

T

j

i

˜ e

j

=

3

¸

j=1

3

¸

i=1

x

i

T

j

i

˜ e

j

=

=

3

¸

j=1

3

¸

i=1

T

j

i

x

i

˜ e

j

=

3

¸

j=1

˜ x

j

˜ e

j

, where ˜ x

j

=

3

¸

i=1

T

j

i

x

i

.

Thus we have calculated the expansion of vector x in the new basis and have

derived the formula relating its new coordinates to its initial ones:

˜ x

j

=

3

¸

i=1

T

j

i

x

i

. (6.2)

This formula is called a transformation formula, or direct transformation

formula. Like (5.7), it can be written in expanded form:

˜ x

1

= T

1

1

x

1

+ T

1

2

x

2

+ T

1

3

x

3

,

˜ x

2

= T

2

1

x

1

+ T

2

2

x

2

+ T

2

3

x

3

,

˜ x

3

= T

3

1

x

1

+ T

3

2

x

2

+ T

3

3

x

3

.

(6.3)

And the transformation formula (6.2) can be written in matrix form as well:

˜ x

1

˜ x

2

˜ x

3

=

T

1

1

T

1

2

T

1

3

T

2

1

T

2

2

T

2

3

T

3

1

T

3

2

T

3

3

x

1

x

2

x

3

. (6.4)

Like (5.7), formula (6.2) can be inverted. Here is the inverse transformation

formula expressing the initial coordinates of vector x through its new coordinates:

x

j

=

3

¸

i=1

S

j

i

˜ x

i

. (6.5)

Exercise 6.1. By analogy with the above calculations derive the inverse trans-

formation formula (6.5) using formula (5.7).

Exercise 6.2. By analogy with (6.3) and (6.4) write (6.5) in expanded form

and in matrix form.

Exercise 6.3. Derive formula (6.5) directly from (6.2) using the concept of the

inverse matrix S = T

−1

.

Note that the direct transformation formula (6.2) uses the inverse transition

matrix T, and the inverse transformation formula (6.5) uses direct transition

matrix S. It’s funny, but it’s really so.

§ 7. WHAT IS THE NOVELTY ABOUT THE VECTORS . . . 17

§ 7. What is the novelty about the vectors that we learned

knowing transformation formula for their coordinates ?

Vectors are too common, too well-known things for one to expect that there

are some novelties about them. However, the novelty is that the method of

their treatment can be generalized and then applied to less customary objects.

Suppose, we cannot visually observe vectors (this is really so for some kinds of

them, see section 1), but suppose we can measure their coordinates in any basis

we choose for this purpose. What then do we know about vectors ? And how

can we tell them from other (non-vectorial) objects ? The answer is in formulas

(6.2) and (6.5). Coordinates of vectors (and only coordinates of vectors) will obey

transformation rules (6.2) and (6.5) under a change of basis. Other objects usually

have a diﬀerent number of numeric parameters related to the basis, and even if

they have exactly three coordinates (like vectors have), their coordinates behave

diﬀerently under a change of basis. So transformation formulas (6.2) and (6.5)

work like detectors, like a sieve for separating vectors from non-vectors. What

are here non-vectors, and what kind of geometrical and/or physical objects of a

non-vectorial nature could exist — these are questions for a separate discussion.

Furthermore, we shall consider only a part of the set of such objects, which are

called tensors.

CHAPTER II

TENSORS IN CARTESIAN COORDINATES.

§ 8. Covectors.

In previous 7 sections we learned the following important fact about vectors:

a vector is a physical object in each basis of our three-dimensional Euclidean

space E represented by three numbers such that these numbers obey certain

transformation rules when we change the basis. These certain transformation rules

are represented by formulas (6.2) and (6.5).

Now suppose that we have some other physical object that is represented by

three numbers in each basis, and suppose that these numbers obey some certain

transformation rules when we change the basis, but these rules are diﬀerent from

(6.2) and (6.5). Is it possible ? One can try to ﬁnd such an object in nature.

However, in mathematics we have another option. We can construct such an

object mentally, then study its properties, and ﬁnally look if it is represented

somehow in nature.

Let’s denote our hypothetical object by a, and denote by a

1

, a

2

, a

3

that

three numbers which represent this object in the basis e

1

, e

2

, e

3

. By analogy

with vectors we shall call them coordinates. But in contrast to vectors, we

intentionally used lower indices when denoting them by a

1

, a

2

, a

3

. Let’s prescribe

the following transformation rules to a

1

, a

2

, a

3

when we change e

1

, e

2

, e

3

to

˜ e

1

, ˜ e

2

, ˜ e

3

:

˜ a

j

=

3

¸

i=1

S

i

j

a

i

, (8.1)

a

j

=

3

¸

i=1

T

i

j

˜ a

i

. (8.2)

Here S and T are the same transition matrices as in case of the vectors in (6.2)

and (6.5). Note that (8.1) is suﬃcient, formula (8.2) is derived from (8.1).

Exercise 8.1. Using the concept of the inverse matrix T = S

−1

derive formula

(8.2) from formula (8.1). Compare exercise 8.1 and exercise 6.3.

Definition 8.1. A geometric object a in each basis represented by a triple

of coordinates a

1

, a

2

, a

3

and such that its coordinates obey transformation rules

(8.1) and (8.2) under a change of basis is called a covector.

Looking at the above considerations one can think that we arbitrarily chose

the transformation formula (8.1). However, this is not so. The choice of the

transformation formula should be self-consistent in the following sense. Let

e

1

, e

2

, e

3

and ˜ e

1

, ˜ e

2

, ˜ e

3

be two bases and let

˜

˜ e

1

,

˜

˜ e

2

,

˜

˜ e

3

be the third basis in the

§ 9. SCALAR PRODUCT OF VECTOR AND COVECTOR. 19

space. Let’s call them basis one, basis two and basis three for short. We can pass

from basis one to basis three directly, see the right arrow in (5.14). Or we can use

basis two as an intermediate basis, see the right arrows in (5.13). In both cases the

ultimate result for the coordinates of a covector in basis three should be the same:

this is the self-consistence requirement. It means that coordinates of a geometric

object should depend on the basis, but not on the way that they were calculated.

Exercise 8.2. Using (5.13) and (5.14), and relying on the results of exer-

cise 5.8 prove that formulas (8.1) and (8.2) yield a self-consistent way of deﬁning the

covector.

Exercise 8.3. Replace S by T in (8.1) and T by S in (8.2). Show that the

resulting formulas are not self-consistent.

What about the physical reality of covectors ? Later on we shall see that

covectors do exist in nature. They are the nearest relatives of vectors. And

moreover, we shall see that some well-known physical objects we thought to be

vectors are of covectorial nature rather than vectorial.

§ 9. Scalar product of vector and covector.

Suppose we have a vector x and a covector a. Upon choosing some basis

e

1

, e

2

, e

3

, both of them have three coordinates: x

1

, x

2

, x

3

for vector x, and

a

1

, a

2

, a

3

for covector a. Let’s denote by

a, x

the following sum:

a, x

=

3

¸

i=1

a

i

x

i

. (9.1)

The sum (9.1) is written in agreement with Einstein’s tensorial notation, see

rule 5.2 in section 5 above. It is a number depending on the vector x and on

the covector a. This number is called the scalar product of the vector x and the

covector a. We use angular brackets for this scalar product in order to distinguish

it from the scalar product of two vectors in E, which is also known as the dot

product.

Deﬁning the scalar product

a, x

**by means of sum (9.1) we used the coordi-
**

nates of vector x and of covector a, which are basis-dependent. However, the value

of sum (9.1) does not depend on any basis. Such numeric quantities that do not

depend on the choice of basis are called scalars or true scalars.

Exercise 9.1. Consider two bases e

1

, e

2

, e

3

and ˜ e

1

, ˜ e

2

, ˜ e

3

, and consider the

coordinates of vector x and covector a in both of them. Relying on transformation

rules (6.2), (6.5), (8.1), and (8.2) prove the equality

3

¸

i=1

a

i

x

i

=

3

¸

i=1

˜ a

i

˜ x

i

. (9.2)

Thus, you are proving the self-consistence of formula (9.1) and showing that the

scalar product

a, x

**given by this formula is a true scalar quantity.
**

20 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

Exercise 9.2. Let α be a real number, let a and b be two covectors, and let

x and y be two vectors. Prove the following properties of the scalar product (9.1):

(1)

a+b, x

=

a, x

+

b, x

;

(2)

αa, x

= α

a, x

;

(3)

a, x +y

=

a, x

+

a, y

;

(4)

a, αx

= α

a, x

.

Exercise 9.3. Explain why the scalar product

a, x

**is sometimes called the
**

bilinear function of vectorial argument x and covectorial argument a. In this ca-

pacity, it can be denoted as f(a, x). Remember our discussion about functions with

non-numeric arguments in section 2.

Important note. The scalar product

a, x

**is not symmetric. Moreover, the
**

formula

a, x

=

x, a

**is incorrect in its right hand side since the ﬁrst argument of scalar product (9.1)
**

by deﬁnition should be a covector. In a similar way, the second argument should

be a vector. Therefore, we never can swap them.

§ 10. Linear operators.

In this section we consider more complicated geometric objects. For the sake

of certainty, let’s denote one of such objects by F. In each basis e

1

, e

2

, e

3

, it is

represented by a square 3 × 3 matrix F

i

j

of real numbers. Components of this

matrix play the same role as coordinates in the case of vectors or covectors. Let’s

prescribe the following transformation rules to F

i

j

:

˜

F

i

j

=

3

¸

p=1

3

¸

q=1

T

i

p

S

q

j

F

p

q

, (10.1)

F

i

j

=

3

¸

p=1

3

¸

q=1

S

i

p

T

q

j

˜

F

p

q

. (10.2)

Exercise 10.1. Using the concept of the inverse matrix T = S

−1

prove that

formula (10.2) is derived from formula (10.1).

If we write matrices F

i

j

and

˜

F

p

q

according to the rule 5.3 (see in section 5), then

(10.1) and (10.2) can be written as two matrix equalities:

˜

F = T F S, F = S

˜

F T. (10.3)

Exercise 10.2. Remember matrix multiplication (we already considered it in

exercises 5.5 and 5.6) and derive (10.3) from (10.1) and (10.2).

Definition 10.1. A geometric object F in each basis represented by some

square matrix F

i

j

and such that components of its matrix F

i

j

obey transformation

rules (10.1) and (10.2) under a change of basis is called a linear operator.

§ 10. LINEAR OPERATORS. 21

Exercise 10.3. By analogy with exercise 8.2 prove the self-consistence of the

above deﬁnition of a linear operator.

Let’s take a linear operator F represented by matrix F

i

j

in some basis e

1

, e

2

, e

3

and take some vector x with coordinates x

1

, x

2

, x

3

in the same basis. Using F

i

j

and x

j

we can construct the following sum:

y

i

=

3

¸

j=1

F

i

j

x

j

. (10.4)

Index i in the sum (10.4) is a free index; it can deliberately take any one of three

values: i = 1, i = 2, or i = 3. For each speciﬁc value of i we get some speciﬁc

value of the sum (10.4). They are denoted by y

1

, y

2

, y

3

according to (10.4). Now

suppose that we pass to another basis ˜ e

1

, ˜ e

2

, ˜ e

3

and do the same things. As a

result we get other three values ˜ y

1

, ˜ y

2

, ˜ y

3

given by formula

˜ y

p

=

3

¸

q=1

˜

F

p

q

˜ x

q

. (10.5)

Exercise 10.4. Relying upon (10.1) and (10.2) prove that the three numbers

y

1

, y

2

, y

3

and the other three numbers ˜ y

1

, ˜ y

2

, ˜ y

3

are related as follows:

˜ y

j

=

3

¸

i=1

T

j

i

y

i

, y

j

=

3

¸

i=1

S

j

i

˜ y

i

. (10.6)

Exercise 10.5. Looking at (10.6) and comparing it with (6.2) and (6.5) ﬁnd

that the y

1

, y

2

, y

3

and ˜ y

1

, ˜ y

2

, ˜ y

3

calculated by formulas (10.4) and (10.5) represent

the same vector, but in diﬀerent bases.

Thus formula (10.4) deﬁnes the vectorial object y, while exercise 10.5 assures

the correctness of this deﬁnition. As a result we have vector y determined by a

linear operator F and by vector x. Therefore, we write

y = F(x) (10.7)

and say that y is obtained by applying linear operator F to vector x. Some people

like to write (10.7) without parentheses:

y = Fx. (10.8)

Formula (10.8) is a more algebraistic form of formula (10.7). Here the action of

operator F upon vector x is designated like a kind of multiplication. There is also

a matrix representation of formula (10.8), in which x and y are represented as

columns:

y

1

y

2

y

3

=

F

1

1

F

1

2

F

1

3

F

2

1

F

2

2

F

2

3

F

3

1

F

3

2

F

3

3

x

1

x

2

x

3

. (10.9)

CopyRight c Sharipov R.A., 2004.

22 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

Exercise 10.6. Derive (10.9) from (10.4).

Exercise 10.7. Let α be some real number and let x and y be two vectors.

Prove the following properties of a linear operator (10.7):

(1) F(x +y) = F(x) +F(y),

(2) F(αx) = αF(x).

Write these equalities in the more algebraistic style introduced by (10.8). Are they

really similar to the properties of multiplication ?

Exercise 10.8. Remember that for the product of two matrices

det(AB) = det A det B. (10.10)

Also remember the formula for det(A

−1

). Apply these two formulas to (10.3) and

derive

det F = det

˜

F. (10.11)

Formula (10.10) means that despite the fact that in various bases linear operator

F is represented by various matrices, the determinants of all these matrices are

equal to each other. Then we can deﬁne the determinant of linear operator F as

the number equal to the determinant of its matrix in any one arbitrarily chosen

basis e

1

, e

2

, e

3

:

det F = det F. (10.12)

Exercise 10.9 (for deep thinking). Square matrices have various attributes:

eigenvalues, eigenvectors, a characteristic polynomial, a rank (maybe you remember

some others). If we study these attributes for the matrix of a linear operator, which

of them can be raised one level up and considered as basis-independent attributes

of the linear operator itself ? Determinant (10.12) is an example of such attribute.

Exercise 10.10. Substitute the unit matrix for F

i

j

into (10.1) and verify that

˜

F

i

j

is also a unit matrix in this case. Interpret this fact.

Exercise 10.11. Let x = e

i

for some basis e

1

, e

2

, e

3

in the space. Substitute

this vector x into (10.7) and by means of (10.4) derive the following formula:

F(e

i

) =

3

¸

j=1

F

j

i

e

j

. (10.13)

Compare (10.13) and (5.7). Discuss the similarities and diﬀerences of these two

formulas. The fact is that in some books the linear operator is determined ﬁrst,

then its matrix is introduced by formula (10.13). Explain why if we know three

vectors F(e

1

), F(e

2

), and F(e

3

), then we can reconstruct the whole matrix of

operator F by means of formula (10.13).

Suppose we have two linear operators F and H. We can apply H to vector x

and then we can apply F to vector H(x). As a result we get

F◦ H(x) = F(H(x)). (10.14)

Here F◦ H is new linear operator introduced by formula (10.14). It is called a

composite operator, and the small circle sign denotes composition.

§ 11. BILINEAR AND QUADRATIC FORMS. 23

Exercise 10.12. Find the matrix of composite operator F◦ H if the matrices

for F and H in the basis e

1

, e

2

, e

3

are known.

Exercise 10.13. Remember the deﬁnition of the identity map in mathematics

(see on-line Math. Encyclopedia) and deﬁne the identity operator id. Find the

matrix of this operator.

Exercise 10.14. Remember the deﬁnition of the inverse map in mathematics

and deﬁne inverse operator F

−1

for linear operator F. Find the matrix of this

operator if the matrix of F is known.

§ 11. Bilinear and quadratic forms.

Vectors, covectors, and linear operators are all examples of tensors (though we

have no deﬁnition of tensors yet). Now we consider another one class of tensorial

objects. For the sake of clarity, let’s denote by a one of such objects. In each

basis e

1

, e

2

, e

3

this object is represented by some square 3 × 3 matrix a

ij

of real

numbers. Under a change of basis these numbers are transformed as follows:

˜ a

ij

=

3

¸

p=1

3

¸

q=1

S

p

i

S

q

j

a

pq

, (11.1)

a

ij

=

3

¸

p=1

3

¸

q=1

T

p

i

T

q

j

˜ a

pq

. (11.2)

Transformation rules (11.1) and (11.2) can be written in matrix form:

˜ a = S

a S, a = T

˜ aT. (11.3)

Here by S

and T

**we denote the transposed matrices for S and T respectively.
**

Exercise 11.1. Derive (11.2) from (11.1), then (11.3) from (11.1) and (11.2).

Definition 11.1. A geometric object a in each basis represented by some

square matrix a

ij

and such that components of its matrix a

ij

obey transformation

rules (11.1) and (11.2) under a change of basis is called a bilinear form.

Let’s consider two arbitrary vectors x and y. We use their coordinates and the

components of bilinear form a in order to write the following sum:

a(x, y) =

3

¸

i=1

3

¸

j=1

a

ij

x

i

y

j

. (11.4)

Exercise 11.2. Prove that the sum in the right hand side of formula (11.4)

does not depend on the basis, i. e. prove the equality

3

¸

i=1

3

¸

j=1

a

ij

x

i

y

j

=

3

¸

p=1

3

¸

q=1

˜ a

pq

˜ x

p

˜ y

q

.

This equality means that a(x, y) is a number determined by vectors x and y

irrespective of the choice of basis. Hence we can treat (11.4) as a scalar function

of two vectorial arguments.

24 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

Exercise 11.3. Let α be some real number, and let x, y, and z be three vectors.

Prove the following properties of function (11.4):

(1) a(x+y, z) = a(x, z)+a(y, z);

(2) a(αx, y) = αa(x, y);

(3) a(x, y+z) = a(x, y)+a(x, z);

(4) a(x, αy) = αa(x, y).

Due to these properties function (10.4) is called a bilinear function or a bilinear

form. It is linear with respect to each of its two arguments.

Note that scalar product (9.1) is also a bilinear function of its arguments.

However, there is a crucial diﬀerence between (9.1) and (11.4). The arguments of

scalar product (9.1) are of a diﬀerent nature: the ﬁrst argument is a covector, the

second is a vector. Therefore, we cannot swap them. In bilinear form (11.4) we

can swap arguments. As a result we get another bilinear function

b(x, y) = a(y, x). (11.5)

The matrices of a and b are related to each other as follows:

b

ij

= a

ji

, b = a

. (11.6)

Definition 11.2. A bilinear form is called symmetric if a(x, y) = a(y, x).

Exercise 11.4. Prove the following identity for a symmetric bilinear form:

a(x, y) =

a(x +y, x +y) −a(x, x) −a(y, y)

2

. (11.7)

Definition 11.3. A quadratic form is a scalar function of one vectorial argu-

ment f(x) produced from some bilinear function a(x, y) by substituting y = x:

f(x) = a(x, x). (11.8)

Without a loss of generality a bilinear function a in (11.8) can be assumed

symmetric. Indeed, if a is not symmetric, we can produce symmetric bilinear

function

c(x, y) =

a(x, y) + a(y, x)

2

, (11.9)

and then from (11.8) due to (11.9) we derive

f(x) = a(x, x) =

a(x, x) + a(x, x)

2

= c(x, x).

This equality is the same as (11.8) with a replaced by c. Thus, each quadratic

function f is produced by some symmetric bilinear function a. And conversely,

comparing (11.8) and (11.7) we get that a is produced by f:

a(x, y) =

f(x +y) −f(x) − f(y)

2

. (11.10)

Formula (11.10) is called the recovery formula. It recovers bilinear function a

from quadratic function f produced in (11.8). Due to this formula, in referring to

a quadratic form we always imply some symmetric bilinear form like the geometric

tensorial object introduced by deﬁnition 11.1.

§ 12. GENERAL DEFINITION OF TENSORS. 25

§ 12. General deﬁnition of tensors.

Vectors, covectors, linear operators, and bilinear forms are examples of tensors.

They are geometric objects that are represented numerically when some basis in

the space is chosen. This numeric representation is speciﬁc to each of them:

vectors and covectors are represented by one-dimensional arrays, linear operators

and quadratic forms are represented by two-dimensional arrays. Apart from the

number of indices, their position does matter. The coordinates of a vector are

numerated by one upper index, which is called the contravariant index. The

coordinates of a covector are numerated by one lower index, which is called the

covariant index. In a matrix of bilinear form we use two lower indices; therefore

bilinear forms are called twice-covariant tensors. Linear operators are tensors

of mixed type; their components are numerated by one upper and one lower

index. The number of indices and their positions determine the transformation

rules, i. e. the way the components of each particular tensor behave under a change

of basis. In the general case, any tensor is represented by a multidimensional array

with a deﬁnite number of upper indices and a deﬁnite number of lower indices.

Let’s denote these numbers by r and s. Then we have a tensor of the type (r, s),

or sometimes the term valency is used. A tensor of type (r, s), or of valency (r, s)

is called an r-times contravariant and an s-times covariant tensor. This is

terminology; now let’s proceed to the exact deﬁnition. It is based on the following

general transformation formulas:

X

i1... ir

j1... js

=

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

S

i1

h1

. . . S

ir

hr

T

k1

j1

. . . T

ks

js

˜

X

h1... hr

k1... ks

, (12.1)

˜

X

i1... ir

j1... js

=

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

T

i1

h1

. . . T

ir

hr

S

k1

j1

. . . S

ks

js

X

h1... hr

k1... ks

. (12.2)

Definition 12.1. A geometric object X in each basis represented by (r + s)-

dimensional array X

i1... ir

j1... js

of real numbers and such that the components of this

array obey the transformation rules (12.1) and (12.2) under a change of basis is

called tensor of type (r, s), or of valency (r, s).

