April 27, 2006 10:13 Book Trim Size For 9in X 6in Field

FIELD THEORY - A Path Integral Approach (Second Edition)
orld Scienti!ic P"#li$hing %o& Pte& Ltd&

http'(()))&)orld$ci#oo*$&co+(ph,$ic$(-./0&ht+l
April 27, 2006 10:13 Book Trim Size for 9in x 6in eld
Chapter 1
Introduction
1.1 Particles and Fields
Classically, there are two kinds of dynamical systems that we en-
counter. First, there is the motion of a particle or a rigid body (with
a nite number of degrees of freedom) which can be described by a
nite number of coordinates. And then, there are physical systems
where the number of degrees of freedom is nondenumerably (non-
countably) innite. Such systems are described by elds. Familiar
examples of classical elds are the electromagnetic elds described by
E(x, t) and B(x, t) or equivalently by the potentials ((x, t), A(x, t)).
Similarly, the motion of a one-dimensional string is also described by
a eld (x, t), namely, the displacement eld. Thus, while the coor-
dinates of a particle depend only on time, elds depend continuously
on some space variables as well. Therefore, a theory described by
elds is usually known as a D + 1 dimensional eld theory where D
represents the number of spatial dimensions on which the eld vari-
ables depend. For example, a theory describing the displacements
of the one-dimensional string would constitute a 1+1 dimensional
eld theory whereas the more familiar Maxwells equations (in four
dimensions) can be regarded as a 3+1 dimensional eld theory. In
this language, then, it is clear that a theory describing the motion of
a particle can be regarded as a special case, namely, we can think of
such a theory as a 0+1 dimensional eld theory.
1
2 Field Theory: A Path Integral Approach
1.2 Metric and Other Notations
In these lectures, we will discuss both non-relativistic as well as rel-
ativistic theories. For the relativistic case, we will use the Bjorken-
Drell convention. Namely, the contravariant coordinates are assumed
to be
x
= (t, x), = 0, 1, 2, 3 , (1.1)

while the covariant coordinates have the form
x
= (t, x) . (1.2)
Here we have assumed the speed of light to be unity (c = 1). The
covariant metric, therefore, follows to have a diagonal form with the
signatures
= (+, , , ) . (1.3)
The inverse or the contravariant metric clearly also has the same
form, namely,
= (+, , , ) . (1.4)
The invariant length is given by
x
2
= x
= t
2
x
2
. (1.5)
The gradients are similarly obtained from Eqs. (1.1) and (1.2) to be
=