Formula (12.2) is derived from (12.1), so it is suﬃcient to remember only one

of them. Let it be the formula (12.1). Though huge, formula (12.1) is easy to

remember. One should strictly follow the rules 5.1 and 5.2 from section 5.

Indices i

1

, . . . , i

r

and j

1

, . . . , j

s

are free indices. In right hand side of the

equality (12.1) they are distributed in S-s and T-s, each having only one entry

and each keeping its position, i. e. upper indices i

1

, . . . , i

r

remain upper and lower

indices j

1

, . . . , j

s

remain lower in right hand side of the equality (12.1).

Other indices h

1

, . . . , h

r

and k

1

, . . . , k

s

are summation indices; they enter the

right hand side of (12.1) pairwise: once as an upper index and once as a lower

index, once in S-s or T-s and once in components of array

˜

X

h1... hr

k1... ks

.

When expressing X

i1... ir

j1... js

through

˜

X

h1... hr

k1... ks

each upper index is served by direct

transition matrix S and produces one summation in (12.1):

X

... iα ...

... ... ...

=

¸

. . .

3

¸

hα=1

. . .

¸

. . . S

iα

hα

. . .

˜

X

... hα ...

... ... ...

. (12.3)

26 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

In a similar way, each lower index is served by inverse transition matrix T and

also produces one summation in formula (12.1):

X

... ... ...

... jα ...

=

¸

. . .

3

¸

kα=1

. . .

¸

. . . T

kα

jα

. . .

˜

X

... ... ...

... kα ...

. (12.4)

Formulas (12.3) and (12.4) are the same as (12.1) and used to highlight how

(12.1) is written. So tensors are deﬁned. Further we shall consider more examples

showing that many well-known objects undergo the deﬁnition 12.1.

Exercise 12.1. Verify that formulas (6.5), (8.2), (10.2), and (11.2) are spe-

cial cases of formula (12.1). What are the valencies of vectors, covectors, linear

operators, and bilinear forms when they are considered as tensors.

Exercise 12.2. Let a

ij

be the matrix of some bilinear form a. Let’s denote

by b

ij

components of inverse matrix for a

ij

. Prove that matrix b

ij

under a change

of basis transforms like matrix of twice-contravariant tensor. Hence it determines

tensor b of valency (2, 0). Tensor b is called a dual bilinear form for a.

§ 13. Dot product and metric tensor.

The covectors, linear operators, and bilinear forms that we considered above

were artiﬁcially constructed tensors. However there are some tensors of natural

origin. Let’s remember that we live in a space with measure. We can measure

distance between points (hence we can measure length of vectors) and we can

measure angles between two directions in our space. Therefore for any two vectors

x and y we can deﬁne their natural scalar product (or dot product):

(x, y) = |x| |y| cos(ϕ), (13.1)

where ϕ is the angle between vectors x and y.

Exercise 13.1. Remember the following properties of the scalar product (13.1):

(1) (x +y, z) = (x, z) + (y, z);

(2) (αx, y) = α(x, y);

(3) (x, y + z) = (x, y) + (x, z);

(4) (x, αy) = α(x, y);

(5) (x, y) = (y, x);

(6) (x, x) 0 and (x, x) = 0 implies x = 0.

These properties are usually considered in courses on analytic geometry or vector

algebra, see Vector Lessons on the Web.

Note that the ﬁrst four properties of the scalar product (13.1) are quite

similar to those for quadratic forms, see exercise 11.3. This is not an occasional

coincidence.

Exercise 13.2. Let’s consider two arbitrary vectors x and y expanded in some

basis e

1

, e

2

, e

3

. This means that we have the following expressions for them:

x =

3

¸

i=1

x

i

e

i

, y =

3

¸

j=1

x

j

e

j

. (13.2)

§ 14. MULTIPLICATION BY NUMBERS AND ADDITION. 27

Substitute (13.2) into (13.1) and using properties (1)– (4) listed in exercise 13.1

derive the following formula for the scalar product of x and y:

(x, y) =

3

¸

i=1

3

¸

j=1

(e

i

, e

j

) x

i

y

j

. (13.3)

Exercise 13.3. Denote g

ij

= (e

i

, e

j

) and rewrite formula (13.3) as

(x, y) =

3

¸

i=1

3

¸

j=1

g

ij

x

i

y

j

. (13.4)

Compare (13.4) with formula (11.4). Consider some other basis ˜ e

1

, ˜ e

2

, ˜ e

3

, denote

˜ g

pq

= (˜ e

p

, ˜ e

q

) and by means of transition formulas (5.7) and (5.11) prove that

matrices g

ij

and ˜ g

pq

are components of a geometric object obeying transformation

rules (11.1) and (11.2) under a change of base. Thus you prove that the Gram

matrix

g

ij

= (e

i

, e

j

) (13.5)

deﬁnes tensor of type (0, 2). This is very important tensor; it is called the metric

tensor. It describes not only the scalar product in form of (13.4), but the whole

geometry of our space. Evidences for this fact are below.

Matrix (13.5) is symmetric due to property (5) in exercise 13.1. Now, comparing

(13.4) and (11.4) and keeping in mind the tensorial nature of matrix (13.5), we

conclude that the scalar product is a symmetric bilinear form:

(x, y) = g(x, y). (13.6)

The quadratic form corresponding to (13.6) is very simple: f(x) = g(x, x) = |x|

2

.

The inverse matrix for (13.5) is denoted by the same symbol g but with upper

indices: g

ij

. It determines a tensor of type (2, 0), this tensor is called dual metric

tensor (see exercise 12.2 for more details).

§ 14. Multiplication by numbers and addition.

Tensor operations are used to produce new tensors from those we already have.

The most simple of them are multiplication by number and addition. If

we have some tensor X of type (r, s) and a real number α, then in some base

e

1

, e

2

, e

3

we have the array of components of tensor X; let’s denote it X

i1... ir

j1... js

.

Then by multiplying all the components of this array by α we get another array

Y

i1... ir

j1... js

= αX

i1... ir

j1... js

. (14.1)

Choosing another base ˜ e

1

, ˜ e

2

, ˜ e

3

and repeating this operation we get

˜

Y

i1... ir

j1... js

= α

˜

X

i1... ir

j1... js

. (14.2)

28 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

Exercise 14.1. Prove that arrays

˜

Y

i1... ir

j1... js

and Y

i1... ir

j1... js

are related to each other

in the same way as arrays

˜

X

i1... ir

j1... js

and X

i1... ir

j1... js

, i. e. according to transformation

formulas (12.1) and (12.2). In doing this you prove that formula (14.1) applied in

all bases produces new tensor Y = αX from initial tensor X.

Formula (14.1) deﬁnes the multiplication of tensors by numbers. In

exercise 14.1 you prove its consistence. The next formula deﬁnes the addition of

tensors:

X

i1... ir

j1... js

+ Y

i1... ir

j1... js

= Z

i1... ir

j1... js

. (14.3)

Having two tensors X and Y both of type (r, s) we produce a third tensor Z of the

same type (r, s) by means of formula (14.3). It’s natural to denote Z = X+Y.

Exercise 14.2. By analogy with exercise 14.1 prove the consistence of formula

(14.3).

Exercise 14.3. What happens if we multiply tensor X by the number zero and

by the number minus one ? What would you call the resulting tensors ?

§ 15. Tensor product.

The tensor product is deﬁned by a more tricky formula. Suppose we have tensor

X of type (r, s) and tensor Y of type (p, q), then we can write:

Z

i1... ir+p

j1... js+q

= X

i1... ir

j1... js

Y

ir+1... ir+p

js+1... js+q

. (15.1)

Formula (15.1) produces new tensor Z of the type (r + p, s + q). It is called the

tensor product of X and Y and denoted Z = X ⊗ Y. Don’t mix the tensor

product and the cross product. They are diﬀerent.

Exercise 15.1. By analogy with exercise 14.1 prove the consistence of formula

(15.1).

Exercise 15.2. Give an example of two tensors such that X⊗Y = Y⊗X.

§ 16. Contraction.

As we have seen above, the tensor product increases the number of indices.

Usually the tensor Z = X⊗Y has more indices than X and Y. Contraction is an

operation that decreases the number of indices. Suppose we have tensor X of the

type (r + 1, s + 1). Then we can produce tensor Z of type (r, s) by means of the

following formula:

Z

i1... ir

j1... js

=

n

¸

ρ=1

X

i1... im−1 ρ im... ir

j1... j

k−1

ρ j

k

... js

. (16.1)

What we do ? Tensor X has at least one upper index and at least one lower index.

We choose the m-th upper index and replace it by the summation index ρ. In the

same way we replace the k-th lower index by ρ. Other r upper indices and s lower

indices are free. They are numerated in some convenient way, say as in formula

(16.1). Then we perform summation with respect to index ρ. The contraction is

over. This operation is called a contraction with respect to m-th upper and

k-th lower indices. Thus, if we have many upper an many lower indices in tensor

X, we can perform various types of contractions to this tensor.

CopyRight c Sharipov R.A., 2004.

§ 18. SOME SPECIAL TENSORS AND SOME USEFUL FORMULAS. 29

Exercise 16.1. Prove the consistence of formula (16.1).

Exercise 16.2. Look at formula (9.1) and interpret this formula as the con-

traction of the tensor product a⊗x. Find similar interpretations for (10.4), (11.4),

and (13.4).

§ 17. Raising and lowering indices.

Suppose that X is some tensor of type (r, s). Let’s choose its α-th lower index:

X

... ... ...

... k ...

. The symbols used for the other indices are of no importance. Therefore,

we denoted them by dots. Then let’s consider the tensor product Y = g ⊗X:

Y

... p q ...

... k ...

= g

pq

X

... ... ...

... k ...

. (17.1)

Here g is the dual metric tensor with the components g

pq

(see section 13 above).

In the next step let’s contract (17.1) with respect to the pair of indices k and q.

For this purpose we replace them both by s and perform the summation:

X

... p ...

... ... ...

=

3

¸

s=1

g

ps

X

... ... ...

... s ...

. (17.2)

This operation (17.2) is called the index raising procedure. It is invertible.

The inverse operation is called the index lowering procedure:

X

... ... ...

... p ...

=

3

¸

s=1

g

ps

X

... s ...

... ... ...

. (17.3)

Like (17.2), the index lowering procedure (17.3) comprises two tensorial operations:

the tensor product and contraction.

§ 18. Some special tensors and some useful formulas.

Kronecker symbol is a well known object. This is a two-dimensional array

representing the unit matrix. It is determined as follows:

δ

i

j

=

1 for i = j,

0 for i = j.

(18.1)

We can determine two other versions of Kronecker symbol:

δ

ij

= δ

ij

=

1 for i = j,

0 for i = j.

(18.2)

Exercise 18.1. Prove that deﬁnition (18.1) is invariant under a change of basis,

if we interpret the Kronecker symbol as a tensor. Show that both deﬁnitions in

(18.2) are not invariant under a change of basis.

Exercise 18.2. Lower index i of tensor (18.1) by means of (17.3). What ten-

sorial object do you get as a result of this operation?

30 CHAPTER II. TENSORS IN CARTESIAN COORDINATES.

Exercise 18.3. Likewise, raise index J in (18.1).

Another well known object is the Levi-Civita symbol. This is a three-

dimensional array determined by the following formula:

jkq

=

jkq

=

**0, if among j, k, q, there are
**

at least two equal num-

bers;

1, if (j k q) is even permuta-

tion of numbers (1 2 3);

−1, if (j k q) is odd permuta-

tion of numbers (1 2 3).

(18.3)

The Levi-Civita symbol (18.3) is not a tensor. However, we can produce two

tensors by means of Levi-Civita symbol. The ﬁrst of them

ω

ijk

=

det(g

ij

)

ijk

(18.4)

is known as the volume tensor. Another one is the dual volume tensor:

ω

ijk

=

det(g

ij

)

ijk

. (18.5)

Let’s take two vectors x and y. Then using (18.4) we can produce covector a:

a

i

=

3

¸

j=1

3

¸

k=1

ω

ijk

x

j

y

k

. (18.6)

Then we can apply index raising procedure (17.2) and produce vector a:

a

r

=

3

¸

i=1

3

¸

j=1

3

¸

k=1

g

ri

ω

ijk

x

j

y

k

. (18.7)

Formula (18.7) is known as formula for the vectorial product (cross product) in

skew-angular basis.

Exercise 18.4. Prove that the vector a with components (18.7) coincides with

cross product of vectors x and y, i. e. a = [x, y].

CHAPTER III

TENSOR FIELDS. DIFFERENTIATION OF TENSORS.

§ 19. Tensor ﬁelds in Cartesian coordinates.

The tensors that we deﬁned in section 12 are free tensors. Indeed, their

components are arrays related to bases, while any basis is a triple of free vectors

(not bound to any point). Hence, the tensors previously considered are also not

bound to any point.

Now suppose we want to bind our tensor to some point in space, then another

tensor to another point and so on. Doing so we can ﬁll our space with tensors, one

per each point. In this case we say that we have a tensor ﬁeld. In order to mark a

point P to which our particular tensor is bound we shall write P as an argument:

X = X(P). (19.1)

Usually the valencies of all tensors composing the tensor ﬁeld are the same. Let

them all be of type (r, s). Then if we choose some basis e

1

, e

2

, e

3

, we can represent

any tensor of our tensor ﬁeld as an array X

i1... ir

j1... js

with r + s indices:

X

i1... ir

j1... js

= X

i1... ir

j1... js

(P). (19.2)

Thus, the tensor ﬁeld (19.1) is a tensor-valued function with argument P being a

point of three-dimensional Euclidean space E, and (19.2) is the basis representation

for (19.1). For each ﬁxed set of numeric values of indices i

1

, . . . , i

r

, j

1

, . . . , j

s

in

(19.2), we have a numeric function with a point-valued argument. Dealing with

point-valued arguments is not so convenient, for example, if we want to calculate

derivatives. Therefore, we need to replace P by something numeric. Remember

that we have already chosen a basis. If, in addition, we ﬁx some point O as an

origin, then we get Cartesian coordinate system in space and hence can represent

P by its radius-vector r

P

=

−→

OP and by its coordinates x

1

, x

2

, x

3

:

X

i1... ir

j1... js

= X

i1... ir

j1... js

(x

1

, x

2

, x

3

). (19.3)

Conclusion 19.1. In contrast to free tensors, tensor ﬁelds are related not to

bases, but to whole coordinate systems (including the origin). In each coordinate

system they are represented by functional arrays, i. e. by arrays of functions (see

(19.3)).

A functional array (19.3) is a coordinate representation of a tensor ﬁeld (19.1).

What happens when we change the coordinate system? Dealing with (19.2), we

32 CHAPTER III. TENSOR FIELDS. DIFFERENTIATION OF TENSORS.

need only to recalculate the components of the array X

i1... ir

j1... js

in the basis by means

of (12.2):

˜

X

i1... ir

j1... js

(P) =

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

T

i1

h1

. . . T

ir

hr

S

k1

j1

. . . S

ks

js

X

h1... hr

k1... ks

(P). (19.4)

In the case of (19.3), we need to recalculate the components of the array X

i1... ir

j1... js

in the new basis

˜

X

i1... ir

j1... js

(˜ x

1

, ˜ x

2

, ˜ x

3

) =

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

T

i1

h1

. . . T

ir

hr

S

k1

j1

. . . S

ks

js

X

h1... hr

k1... ks

(x

1

, x

2

, x

3

), (19.5)

using (12.2), and we also need to express the old coordinates x

1

, x

2

, x

3

of the

point P in right hand side of (19.5) through new coordinates of the same point:

x

1

= x

1

(˜ x

1

, ˜ x

2

, ˜ x

3

),

x

2

= x

2

(˜ x

1

, ˜ x

2

, ˜ x

3

),

x

3

= x

3

(˜ x

1

, ˜ x

2

, ˜ x

3

).

(19.6)

Like (12.2), formula (19.5) can be inverted by means of (12.1):

X

i1... ir

j1... js

(x

1

, x

2

, x

3

) =

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

S

i1

h1

. . . S

ir

hr

T

k1

j1

. . . T

ks

js

˜

X

h1... hr

k1... ks

(˜ x

1

, ˜ x

2

, ˜ x

3

). (19.7)

But now, apart from (19.7), we should have inverse formulas for (19.6) as well:

˜ x

1

= x

1

(x

1

, x

2

, x

3

),

˜ x

2

= x

2

(x

1

, x

2

, x

3

),

˜ x

3

= x

3

(x

1

, x

2

, x

3

).

(19.8)

THe couple of formulas (19.5) and (19.6), and another couple of formulas (19.7)

and (19.8), in the case of tensor ﬁelds play the same role as transformation

formulas (12.1) and (12.2) in the case of free tensors.

§ 20. Change of Cartesian coordinate system.

Note that formulas (19.6) and (19.8) are written in abstract form. They only

indicate the functional dependence of new coordinates of the point P from old

ones and vice versa. Now we shall specify them for the case when one Cartesian

coordinate system is changed to another Cartesian coordinate system. Remember

that each Cartesian coordinate system is determined by some basis and some ﬁxed

point (the origin). We consider two Cartesian coordinate systems. Let the origins

of the ﬁrst and second systems be at the points O and

˜

O, respectively. Denote by

e

1

, e

2

, e

3

the basis of the ﬁrst coordinate system, and by ˜ e

1

, ˜ e

2

, ˜ e

3

the basis of

the second coordinate system (see Fig. 7 below).

Let P be some point in the space for whose coordinates we are going to

derive the specializations of formulas (19.6) and (19.8). Denote by r

P

and ˜r

P

the

§ 20. CHANGE OF CARTESIAN COORDINATE SYSTEM. 33

radius-vectors of this point in our two

coordinate systems. Then r

P

=

−→

OP and

˜r

P

=

−→

˜

OP. Hence

r

P

=

−→

O

˜

O + ˜r

P

. (20.1)

Vector

−→

O

˜

O determines the origin shift

from the old to the new coordinate sys-

tem. We expand this vector in the basis

e

1

, e

2

, e

3

:

a =

−→

O

˜

O =

3

¸

i=1

a

i

e

i

. (20.2)

Radius-vectors r

P

and ˜r

P

are expanded

in the bases of their own coordinate sys-

tems:

r

P

=

3

¸

i=1

x

i

e

i

,

˜r

P

=

3

¸

i=1

˜ x

i

˜ e

i

,

(20.3)

Exercise 20.1. Using (20.1), (20.2), (20.3), and (5.7) derive the following for-

mula relating the coordinates of the point P in the two coordinate systems in Fig. 7:

x

i

= a

i

+

3

¸

j=1

S

i

j

˜ x

j

. (20.4)

Compare (20.4) with (6.5). Explain the diﬀerences in these two formulas.

Exercise 20.2. Derive the following inverse formula for (20.4):

˜ x

i

= ˜ a

i

+

3

¸

j=1

T

i

j

x

j

. (20.5)

Prove that a

i

in (20.4) and ˜ a

i

in (20.5) are related to each other as follows:

˜ a

i

= −

3

¸

j=1

T

i

j

a

j

, a

i

= −

3

¸

j=1

S

i

j

˜ a

j

. (20.6)

Compare (20.6) with (6.2) and (6.5). Explain the minus signs in these formulas.

Formula (20.4) can be written in the following expanded form:

x

1

= S

1

1

˜ x

1

+ S

1

2

˜ x

2

+ S

1

3

˜ x

3

+ a

1

,

x

2

= S

2

1

˜ x

1

+ S

2

2

˜ x

2

+ S

2

3

˜ x

3

+ a

2

,

x

3

= S

3

1

˜ x

1

+ S

3

2

˜ x

2

+ S

3

3

˜ x

3

+ a

3

.

(20.7)

34 CHAPTER III. TENSOR FIELDS. DIFFERENTIATION OF TENSORS.

This is the required specialization for (19.6). In a similar way we can expand

(20.5):

˜ x

1

= T

1

1

x

1

+ T

1

2

x

2

+ T

1

3

x

3

+ ˜ a

1

,

˜ x

2

= T

2

1

x

1

+ T

2

2

x

2

+ T

2

3

x

3

+ ˜ a

2

,

˜ x

3

= T

3

1

x

1

+ T

3

2

x

2

+ T

3

3

x

3

+ ˜ a

3

.

(20.8)

This is the required specialization for (19.8). Formulas (20.7) and (20.8) are used

to accompany the main transformation formulas (19.5) and (19.7).

§ 21. Diﬀerentiation of tensor ﬁelds.

In this section we consider two diﬀerent types of derivatives that are usually

applied to tensor ﬁelds: diﬀerentiation with respect to spacial variables x

1

, x

2

, x

3

and diﬀerentiation with respect to external parameters other than x

1

, x

2

, x

3

, if

they are present. The second type of derivatives are simpler to understand. Let’s

consider them to start. Suppose we have tensor ﬁeld X of type (r, s) and depending

on the additional parameter t (for instance, this could be a time variable). Then,

upon choosing some Cartesian coordinate system, we can write

∂X

i1... ir

j1... js

∂t

= lim

h→0

X

i1... ir

j1... js

(t + h, x

1

, x

2

, x

3

) −X

i1... ir

j1... js

(t, x

1

, x

2

, x

3

)

h

. (21.1)

The left hand side of (21.1) is a tensor since the fraction in right hand side is

constructed by means of tensorial operations (14.1) and (14.3). Passing to the limit

h → 0 does not destroy the tensorial nature of this fraction since the transition

matrices S and T in (19.5), (19.7), (20.7), (20.8) are all time-independent.

Conclusion 21.1. Diﬀerentiation with respect to external parameters (like t

in (21.1)) is a tensorial operation producing new tensors from existing ones.

Exercise 21.1. Give a more detailed explanation of why the time derivative

(21.1) represents a tensor of type (r, s).

Now let’s consider the spacial derivative of tensor ﬁeld X, i. e. its derivative

with respect to a spacial variable, e. g. with respect to x

1

. Here we also can write

∂X

i1... ir

j1... js

∂x

1

= lim

h→0

X

i1... ir

j1... js

(x

1

+ h, x

2

, x

3

) −X

i1... ir

j1... js

(x

1

, x

2

, x

3

)

h

, (21.2)

but in numerator of the fraction in the right hand side of (21.2) we get the

diﬀerence of two tensors bound to diﬀerent points of space: to the point P with

coordinates x

1

, x

2

, x

3

and to the point P

with coordinates x

1

+ h, x

2

, x

3

. To

which point should be attributed the diﬀerence of two such tensors ? This is not

clear. Therefore, we should treat partial derivatives like (21.2) in a diﬀerent way.

Let’s choose some additional symbol, say it can be q, and consider the partial

derivative of X

i1... ir

j1... js

with respect to the spacial variable x

q

:

Y

i1... ir

q j1... js

=

∂X

i1... ir

j1... js

∂x

q

. (21.3)

Partial derivatives (21.2), taken as a whole, form an (r + s + 1)-dimensional array

§ 22. GRADIENT, DIVERGENCY, AND ROTOR . . . 35

with one extra dimension due to index q. We write it as a lower index in Y

i1... ir

q j1... js

due to the following theorem concerning (21.3).

Theorem 21.1. For any tensor ﬁeld X of type (r, s) partial derivatives (21.3)

with respect to spacial variables x

1

, x

2

, x

3

in any Cartesian coordinate system

represent another tensor ﬁeld Y of the type (r, s + 1).

Thus diﬀerentiation with respect to x

1

, x

2

, x

3

produces new tensors from

already existing ones. For the sake of beauty and convenience this operation is

denoted by the nabla sign: Y = ∇X. In index form this looks like

Y

i1... ir

q j1... js

= ∇

q

X

i1... ir

j1... js

. (21.4)

Simplifying the notations we also write:

∇

q

=

∂

∂x

q

. (21.5)

Warning 21.1. Theorem 21.1 and the equality (21.5) are valid only for Carte-

sian coordinate systems. In curvilinear coordinates things are diﬀerent.

Exercise 21.2. Prove theorem 21.1. For this purpose consider another Carte-

sian coordinate system ˜ x

1

, ˜ x

2

, ˜ x

3

related to x

1

, x

2

, x

3

via (20.7) and (20.8). Then

in the new coordinate system consider the partial derivatives

˜

Y

i1... ir

q j1... js

=

∂

˜

X

i1... ir

j1... js

∂˜ x

q

(21.6)

and derive relationships binding (21.6) and (21.3).

§ 22. Gradient, divergency, and rotor.

Laplace and d’Alambert operators.

The tensorial nature of partial derivatives established by theorem 21.1 is a very

useful feature. We can apply it to extend the scope of classical operations of

vector analysis. Let’s consider the gradient, grad = ∇. Usually the gradient

operator is applied to scalar ﬁelds, i. e. to functions ϕ = ϕ(P) or ϕ = ϕ(x

1

, x

2

, x

3

)

in coordinate form:

a

q

= ∇

q

ϕ =

∂ϕ

∂x

q

. (22.1)

Note that in (22.1) we used a lower index q for a

q

. This means that a = gradϕ

is a covector. Indeed, according to theorem 21.1, the nabla operator applied to

a scalar ﬁeld, which is tensor ﬁeld of type (0, 0), produces a tensor ﬁeld of type

(0, 1). In order to get the vector form of the gradient one should raise index q:

a

q

=

3

¸

i=1

g

qi

a

i

=

3

¸

i=1

g

qi

∇

i

ϕ. (22.2)

Let’s write (22.2) in the form of a diﬀerential operator (without applying to ϕ):

∇

q

=

3

¸

i=1

g

qi

∇

i

. (22.3)

CopyRight c Sharipov R.A., 2004.

36 CHAPTER III. TENSOR FIELDS. DIFFERENTIATION OF TENSORS.

In this form the gradient operator (22.3) can be applied not only to scalar ﬁelds,

but also to vector ﬁelds, covector ﬁelds and to any other tensor ﬁelds.

Usually in physics we do not distinguish between the vectorial gradient ∇

q

and

the covectorial gradient ∇

q

because we use orthonormal coordinates with ONB as

a basis. In this case dual metric tensor is given by unit matrix (g

ij

= δ

ij

) and

components of ∇

q

and ∇

q

coincide by value.

Divergency is the second diﬀerential operation of vector analysis. Usually it

is applied to a vector ﬁeld and is given by formula:

div X =

3

¸

i=1

∇

i

X

i

. (22.4)

As we see, (22.4) is produced by contraction (see section 16) from tensor ∇

q

X

i

.

Therefore we can generalize formula (22.4) and apply divergency operator to

arbitrary tensor ﬁeld with at least one upper index:

(div X)

... ... ...

... ... ...

=

3

¸

s=1

∇

s

X

... .s. ...

... ... ...

. (22.5)

The Laplace operator is deﬁned as the divergency applied to a vectorial gradient

of something, it is denoted by the triangle sign: = div grad. From (22.3) and

(22.5) for Laplace operator we derive the following formula:

=

3

¸

i=1

3

¸

j=1

g

ij

∇

i

∇

j

. (22.6)

Denote by the following diﬀerential operator:

=

1

c

2

∂

2

∂t

2

−. (22.7)

Operator (22.7) is called the d’Alambert operator or wave operator. In

general relativity upon introducing the additional coordinate x

0

= c t one usually

rewrites the d’Alambert operator in a form quite similar to (22.6) (see my book

[5], it is free for download from http://samizdat.mines.edu/).

And ﬁnally, let’s consider the rotor operator or curl operator (the term

“rotor” is derived from “rotation” so that “rotor” and “curl” have approximately

the same meaning). The rotor operator is usually applied to a vector ﬁeld and

produces another vector ﬁeld: Y = rot X. Here is the formula for the r-th

coordinate of rot X:

(rot X)

r

=

3

¸

i=1

3

¸

j=1

3

¸

k=1

g

ri

ω

ijk

∇

j

X

k

. (22.8)

The volume tensor ω in (22.8) is given by formula (18.4), while the vectorial

gradient operator ∇

j

is deﬁned in (22.3).

§ 22. GRADIENT, DIVERGENCY, AND ROTOR . . . 37

Exercise 22.1. Formula (22.8) can be generalized for the case when X is an

arbitrary tensor ﬁeld with at least one upper index. By analogy with (22.5) suggest

your version of such a generalization.

Note that formulas (22.6) and (22.8) for the Laplace operator and for the rotor

are diﬀerent from those that are commonly used. Here are standard formulas:

=

∂

∂x

1

2

+

∂

∂x

2

2

+

∂

∂x

3

2

, (22.9)

rot X = det

e

1

e

2

e

3

∂

∂x

1

∂

∂x

2

∂

∂x

3

X

1

X

2

X

3

. (22.10)

The truth is that formulas (22.6) and (22.8) are written for a general skew-angular

coordinate system with a SAB as a basis. The standard formulas (22.10) are valid

only for orthonormal coordinates with ONB as a basis.

Exercise 22.2. Show that in case of orthonormal coordinates, when g

ij

= δ

ij

,

formula (22.6) for the Laplace operator reduces to the standard formula (22.9).

The coordinates of the vector rot X in a skew-angular coordinate system are

given by formula (22.8). Then for vector rot X itself we have the expansion:

rot X =

3

¸

r=1

(rot X)

r

e

r

. (22.11)

Exercise 22.3. Substitute (22.8) into (22.11) and show that in the case of a

orthonormal coordinate system the resulting formula (22.11) reduces to (22.10).

CHAPTER IV

TENSOR FIELDS IN CURVILINEAR COORDINATES.

§ 23. General idea of curvilinear coordinates.

What are coordinates, if we forget for a moment about radius-vectors, bases

and axes ? What is the pure idea of coordinates ? The pure idea is in representing

points of space by triples of numbers. This means that we should have one to

one map P (y

1

, y

2

, y

3

) in the whole space or at least in some domain, where

we are going to use our coordinates y

1

, y

2

, y

3

. In Cartesian coordinates this

map P (y

1

, y

2

, y

3

) is constructed by means of vectors and bases. Arranging

other coordinate systems one can use other methods. For example, in spherical

coordinates y

1

= r is a distance from the point P to the center of sphere, y

2

= θ

and y

3

= ϕ are two angles. By the way, spherical coordinates are one of the

simplest examples of curvilinear coordinates. Furthermore, let’s keep in mind

spherical coordinates when thinking about more general and hence more abstract

curvilinear coordinate systems.

§ 24. Auxiliary Cartesian coordinate system.

Now we know almost everything about Cartesian coordinates and almost noth-

ing about the abstract curvilinear coordinate system y

1

, y

2

, y

3

that we are going

to study. Therefore, the best idea is to represent each point P by its radius-

vector r

P

in some auxiliary Cartesian coordinate system and then consider a map

r

P

(y

1

, y

2

, y

3

). The radius-vector itself is represented by three coordinates in

the basis e

1

, e

2

, e

3

of the auxiliary Cartesian coordinate system:

r

P

=

3

¸

i=1

x

i

e

i

. (24.1)

Therefore, we have a one-to-one map (x

1

, x

2

, x

3

) (y

1

, y

2

, y

3

). Hurrah! This is a

numeric map. We can treat it numerically. In the left direction it is represented

by three functions of three variables:

x

1

= x

1

(y

1

, y

2

, y

3

),

x

2

= x

2

(y

1

, y

2

, y

3

),

x

3

= x

3

(y

1

, y

2

, y

3

).

(24.2)

In the right direction we again have three functions of three variables:

y

1

= y

1

(x

1

, x

2

, x

3

),

y

2

= y

2

(x

1

, x

2

, x

3

),

y

3

= y

3

(x

1

, x

2

, x

3

).

(24.3)

§ 25. COORDINATE LINES AND THE COORDINATE GRID. 39

Further we shall assume all functions in (24.2) and (24.3) to be diﬀerentiable and

consider their partial derivatives. Let’s denote

S

i

j

=

∂x

i

∂y

j

, T

i

j

=

∂y

i

∂x

j

. (24.4)

Partial derivatives (24.4) can be arranged into two square matrices S and T

respectively. In mathematics such matrices are called Jacobi matrices. The

components of matrix S in that form, as they are deﬁned in (24.4), are functions

of y

1

, y

2

, y

3

. The components of T are functions of x

1

, x

2

, x

3

:

S

i

j

= S

i

j

(y

1

, y

2

, y

3

), T

i

j

= T

i

j

(x

1

, x

2

, x

3

). (24.5)

However, by substituting (24.3) into the arguments of S

i

j

, or by substituting (24.2)

into the arguments of T

i

j

, we can make them have a common set of arguments:

S

i

j

= S

i

j

(x

1

, x

2

, x

3

), T

i

j

= T

i

j

(x

1

, x

2

, x

3

), (24.6)

S

i

j

= S

i

j

(y

1

, y

2

, y

3

), T

i

j

= T

i

j

(y

1

, y

2

, y

3

), (24.7)

When brought to the form (24.6), or when brought to the form (24.7) (but not in

form of (24.5)), matrices S and T are inverse of each other:

T = S

−1

. (24.8)

This relationship (24.8) is due to the fact that numeric maps (24.2) and (24.3) are

inverse of each other.

Exercise 24.1. You certainly know the following formula:

df(x

1

(y), x

2

(y), x

3

(y))

dy

=

3

¸

i=1

f

i

(x

1

(y), x

2

(y), x

3

(y))

dx

i

(y)

dy

, where f

i

=

∂f

∂x

i

.

It’s for the diﬀerentiation of composite function. Apply this formula to functions

(24.2) and derive the relationship (24.8).

§ 25. Coordinate lines and the coordinate grid.

Let’s substitute (24.2) into (24.1) and take into account that (24.2) now assumed

to contain diﬀerentiable functions. Then the vector-function

R(y

1

, y

2

, y

3

) = r

P

=

3

¸

i=1

x

i

(y

1

, y

2

, y

3

) e

i

(25.1)

is a diﬀerentiable function of three variables y

1

, y

2

, y

3

. The vector-function

R(y

1

, y

2

, y

3

) determined by (25.1) is called a basic vector-function of a curvi-

linear coordinate system. Let P

0

be some ﬁxed point of space given by its

curvilinear coordinates y

1

0

, y

2

0

, y

3

0

. Here zero is not the tensorial index, we use it

in order to emphasize that P

0

is ﬁxed point, and that its coordinates y

1

0

, y

2

0

, y

3

0

40 CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES.

are three ﬁxed numbers. In the next step let’s undo one of them, say ﬁrst one, by

setting

y

1

= y

1

0

+ t, y

2

= y

2

0

, y

3

= y

3

0

. (25.2)

Substituting (25.2) into (25.1) we get a vector-function of one variable t:

R

1

(t) = R(y

1

0

+ t, y

2

0

, y

3

0

), (25.3)

If we treat t as time variable (though it

may have a unit other than time), then

(25.3) describes a curve (the trajectory

of a particle). At time instant t = 0

this curve passes through the ﬁxed point

P

0

. Same is true for curves given by two

other vector-functions similar to (25.4):

R

2

(t) = R(y

1

0

, y

2

0

+ t, y

3

0

), (25.4)

R

3

(t) = R(y

1

0

, y

2

0

, y

3

0

+ t). (25.5)

This means that all three curves given by

vector-functions (25.3), (25.4), and (25.5)

are intersected at the point P

0

as shown

on Fig. 8. Arrowheads on these lines

indicate the directions in which parameter t increases. Curves (25.3), (25.4),

and (25.5) are called coordinate lines. They are subdivided into three families.

Curves within one family do not intersect each other. Curves from diﬀerent

families intersect so that any regular point of space is an intersection of exactly

three coordinate curves (one per family).

Coordinate lines taken in whole form

a coordinate grid. This is an inﬁnitely

dense grid. But usually, when drawing,

it is represented as a grid with ﬁnite

density. On Fig. 9 the coordinate grid

of curvilinear coordinates is compared to

that of the Cartesian coordinate system.

Another example of coordinate grid is

on Fig. 2. Indeed, meridians and par-

allels are coordinate lines of a spherical

coordinate system. The parallels do not

intersect, but the meridians forming one

family of coordinate lines do intersect at the North and at South Poles. This

means that North and South Poles are singular points for spherical coordinates.

Exercise 25.1. Remember the exact deﬁnition of spherical coordinates and

ﬁnd all singular points for them.

§ 26. MOVING FRAME OF CURVILINEAR COORDINATES. 41

§ 26. Moving frame of curvilinear coordinates.

Let’s consider the three coordinate lines shown on Fig. 8 again. And let’s ﬁnd

tangent vectors to them at the point P

0

. For this purpose we should diﬀerentiate

vector-functions (25.3), (25.4), and (25.5) with respect to the time variable t and

then substitute t = 0 into the derivatives:

E

i

=

dR

i

dt

t=0

=

∂R

∂y

i

at the point P0

. (26.1)

Now let’s substitute the expansion (25.1) into (26.1) and remember (24.4):

E

i

=

∂R

∂y

i

=

3

¸

j=1

∂x

j

∂y

i

e

j

=

3

¸

j=1

S

j

i

e

j

. (26.2)

All calculations in (26.2) are still in reference to the point P

0

. Though P

0

is a

ﬁxed point, it is an arbitrary ﬁxed point. Therefore, the equality (26.2) is valid at

any point. Now let’s omit the intermediate calculations and write (26.2) as

E

i

=

3

¸

i=1

S

j

i

e

j

. (26.3)

And then compare (26.3) with (5.7). They are strikingly similar, and det S = 0

due to (24.8). Formula (26.3) means that tangent vectors to coordinate lines

E

1

, E

2

, E

3

form a basis (see Fig. 10),

matrices (24.4) are transition matrices

to this basis and back to the Cartesian

basis.

Despite obvious similarity of the for-

mulas (26.3) and (5.7), there is some

crucial diﬀerence of basis E

1

, E

2

, E

3

as

compared to e

1

, e

2

, e

3

. Vectors E

1

, E

2

,

E

3

are not free. They are bound to that

point where derivatives (24.4) are calcu-

lated. And they move when we move this

point. For this reason basis E

1

, E

2

, E

3

is called moving frame of the curvi-

linear coordinate system. During their

motion the vectors of the moving frame

E

1

, E

2

, E

3

are not simply translated

from point to point, they can change

their lengths and the angles they form with each other. Therefore, in general

the moving frame E

1

, E

2

, E

3

is a skew-angular basis. In some cases vectors

E

1

, E

2

, E

3

can be orthogonal to each other at all points of space. In that case we

say that we have an orthogonal curvilinear coordinate system. Most of the well

known curvilinear coordinate systems are orthogonal, e. g. spherical, cylindrical,

elliptic, parabolic, toroidal, and others. However, there is no curvilinear coordinate

system with the moving frame being ONB! We shall not prove this fact since it

leads deep into diﬀerential geometry.

42 CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES.

§ 27. Dynamics of moving frame.

Thus, we know that the moving frame moves. Let’s describe this motion

quantitatively. According to (24.5) the components of matrix S in (26.3) are

functions of the curvilinear coordinates y

1

, y

2

, y

3

. Therefore, diﬀerentiating E

i

with respect to y

j

we should expect to get some nonzero vector ∂E

i

/∂y

j

. This

vector can be expanded back in moving frame E

1

, E

2

, E

3

. This expansion is

written as

∂E

i

∂y

j

=

3

¸

k=1

Γ

k

ij

E

k

. (27.1)

Formula (27.1) is known as the derivational formula. Coeﬃcients Γ

k

ij

in (27.1)

are called Christoﬀel symbols or connection components.

Exercise 27.1. Relying upon formula (25.1) and (26.1) draw the vectors of the

moving frame for cylindrical coordinates.

Exercise 27.2. Do the same for spherical coordinates.

Exercise 27.3. Relying upon formula (27.1) and results of exercise 27.1 calcu-

late the Christoﬀel symbols for cylindrical coordinates.

Exercise 27.4. Do the same for spherical coordinates.

Exercise 27.5. Remember formula (26.2) from which you derive

E

i

=

∂R

∂y

i

. (27.2)

Substitute (27.2) into left hand side of the derivational formula (27.1) and relying on

the properties of mixed derivatives prove that the Christoﬀel symbols are symmetric

with respect to their lower indices: Γ

k

ij

= Γ

k

ji

.

Note that Christoﬀel symbols Γ

k

ij

form a three-dimensional array with one

upper index and two lower indices. However, they do not represent a tensor. We

shall not prove this fact since it again leads deep into diﬀerential geometry.

§ 28. Formula for Christoﬀel symbols.

Let’s take formula (26.3) and substitute it into both sides of (27.1). As a result

we get the following equality for Christoﬀel symbols Γ

k

ij

:

3

¸

q=1

∂S

q

i

∂y

j

e

q

=

3

¸

k=1

3

¸

q=1

Γ

k

ij

S

q

k

e

q

. (28.1)

Cartesian basis vectors e

q

do not depend on y

j

; therefore, they are not diﬀeren-

tiated when we substitute (26.3) into (27.1). Both sides of (28.1) are expansions

in the base e

1

, e

2

, e

3

of the auxiliary Cartesian coordinate system. Due to the

uniqueness of such expansions we have the following equality derived from (28.1):

∂S

q

i

∂y

j

=

3

¸

k=1

Γ

k

ij

S

q

k

. (28.2)

CopyRight c Sharipov R.A., 2004.

§ 29. TENSOR FIELDS IN CURVILINEAR COORDINATES. 43

Exercise 28.1. Using concept of the inverse matrix ( T = S

−1

) derive the

following formula for the Christoﬀel symbols Γ

k

ij

from (28.2):

Γ

k

ij

=

3

¸

q=1

T

k

q

∂S

q

i

∂y

j

. (28.3)

Due to (24.4) this formula (28.3) can be transformed in the following way:

Γ

k

ij

=

3

¸

q=1

T

k

q

∂S

q

i

∂y

j

=

3

¸

q=1

T

k

q

∂

2

x

q

∂y

i

∂y

j

=

3

¸

q=1

T

k

q

∂S

q

j

∂y

i

. (28.4)

Formulas (28.4) are of no practical use because they express Γ

k

ij

through an exter-

nal thing like transition matrices to and from the auxiliary Cartesian coordinate

system. However, they will help us below in understanding the diﬀerentiation of

tensors.

§ 29. Tensor ﬁelds in curvilinear coordinates.

As we remember, tensors are geometric objects related to bases and represented

by arrays if some basis is speciﬁed. Each curvilinear coordinate system provides us

a numeric representation for points, and in addition to this it provides the basis.

This is the moving frame. Therefore, we can refer tensorial objects to curvilinear

coordinate systems, where they are represented as arrays of functions:

X

i1... ir

j1... js

= X

i1... ir

j1... js

(y

1

, y

2

, y

3

). (29.1)

We also can have two curvilinear coordinate systems and can pass from one to

another by means of transition functions:

˜ y

1

= ˜ y

1

(y

1

, y

2

, y

3

),

˜ y

2

= ˜ y

2

(y

1

, y

2

, y

3

),

˜ y

3

= ˜ y

3

(y

1

, y

2

, y

3

),

y

1

= y

1

(˜ y

1

, ˜ y

2

, ˜ y

3

),

y

2

= y

2

(˜ y

1

, ˜ y

2

, ˜ y

3

),

y

3

= y

3

(˜ y

1

, ˜ y

2

, ˜ y

3

).