x

t
,
, (1.6)
=

x

t
,
, (1.7)
so that the DAlembertian takes the form
=
=

2
t
2

2
. (1.8)
Introduction 3
1.3 Functionals
It is evident that in dealing with dynamical systems, we are dealing
with functions of continuous variables. In fact, most of the times,
we are really dealing with functions of functions which are otherwise
known as functionals. If we are considering the motion of a particle
in a potential in one dimension, then the Lagrangian is given by
L(x, x) =
1
2
m x
2
V (x), (1.9)
where x(t) and x(t) denote the coordinate and the velocity of the
particle and the simplest functional we can think of is the action
functional dened as
S[x] =
t
f
t
i
dt L(x, x) . (1.10)
Note that unlike a function whose value depends on a particular
point in the coordinate space, the value of the action depends on
the entire trajectory along which the integration is carried out. For
dierent paths connecting the initial and the nal points, the value
of the action functional will be dierent.
Thus, a functional has the generic form
F[f] =
dx F(f(x)), (1.11)
where, for example, we may have
F(f(x)) = (f(x))
n
. (1.12)
Sometimes, one loosely also says that F(f(x)) is a functional. The
notion of a derivative can be extended to the case of functionals in a
natural way through the notion of generalized functions. Thus, one
denes the functional derivative or the Gateaux derivative from the
linear functional
F
[v] =
d
d
F[f +v]
=0
=
dx
F[f]
f(x)
v(x) . (1.13)
Equivalently, from the working point of view, this simply corresponds
to dening
F(f(x))
f(y)
= lim
0
F(f(x) +(x y)) F(f(x))
. (1.14)
It now follows from Eq. (1.14) that
f(x)
f(y)
= (x y) . (1.15)
The functional derivative satises all the properties of a deriva-
tive, namely, it is linear and associative,
f(x)
(F
1
[f] +F
2
[f]) =
F
1
[f]
f(x)
+
F
2
[f]
f(x)
,
f(x)
(F
1
[f]F
2
[f]) =
F
1
[f]
f(x)
F
2
[f] +F
1
[f]
F
2
[f]
f(x)
. (1.16)
It also satises the chain rule of dierentiation. Furthermore, we now
see that given a functional F[f], we can Taylor expand it in the form
F[f] =
dx P
0
(x) +
dx
1
dx
2
P
1
(x
1
, x
2
) f(x
2
)
+
dx
1
dx
2
dx
3
P
2
(x
1
, x
2
, x
3
) f(x
2
)f(x
3
) + , (1.17)
where
P
0
(x) = F(f(x))|
f(x)=0
,
P
1
(x
1
, x
2
) =
F(f(x
1
))
f(x
2
)
f(x)=0
, (1.18)
P
2
(x
1
, x
2
, x
3
) =
1
2!
2
F(f(x
1
))
f(x
2
)f(x
3
)
f(x)=0
,
and so on.
As simple examples, let us calculate a few particular functional
derivatives.
Introduction 5
(i) Let
F[f] =
dy F(f(y)) =
dy (f(y))
n
, (1.19)
where n denotes a positive integer. Then,
F(f(y))
f(x)
= lim
0
F(f(y) +(y x)) F(f(y))
= lim
0
(f(y) +(y x))
n
(f(y))
n
= lim
0
(f(y))
n
+n(f(y))
n1
(y x) +O(
2
) (f(y))
n
= n(f(y))
n1
(y x) . (1.20)
Therefore, we obtain
F[f]
f(x)
=
dy
F(f(y))
f(x)
=
dy n(f(y))
n1
(y x)
= n(f(x))
n1
. (1.21)
(ii) Let us next consider the one-dimensional action in Eq. (1.10)
S[x] =
t
f
t
i
dt
L(x(t
), x(t
)), (1.22)
with
L(x(t), x(t)) =
1
2
m( x(t))
2
V (x(t))
= T( x(t)) V (x(t)) . (1.23)
In a straightforward manner, we obtain
V (x(t
))
x(t)
= lim
0
V (x(t
) +(t
t)) V (x(t
))
= V
(x(t
))(t
t) , (1.24)
where we have dened
V
(x(t
)) =
V (x(t
))
x(t
)
.
Similarly,
T( x(t
))
x(t)
= lim
0
T( x(t
) +
d
dt
(t
t)) T( x(t
))
= m x(t
)
d
dt
(t
t) . (1.25)
It is clear now that
L(x(t
), x(t
))
x(t)
=
(T( x(t
)) V (x(t
)))
x(t)
= m x(t
)
d
dt
(t
t) V
(x(t
))(t
t) . (1.26)
Consequently, in this case, we obtain for t
i
t t
f
S[x]
x(t)
=
t
f
t
i
dt
L(x(t
), x(t
))
x(t)
=
t
f
t
i
dt
(m x(t
)
d
dt
(t
t) V
(x(t
))(t
t))
= m x(t) V
(x(t))
=
d
dt
L(x(t), x(t))
x(t)
+
L(x(t), x(t))
x(t)
. (1.27)
The right hand side is, of course, reminiscent of the Euler-Lagrange
equation. In fact, we note that
S[x]
x(t)
=
d
dt
L
x(t)
+
L
x(t)
= 0 , (1.28)
gives the Euler-Lagrange equation as a functional extremum of the
action. This is nothing other than the principle of least action ex-
pressed in a compact notation in the language of functionals.
Introduction 7
1.4 Review of Quantum Mechanics
In this section, we will describe very briey the essential features of
quantum mechanics assuming that the readers are familiar with the
subject. The conventional approach to quantum mechanics starts
with the Hamiltonian formulation of classical mechanics and pro-
motes observables to non-commuting operators. The dynamics, in
this case, is given by the time-dependent Schr odinger equation
i
|(t)
t
= H|(t) , (1.29)
where H denotes the Hamiltonian operator of the system. Equiva-
lently, in the one dimensional case, the wave function of a particle
satises
i
(x, t)
t
= H(x)(x, t)
=