(29.2)

If we call ˜ y

1

, ˜ y

2

, ˜ y

3

the new coordinates, and y

1

, y

2

, y

3

the old coordinates, then

transition matrices S and T are given by the following formulas:

S

i

j

=

∂y

i

∂˜ y

j

, T

i

j

=

∂˜ y

i

∂y

j

. (29.3)

They relate moving frames of two curvilinear coordinate systems:

˜

E

i

=

3

¸

j=1

S

j

i

E

j

, E

j

=

3

¸

i=1

T

i

j

˜

E

i

. (29.4)

44 CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES.

Exercise 29.1. Derive (29.3) from (29.4) and (29.2) using some auxiliary Carte-

sian coordinates with basis e

1

, e

2

, e

3

as intermediate coordinate system:

(E

1

, E

2

, E

3

)

S

←−−−−

−−−−→

T

(e

1

, e

2

, e

3

)

˜

S

−−−−→

←−−−−

˜

T

(

˜

E

1

,

˜

E

2

,

˜

E

3

) (29.5)

Compare (29.5) with (5.13) and explain diﬀerences you have detected.

Transformation formulas for tensor ﬁelds for two curvilinear coordinate systems

are the same as in (19.4) and (19.5):

˜

X

i1... ir

j1... js

(˜ y

1

, ˜ y

2

, ˜ y

3

) =

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

T

i1

h1

. . . T

ir

hr

S

k1

j1

. . . S

ks

js

X

h1... hr

k1... ks

(y

1

, y

2

, y

3

), (29.6)

X

i1... ir

j1... js

(y

1

, y

2

, y

3

) =

3

¸

...

3

¸

h1, ... , hr

k1, ... , ks

S

i1

h1

. . . S

ir

hr

T

k1

j1

. . . T

ks

js

˜

X

h1... hr

k1... ks

(˜ y

1

, ˜ y

2

, ˜ y

3

). (29.7)

But formulas (19.6) and (19.8) should be replaced by (29.2).

§ 30. Diﬀerentiation of tensor ﬁelds

in curvilinear coordinates.

We already know how to diﬀerentiate tensor ﬁelds in Cartesian coordinates (see

section 21). We know that operator ∇ produces tensor ﬁeld of type (r, s + 1)

when applied to a tensor ﬁeld of type (r, s). The only thing we need now is to

transform ∇ to a curvilinear coordinate system. In order to calculate tensor ∇X in

curvilinear coordinates, let’s ﬁrst transform X into auxiliary Cartesian coordinates,

then apply ∇, and then transform ∇X back into curvilinear coordinates:

X

h1... hr

k1... ks

(y

1

, y

2

, y

3

)

S,T

−−−−→ X

h1... hr

k1... ks

(x

1

, x

2

, x

3

)

∇p

∇q=∂/∂x

q

∇

p

X

i1... ir

j1... js

(y

1

, y

2

, y

3

)

T,S

←−−−− ∇

q

X

h1... hr

k1... ks

(x

1

, x

2

, x

3

)

(30.1)

Matrices (24.4) are used in (30.1). From (12.3) and (12.4) we know that the

transformation of each index is a separate multiplicative procedure. When applied

to the α-th upper index, the whole chain of transformations (30.1) looks like

∇

p

X

... iα ...

... ... ...

=

3

¸

q=1

S

q

p

. . .

3

¸

hα=1

T

iα

hα

. . . ∇

q

. . .

3

¸

mα=1

S

hα

mα

. . . X

... mα ...

... ... ...

. (30.2)

Note that ∇

q

= ∂/∂x

q

is a diﬀerential operator and due to (24.4) we have

3

¸

q=1

S

q

p

∂

∂x

q

=

∂

∂y

p

. (30.3)

§ 30. DIFFERENTIATION OF TENSOR FIELDS IN CURVILINEAR COORDINATES. 45

Any diﬀerential operator when applied to a product produces a sum with as

many summands as there were multiplicand in the product. Here is the summand

produced by term S

hα

mα

in formula (30.2):

∇

p

X

... iα ...

... ... ...

= . . . +

3

¸

mα=1

3

¸

hα=1

T

iα

hα

S

hα

mα

∂y

p

X

... mα ...

... ... ...

+ . . . . (30.4)

Comparing (30.4) with (28.3) or (28.4) we can transform it into the following

equality:

∇

p

X

... iα ...

... ... ...

= . . . +

3

¸

mα=1

Γ

iα

pmα

X

... mα ...

... ... ...

+ . . . . (30.5)

Now let’s consider the transformation of the α-th lower index in (30.1):

∇

p

X

... ... ...

... jα ...

=

3

¸

q=1

S

q

p

. . .

3

¸

kα=1

S

kα

jα

. . . ∇

q

. . .

3

¸

nα=1

T

nα

kα

. . . X

... ... ...

... nα ...

. (30.6)

Applying (30.3) to (30.6) with the same logic as in deriving (30.4) we get

∇

p

X

... ... ...

... jα ...

= . . . +

3

¸

nα=1

3

¸

kα=1

S

kα

jα

T

nα

kα

∂y

p

X

... ... ...

... nα ...

+ . . . . (30.7)

In order to simplify (30.7) we need the following formula derived from (28.3):

Γ

k

ij

= −

3

¸

q=1

S

q

i

∂T

k

q

∂y

j

. (30.8)

Applying (30.8) to (30.7) we obtain

∇

p

X

... ... ...

... jα ...

= . . . −

3

¸

nα=1

Γ

nα

pjα

X

... ... ...

... nα ...

+ . . . . (30.9)

Now we should gather (30.5), (30.9), and add the term produced when ∇

q

in (30.2)

(or equivalently in (30.4)) acts upon components of tensor X. As a result we get

the following general formula for ∇

p

X

i1... ir

j1... js

:

∇

p

X

i1... ir

j1... js

=

∂X

i1... ir

j1... js

∂y

p

+

r

¸

α=1

3

¸

mα=1

Γ

iα

pmα

X

i1... mα ... ir

j1... ... ... js

−

−

s

¸

α=1

3

¸

nα=1

Γ

nα

pjα

X

i1... ... ... ir

j1... nα ... js

.

(30.10)

The operator ∇

p

determined by this formula is called the covariant derivative.

Exercise 30.1. Apply the general formula (30.10) to a vector ﬁeld and calculate

the covariant derivative ∇

p

X

q

.

46 CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES.

Exercise 30.2. Apply the general formula (30.10) to a covector ﬁeld and cal-

culate the covariant derivative ∇

p

X

q

.

Exercise 30.3. Apply the general formula (30.10) to an operator ﬁeld and ﬁnd

∇

p

F

q

m

. Consider special case when ∇

p

is applied to the Kronecker symbol δ

q

m

.

Exercise 30.4. Apply the general formula (30.10) to a bilinear form and ﬁnd

∇

p

a

qm

.

Exercise 30.5. Apply the general formula (30.10) to a tensor product a ⊗ x

for the case when x is a vector and a is a covector. Verify formula ∇(a ⊗ x) =

∇a ⊗x +a ⊗∇x.

Exercise 30.6. Apply the general formula (30.10) to the contraction C(F) for

the case when F is an operator ﬁeld. Verify the formula ∇C(F) = C(∇F).

Exercise 30.7. Derive (30.8) from (28.3).

§ 31. Concordance of metric and connection.

Let’s remember that we consider curvilinear coordinates in Euclidean space E.

In this space we have the scalar product (13.1) and the metric tensor (13.5).

Exercise 31.1. Transform the metric tensor (13.5) to curvilinear coordinates

using transition matrices (24.4) and show that here it is given by formula

g

ij

= (E

i

, E

j

). (31.1)

In Cartesian coordinates all components of the metric tensor are constant since

the basis vectors e

1

, e

2

, e

3

are constant. The covariant derivative (30.10) in

Cartesian coordinates reduces to diﬀerentiation ∇

p

= ∂/∂x

p

. Therefore,

∇

p

g

ij

= 0. (31.2)

But ∇g is a tensor. If all of its components in some coordinate system are

zero, then they are identically zero in any other coordinate system (explain why).

Therefore the identity (31.2) is valid in curvilinear coordinates as well.

Exercise 31.2. Prove (31.2) by direct calculations using formula (27.1).

The identity (31.2) is known as the concordance condition for the metric g

ij

and connection Γ

k

ij

. It is very important for general relativity.

Remember that the metric tensor enters into many useful formulas for the

gradient, divergency, rotor, and Laplace operator in section 22. What is important

is that all of these formulas remain valid in curvilinear coordinates, with the only

diﬀerence being that you should understand that ∇

p

is not the partial derivative

∂/∂x

p

, but the covariant derivative in the sense of formula (30.10).

Exercise 31.3. Calculate rot A, div H, gradϕ (vectorial gradient) in cylindrical

and spherical coordinates.

Exercise 31.4. Calculate the Laplace operator ϕ applied to the scalar ﬁeld

ϕ in cylindrical and in spherical coordinates.

REFERENCES.

1. Heﬀeron J., Linear algebra, Electronic textbook, free for downloading from Web site of Saint

Michael’s College, Colchester, VM 05439, USA; Download [PDF] or [PS] ﬁle.

2. Lehnen A. P., An elementary introduction to logic and set theory, On-line resource, Madison

Area Technical College, Madison, WI 53704, USA.

3. Konstantopoulos T., Basic background for a course of information and cryptography, On-

line materials, Feb. 2000, Electrical & Computer Engineering Dept., University of Texas at

Austin, Austin, Texas 78712, USA.

4. Vaughan-Lee M., B2 ring theory preliminaries, On-line lecture materials, Sept. 2000, Univer-

sity of Oxford, Math. Institute, Oxford, OX1 3LB, UK.

5. Sharipov R. A., Classical electrodynamics and theory of relativity, Bashkir State University,

Ufa, Russia, 1997; English tr., 2003, physics/0311011 in Electronic Archive http://arXiv.org.

2

MSC 97U20 PACS 01.30.Pp R. A. Sharipov. Quick Introduction to Tensor Analysis: lecture notes. Freely distributed on-line. Is free for individual use and educational purposes. Any commercial use without written consent from the author is prohibited. This book was written as lecture notes for classes that I taught to undergraduate students majoring in physics in February 2004 during my time as a guest instructor at The University of Akron, which was supported by Dr. Sergei F. Lyuksyutov’s grant from the National Research Council under the COBASE program. These 4 classes have been taught in the frame of a regular Electromagnetism course as an introduction to tensorial methods. I wrote this book in a ”do-it-yourself” style so that I give only a draft of tensor theory, which includes formulating deﬁnitions and theorems and giving basic ideas and formulas. All other work such as proving consistence of deﬁnitions, deriving formulas, proving theorems or completing details to proofs is left to the reader in the form of numerous exercises. I hope that this style makes learning the subject really quick and more eﬀective for understanding and memorizing. I am grateful to Department Chair Prof. Robert R. Mallik for the opportunity to teach classes and thus to be involved fully in the atmosphere of an American university. I am also grateful to Mr. M. Boiwka (mboiwka@hotmail.com) Mr. A. Calabrese (ajc10@uakron.edu) Mr. J. Comer (funnybef@lycos.com) Mr. A. Mozinski (arm5@uakron.edu) Mr. M. J. Shepard (sheppp2000@yahoo.com) for attending my classes and reading the manuscript of this book. I would like to especially acknowledge and thank Mr. Jeﬀ Comer for correcting the grammar and wording in it. Contacts to author. Oﬃce: Mathematics Department, Bashkir State University, 32 Frunze street, 450074 Ufa, Russia Phone: 7-(3472)-23-67-18 Fax: 7-(3472)-23-67-74 Home: 5 Rabochaya street, 450003 Ufa, Russia Phone: 7-(917)-75-55-786 E-mails: R Sharipov@ic.bashedu.ru, r-sharipov@mail.ru, ra sharipov@hotmail.com, URL: http:/ /www.geocities.com/r-sharipov

CopyRight c Sharipov R.A., 2004

CONTENTS.

CONTENTS. ............................................................................................... 3. CHAPTER I. PRELIMINARY INFORMATION. .......................................... 4. § § § § § § § 1. 2. 3. 4. 5. 6. 7. Geometrical and physical vectors. ........................................................... 4. Bound vectors and free vectors. .............................................................. 5. Euclidean space. .................................................................................... 8. Bases and Cartesian coordinates. ........................................................... 8. What if we need to change a basis ? ...................................................... 12. What happens to vectors when we change the basis ? ............................ 15. What is the novelty about vectors that we learned knowing transformation formula for their coordinates ? ....................................... 17.

CHAPTER II. TENSORS IN CARTESIAN COORDINATES. ..................... 18. § § § § § § § § § § § 8. Covectors. ........................................................................................... 9. Scalar product of vector and covector. .................................................. 10. Linear operators. ............................................................................... 11. Bilinear and quadratic forms. ............................................................. 12. General deﬁnition of tensors. .............................................................. 13. Dot product and metric tensor. .......................................................... 14. Multiplication by numbers and addition. ............................................. 15. Tensor product. ................................................................................. 16. Contraction. ...................................................................................... 17. Raising and lowering indices. .............................................................. 18. Some special tensors and some useful formulas. ................................... 18. 19. 20. 23. 25. 26. 27. 28. 28. 29. 29.

CHAPTER III. TENSOR FIELDS. DIFFERENTIATION OF TENSORS. ... 31. § § § § 19. 20. 21. 22. Tensor ﬁelds in Cartesian coordinates. ................................................ Change of Cartesian coordinate system. .............................................. Diﬀerentiation of tensor ﬁelds. ............................................................ Gradient, divergency, and rotor. Laplace and d’Alambert operators. ..... 31. 32. 34. 35.

CHAPTER IV. TENSOR FIELDS IN CURVILINEAR COORDINATES. ..... 38. § § § § § § § § § 23. 24. 25. 26. 27. 28. 29. 30. 31. General idea of curvilinear coordinates. ............................................... Auxiliary Cartesian coordinate system. ............................................... Coordinate lines and the coordinate grid. ............................................ Moving frame of curvilinear coordinates. ............................................. Dynamics of moving frame. ................................................................ Formula for Christoﬀel symbols. ......................................................... Tensor ﬁelds in curvilinear coordinates. ............................................... Diﬀerentiation of tensor ﬁelds in curvilinear coordinates. ...................... Concordance of metric and connection. ............................................... 38. 38. 39. 41. 42. 42. 43. 44. 46.

REFERENCES. ......................................................................................... 47.

CHAPTER I

PRELIMINARY INFORMATION.

§ 1. Geometrical and physical vectors. Vector is usually understood as a segment of straight line equipped with an arrow. Simplest example is displacement vector a. Say its length is 4 cm, i. e. |a| = 4 cm. You can draw it on the paper as shown on Fig. 1a. Then it means that point B is 4 cm apart from the point A in the direction pointed to by vector a. However, if

you take velocity vector v for a stream in a brook, you cannot draw it on the paper immediately. You should ﬁrst adopt a scaling convention, for example, saying that 1 cm on paper represents 1 cm/sec (see Fig. 1b). Conclusion 1.1. Vectors with physical meaning other than displacement vectors have no unconditional geometric presentation. Their geometric presentation is conventional; it depends on the scaling convention we choose. Conclusion 1.2. There are plenty of physical vectors, which are not geometrically visible, but can be measured and then drawn as geometric vectors. One can consider unit vector m. Its length is equal to unity not 1 cm, not 1 km, not 1 inch, and not 1 mile, but simply number 1: |m| = 1. Like physical vectors, unit vector m cannot be drawn without adopting some scaling convention. The concept of a unit vector is a very convenient one. By

BOUND VECTORS AND FREE VECTORS. acceleration. force.§ 2. etc. Conclusion 1.3. Though it is usually drawn attached to the North pole. All displacement vectors are bound ones. Along with geometrical and physical vectors one can imagine vectors whose length is a number with no unit of measure. They are bound to those points whose displacement they represent. . Bound vectors and free vectors. e. vector of angular velocity ω for Earth rotation about its axis. we can translate this vector to any point along any path provided we keep its length and direction unchanged. g. Free vectors are usually those representing global physical parameters. 5 multiplying m to various scalar quantities. This vector produces the Coriolis force aﬀecting water streams in small rivers and in oceans around the world. we can produce vectorial quantities of various physical nature: velocity. § 2. torque.

Let y = f (x) be a function with a nonnumeric argument. and α = 0. Let’s denote it VA . The next example illustrates the concept of a vector ﬁeld. But there it loses its original purpose.1. (2. Certainly. What is a scalar ﬁeld ? Suggest an appropriate deﬁnition by analogy with deﬁnition 2. Suggest your version of such a generalization. in the sky. (2) we can multiply any one of them by any real number α ∈ R. The second argument of function (2. Remember the parallelogram method for adding two vectors (draw picture). Let v be the value of function (2. Let A be some ﬁxed point (on the ground. We can perform certain algebraic operations over the vectors from VA : (1) we can add any two of them. 3). Values of a function (2. under the ground. How it is represented geometrically ? .3. Remember how vectors are multiplied by a real number α. Exercise 2. There exist functions with non-numeric arguments and nonnumeric values. it is called a time-dependent vector ﬁeld. Then vector v is a bound vector. Exercise 2. Thus we have a function v = v(t. Consider three cases: α > 0.1) is not numeric.1) at the point A in a river.1. answer is negative.1. Conclusion 2. Hence. which is to mark the water velocity at the point A. A vector-valued function with point argument is called vector ﬁeld. Exercise 2. α < 0. It is geometric object — a point. However. wherever).1) are also not numeric: they are vectors. or in outer space. Remember what the zero vector is. They constitute an inﬁnite set.6 CHAPTER I. For each point P in the water the velocity of the water jet passing through this point is deﬁned. Consider the water ﬂow in a river at some ﬁxed instant of time t.1) Its ﬁrst argument is time variable t. remember this problem and return to it later when you gain more experience. in some cases one can extend the deﬁnition of continuity and the deﬁnition of derivatives in a way applicable to some functions with non-numeric arguments. P ).2 (for deep thinking). PRELIMINARY INFORMATION. one can translate it to the point B on the bank of the river (see Fig. If no versions. Can it be continuous ? Can it be diﬀerentiable ? In general. Consider all vectors of some physical nature bound to this point (say all force vectors). Definition2. If it has an additional argument t.1. These operations are called linear operations and VA is called a linear vector space. It represents the velocity of the water jet at the point A. it is bound to point A.

visit Web page of Jim Heﬀeron http:/ /joshua.4. quotient groups. etc.2) The free vectors forming this set (2.A. we can replicate it by parallel translations that produce inﬁnitely many copies of it (see Fig. This idea.§ 2. quotient sets. then have a look to the references [2].smcvt. Exercise 2. Let’s denote by V the set of all free vectors. For the sake of simplicity further we shall denote it by the same letter V as original set (2. 2004. each point A of our geometric space is not so simple. taken as they are. or quotient set. Keep this book for further references. Let’s denote it as Cl(a). 7 Exercise 2. Then consider the set of all such objects. but remember that they are freely available on demand. Thus. This quotient set V / ∼ satisﬁes the deﬁnition of linear vector space. This set is called a factor-set. even if it is a point in a vacuum. It is denoted as V / ∼ . the class of vector a.2).edu Conclusion 2. [3]. Do you remember the exact mathematical deﬁnition of a linear vector space ? If yes. say it can be vector ˜ a. you can acknowledge the author by sending him e-mail: jim@joshua.edu/linearalgebra/ and download his book [1]. Indeed. do not form a linear vector space. If you ﬁnd it useful. where each point of vacuum space is treated as a container for various physical ﬁelds. All these clones of vector a form a class. Have you heard about binary relations. [4].2.5. Then V is union of vector spaces VA associated with all points A in space: V = A∈E VA . from which it is produced by the operation of factorization.. 4). you shouldn’t read all of these references. Free vectors. . a Let’s treat Cl(a) as a whole unit. quotient rings and so on ? If yes. try to remember strict mathematical deﬁnitions for them. as one indivisible object. if we have a vector a and if it is a free vector. is popular in modern physics. Certainly.2) are too numerous: we should work to make them conﬁne the deﬁnition of a linear vector space. (2. tensor bundles. However. CopyRight c Sharipov R. we can choose any other vector of this class as a representative. write it. Mathematically it is realized in the concept of bundles: vector bundles. Vector a is a representative of its class. BOUND VECTORS AND FREE VECTORS. It can be equipped with linear vector spaces of various natures (such as a space of force vectors in the above example). Then we have Cl(a) = Cl(˜ ). If not.smcvt. If no.

Furthermore. Therefore. Properties of our space were ﬁrst systematically described by Euclid. These three vectors can be bound to some point O common to all of them. 20-th axiom. Visit the following NonEuclidean Geometry web-site and read a few words about non-Euclidean geometry and the role of Euclid’s 5-th postulate in its discovery. is most famous. Then each point −→ B ∈ E can be uniquely identiﬁed with the displacement vector rB = OB. e3 form a non-coplanar triple of vectors if they cannot be laid onto the plane by parallel translations. What is our geometric space ? Is it a linear vector space ? By no means. axioms of order. Thus. we should remember that such identiﬁcation is not unique: it depends on our choice of the point O for the origin. Let’s choose one of them. Bases and Cartesian coordinates. passing from the vectors to their classes. Euclidean space. not by vectors.1. They also can be treated as free vectors without any deﬁnite binding point. Exercise 3. dealing with the Euclidean space E. Then. it is called Euclidean space and denoted by E. 3. It is formed by points. axioms of congruence. axioms of continuity. it makes no diﬀerence. we can identify V with the space of free vectors. while O is called origin. Usually nobody remembers all 20 of these axioms by heart. We say that three vectors e1 . e2 . However. we shall rely only on common sense and on our geometric intuition. § 4. or they can be bound to diﬀerent points in the space.8 CHAPTER I. In memory of Euclid they are subdivided into 5 groups: (1) (2) (3) (4) (5) axioms of incidence. even me. It is called the radius-vector of the point B. which is also known as 5-th postulate. This identiﬁcation is a convenient tool in studying E without referring to Euclidean axioms. denote it by O and consider the vector space VO composed by displacement vectors. Definition 4.1. . Passing from points to their radius-vectors we identify E with the linear vector space VO . the Greek mathematician of antiquity. though I wrote a textbook on the foundations of Euclidean geometry in 1998. PRELIMINARY INFORMATION. However. E is composed by points. Euclid suggested 5 axioms (5 postulates) to describe E. his statements were not satisfactorily strict from a modern point of view. axiom of parallels. Currently E is described by 20 axioms.

γ are scalars. e2 . e3 ⊥ e 1 . but lengths of basis vectors are not speciﬁed. Formulate the deﬁnitions of bases on a plane and on a straight line by analogy with deﬁnition 4. e3 are bound to a common point O as shown on Fig. Now from (4.2. Assume that vectors e1 . Let’s choose some basis e1 . |e1 | = 1. Any non-coplanar ordered triple of vectors e1 . e2 . e3 in E. we get −→ − → −→ − → a = OD = OA + OB + OC.2) (4. For this basis neither angles nor lengths are speciﬁed.3) and (4. e2 . |e2 | = 1. 9 Definition 4. −→ − − → OE1 OA. As we shall see below. and a beginning at the same point O. Below we distinguish three types of bases: orthonormal basis (ONB). e3 . In the general case this is a skewangular basis. −→ − e3 = OE3 . Drawing additional lines and vectors as shown on Fig. 6. (4.4) we obtain a = α e1 + β e2 + γ e 3 .5) . e2 . 6 below. the three conditions (4. Let a be some arbitrary vector. β. −→ − − → OE3 OC (4. This vector also can be translated to the point O.3) Then from the following obvious relationships −→ − e1 = OE1 . For orthogonal basis. (4.§ 4. due to its asymmetry SAB can reveal a lot of features that are hidden in symmetric ONB.2. And skew-angular basis is the most general case. Otherwise they can be brought to this position by means of parallel translations. we derive − → OA = α e1 . Exercise 4.1) are fulﬁlled. −→ OB = β e2 .1) where α. → − OC = γ e3 . (4. e2 ⊥ e 3 . − → −→ − OE2 OB. orthogonal basis (OB).1. Orthonormal basis is formed by three mutually perpendicular unit vectors: e1 ⊥ e 2 . BASES AND CARTESIAN COORDINATES. e3 is called a basis in our geometric space E.4) −→ − e2 = OE2 . and skew-angular basis (SAB). |e3 | = 1. As a result we have four vectors e1 .

Explain why α. β. γ are coordinates of vector a in this basis.2. They are known as Einstein’s tensorial notations. we can associate vectors with columns of numbers: a1 a2 .5) as follows 3 a = a 1 e1 + a 2 e2 + a 3 e3 = i=1 ai ei . e3 (no matter ONB.6) denoting α = a1 . Give exact mathematical statements for these concepts. β. a2 is not a squared.10 CHAPTER I. e2 . Hint: remember what linear dependence and linear independence are. Formula (4. Usage of upper indices and the implicit summation rule were suggested by Einstein. OB. b3 a ←→ b ←→ (4.7) . e2 . or SAB). Exercise 4. 6 are drawn. while α.3.6) with power exponentials. Further we shall write formula (4. e3 . Once we have chosen the basis e1 . Apply them to exercise 4. β = a2 . and a3 is not a cubed. Exercise 4. PRELIMINARY INFORMATION.5) is known as the expansion of vector a in the basis e1 . for what reasons. Don’t confuse upper indices in (4. a1 here is not a. (4.3. a3 b1 b2 . Explain how. and in what order additional lines on Fig. and γ = a3 . and γ are uniquely determined by vector a.

Exercise 4. Remember the exact mathematical deﬁnition for the real arithmetic vector space Rn . Definition 4. The reason will be clear when we consider curvilinear coordinates.7) are called vector-columns. Coordinates of a point are also speciﬁed by upper indices since they are coordinates of the radiusvector for that point. The idea of cpeciﬁcation of geometric objects by means of coordinates was ﬁrst raised by French mathematician and philosopher Ren´ Descartes (1596-1650). Then. So. a3 + b 3 α a ←→ α α a1 = α a2 . and a3 . α a3 Columns of numbers framed by matrix delimiters like those in (4. reducing them to arithmetic operations with numbers: a + b ←→ a1 b1 2 a + b2 a3 b3 a1 a2 a3 = a1 + b 1 a2 + b 2 .§ 4. Indeed. a2 . a3 ) is quite an acceptable notation for the point A with coordinates a1 . if we have an origin O. where n is a positive integer.1.4. BASES AND CARTESIAN COORDINATES. 11 We can then produce algebraic operations with vectors. we get three numbers that are called the Cartesian coordinates of A. e Cartesian coordinates are named in memory of him. . writing A(a1 . The Cartesian coordinate system is a basis complemented with some ﬁxed point that is called the origin. They form linear vector spaces. a2 . having expanded this vector in a basis. they are usually not written in a column. unlike coordinates of vectors. However. then we can associate each point A of our space − → with its radius-vector rA = OA.

Let’s take the ﬁrst vector ˜ ˜ of the new basis e1 . it is ˜ nothing.5) are called transition formulas. A better way is to add an extra index to S in (5. e3 .3).1). (5. vector e1 can be expanded in the old basis e1 . e3 : 3 ˜ e1 = S 1 e1 + S 2 e2 + S 3 e3 = j=1 S j ej . PRELIMINARY INFORMATION. e2 . Instead of (5. (5. (5. e2 . Suppose we have a basis e1 . 1 2 3 ˜ e3 = S3 e1 + S3 e2 + S3 e3 . formulas (5. let’s call it the old basis. and ˜ ˜ ˜ suppose we want to change it to a new one e1 . but quite an ordinary vector of space.4) And for third vector as well: 3 1 2 3 ˜ e3 = S3 e1 + S3 e2 + S3 e3 = j=1 j S3 e j . Then we can take another vector e2 and also expand it in the old basis. We use a left curly bracket to denote their union: 1 2 3 ˜ e1 = S1 e1 + S1 e2 + S1 e3 . What if we need to change a basis ? Why could we need to change a basis ? There may be various reasons: we may dislike initial basis because it is too symmetric like ONB.1) ˜ Compare (5. this is not the best decision. § 5. or too asymmetric like SAB. Maybe we are completely satisﬁed. This is the lower index coinciding with the number of the vector: 3 1 2 3 ˜ e1 = S 1 e 1 + S 1 e 2 + S 1 e 3 = j=1 j S1 e j (5. for the second vector e2 we write a formula similar to (5. But what letter should we choose for denoting the coeﬃcients of this expansion ? We can choose another letter. e3 . 1 2 3 ˜ e2 = S2 e1 + S2 e2 + S2 e3 .5) When considered jointly. (5.6). say the letter “R”: 3 ˜ e2 = R 1 e1 + R 2 e2 + R 3 e3 = j=1 R j ej .4).3): 3 1 2 3 ˜ e2 = S2 e1 + S2 e2 + S2 e3 = j=1 j S2 e j . and (5. Indeed. it is used for highlighting only. (5. vectors e1 and e2 diﬀer only in number.12 CHAPTER I. (5. while for their coordinates we use diﬀerent letters. but the wisdom is that looking on how something changes we can learn more about this thing than if we observe it in a static position.3) Color is of no importance.2). Being isolated from the other two vectors e2 and e3 .1) and (4. e2 .6) . In this capacity.2) ˜ ˜ However.

However. Proposing this rule 5. ˜ Exercise 5.7).3). constitute ﬁrst column in matrix (5. In correctly written tensorial formulas free indices are written on the same level (upper or lower) in both sides of the equality. I don’t like omitting sums.1. It is present only in right hand side of formula (5. Einstein also suggested not to write the summation j ˜ symbols at all.7) Here index i runs over the range of integers from 1 to 3. 2. It is summation index.2. while lower one is a column number. This is a general rule.8). In other words.3) for this case ? Returning to transition formulas (5. while the second index is a column number. Formula (5. What happens if e1 = e1 ? What are the numeric values of 1 2 3 coeﬃcients S1 . If indices are on diﬀerent levels (one upper and one lower). WHAT IF WE NEED TO CHANGE A BASIS ? 13 We also can write transition formulas (5. Note that according to this rule 5. S1 .6) turn into columns in matrix (5.2. Newcomers from other branches of science would have diﬃculties in understanding formulas with implicit summation. This is also general rule for tensorial formulas. So lines of formula (5. It breaks the integrity of notations in science.8) Matrix S is called a transition matrix or direct transition matrix since we use it in passing from old basis to new one. Look at index i in formula (5.7) note that coeﬃcients in them are parameterized by two indices running independently over the range of integer numbers from 1 to 3.8). Now look at index j. Each free index has only one entry in each side of the equality. Rule 5. Rule 5. it can freely take any numeric value from its range: 1.3. In writing such matrices like S the following rule applies.7).1.6) in a more symbolic form 3 ˜ ei = j=1 j Si ej . Note that i is the lower index in both sides of formula (5. (5. It is a free index. It would be worthwhile to remember this fact. Many physicists (especially those in astrophysics) prefer writing tensorial formulas in this way.6) and (5. In correctly written tensorial formulas each summation index should have exactly two entries: one upper entry and one lower entry. For any double indexed array with indices on the same level (both upper or both lower) the ﬁrst index is a row number. then the upper index is a row number.7) then would look like ei = Si ej with implicit summation with respect to the double index j. they form a two-dimensional array that usually is represented as a table or as a matrix: 1 S1 1 S2 2 S2 3 S2 1 S3 2 S3 3 S3 S= 2 S1 3 S1 (5. and it has exactly two entries (apart from that j = 1 under the sum symbol): one in the upper level and one in the lower level. Rule 5. coeﬃcients of formula (5.3. which are written in line. .§ 5.7). and S1 in formula (5. or 3.

8). MathCad.10): 3 ei = j=1 ˜ Tij ej . e3 = e3 − e1 .2.6).10) and (5. e2 . in writing the transition formulas (5. e3 as the old basis. ˜ ˜ ˜ Now let’s swap bases.9) are exactly the ﬁrst. This means that we are going to consider e1 . remember the algorithm for calculating A−1 ). − − − T S (5. The inverse transition matrix T in (5. e2 . e3 as a column of its coordinates in the old basis just like it was done for a and b in formula (4. e2 . T = S −1 .A. e3 = e2 . (5. e.3. 3 S2 e3 ←→ 2 S3 .10) 1 2 3 ˜ ˜ ˜ e3 = T3 e1 + T3 e2 + T3 e3 . the second. 2004. All of the above stuﬀ applies to this situation. e2 . (5. e2 = e2 . e2 = e1 . 1 2 3 ˜ ˜ ˜ e2 = T2 e1 + T2 e2 + T2 e3 . Explain why the next case is impossible: ˜ ˜ ˜ (3) e1 = e1 − e2 . e3 = e2 . e3 as the new basis. Consider also the following two cases and write the transition matrices for each of them: ˜ ˜ ˜ (1) e1 = e1 . e1 . . e3 ) ←−− (˜1 . e2 = e2 − e3 . What is the inverse matrix ? Remember the deﬁnition. e3 ). e2 . This is the easiest way to remember the structure of matrix S. let’s use another letter for the coeﬃcients. Exercise 5.7) above 1 S1 1 S2 1 S3 e1 ←→ 2 S1 . However. e2 = e3 . PRELIMINARY INFORMATION. and e3 = e3 ? Find the transition matrix for this case.11).9) then these columns (5. Here is the short symbolic version of transition formulas (5. and study the inverse transition. 3 S1 e2 ←→ 2 S2 . CopyRight c Sharipov R. By tradition here the letter “T” is used: 1 2 3 ˜ ˜ ˜ e1 = T1 e1 + T1 e2 + T1 e3 . or any other..1. How is the inverse matrix A−1 calculated if A is known ? (Don’t say that you use a computer package like Maple. What happens if e1 = e1 . ˜ ˜ ˜ If we represent each vector of the new basis e1 .12) is the inverse matrix for the direct transition matrix S. i. ˜ ˜ ˜ (2) e1 = e3 .14 CHAPTER I.11) Denote by T the transition matrix constructed on the base of (5. 3 S3 (5. It is called the inverse transition matrix when compared to the direct transition matrix S: − − − → e ˜ ˜ (e1 .12) Theorem 5. and the third columns in matrix (5. ˜ ˜ ˜ Exercise 5.

14) ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ Exercise 5. or lower. e3 .§ 6. Suppose you have a rectangular 5 × 3 matrix A and another rectangular matrix B which is 4 × 5. e2 . But their coordinates. (6. and suppose that C = A B.7.14) prove that S = S S and T = T T . What happens to vectors when we change the basis ? The answer to this question is very simple. Let’s study how they change. e3 ). e2 . e2 . this process is described by transition formula (5.6. How is it usually calculated ? Can you calculate det(A−1 ) if det A is already known ? Exercise 5. WHAT HAPPENS TO VECTORS WHEN WE CHANGE THE BASIS ? 15 Exercise 5. And they change if we change the basis. e2 .11): 3 ei = j=1 ˜ Tij ej . e2 . e2 . e3 ) −− −− (˜1 .6). e2 . And let’s consider the transition matrices relating them: − → e ˜ ˜ − − − − → ˜ ˜ ˜ (e1 . they depend on our choice of basis. ←− − − − ˜ ˜ ˜ − T ˜ T S ˜ S (5. ˜ ˜ ˜ Let’s consider three bases: basis one e1 . Suppose that A and B are two rectangular matrices. and basis ˜ ˜ ˜ ˜ ˜ ˜ three e1 . Rewrite this formula for the case when the components of B are denoted by B pq . e3 ) ← − − ˜ ˜ ˜ −− ˜ ˜ ˜ ˜ ˜ T (5. Suppose we have some vector x expanded in the basis e1 .4.13) ˜ ˜ ˜ ˜ Denote by S and T transition matrices relating basis one with basis three in (5.5. . e3 . −− (e1 .1) ˜ ˜ ˜ Then we keep vector x and change the basis e1 .13): ˜ ˜ S − − → (e1 . or mixed) you would use for components of C in the last case (see rules 5. e3 to another basis e1 . e3 . e3 : 3 x = x 1 e1 + x2 e2 + x3 e3 = i=1 xi e i . e2 .2 of Einstein’s tensorial notation). Exercise 5. Give some examples of matrix multiplication that are consistent with Einstein’s tensorial notation and those that are not (please. Remember the formula for the components in matrix C if the components of A and B are known (they are denoted by Aij and Bpq ). § 6. e3 ). Remember what is the determinant of a matrix. e2 . basis two e1 . For matrices S and T in (5. Really nothing ! Vectors do not need a basis for their being. Which indices (upper. e2 . do not use examples that are already considered in exercise 5.8. e3 . As we already learned.1.1 and 5. e2 . Apply this result for proving theorem 5. e3 ) ←−− (e1 . What is matrix multiplication ? Remember how it is deﬁned. Which of these two products A B or B A you can calculate ? Exercise 5.

Derive formula (6. ˜ (6. Exercise 6. 2 2 2 x 2 = T 1 x1 + T 2 x2 + T 3 x3 .2) uses the inverse transition matrix T . It’s funny. ˜ 3 3 3 3 x = T 1 x1 + T 2 x2 + T 3 x3 .2) can be written in matrix form as well: x1 x2 .2. Let’s substitute this formula into (6. Note that the direct transformation formula (6. Like (5. (6.7). it can be written in expanded form: 1 1 1 1 2 1 3 ˜ x = T1 x + T2 x + T3 x . . but it’s really so. Thus we have calculated the expansion of vector x in the new basis and have derived the formula relating its new coordinates to its initial ones: 3 xj = ˜ i=1 Tij xi .5) uses direct transition matrix S.1) for ei : 3 3 3 3 3 3 x= i=1 3 x i j=1 3 Tij Tij x i ej = i=1 j=1 3 j x i Tij ˜ ej = j=1 i=1 j ˜ xi Tij ej = 3 = j=1 i=1 ˜ ej = j=1 x ej .7).3) and (6.5) directly from (6.7).4) write (6.2) using the concept of the inverse matrix S = T −1 . and the inverse transformation formula (6.3) And the transformation formula (6.5) Exercise 6.2) can be inverted.16 CHAPTER I. PRELIMINARY INFORMATION.5) using formula (5. Exercise 6. By analogy with (6.1.4) Like (5. ˜ x1 ˜ x2 ˜ x ˜ 3 1 T1 2 = T1 3 T1 1 T2 2 T2 3 T2 1 T3 2 T3 3 T3 (6.2) This formula is called a transformation formula.3. x 3 (6. formula (6. By analogy with the above calculations derive the inverse transformation formula (6. Here is the inverse transformation formula expressing the initial coordinates of vector x through its new coordinates: 3 xj = i=1 j Si x i . where x = ˜ ˜ ˜ i=1 Tij xi .5) in expanded form and in matrix form. or direct transformation formula.

Coordinates of vectors (and only coordinates of vectors) will obey transformation rules (6.2) and (6. However. we cannot visually observe vectors (this is really so for some kinds of them. What is the novelty about the vectors that we learned knowing transformation formula for their coordinates ? Vectors are too common. too well-known things for one to expect that there are some novelties about them. So transformation formulas (6. and even if they have exactly three coordinates (like vectors have). their coordinates behave diﬀerently under a change of basis.2) and (6.5) work like detectors.5) under a change of basis. Other objects usually have a diﬀerent number of numeric parameters related to the basis.2) and (6. which are called tensors. Furthermore. the novelty is that the method of their treatment can be generalized and then applied to less customary objects. but suppose we can measure their coordinates in any basis we choose for this purpose. . 17 § 7. WHAT IS THE NOVELTY ABOUT THE VECTORS . like a sieve for separating vectors from non-vectors. see section 1). we shall consider only a part of the set of such objects.§ 7. Suppose. and what kind of geometrical and/or physical objects of a non-vectorial nature could exist — these are questions for a separate discussion. . . What are here non-vectors. What then do we know about vectors ? And how can we tell them from other (non-vectorial) objects ? The answer is in formulas (6.5).