2
2m
2
x
2
+V (x)
(x, t) , (1.30)
where we have identied
(x, t) = x|(t) , (1.31)
with |x denoting the coordinate basis states. This, then, denes the
time evolution of the system.
The main purpose behind solving the Schr odinger equation lies
in determining the time evolution operator which generates the time
translation of the system. Namely, the time evolution operator trans-
forms the quantum mechanical state at an earlier time t
2
to a future
time t
1
as
|(t
1
) = U(t
1
, t
2
)|(t
2
) . (1.32)
Clearly, for a time independent Hamiltonian, we see from Eq. (1.29)
(the Schr odinger equation) that for t
1
> t
2
,
U(t
1
, t
2
) = e
(t
1
t
2
)H
. (1.33)
More explicitly, we can write
U(t
1
, t
2
) = (t
1
t
2
)e
(t
1
t
2
)H
. (1.34)
It is obvious that the time evolution operator is nothing other than
the Greens function for the time dependent Schr odinger equation
and satises
i

t
1
H
U(t
1
, t
2
) = i(t
1
t
2
) . (1.35)
Determining this operator is equivalent to nding its matrix elements
in a given basis. Thus, for example, in the coordinate basis dened
by
X|x = x|x , (1.36)
we can write
x
1
|U(t
1
, t
2
)|x
2
= U(t
1
, x
1
; t
2
, x
2
) . (1.37)
If we know the function U(t
1
, x
1
; t
2
, x
2
) completely, then the time
evolution of the wave function can be written as
(x
1
, t
1
) =
dx
2
U(t
1
, x
1
; t
2
, x
2
)(x
2
, t
2
) . (1.38)
It is interesting to note that the dependence on the intermediate
times drops out in the above equation as can be easily checked.
Our discussion has been within the framework of the Schr- odinger
picture so far where the quantum states |(t) carry time dependence
while the operators are time independent. On the other hand, in the
Heisenberg picture, where the quantum states are time independent,
using Eq. (1.32) we can identify
|
H
= |(t = 0)
S
= |(t = 0)
= e
i
tH
|(t) = e
i
tH
|(t)
S
. (1.39)
Introduction 9
In this picture, the operators carry all the time dependence. For
example, the coordinate operator in the Heisenberg picture is related
to the coordinate operator in the Schr odinger picture through the
relation
X
H
(t) = e
i
tH
Xe
tH
. (1.40)
The eigenstates of this operator satisfying
X
H
(t)|x, t
H
= x|x, t
H
, (1.41)
are then easily seen to be related to the coordinate basis in the
Schr odinger picture through
|x, t
H
= e
i
tH
|x . (1.42)
It is clear now that for t
1
> t
2
we can write
H
x
1
, t
1
|x
2
, t
2
H
= x
1
|e
t
1
H
e
i
t
2
H
|x
2
= x
1
|e
(t
1
t
2
)H
|x
2
= x
1
|U(t
1
, t
2
)|x
2
= U(t
1
, x
1
; t
2
, x
2
) . (1.43)
This shows that the matrix elements of the time evolution opera-
tor are nothing other than the time ordered transition amplitudes
between the coordinate basis states in the Heisenberg picture.
Finally, there is the interaction picture where both the quantum
states as well as the operators carry partial time dependence. With-
out going into any technical detail, let us simply note here that the
interaction picture is quite useful in the study of nontrivially inter-
acting theories. In any case, the goal of the study of quantum me-
chanics in any of these pictures is to construct the matrix elements of
the time evolution operator which as we have seen can be identied
with transition amplitudes between the coordinate basis states in the
Heisenberg picture.
1.5 References
Das, A., Lectures on Quantum Mechanics, Hindustan Book
Agency.
Dirac, P. A. M., Principles of Quantum Mechanics, Oxford Univ.
Press.
Schi, L. I., Quantum Mechanics, McGraw-Hill Publishing.

April 27, 2006 10:13 Book Trim Size For 9in X 6in Field

Uploaded by

Copyright:

Available Formats

April 27, 2006 10:13 Book Trim Size For 9in X 6in Field

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

April 27, 2006 10:13 Book Trim Size For 9in X 6in Field

Uploaded by

Copyright:

Available Formats

FIELD THEORY - A Path Integral Approach (Second Edition)

orld Scienti!ic P"#li$hing %o& Pte& Ltd&

= (t, x), = 0, 1, 2, 3 , (1.1)

You might also like