In previous 7 sections we learned the following important fact about vectors: a vector is a physical object in each basis of our three-dimensional Euclidean space E represented by three numbers such that these numbers obey certain transformation rules when we change the basis. and suppose that these numbers obey some certain transformation rules when we change the basis.1) (8. ˜ Here S and T are the same transition matrices as in case of the vectors in (6. a2 . Let’s prescribe the following transformation rules to a1 . in mathematics we have another option. we intentionally used lower indices when denoting them by a1 . e2 . A geometric object a in each basis represented by a triple of coordinates a1 . Definition 8. By analogy with vectors we shall call them coordinates. These certain transformation rules are represented by formulas (6. Is it possible ? One can try to ﬁnd such an object in nature. e3 be two bases and let e1 . Compare exercise 8.1).1) and (8. Now suppose that we have some other physical object that is represented by three numbers in each basis. e2 . However.2) is derived from (8.1. a3 that three numbers which represent this object in the basis e1 . e3 : 3 aj = ˜ i=1 3 i Sj a i . e2 . then study its properties. § 8. e3 to ˜ ˜ ˜ e1 . Exercise 8. Looking at the above considerations one can think that we arbitrarily chose the transformation formula (8. Let ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ e1 . e3 .CHAPTER II TENSORS IN CARTESIAN COORDINATES.2) and (6. Using the concept of the inverse matrix T = S −1 derive formula (8. But in contrast to vectors.2) under a change of basis is called a covector.1). a2 . a3 and such that its coordinates obey transformation rules (8. a3 . The choice of the transformation formula should be self-consistent in the following sense. Let’s denote our hypothetical object by a. e3 and e1 . a2 . and denote by a1 .1 and exercise 6.2) from formula (8. e2 . Note that (8.2) and (6. We can construct such an object mentally. Covectors.1) is suﬃcient. e3 be the third basis in the .3. e2 . but these rules are diﬀerent from (6.1).1. formula (8. and ﬁnally look if it is represented somehow in nature. However.5).5). this is not so. a3 when we change e1 . (8.2) aj = i=1 Tji ai .2) and (6.5). a2 . e2 .

e2 .1) The sum (9. We can pass from basis one to basis three directly. This number is called the scalar product of the vector x and the covector a.3. x3 for vector x. Using (5. We use angular brackets for this scalar product in order to distinguish it from the scalar product of two vectors in E. Consider two bases e1 . ˜ ˜ ˜ Exercise 9.2 in section 5 above. x the following sum: 3 a. x2 . What about the physical reality of covectors ? Later on we shall see that covectors do exist in nature.1) we used the coordinates of vector x and of covector a. Exercise 8. the value of sum (9. It is a number depending on the vector x and on the covector a. SCALAR PRODUCT OF VECTOR AND COVECTOR. . see the right arrows in (5. 19 space. Such numeric quantities that do not depend on the choice of basis are called scalars or true scalars. However. we shall see that some well-known physical objects we thought to be vectors are of covectorial nature rather than vectorial. both of them have three coordinates: x1 .14). e3 . Exercise 8. In both cases the ultimate result for the coordinates of a covector in basis three should be the same: this is the self-consistence requirement. Upon choosing some basis e1 . x by means of sum (9. (8. x = i=1 a i xi . and (8.2). (6. e2 . Replace S by T in (8.1) and (8. basis two and basis three for short. § 9. e2 .2) Thus. and consider the coordinates of vector x and covector a in both of them.§ 9.1) is written in agreement with Einstein’s tensorial notation. and relying on the results of exercise 5.1. Let’s denote by a.2) prove the equality 3 3 a i xi = i=1 i=1 a i xi .14).8 prove that formulas (8. e3 and e1 .1).2.13). which is also known as the dot product.1) and showing that the scalar product a. And moreover. Deﬁning the scalar product a. (9.1) does not depend on any basis. Relying on transformation rules (6.2). see rule 5. you are proving the self-consistence of formula (9. e3 .2) yield a self-consistent way of deﬁning the covector. see the right arrow in (5. but not on the way that they were calculated. a2 . which are basis-dependent. It means that coordinates of a geometric object should depend on the basis. Suppose we have a vector x and a covector a. Let’s call them basis one.5). ˜ ˜ (9. x given by this formula is a true scalar quantity.13) and (5. and a1 .1) and T by S in (8. Or we can use basis two as an intermediate basis. a3 for covector a. They are the nearest relatives of vectors. Show that the resulting formulas are not self-consistent. Scalar product of vector and covector.

x . x + b.20 CHAPTER II. y . ˜ F = S F T. Remember our discussion about functions with non-numeric arguments in section 2.5 and 5.1. Let α be a real number. p=1 q=1 3 3 i ˜p Sp Tjq Fq . (4) a. (3) a.3 (see in section 5). ˜p If we write matrices Fji and Fq according to the rule 5.1. p=1 q=1 ˜ Fji = Fji = (10. Important note.2). x .2. Using the concept of the inverse matrix T = S −1 prove that formula (10. Exercise 9. Components of this matrix play the same role as coordinates in the case of vectors or covectors. A geometric object F in each basis represented by some square matrix Fji and such that components of its matrix Fji obey transformation rules (10.3. and let x and y be two vectors. Let’s prescribe the following transformation rules to Fji : 3 3 i q p Tp S j F q .2.2) Exercise 10. x). e2 . § 10. Exercise 9. Explain why the scalar product a.1) (10. a is incorrect in its right hand side since the ﬁrst argument of scalar product (9.3) Exercise 10.2) can be written as two matrix equalities: ˜ F = T F S. x = x. x + a. x = α a.2) under a change of basis is called a linear operator. let a and b be two covectors. Therefore.1) and (10. x + y = a. In this capacity. In a similar way. it is represented by a square 3 × 3 matrix Fji of real numbers. the formula a. The scalar product a. Definition 10. it can be denoted as f (a. x . Moreover. x is sometimes called the bilinear function of vectorial argument x and covectorial argument a. TENSORS IN CARTESIAN COORDINATES. α x = α a. e3 .1): (1) a + b. x is not symmetric.1). then (10. the second argument should be a vector.1) and (10.1) and (10. let’s denote one of such objects by F. Linear operators. we never can swap them. In this section we consider more complicated geometric objects. (10. Prove the following properties of the scalar product (9. (2) α a.3) from (10.6) and derive (10. Remember matrix multiplication (we already considered it in exercises 5. In each basis e1 .2) is derived from formula (10. x = a.1) by deﬁnition should be a covector. For the sake of certainty. .

i = 2.5) ﬁnd that the y 1 .8). e3 and take some vector x with coordinates x1 . x3 in the same basis. (10. y 2 ..6) and comparing it with (6. in which x and y are represented as columns: 1 1 1 x1 F1 F2 F3 y1 y2 y 3 = 2 F1 3 F1 2 F2 3 F2 2 F3 3 F3 x2 . They are denoted by y 1 .2 prove the self-consistence of the above deﬁnition of a linear operator.5) represent ˜ ˜ ˜ the same vector. Some people like to write (10. y 2 . e2 . Therefore. x 3 (10. y 3 are related as follows: ˜ ˜ ˜ 3 3 yj = ˜ i=1 Tij y i . y 3 calculated by formulas (10.1) and (10. Here the action of operator F upon vector x is designated like a kind of multiplication. y 2 . 2004. yj = i=1 j Si y i .4) and (10.8) Formula (10.3.4) is a free index. e3 and do the same things. e2 .2) and (6.7). y 2 . y 3 and the other three numbers y 1 . ˜ (10. while exercise 10.4) deﬁnes the vectorial object y. Now ˜ ˜ ˜ suppose that we pass to another basis e1 .7) without parentheses: y = F x. Let’s take a linear operator F represented by matrix Fji in some basis e1 . y 2 .4. 21 Exercise 10. but in diﬀerent bases. Looking at (10.4). There is also a matrix representation of formula (10. we write y = F(x) (10. or i = 3.5) Exercise 10.6) Exercise 10.8) is a more algebraistic form of formula (10. Thus formula (10.5 assures the correctness of this deﬁnition. y 3 and y 1 .7) and say that y is obtained by applying linear operator F to vector x.§ 10. (10. (10. By analogy with exercise 8. LINEAR OPERATORS.9) CopyRight c Sharipov R. it can deliberately take any one of three values: i = 1.4) Index i in the sum (10. x2 .A. For each speciﬁc value of i we get some speciﬁc value of the sum (10. y 3 according to (10. y 3 given by formula ˜ ˜ ˜ 3 ˜ yp = q=1 ˜p ˜ Fq x q .5. As a result we have vector y determined by a linear operator F and by vector x.4). y 2 . . Using Fji and xj we can construct the following sum: 3 yi = j=1 Fji xj . Relying upon (10.2) prove that the three numbers y 1 . As a result we get other three values y 1 .

then its matrix is introduced by formula (10.10. Prove the following properties of a linear operator (10. (10. We can apply H to vector x and then we can apply F to vector H(x). Let α be some real number and let x and y be two vectors. Square matrices have various attributes: eigenvalues.9) from (10. and F(e3 ).11) Formula (10. a characteristic polynomial. As a result we get F ◦ H(x) = F(H(x)). TENSORS IN CARTESIAN COORDINATES. (10.11. Exercise 10.8).13) and (5. Fj Exercise 10.12) Exercise 10. Then we can deﬁne the determinant of linear operator F as the number equal to the determinant of its matrix in any one arbitrarily chosen basis e1 .10) means that despite the fact that in various bases linear operator F is represented by various matrices.10) Also remember the formula for det(A−1 ). Substitute the unit matrix for Fji into (10. a rank (maybe you remember some others). Exercise 10. Substitute this vector x into (10. (10.1) and verify that ˜ i is also a unit matrix in this case. Exercise 10. Apply these two formulas to (10. Are they really similar to the properties of multiplication ? Exercise 10. Write these equalities in the more algebraistic style introduced by (10.14).12) is an example of such attribute.22 CHAPTER II. Derive (10. and the small circle sign denotes composition.14) Here F ◦ H is new linear operator introduced by formula (10. If we study these attributes for the matrix of a linear operator. Suppose we have two linear operators F and H. (2) F(α x) = α F(x). e3 in the space. Interpret this fact. which of them can be raised one level up and considered as basis-independent attributes of the linear operator itself ? Determinant (10. Let x = ei for some basis e1 .6. eigenvectors.7): (1) F(x + y) = F(x) + F(y).4) derive the following formula: 3 F(ei ) = j=1 Fij ej .7) and by means of (10.7). e3 : det F = det F. the determinants of all these matrices are equal to each other. e2 . It is called a composite operator. F(e2 ). e2 .3) and derive ˜ det F = det F .7. then we can reconstruct the whole matrix of operator F by means of formula (10. Remember that for the product of two matrices det(A B) = det A det B.13).13). Discuss the similarities and diﬀerences of these two formulas.13) Compare (10.8. (10. (10. Explain why if we know three vectors F(e1 ).9 (for deep thinking). The fact is that in some books the linear operator is determined ﬁrst. .4).

A geometric object a in each basis represented by some square matrix aij and such that components of its matrix aij obey transformation rules (11. y) is a number determined by vectors x and y irrespective of the choice of basis.2) Transformation rules (11.2). covectors. Now we consider another one class of tensorial objects. and linear operators are all examples of tensors (though we have no deﬁnition of tensors yet). Exercise 10. ˜ (11.13. § 11. p=1 q=1 3 3 aij = ˜ (11. Find the matrix of this operator if the matrix of F is known. Find the matrix of composite operator F ◦ H if the matrices for F and H in the basis e1 .2) from (11. e2 . Exercise 10. i. Vectors.4) as a scalar function of two vectorial arguments. Under a change of basis these numbers are transformed as follows: 3 3 p q Si Sj apq . For the sake of clarity. Exercise 11.1. e2 .12.1). Encyclopedia) and deﬁne the identity operator id. Hence we can treat (11. . let’s denote by a one of such objects. prove the equality 3 3 3 3 aij x y = i=1 j=1 p=1 q=1 i j apq xp y q . e3 this object is represented by some square 3 × 3 matrix aij of real numbers.§ 11.1. Prove that the sum in the right hand side of formula (11. In each basis e1 .1) and (11.2.2) can be written in matrix form: a = S a S.1) and (11.1) aij = p=1 q=1 Tip Tjq apq . Bilinear and quadratic forms. 23 Exercise 10.3) we denote the transposed matrices for S and T respectively. Remember the deﬁnition of the inverse map in mathematics and deﬁne inverse operator F−1 for linear operator F.4) does not depend on the basis. e3 are known. ˜ Here by S and T a = T a T. We use their coordinates and the components of bilinear form a in order to write the following sum: 3 3 a(x. Remember the deﬁnition of the identity map in mathematics (see on-line Math.1) and (11. Definition 11. Let’s consider two arbitrary vectors x and y.4) Exercise 11. e.2) under a change of basis is called a bilinear form. Derive (11. BILINEAR AND QUADRATIC FORMS.14. ˜ ˜ ˜ This equality means that a(x. Find the matrix of this operator. ˜ (11.3) from (11. y) = i=1 j=1 aij xi y j . then (11. (11.

Prove the following identity for a symmetric bilinear form: a(x. Thus. And conversely. z) = a(x. Due to these properties function (10. It recovers bilinear function a from quadratic function f produced in (11. (11.24 CHAPTER II. x) = c(x. y+z) = a(x.8) Without a loss of generality a bilinear function a in (11. As a result we get another bilinear function b(x. Exercise 11. Note that scalar product (9. y) = α a(x.4) is called a bilinear function or a bilinear form. we can produce symmetric bilinear function a(x. y) = . comparing (11. x).9) we derive f (x) = a(x. Due to this formula. x).10) is called the recovery formula. However. x) c(x. y)+a(x. b=a . 2 (11.2. . the second is a vector.3. 2 (11. y) = a(y.4). (1) a(x+y. α y) = α a(x. we cannot swap them. Indeed. if a is not symmetric. y) by substituting y = x: f (x) = a(x. TENSORS IN CARTESIAN COORDINATES.10) Formula (11. y.8) due to (11. Prove the following properties of function (11. y) . y).1. (11. y) = f (x + y) − f (x) − f (y) . (11. (2) a(α x.8) and (11.5) Definition 11. x) + a(x. z)+a(y. and z be three vectors. The matrices of a and b are related to each other as follows: bij = aji . y) = a(x + y.9) 2 and then from (11. x) = a(x. y) + a(y.8) with a replaced by c. z). y) = a(y. x).4) we can swap arguments. In bilinear form (11. z). x + y) − a(x.7) Definition 11.6) (11. Exercise 11. It is linear with respect to each of its two arguments.8) can be assumed symmetric. Let α be some real number.3.7) we get that a is produced by f : a(x. there is a crucial diﬀerence between (9. in referring to a quadratic form we always imply some symmetric bilinear form like the geometric tensorial object introduced by deﬁnition 11. y). A bilinear form is called symmetric if a(x.1) are of a diﬀerent nature: the ﬁrst argument is a covector. The arguments of scalar product (9. and let x.1) is also a bilinear function of its arguments. (4) a(x. A quadratic form is a scalar function of one vectorial argument f (x) produced from some bilinear function a(x. 2 This equality is the same as (11.8). Therefore.4.1) and (11.4): (3) a(x. each quadratic function f is produced by some symmetric bilinear function a. x) − a(y. x).

. X.. h index.. s). i dimensional array Xj1 . or of valency (r. ks each upper index is served by direct transition matrix S and produces one summation in (12. ir remain upper and lower indices j1 . Indices i1 . ks are summation indices. In right hand side of the equality (12.. therefore bilinear forms are called twice-covariant tensors. General deﬁnition of tensors. s 1 1 r (12. . . Other indices h1 .. .. .. or sometimes the term valency is used. Sjss Xk1 . their components are numerated by one upper and one lower index.. . . ksr . Linear operators are tensors of mixed type. jr = s (12.1 and 5. the way the components of each particular tensor behave under a change of basis... once in S-s or T -s and once in components of array Xk1 . A geometric object X in each basis represented by (r + s)i1 .. . . Thr Sj11 ..1). . s). GENERAL DEFINITION OF TENSORS. . A tensor of type (r... .. i ˜ .. In a matrix of bilinear form we use two lower indices. js through Xk1 . .. ksr .. Shr Tjk1 . . so it is suﬃcient to remember only one of them. .2) Definition 12. i Xj1 . and bilinear forms are examples of tensors. h Sh1 . s). .. s) is called an r-times contravariant and an s-times covariant tensor.. i.. . . . = . e. 25 § 12. . .. hr k1 . e... .. i... upper indices i1 . i Xj1 . hr ˜ When expressing Xj1 .. ir h1 . ksr . . i1 .. . ..... Let it be the formula (12.1): 3 . .. hα . .. s). α (12.. ks h1 . . h k i i k Th1 . which is called the contravariant index.1. They are geometric objects that are represented numerically when some basis in the space is chosen.. h1 ... . .1). jr = s . jr of real numbers and such that the components of this s array obey the transformation rules (12. One should strictly follow the rules 5. hr and k1 . The coordinates of a covector are numerated by one lower index...... Then we have a tensor of the type (r.1) ˜ i1 . js remain lower in right hand side of the equality (12. hα =1 . linear operators and quadratic forms are represented by two-dimensional arrays. .2 from section 5. Tjks Xk1 ..2) under a change of basis is called tensor of type (r... js are free indices. covectors. . . . . . . The coordinates of a vector are numerated by one upper index.. hr k1 .. their position does matter. Vectors. This is terminology. which is called the covariant index.. iα . . Formula (12. . or of valency (r.1) and (12. formula (12.. 1 r 3 i i ˜ h1 . ....1) they are distributed in S-s and T -s. h1 .. any tensor is represented by a multidimensional array with a deﬁnite number of upper indices and a deﬁnite number of lower indices. In the general case.. Let’s denote these numbers by r and s.3) ... . . linear operators. . The number of indices and their positions determine the transformation rules... Shα .2) is derived from (12.. . ..§ 12. . ks 3 3 . X. Though huge.. they enter the right hand side of (12. .. ir and j1 ... now let’s proceed to the exact deﬁnition.. It is based on the following general transformation formulas: 3 i1 . This numeric representation is speciﬁc to each of them: vectors and covectors are represented by one-dimensional arrays.. . . .. Apart from the number of indices.1) is easy to remember.1).1) pairwise: once as an upper index and once as a lower ˜ h1 . each having only one entry and each keeping its position.

(10.2) . covectors.. In a similar way. (13.. Remember the following properties of the scalar product (13. x) 0 and (x.5).. .. T jα . α y) = α (x. . y). Tensor b is called a dual bilinear form for a.26 CHAPTER II.. and bilinear forms when they are considered as tensors. x). y) = (y.2. However there are some tensors of natural origin. 0). . y) = |x| |y| cos(ϕ). each lower index is served by inverse transition matrix T and also produces one summation in formula (12. Verify that formulas (6. Hence it determines tensor b of valency (2.3) and (12. where ϕ is the angle between vectors x and y. and (11. . (6) (x.1) are quite similar to those for quadratic forms. z). (3) (x. TENSORS IN CARTESIAN COORDINATES.. § 13..1) is written. . z) = (x... (8.3.. (4) (x.4) are the same as (12. Therefore for any two vectors x and y we can deﬁne their natural scalar product (or dot product): (x. Exercise 13. The covectors.1): 3 . jα .1... Let’s remember that we live in a space with measure.4) Formulas (12..1).2). see exercise 11. y). X.. (12. Let aij be the matrix of some bilinear form a..1) and used to highlight how (12.. Dot product and metric tensor.. (13.2. Note that the ﬁrst four properties of the scalar product (13. . linear operators. y= j=1 xj e j .. This means that we have the following expressions for them: 3 3 x= i=1 xi e i . y + z) = (x. .. z) + (y.2) are special cases of formula (12. We can measure distance between points (hence we can measure length of vectors) and we can measure angles between two directions in our space. kα =1 . y) + (x. X.. kα ˜ . e3 . linear operators.1. Prove that matrix bij under a change of basis transforms like matrix of twice-contravariant tensor. see Vector Lessons on the Web. kα . . Further we shall consider more examples showing that many well-known objects undergo the deﬁnition 12. Exercise 12.1) (5) (x. y) = α (x. z).2). . Let’s denote by bij components of inverse matrix for aij . So tensors are deﬁned.. Exercise 13...1. This is not an occasional coincidence. (2) (α x. What are the valencies of vectors. e2 .1): (1) (x + y. These properties are usually considered in courses on analytic geometry or vector algebra. = . Let’s consider two arbitrary vectors x and y expanded in some basis e1 . . and bilinear forms that we considered above were artiﬁcially constructed tensors. Exercise 12. x) = 0 implies x = 0..

e3 and repeating this operation we get ˜ .2) .5) is symmetric due to property (5) in exercise 13. The most simple of them are multiplication by number and addition. Evidences for this fact are below. i Yji11.2) under a change of base. jr . then in some base i1 . but the whole geometry of our space. Denote gij = (ei . s Then by multiplying all the components of this array by α we get another array ...4) ˜ ˜ ˜ Compare (13.5) deﬁnes tensor of type (0.. Tensor operations are used to produce new tensors from those we already have.6) The quadratic form corresponding to (13. It determines a tensor of type (2.. x) = |x|2 ..1 derive the following formula for the scalar product of x and y: 3 3 (x.1) and (11. y). r ˜ i1 ..5).4) with formula (11. s) and a real number α. e2 ..3) Exercise 13... Matrix (13. Now.4) and keeping in mind the tensorial nature of matrix (13. ej ) (13. e3 . Thus you prove that the Gram matrix gij = (ei . The inverse matrix for (13. jr . (13.11) prove that ˜ e ˜ matrices gij and gpq are components of a geometric object obeying transformation ˜ rules (11. It describes not only the scalar product in form of (13..3.§ 14. § 14. y) = i=1 j=1 gij xi y j . Multiplication by numbers and addition.. let’s denote it Xj1 . ej ) and rewrite formula (13...5) is denoted by the same symbol g but with upper indices: g ij . i Yji11. e2 .. (13. jis = α Xj1 .1) and using properties (1)– (4) listed in exercise 13. denote gpq = (˜p . i e1 . y) = i=1 j=1 (ei . 0).6) is very simple: f (x) = g(x.4) and (11..4). jis = α Xj1 . e3 we have the array of components of tensor X.7) and (5...2 for more details).. it is called the metric tensor. jr . e2 . comparing (13. Consider some other basis e1 . eq ) and by means of transition formulas (5.3) as 3 3 (x. 27 Substitute (13.. s (14. (13. we conclude that the scalar product is a symmetric bilinear form: (x..4). If we have some tensor X of type (r.1) ˜ ˜ ˜ Choosing another base e1 . y) = g(x.1. This is very important tensor. ej ) xi y j .2) into (13. r i1 . 2). MULTIPLICATION BY NUMBERS AND ADDITION. s (14. this tensor is called dual metric tensor (see exercise 12.

Exercise 14. i Zj1 . then we can write: r+1 r+p i1 . r ˜ .. jm−1ρ jk . Exercise 15. if we have many upper an many lower indices in tensor X.. § 16.. k−1 i ... Thus.1) applied in all bases produces new tensor Y = α X from initial tensor X.. They are numerated in some convenient way. The contraction is over. Other r upper indices and s lower indices are free..1. say as in formula (16.28 CHAPTER II.. By analogy with exercise 14. jis and Yji11. The tensor product is deﬁned by a more tricky formula. In doing this you prove that formula (14.. jr Yjs+1 ... ir ˜ in the same way as arrays Xj1 ..3..1).. ir i1 . Contraction is an operation that decreases the number of indices. jis = Zj1 .. . Don’t mix the tensor product and the cross product.. js and Xj1 . s) we produce a third tensor Z of the same type (r. 2004. In exercise 14. i ...1 prove the consistence of formula (14. TENSORS IN CARTESIAN COORDINATES.. i 1 Zj1 .. e. Contraction. Then we perform summation with respect to index ρ.1) deﬁnes the multiplication of tensors by numbers..... s + q).. jsr . Suppose we have tensor X of the type (r + 1.. q). Give an example of two tensors such that X ⊗ Y = Y ⊗ X. By analogy with exercise 14. jr . Prove that arrays Yji11.2. We choose the m-th upper index and replace it by the summation index ρ. i i1 .1 prove the consistence of formula (15..1.1 you prove its consistence.. (14. according to transformation formulas (12.3).. jr + Yji11.. s s+q i .1) and (12... CopyRight c Sharipov R. we can perform various types of contractions to this tensor.. Exercise 14. It’s natural to denote Z = X + Y. In the same way we replace the k-th lower index by ρ. i. js . Suppose we have tensor X of type (r... They are diﬀerent.. s) by means of formula (14.1). What happens if we multiply tensor X by the number zero and by the number minus one ? What would you call the resulting tensors ? § 15.. s) by means of the following formula: n i1 .. r Xj1 .2. s) and tensor Y of type (p. Usually the tensor Z = X ⊗ Y has more indices than X and Y. This operation is called a contraction with respect to m-th upper and k-th lower indices. r Exercise 14.. Tensor product..3) s s Having two tensors X and Y both of type (r..1) Formula (15. Formula (14. jr = s ρ=1 1 m Xj1 . i (15...2).1) produces new tensor Z of the type (r + p. . the tensor product increases the number of indices.. jr+p = Xj1 ... It is called the tensor product of X and Y and denoted Z = X ⊗ Y. s + 1).. jis are related to each other i1 ..1) What we do ? Tensor X has at least one upper index and at least one lower index. js+q .. i (16. i ρ i . Then we can produce tensor Z of type (r. Exercise 15. i i .3).. As we have seen above. The next formula deﬁnes the addition of tensors: i1 ....A.

. p q . X. k .. if we interpret the Kronecker symbol as a tensor..3) comprises two tensorial operations: the tensor product and contraction..2..§ 18. Suppose that X is some tensor of type (r.... Prove the consistence of formula (16.4). For this purpose we replace them both by s and perform the summation: 3 .. (11.. Y.... s=1 .4).... p .. It is determined as follows: i δj = 1 0 for i = j. . for i = j.. In the next step let’s contract (17..1) and interpret this formula as the contraction of the tensor product a ⊗ x...2)....2) are not invariant under a change of basis. Find similar interpretations for (10.1) is invariant under a change of basis.. Therefore. Raising and lowering indices..3). SOME SPECIAL TENSORS AND SOME USEFUL FORMULAS. . s .1. (17. (17. Then let’s consider the tensor product Y = g ⊗ X: ... .. k .1.... ..1).. . Look at formula (9... This is a two-dimensional array representing the unit matrix... s gps X... § 17. . (17.. Some special tensors and some useful formulas.. k . . (18.2) is called the index raising procedure.. for i = j.. . § 18. . X. It is invertible. The inverse operation is called the index lowering procedure: 3 . Exercise 18. s).. we denoted them by dots. p X.. and (13.... What tensorial object do you get as a result of this operation ? ... g ps X... Prove that deﬁnition (18..1) by means of (17...2. = g pq X.. (18. . Let’s choose its α-th lower index: . .2) This operation (17.. Exercise 16.... 29 Exercise 16... . ... the index lowering procedure (17. Lower index i of tensor (18.3) Like (17..1) Here g is the dual metric tensor with the components g pq (see section 13 above). Show that both deﬁnitions in (18..2) Exercise 18.1) with respect to the pair of indices k and q.4). Kronecker symbol is a well known object. .. The symbols used for the other indices are of no importance.. .1) We can determine two other versions of Kronecker symbol: δ ij = δij = 1 0 for i = j. = s=1 .. = ..

i.6) Then we can apply index raising procedure (17. −1. Exercise 18.1). (18.3. The ﬁrst of them ωijk = det(gij ) ijk (18. k. Then using (18. Prove that the vector a with components (18. e.3) is not a tensor. q. Another well known object is the Levi-Civita symbol.2) and produce vector a: 3 3 3 ar = i=1 j=1 k=1 g ri ωijk xj y k . However. raise index J in (18.5) Let’s take two vectors x and y.3) The Levi-Civita symbol (18. we can produce two tensors by means of Levi-Civita symbol. there are 0. (18. TENSORS IN CARTESIAN COORDINATES. Exercise 18.4) we can produce covector a: 3 3 ai = j=1 k=1 ωijk xj y k . a = [x.7) Formula (18. y]. This is a three- jkq = jkq (18. Likewise. Another one is the dual volume tensor: ω ijk = det(g ij ) ijk .4) is known as the volume tensor. at least two equal num bers.30 CHAPTER II. tion of numbers (1 2 3). if (j k q) is odd permuta tion of numbers (1 2 3). if (j k q) is even permuta= 1. dimensional array determined by the following formula: if among j.4.7) is known as formula for the vectorial product (cross product) in skew-angular basis. .7) coincides with cross product of vectors x and y. (18.

and (19. e2 ..1) is a tensor-valued function with argument P being a point of three-dimensional Euclidean space E.CHAPTER III TENSOR FIELDS. Dealing with point-valued arguments is not so convenient. we have a numeric function with a point-valued argument. .1). .. the tensor ﬁeld (19. (19. . Now suppose we want to bind our tensor to some point in space. . A functional array (19..2). In each coordinate system they are represented by functional arrays. by arrays of functions (see (19. Doing so we can ﬁll our space with tensors. Hence..1) Usually the valencies of all tensors composing the tensor ﬁeld are the same. Tensor ﬁelds in Cartesian coordinates. s). tensor ﬁelds are related not to bases. Indeed. x3 : i1 . i any tensor of our tensor ﬁeld as an array Xj1 .2) Thus. then we get Cartesian coordinate system in space and hence can represent − → P by its radius-vector rP = OP and by its coordinates x1 .. but to whole coordinate systems (including the origin).. one per each point. for example..2). Then if we choose some basis e1 . . s s (19.. i Xj1 . e. i i1 . ir . In order to mark a point P to which our particular tensor is bound we shall write P as an argument: X = X(P ). jr = Xj1 .3)). x2 .2) is the basis representation for (19. js in (19. e3 ..... s s (19. jr (P ). What happens when we change the coordinate system ? Dealing with (19. . jr = Xj1 . jr (x1 . i Xj1 .1. if we want to calculate derivatives. For each ﬁxed set of numeric values of indices i1 .. The tensors that we deﬁned in section 12 are free tensors. . we ﬁx some point O as an origin. DIFFERENTIATION OF TENSORS. we . j1 . In this case we say that we have a tensor ﬁeld.1). In contrast to free tensors..3) is a coordinate representation of a tensor ﬁeld (19. then another tensor to another point and so on.. Remember that we have already chosen a basis. i i1 . If. x3 ). i. . x2 . Let them all be of type (r.. Therefore. we can represent i1 . jr with r + s indices: s i1 . in addition. we need to replace P by something numeric.. while any basis is a triple of free vectors (not bound to any point). their components are arrays related to bases.. the tensors previously considered are also not bound to any point.. § 19.3) Conclusion 19..

and by e1 .7) THe couple of formulas (19. jr (x1 . x2 ..4) ˜ i1 . x3 ). h Th1 .8).. .. e3 the basis of the ﬁrst coordinate system. ˜ x2 = x2 (x1 . Sjss Xk1 . Shr Tjk1 . and another couple of formulas (19.. 1 r i i k k h1 . x ˜ ˜ i i ˜ h1 . x3 )... jr s in the new basis 3 3 . x2 . h1 ..2). ..32 CHAPTER III. ksr (P ). x2 . .7). Denote by rP and ˜P the r . . ksr (˜1 . ˜ (19. We consider two Cartesian coordinate systems.8) are written in abstract form.2): 3 3 .. ks i i k k h1 . x3 ). .. But now. . .. ksr (x1 . x2 . h x ˜ ˜ Sh1 . Change of Cartesian coordinate system.8).. x3 ) = s . respectively. hr k1 . x . They only indicate the functional dependence of new coordinates of the point P from old ones and vice versa. 1 r ˜ i1 .5) can be inverted by means of (12... .1) and (12.. ks i1 . DIFFERENTIATION OF TENSORS. . . h1 .. Remember that each Cartesian coordinate system is determined by some basis and some ﬁxed point (the origin). x )..6) as well: 1 x = x1 (x1 .2).. ˜ Note that formulas (19.7) and (19. Let the origins ˜ of the ﬁrst and second systems be at the points O and O. .. . i Xj1 . ks 3 using (12.5) Like (12. x3 )... i Xj1 .. Thr Sj11 . e2 ... x3 of the point P in right hand side of (19. x2 . we should have inverse formulas for (19. hr k1 . . e3 the basis of the second coordinate system (see Fig.5) and (19.. . Tjks Xk1 .8) 3 3 1 2 3 x = x (x .. . x2 .. x2 ...... . formula (19.. i1 .5) through new coordinates of the same point: 1 x ˜ ˜ x = x1 (˜1 . Thr Sj11 .. x3 ) = s (19. x ˜ ˜ 3 3 1 2 3 x = x (˜ ..3). jr (˜1 ... we need to recalculate the components of the array X j1 .6) x = x (˜ ..6) and (19. x3 ).. x . Sjss Xk1 . . . in the case of tensor ﬁelds play the same role as transformation formulas (12. 2 2 1 (19. x2 . jr in the basis by means s of (12. and we also need to express the old coordinates x1 . Now we shall specify them for the case when one Cartesian coordinate system is changed to another Cartesian coordinate system. apart from (19. 7 below)..6) and (19. s 1 1 r (19. i In the case of (19. . hr k1 . Denote by ˜ ˜ ˜ e1 . .1): 3 i1 .. x ).. .2) in the case of free tensors. Let P be some point in the space for whose coordinates we are going to derive the specializations of formulas (19.... § 20.. e2 . h Th1 . .. . i need only to recalculate the components of the array Xj1 . TENSOR FIELDS. x2 . jr (P ) = s (19. ... i x ˜ ˜ Xj1 .6).. x3 ). h1 .

Derive the following inverse formula for (20. Exercise 20.3) x ei .2) and (6. ˜ ˜ i=1 i ˜P = r Exercise 20. 2 2 2 x2 = S 1 x 1 + S 2 x 2 + S 3 x 3 + a 2 . Then rP = OP and − → ˜ ˜P = OP .1).5) ˜ Prove that ai in (20. 33 radius-vectors of this point in our two − → coordinate systems.§ 20. Formula (20. (20. CHANGE OF CARTESIAN COORDINATE SYSTEM. ˜ (20. (20.4) can be written in the following expanded form: 1 1 1 1 2 1 3 1 ˜ ˜ ˜ x = S1 x + S2 x + S3 x + a .2.1) − → ˜ Vector OO determines the origin shift from the old to the new coordinate system. We expand this vector in the basis e1 . i=1 (20.6) with (6.6) Compare (20.4): 3 xi = a i + ˜ ˜ j=1 Tji xj .2) Radius-vectors rP and ˜P are expanded r in the bases of their own coordinate systems: 3 rP = i=1 3 xi e i .2). Explain the diﬀerences in these two formulas. ˜ ˜ ˜ 3 3 1 3 2 3 3 ˜ ˜ ˜ x = S1 x + S2 x + S3 x + a 3 . ˜ (20.7) derive the following formula relating the coordinates of the point P in the two coordinate systems in Fig. Using (20. ai = − j=1 i Sj a j . 7: 3 xi = a i + j=1 i Sj x j . e3 : − → ˜ a = OO = 3 ai ei . and (5.5). (20. Explain the minus signs in these formulas. (20. (20. Hence r − → ˜ r rP = OO + ˜P .3).7) .5) are related to each other as follows: 3 3 ai = − ˜ j=1 Tji aj .5). e2 .4) Compare (20.4) with (6.1.4) and ai in (20. (20.

jr (x1 ..1. x3 and to the point P with coordinates x1 + h. g. its derivative with respect to a spacial variable... jr Xj1 .3) Partial derivatives (21. jr Xj1 ... i i1 . x2 . In a similar way we can expand (20. Let’s consider them to start... and consider the partial i1 .2) but in numerator of the fraction in the right hand side of (21. x1 . jr (t. x2 . x2 .. say it can be q. (20. this could be a time variable). r s = j j i1 . jr (x1 + h. The second type of derivatives are simpler to understand. ˜ ˜ In this section we consider two diﬀerent types of derivatives that are usually applied to tensor ﬁelds: diﬀerentiation with respect to spacial variables x1 .. Here we also can write i1 .1... Now let’s consider the spacial derivative of tensor ﬁeld X. (19.1) is a tensor since the fraction in right hand side is constructed by means of tensorial operations (14. DIFFERENTIATION OF TENSORS. i i1 .8) 3 3 1 3 2 3 3 3 x = T1 x + T2 x + T3 x + a . jr s . x3 ) − Xj1 . x2 .. taken as a whole. i. To which point should be attributed the diﬀerence of two such tensors ? This is not clear. upon choosing some Cartesian coordinate system. x2 .7) and (20. x3 . x2 . i derivative of Xj1 .34 CHAPTER III.2) in a diﬀerent way. x1 . § 21. x2 .8).7). Suppose we have tensor ﬁeld X of type (r. we can write i1 .7). This is the required specialization for (19. Passing to the limit h → 0 does not destroy the tensorial nature of this fraction since the transition matrices S and T in (19. Diﬀerentiation with respect to external parameters (like t in (21. s)... s) and depending on the additional parameter t (for instance.. x3 and diﬀerentiation with respect to external parameters other than x1 .5) and (19. Give a more detailed explanation of why the time derivative (21. x3 ) s s s = lim .7). h→0 ∂x1 h (21.6). Formulas (20..1)) is a tensorial operation producing new tensors from existing ones.5). x2 . x3 ) − Xj1 . Therefore.8) are used to accompany the main transformation formulas (19...2) we get the diﬀerence of two tensors bound to diﬀerent points of space: to the point P with coordinates x1 .. if they are present. i i1 ...1) represents a tensor of type (r. ∂xq (21.1) The left hand side of (21. x3 ) s s s = lim .8) are all time-independent.. jr (t + h.. form an (r + s + 1)-dimensional array .1) and (14. Exercise 21.2).3). Diﬀerentiation of tensor ﬁelds. (20. TENSOR FIELDS.. Then. jr with respect to the spacial variable xq : s . i ∂Xj1 . 2 2 2 ˜ x 2 = T 1 x1 + T 2 x2 + T 3 x3 + a 2 . i ∂Xj1 . ˜ (20... x3 . h→0 ∂t h (21.... i Yqi11 ... e. i ∂Xj1 .. i i1 . Let’s choose some additional symbol.. Conclusion 21. e. with respect to x1 .5): 1 1 1 1 2 1 3 ˜ ˜1 x = T1 x + T2 x + T3 x + a .. we should treat partial derivatives like (21... This is the required specialization for (19.

. In order to get the vector form of the gradient one should raise index q: 3 3 aq = i=1 g qi ai = i=1 g qi i ϕ. jr s ∂ xq ˜ (21. i with one extra dimension due to index q. Thus diﬀerentiation with respect to x1 .6) and (21..5) are valid only for Cartesian coordinate systems. 0). (22.. according to theorem 21. grad = . 2004. r s = j j i1 . .1 and the equality (21... Exercise 21.3) CopyRight c Sharipov R. .5) Warning 21. i Yqi11 . Prove theorem 21. Then ˜ ˜ ˜ in the new coordinate system consider the partial derivatives ˜ . x2 . For the sake of beauty and convenience this operation is denoted by the nabla sign: Y = X. x3 produces new tensors from already existing ones. ir q Xj1 . GRADIENT..8). i Yqi11 . x2 . Theorem 21. produces a tensor ﬁeld of type (0. r s = j j ˜ i1 . For any tensor ﬁeld X of type (r.... The tensorial nature of partial derivatives established by theorem 21.4) Simplifying the notations we also write: q = ∂ . js . x3 ) in coordinate form: ∂ϕ aq = q ϕ = . In curvilinear coordinates things are diﬀerent. i ∂ Xj1 . which is tensor ﬁeld of type (0. (22. DIVERGENCY.. § 22.3) with respect to spacial variables x1 . x3 via (20... Let’s consider the gradient. to functions ϕ = ϕ(P ) or ϕ = ϕ(x1 . e.7) and (20. This means that a = grad ϕ is a covector.....2. (22.1) we used a lower index q for aq .§ 22. Indeed. AND ROTOR ...A. x3 related to x1 . r s j j due to the following theorem concerning (21. s) partial derivatives (21. i.1.. s + 1).3).. (21. In index form this looks like .1.. Gradient. the nabla operator applied to a scalar ﬁeld. ∂xq (21. Usually the gradient operator is applied to scalar ﬁelds. We can apply it to extend the scope of classical operations of vector analysis. x2 .6) and derive relationships binding (21. x2 .2) in the form of a diﬀerential operator (without applying to ϕ): 3 q = i=1 g qi i. and rotor.3). x3 in any Cartesian coordinate system represent another tensor ﬁeld Y of the type (r.1. divergency. 1). 35 . . For this purpose consider another Cartesian coordinate system x1 .1 is a very useful feature.1. Theorem 21.1) ∂xq Note that in (22.2) Let’s write (22. Laplace and d’Alambert operators. We write it as a lower index in Yqi11 . x2 .

it is denoted by the triangle sign: = div grad. .. (22. . DIFFERENTIATION OF TENSORS. From (22.3)..8) is given by formula (18.7) is called the d’Alambert operator or wave operator.4) is produced by contraction (see section 16) from tensor q X i .4). Therefore we can generalize formula (22.6) (see my book [5]. The rotor operator is usually applied to a vector ﬁeld and produces another vector ﬁeld: Y = rot X. . but also to vector ﬁelds.3) and (22. = s=1 .8) The volume tensor ω in (22.... s X.. In this form the gradient operator (22. (22. (22.... In this case dual metric tensor is given by unit matrix (g ij = δ ij ) and components of q and q coincide by value. . . .36 CHAPTER III. it is free for download from http:/ /samizdat. Usually in physics we do not distinguish between the vectorial gradient q and the covectorial gradient q because we use orthonormal coordinates with ONB as a basis. Divergency is the second diﬀerential operation of vector analysis. (div X).. (22. (22.. covector ﬁelds and to any other tensor ﬁelds. .mines.. And ﬁnally. let’s consider the rotor operator or curl operator (the term “rotor” is derived from “rotation” so that “rotor” and “curl” have approximately the same meaning)..5) for Laplace operator we derive the following formula: 3 3 = i=1 j=1 g ij i j. (22. .5) The Laplace operator is deﬁned as the divergency applied to a vectorial gradient of something. while the vectorial gradient operator j is deﬁned in (22.. .3) can be applied not only to scalar ﬁelds.4) and apply divergency operator to arbitrary tensor ﬁeld with at least one upper index: 3 . TENSOR FIELDS.7) Operator (22...s... Here is the formula for the r-th coordinate of rot X: 3 3 3 (rot X) = i=1 j=1 k=1 r g ri ωijk j X k. Usually it is applied to a vector ﬁeld and is given by formula: 3 i div X = i=1 iX ..6) Denote by the following diﬀerential operator: = 1 ∂2 − c2 ∂t2 . In general relativity upon introducing the additional coordinate x0 = c t one usually rewrites the d’Alambert operator in a form quite similar to (22..edu/). .4) As we see...

10) The truth is that formulas (22.8) can be generalized for the case when X is an arbitrary tensor ﬁeld with at least one upper index.8).1.6) and (22. . Show that in case of orthonormal coordinates. Exercise 22. .6) and (22.8) into (22. .10) are valid only for orthonormal coordinates with ONB as a basis.6) for the Laplace operator reduces to the standard formula (22.2. The standard formulas (22.3. AND ROTOR .11) reduces to (22. By analogy with (22.9) rot X = det ∂ ∂x1 X1 .§ 22.8) are written for a general skew-angular coordinate system with a SAB as a basis. Note that formulas (22.5) suggest your version of such a generalization. DIVERGENCY. The coordinates of the vector rot X in a skew-angular coordinate system are given by formula (22.11) and show that in the case of a orthonormal coordinate system the resulting formula (22.10). when g ij = δ ij . (22. (22. Substitute (22. Here are standard formulas: = ∂ ∂x1 2 + ∂ ∂x2 e1 2 + e2 ∂ ∂x2 X2 ∂ ∂x3 e3 ∂ ∂x3 X3 2 . 37 Exercise 22. (22. formula (22. Then for vector rot X itself we have the expansion: 3 rot X = r=1 (rot X)r er . GRADIENT.8) for the Laplace operator and for the rotor are diﬀerent from those that are commonly used.11) Exercise 22.9). Formula (22.

y 3 ) is constructed by means of vectors and bases. Hurrah! This is a numeric map. In Cartesian coordinates this map P (y 1 . (24. let’s keep in mind spherical coordinates when thinking about more general and hence more abstract curvilinear coordinate systems. bases and axes ? What is the pure idea of coordinates ? The pure idea is in representing points of space by triples of numbers. the best idea is to represent each point P by its radiusvector rP in some auxiliary Cartesian coordinate system and then consider a map rP (y 1 . e2 . x3 ) (y 1 . in spherical coordinates y 1 = r is a distance from the point P to the center of sphere. e3 of the auxiliary Cartesian coordinate system: 3 rP = i=1 xi e i . x3 ). In the left direction it is represented by three functions of three variables: x1 = x1 (y 1 . if we forget for a moment about radius-vectors. Arranging other coordinate systems one can use other methods. y 2 . y 2 .3) . y 3 ). Furthermore. x2 . x2 = x2 (y 1 . 3 y = y 3 (x1 . x2 . Now we know almost everything about Cartesian coordinates and almost nothing about the abstract curvilinear coordinate system y 1 . y 3 ) in the whole space or at least in some domain. Therefore. x2 . y 2 . y 3 ). y 3 ). y 2 . Auxiliary Cartesian coordinate system. This means that we should have one to one map P (y 1 . y 2 . y 3 ).CHAPTER IV TENSOR FIELDS IN CURVILINEAR COORDINATES. The radius-vector itself is represented by three coordinates in the basis e1 . y 3 . Therefore. y 2 = y 2 (x1 . What are coordinates. (24. y ). spherical coordinates are one of the simplest examples of curvilinear coordinates.2) 3 3 1 2 3 x = x (y . we have a one-to-one map (x1 . We can treat it numerically. y 2 . § 23. § 24. x3 ). y 2 . x2 . y 3 that we are going to study. (24. By the way. y 2 . y . where we are going to use our coordinates y 1 .1) In the right direction we again have three functions of three variables: 1 y = y 1 (x1 . x3 ). General idea of curvilinear coordinates. y 2 = θ and y 3 = ϕ are two angles. For example.

y 2 . y 3 ). x2 . we use it 1 2 3 in order to emphasize that P0 is ﬁxed point. or when brought to the form (24. Tji = Tji (y 1 .5) i However. The components of T are functions of x1 . dy ∂xi It’s for the diﬀerentiation of composite function. When brought to the form (24. y0 . x3 (y)) = dy 3 fi (x1 (y). y0 . The components of matrix S in that form.5)). y 2 . and that its coordinates y0 .4) can be arranged into two square matrices S and T respectively. x3 ). x2 (y).3) into the arguments of Sj .6) (24. where fi = .4) Partial derivatives (24. x2 . y 2 . y 3 ) determined by (25.2) i into the arguments of Tj .1) is called a basic vector-function of a curvilinear coordinate system.2) and derive the relationship (24. y 3 . y 3 ) = rP = i=1 xi (y 1 .7) (but not in form of (24. ∂y j Tji = ∂y i . or by substituting (24.3) are inverse of each other. y 3 ). (24. are functions of y 1 . The vector-function R(y 1 . y 2 . y 2 .3) to be diﬀerentiable and consider their partial derivatives. by substituting (24.2) now assumed to contain diﬀerentiable functions.6). Let’s substitute (24. x2 . y 3 . Let’s denote i Sj = ∂xi . x3 ). (24. y 3 ) ei (25. x3 ). x2 . Coordinate lines and the coordinate grid. matrices S and T are inverse of each other: T = S −1 . y 3 ). Tji = Tji (x1 .7) = i Sj (y 1 . we can make them have a common set of arguments: i i Sj = Sj (x1 .2) and (24. In mathematics such matrices are called Jacobi matrices.2) and (24. § 25. x2 (y).8). y0 . 39 Further we shall assume all functions in (24. y 2 . x3 : i i Sj = Sj (y 1 .4). Exercise 24. Here zero is not the tensorial index. Let P0 be some ﬁxed point of space given by its 1 2 3 curvilinear coordinates y0 . y 2 . i Sj Tji = Tji (x1 . y0 .2) into (24. x3 (y)) i=1 dxi (y) ∂f . COORDINATE LINES AND THE COORDINATE GRID.§ 25.1) is a diﬀerentiable function of three variables y 1 . Apply this formula to functions (24. You certainly know the following formula: df (x1 (y).1. as they are deﬁned in (24. (24.8) is due to the fact that numeric maps (24. y 2 .8) This relationship (24.1) and take into account that (24. Then the vector-function 3 R(y 1 . ∂xj (24.

3). are three ﬁxed numbers. The parallels do not intersect. Remember the exact deﬁnition of spherical coordinates and ﬁnd all singular points for them.2) Substituting (25. y0 + t). 9 the coordinate grid of curvilinear coordinates is compared to that of the Cartesian coordinate system. 2 y 2 = y0 . (25.5) are called coordinate lines. On Fig. They are subdivided into three families. by setting 1 y 1 = y0 + t. But usually. but the meridians forming one family of coordinate lines do intersect at the North and at South Poles. (25. 2. 3 y 3 = y0 . Another example of coordinate grid is on Fig. At time instant t = 0 this curve passes through the ﬁxed point P0 . (25. Coordinate lines taken in whole form a coordinate grid. Same is true for curves given by two other vector-functions similar to (25. 2 3 1 R3 (t) = R(y0 . TENSOR FIELDS IN CURVILINEAR COORDINATES.1) we get a vector-function of one variable t: 1 2 3 R1 (t) = R(y0 + t. y0 ). then (25. y0 ). (25. y0 + t.3) describes a curve (the trajectory of a particle).4).2) into (25.5) are intersected at the point P0 as shown on Fig. it is represented as a grid with ﬁnite density. Curves (25.3). meridians and parallels are coordinate lines of a spherical coordinate system. 8. In the next step let’s undo one of them. y0 . This means that North and South Poles are singular points for spherical coordinates.5) This means that all three curves given by vector-functions (25. Curves from diﬀerent families intersect so that any regular point of space is an intersection of exactly three coordinate curves (one per family). . and (25.3) If we treat t as time variable (though it may have a unit other than time).4) (25.4): 1 2 3 R2 (t) = R(y0 . say ﬁrst one. This is an inﬁnitely dense grid. y0 .40 CHAPTER IV. Indeed. when drawing. Arrowheads on these lines indicate the directions in which parameter t increases. (25.4). and (25. Curves within one family do not intersect each other. Exercise 25.1.

However. the equality (26.1) and remember (24. For this reason basis E1 . For this purpose we should diﬀerentiate vector-functions (25.4).7). parabolic. (25.8). During their motion the vectors of the moving frame E1 . elliptic. E2 . Therefore. Though P0 is a ﬁxed point. e2 . E2 . Let’s consider the three coordinate lines shown on Fig. E3 can be orthogonal to each other at all points of space. Therefore. And let’s ﬁnd tangent vectors to them at the point P0 . E3 form a basis (see Fig. there is some crucial diﬀerence of basis E1 . Moving frame of curvilinear coordinates.4) are transition matrices to this basis and back to the Cartesian basis. and det S = 0 due to (24. They are strikingly similar. . Despite obvious similarity of the formulas (26.2) All calculations in (26. there is no curvilinear coordinate system with the moving frame being ONB ! We shall not prove this fact since it leads deep into diﬀerential geometry. Formula (26.§ 26. they can change their lengths and the angles they form with each other. E2 . e3 . 41 § 26. In that case we say that we have an orthogonal curvilinear coordinate system. They are bound to that point where derivatives (24. it is an arbitrary ﬁxed point. (26. In some cases vectors E1 . toroidal. E3 is called moving frame of the curvilinear coordinate system. cylindrical. Now let’s omit the intermediate calculations and write (26. E2 .5) with respect to the time variable t and then substitute t = 0 into the derivatives: Ei = dRi dt = t=0 ∂R ∂y i . E2 . E2 .4) are calculated.2) as 3 Ei = i=1 j Si e j .3). and others. MOVING FRAME OF CURVILINEAR COORDINATES. e. and (25. E2 .7). 8 again.2) are still in reference to the point P0 .3) and (5. Most of the well known curvilinear coordinate systems are orthogonal.3) means that tangent vectors to coordinate lines E1 . at the point P0 (26. E3 as compared to e1 . E3 are not free. And they move when we move this point. 10). E3 are not simply translated from point to point.3) And then compare (26. j=1 (26. Vectors E1 . spherical. in general the moving frame E1 .4): Ei = ∂R = ∂y i 3 j=1 ∂xj ej = ∂y i 3 j Si e j .1) into (26. matrices (24. g.2) is valid at any point.1) Now let’s substitute the expansion (25. E3 is a skew-angular basis.3) with (5.

As a result we get the following equality for Christoﬀel symbols Γk : ij 3 q ∂Si eq = ∂y j 3 3 q Γk Sk e q .1) ij ∂y j k=1 Formula (27.5) the components of matrix S in (26.4.1 calculate the Christoﬀel symbols for cylindrical coordinates. Exercise 27.1).1.2) into left hand side of the derivational formula (27. therefore. E2 . they do not represent a tensor. Let’s describe this motion quantitatively. According to (24.2) from which you derive Ei = ∂R .3) into (27. y 2 .3) and substitute it into both sides of (27. We shall not prove this fact since it again leads deep into diﬀerential geometry. y 3 . Exercise 27.1): q ∂Si = ∂y j 3 q Γk Sk .1) and results of exercise 27. This vector can be expanded back in moving frame E1 . Therefore.1) draw the vectors of the moving frame for cylindrical coordinates. ∂y i (27. This expansion is written as 3 ∂Ei = Γk E k .1) ij are called Christoﬀel symbols or connection components. ij k=1 q=1 (28. Exercise 27. .1) and (26.2) Substitute (27. diﬀerentiating Ei with respect to y j we should expect to get some nonzero vector ∂Ei /∂y j . E3 .2. Dynamics of moving frame.3) are functions of the curvilinear coordinates y 1 .1) is known as the derivational formula. Thus.. Due to the uniqueness of such expansions we have the following equality derived from (28. we know that the moving frame moves. Remember formula (26. However.1) and relying on the properties of mixed derivatives prove that the Christoﬀel symbols are symmetric with respect to their lower indices: Γk = Γk . ij ji Note that Christoﬀel symbols Γk form a three-dimensional array with one ij upper index and two lower indices. Exercise 27. Do the same for spherical coordinates. e3 of the auxiliary Cartesian coordinate system.3. ij k=1 (28.5.A. 2004. Exercise 27.1) are expansions in the base e1 . they are not diﬀerentiated when we substitute (26. Relying upon formula (25.2) CopyRight c Sharipov R. § 27.1).42 CHAPTER IV. Relying upon formula (27. § 28. (27.1) q=1 Cartesian basis vectors eq do not depend on y j . Coeﬃcients Γk in (27. Do the same for spherical coordinates. TENSOR FIELDS IN CURVILINEAR COORDINATES. Both sides of (28. Let’s take formula (26. Formula for Christoﬀel symbols. e2 .

y 3 the old coordinates.. and y 1 . y 2 . y 3 ). and in addition to this it provides the basis. jr (y 1 . ˜ 2 y = y 2 (y 1 . As we remember.4) .1) We also can have two curvilinear coordinate systems and can pass from one to another by means of transition functions: 1 ˜ y = y 1 (y 1 .4) are of no practical use because they express Γk through an exterij nal thing like transition matrices to and from the auxiliary Cartesian coordinate system. y 3 ). y ˜ ˜ 3 y = y 3 (˜1 . y 3 ). y 3 ).4) this formula (28. y 2 . y 2 ...3) can be transformed in the following way: 3 Γk = ij q=1 k Tq q 3 3 q 2 q ∂Sj ∂Si k ∂ x k = Tq = Tq . Ej = i=1 ˜ Tji Ei . Using concept of the inverse matrix ( T = S −1 ) derive the following formula for the Christoﬀel symbols Γk from (28.3) Due to (24. Therefore..§ 29.. Each curvilinear coordinate system provides us a numeric representation for points. y 3 ).3) They relate moving frames of two curvilinear coordinate systems: 3 3 ˜ Ei = j=1 Sij Ej . (29.4) Formulas (28. TENSOR FIELDS IN CURVILINEAR COORDINATES.2) If we call y 1 . y ). y 2 . y . ∂y j (29. ∂ yj ˜ Tji = ∂ yi ˜ . ˜ ˜ 1 y ˜ ˜ y = y 1 (˜1 . we can refer tensorial objects to curvilinear coordinate systems. y 2 . ∂y j ∂y i ∂y j ∂y i q=1 q=1 (28. y ˜ ˜ (29. tensors are geometric objects related to bases and represented by arrays if some basis is speciﬁed. they will help us below in understanding the diﬀerentiation of tensors. Tensor ﬁelds in curvilinear coordinates. y 3 the new coordinates. ˜ ˜ 3 y = y 3 (y 1 . s s (29. ∂y j (28. where they are represented as arrays of functions: i1 . then ˜ ˜ ˜ transition matrices S and T are given by the following formulas: i Sj = ∂y i .. 2 2 1 2 3 y = y (˜ .2): ij 3 Γk = ij q=1 k Tq q ∂Si . This is the moving frame. 43 Exercise 28. § 29.1. However.. y 3 ). y 2 .. i i1 . i Xj1 . y 2 . y 2 . jr = Xj1 .

..2) Note that q = ∂/∂xq is a diﬀerential operator and due to (24.. y 2 . then apply . .. x .. x ) Matrices (24. y . h1 . y2 . h Xk1 . . y 2 . let’s ﬁrst transform X into auxiliary Cartesian coordinates. . . . Thr Sj11 .3) from (29.3) and (12.. ... The only thing we need now is to transform to a curvilinear coordinate system.7) But formulas (19. i Xj1 .. E3 ) − − − → ←− − T ˜ T S ˜ S (29.. α q . ksr (x1 .1. Diﬀerentiation of tensor ﬁelds in curvilinear coordinates. y ) ←−− −− T...1)... jr (y1 . .S h1 . X.. .. . .... m Smα .. E3 ) ←−− (e1 ... Tjks Xk1 .6) (29... E2 . . y 3 ). y 2 ..4) and (29. ksr (y 1 . q ∂x ∂y p (30. e3 as intermediate coordinate system: − − − − → ˜ ˜ ˜ − (E1 . .. α (30. .6) and (19. i y ˜ ˜ Xj1 .4) are used in (30. Sjss Xk1 . § 30. In order to calculate tensor X in curvilinear coordinates. y 2 . hr 1 2 3 q Xk1 .2) using some auxiliary Cartesian coordinates with basis e1 . s).5): 3 3 . p X. 1 r ˜ i1 . . .44 CHAPTER IV.. 1 s 1 r i i k k h1 . We already know how to diﬀerentiate tensor ﬁelds in Cartesian coordinates (see section 21). .. h Xk1 .. . mα =1 h . .. iα . ..5) Compare (29.4) we know that the transformation of each index is a separate multiplicative procedure.. and then transform X back into curvilinear coordinates: h1 .. ir 1 2 3 p Xj1 . hr k1 .α .. x2 .. x3 ) =∂/∂xq q (30. ks (x .4) we have 3 q Sp q=1 ∂ ∂ = .. . Shr Tjk1 . E2 ... q=1 hα =1 3 = Thiα . hr k1 . e3 ) −− −− (E1 . ks 3 i i ˜ h1 . ... ...T i1 .. ksr (y 1 . e2 . h y ˜ ˜ Sh1 .. y 3 ).. When applied to the α-th upper index. . ks 3 i1 . From (12.. We know that operator produces tensor ﬁeld of type (r. Exercise 29.. s + 1) when applied to a tensor ﬁeld of type (r.1) looks like 3 . TENSOR FIELDS IN CURVILINEAR COORDINATES... h Th1 .13) and explain diﬀerences you have detected.2). h1 . .. y3 ) = s ..8) should be replaced by (29.5) with (5.3) . jr (˜1 . Derive (29.4) and (19.. . .. js (y .. 3 q Sp .... Transformation formulas for tensor ﬁelds for two curvilinear coordinate systems are the same as in (19.1) h1 . the whole chain of transformations (30... e2 ... y 3 ) − − → −− p S.... . ksr (˜1 . y 3 ) = s (29...

.. ....3) or (28..8) Applying (30... p X. .. = . . + . i ∂Xj1 ... ... 45 Any diﬀerential operator when applied to a product produces a sum with as many summands as there were multiplicand in the product. p X. iα .....5) Now let’s consider the transformation of the α-th lower index in (30. .. .. .. ......10) − α=1 nα =1 Γnα pjα i1 . ..3): 3 Γk = − ij q=1 q Si k ∂Tq .. α X +.2) (or equivalently in (30.. jα .. nα ... jr : s i1 . q 3 = ...... . and add the term produced when q in (30.. . .. jr i1 . mα ..α . X. (30. ....9).1. . . . ......5). .... ... 3 q Sp .... + ....... ..6) with the same logic as in deriving (30..... . .. js = r 3 i1 .. p X. = . Here is the summand h produced by term Smα in formula (30.... + mα =1 iα .... . ..... jis − .. ...4) we get 3 .. Γnα X... p X..7) In order to simplify (30. ... ∂y p .. . r Xj1 ...§ 30. .. + .6) Applying (30. The operator p determined by this formula is called the covariant derivative..2): α 3 ..... p X.. . i the following general formula for p Xj1 ..... (30.− nα =1 ..10) to a vector ﬁeld and calculate the covariant derivative p X q .4) Comparing (30. .4) we can transform it into the following equality: 3 .. + nα Tkα X ........ Exercise 30.. . (30. . As a result we get i1 . ∂y p α=1 m =1 α s 3 (30. . .. ir p Xj1 . pjα (30. ∂y p . . = . DIFFERENTIATION OF TENSOR FIELDS IN CURVILINEAR COORDINATES.4)) acts upon components of tensor X...7) we obtain 3 .... (30. r s iα + Γpmα Xj1 . + mα =1 hα =1 3 iα Thα h Smα .9) Now we should gather (30. .. m Γpmα X.. q=1 kα =1 kα Sjα .. .. . (30. 3 kα Sjα nα =1 kα =1 = .. . jα .. nα ... Tkα .. ∂y j (30. . nα .. mα ... .. Apply the general formula (30. ....... iα .4) with (28. nα =1 nα .7) we need the following formula derived from (28. ..8) to (30.1): 3 .. jα . nα .3) to (30.. jis ....

Exercise 30. (31. div H. Apply the general formula (30. The identity (31. grad ϕ (vectorial gradient) in cylindrical and spherical coordinates. Let’s remember that we consider curvilinear coordinates in Euclidean space E. Calculate rot A.10). Apply the general formula (30. § 31.8) from (28. divergency.1) In Cartesian coordinates all components of the metric tensor are constant since the basis vectors e1 .2) is known as the concordance condition for the metric gij and connection Γk .10) to a bilinear form and ﬁnd p aqm . ϕ applied to the scalar ﬁeld . Apply the general formula (30. Apply the general formula (30.2. ij Remember that the metric tensor enters into many useful formulas for the gradient. e2 . Exercise 30. Exercise 30.4.10) to a covector ﬁeld and calculate the covariant derivative p Xq .4.2.2) But g is a tensor. but the covariant derivative in the sense of formula (30. Exercise 30. e3 are constant. What is important is that all of these formulas remain valid in curvilinear coordinates.5) to curvilinear coordinates using transition matrices (24.10) to the contraction C(F) for the case when F is an operator ﬁeld. Exercise 31.10) to a tensor product a ⊗ x for the case when x is a vector and a is a covector. with the only diﬀerence being that you should understand that p is not the partial derivative ∂/∂xp .6.3).10) to an operator ﬁeld and ﬁnd q q Fm . rotor.10) in Cartesian coordinates reduces to diﬀerentiation p = ∂/∂xp .46 CHAPTER IV.5. In this space we have the scalar product (13. Consider special case when p is applied to the Kronecker symbol δm . p gij = 0.3. If all of its components in some coordinate system are zero.1. Apply the general formula (30. Therefore.1).2) is valid in curvilinear coordinates as well.1) and the metric tensor (13. then they are identically zero in any other coordinate system (explain why). Derive (30. p Exercise 30. Concordance of metric and connection. Exercise 30. Verify the formula C(F) = C( F). Ej ). Prove (31.4) and show that here it is given by formula gij = (Ei . (31.5).3. Verify formula (a ⊗ x) = a ⊗ x + a ⊗ x. Calculate the Laplace operator ϕ in cylindrical and in spherical coordinates. and Laplace operator in section 22. TENSOR FIELDS IN CURVILINEAR COORDINATES. It is very important for general relativity.7. Exercise 31.2) by direct calculations using formula (27. Transform the metric tensor (13. Exercise 31. The covariant derivative (30. Therefore the identity (31. Exercise 31.

. Russia. Austin. On-line lecture materials. 3.. Feb. University of Texas at Austin. P. Electrical & Computer Engineering Dept. free for downloading from Web site of Saint Michael’s College. Madison. 4. WI 53704. Oxford. . Sept. 2000. 1. Colchester.org... B2 ring theory preliminaries. Basic background for a course of information and cryptography. 2000. Texas 78712. English tr.. Sharipov R. Download [PDF] or [PS] ﬁle. Math. VM 05439. Classical electrodynamics and theory of relativity. 1997. USA. An elementary introduction to logic and set theory. 5. University of Oxford. Ufa. UK. Heﬀeron J.. A. physics/0311011 in Electronic Archive http:/ /arXiv. 2.. Vaughan-Lee M. Linear algebra. Bashkir State University. Online materials. USA. USA.REFERENCES. On-line resource. OX1 3LB. Madison Area Technical College. 2003. Konstantopoulos T. Electronic textbook. Lehnen A. Institute.

Sign up to vote on this title

UsefulNot useful- tensors
- Tensor Calculus
- Matrix and Tensor Calculus - Aristotle D. Michal
- Tensor Analysis With Applications in Mechanics
- Fer King Hoff
- Introduction to Tensor Calculus & Continuum Mechanics
- Vector and Tensor Analysis
- Vector and Tensor Analysis 1950
- 1948-Vector and Tensor Analysis
- Functional and Structural Tensor Analysis for Engineers
- Tensor Analysis Sokolnikoff
- 57253677-Schaum-s-Tensor-Calculus-238.pdf
- Tensor
- Bishop, Goldberg - Tensor Analysis on Manifolds(dover 1980)(288s).pdf
- A Brief on Tensor Analysis
- Binder 1
- Tensor Calculus
- tensor analysis
- tensor
- Tensor Analysis for Students of Physics and Engineering. Read 02.24.2008
- Vector and Tensor Analysis- A.Borisenko I.Tarapov.pdf
- 3-tensor
- A tensor
- B00H44OZ8S(Onlinepdfbooks.blogspot.com)
- Battle Circle 02 Var the Stick
- Battle Circle 03 Neq the Sword
- [Ta-Pei Cheng] Relativity, Gravitation, And Cosmol(BookFi.org)
- National Geographic Deutschland 2012-07
- Current Trends in Arithmetical Algebraic Geometry - K. Ribet
- tensor analysis

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.