6.

Second Quantization and Quantum Field Theory

6.0. Preliminary
6.1. The Occupation Number Representation
6.2. Field Operators and Observables
6.3. Equation of Motion and Lagrangian Formalism for Field Operators
6.4. Second Quantization for Fermions
6.0. Preliminary
Second quantization is an indispensable tool for the study of systems with variable
numbers of particles. Some notable examples of the latter are
1. High energy scattering and decay processes.
2. Relativistic systems.
3. Many body systems (not necessarily relativistic).
The term 1
st
quantization refers to the physical fact that certain classically continuous
quantities, such as energy and angular momentum, can take on only discrete values in
the quantum regime. In a purely mathematical sense, 2
nd
quantization is simply a
transformation to the number representation so that all dynamical properties of the
system can be obtained by counting the numbers of the 1-particle states being
occupied at each instance of time. Here, the quantization refers to the discreteness
of the counting. However, a more physical interpretation is to take 2
nd
quantization
as adopting the viewpoint that all quantum processes can be described in terms of the
exchange of real or virtual particles. For example, the electromagnetic interaction
can be described in terms of the exchange of real or virtual photons.
The technique is particular useful when dealing with systems near their ground states.
In which case, the problem is reduced to the study of only a few weakly interacting
quasi-particles, or elementary excitations, so that standard perturbation
techniques can be applied. Thus, the art is in the identification of a ground state
that
1. can be described in known, preferably simple, mathematical terms.
2. contains as much of the interactions as possible.
6.1. The Occupation Number Representation
6.1.1. Many Body System
6.1.2. Number Representation: States
6.1.3. Creation and Annihilation Operators
6.1.4. Commutation Relations
6.1.5. Number Representation: Operators
6.1.6. Summary
6.1.1. Many Body System
Consider a system of N identical particles. Let

k be a set of complete,
orthonormal, 1-particle states that satisfy the boundary conditions of the system.
Using particle labels as subscripts, the set

1 N
k k of all direct products of
N 1-particle states is an orthonormal basis for the system.
Owing to the uncertainty principle, identical particles in quantum mechanics are
indistinguishable. This means the state , , , ,
i j
k k in which particles i and j
are in states k
i
and k
j
, respectively, is equivalent to state , , , ,
j i
k k obtained
by interchanging the particles i and j. Thus,
, , , , , , , ,
i j j i
k k k k = , i j
where is some complex number. Obviously, exchanging the same pair of particles
twice brings back the original state so that
2
1 = 1 =
Particles with 1 = + ( 1 = ) are called bosons (fermions) since they obey
Bose-Einstein (Fermi-Dirac) statistics. It was found experimentally that all
particles with integer and half-integer spins are bosons and fermions, respectively.
According to the spin-statistics theoremin quantum field theory, the reason for this
association is causality [see Chap 7]. In terms of the 1-particle basis, we can write
, )
, ) , ) 1 1
, ,
P
N P P N
P
k k C k k

for
bosons
fermions
where P denotes a permutation
, ) , ) , ) , ) 1, , 1, , 1 , , N P N P P N = 1
]

with
, )
1
1
P
+

if P consists of an
even
odd
number of transpositions (exchanges)
The orthonormality of

k can be used to show that
1. Normalization of
1
, ,
N
k k gives
!
!
j
j
n
C
N

=
[
, where
j
n is the number of
particles in state
j
k and j runs through the distinct states in
1
, ,
N
k k so that
j
j
N n =

. Note that for fermions, 0,1

j
n = .
2.
1
, ,
N
k k is orthonormal in the sense that

1 1 '
1 1 ' , , , , ,
, , , ,
N N
N N k k k k
k k k k
' '
' ' =

(6.1)
where

1 1 '
, , , , ,
1
0
N N
k k k k

' '

=

if

1 1 '
, , , ,
N N
k k P k k
otherwise
' ' =
Note that 2 states with N N' = are always orthogonal.
6.1.2. Number Representation: States
Consider a set of complete, orthonormal 1-particle (1-P) basis. For the sake of
clarity, we shall assume the quantum numbers to be discrete. (Results for the
continuous case can be obtained by some limiting procedure). To begin, we arrange
the 1-P states by some rule into a unique sequence 0,1, 2, of monotonically
increasing energy so that 0 is always the 1-P ground state. For example, the one
electron spinless states nlm in a hydrogenic atom can be arranged as 0 100 ,
1 11 1 , 2 110 , 3 111 , .
The basis in the number (n-) representation consists of all the eigenstates of the
number operator n

:
0 1 0 1
, , , , , , , , n n n n n n n n


where n

is the number of particles in the 1-particle state . We shall assume

orthonormality:
0 0 1 1
0 1 0 1
, , , , , , , ,
n n n n n n
n n n n n n




(6.2)
Completeness of the basis means the state vector of the system can be written as

0 1
0 1 0 1
, , , ,
, , , , , , , ,
n n n
t n n n n n n t

0 1
0 1
0 1 , , , ,
, , , ,
, , , ,
n n n
n n n
n t n n

(6.3)
6.1.3. Creation and Annihilation Operators
The conjugate variables in the n-representation are the annihilation operators a

and
the creation operators a

. By definition
n

0 1 0 1
, , , , , , , , n n n n n n n n



0 1 0 1
, , , , , , , 1, a n n n A n n n n



0 1 0 1
, , , , , , , 1, a n n n C n n n n

where A(n

) and C(n

) are normalization constants to be determined. Thus,

0 1 0 1
, , , , , , , , n n n n a a n n n

2
0 1 0 1
, , , 1, , , , 1, A n n n n n n n



2
A n

Assuming A to be real, we have

a
A n n

Similarly,


2
0 1 0 1
, , , , , , , , n n n a a n n n C n

1 A n C n


Assuming C to be real, we have
either 0 C n

or 1 1
a
C n A n n


For bosons, there is no upper limit to n

. Therefore, 0 C n

for all n

and
0 1 0 1
, , , , , , , 1, a n n n n n n n


0 1 0 1
, , , , 1 , , , 1, a n n n n n n n

For fermions, the antisymmetry means 0,1 n

. Thus, 0 1 C and 1 0 C
and we can write

0 1 0 1 1 1
, , , , , , , ,
S
a n n n n n n a n n

where
1
0
S n

counts the transpositions required to move a

to the proper
position.

0 1 0 1
, , , , , , , 1,
S
a n n n n n n n

0 1 0 1
, , , , 1 , , , 1,
S
a n n n n n n n

0 1
1 , , ,1 ,
S
n n n n

0 1
1 , , , 1,
S
n n n n

where the various expressions for a

all made use of the fact that 0,1 n

.
Finally, we mention that the completeness of the basis
0 1
, , , , n n n

is with
respect to the Fock space, i.e., all many particle states that can be constructed from the
1-P states. However, there exists many particle states that cannot be constructed in
this manner. The most notorious example is the BCS superconducting state, which
is a condensate (in momentum space) of Cooper pairs.
6.1.4. Commutation Relations
Before proceeding further, we must establish a way for exchange symmetries to be
built into
0 1
, , , , n n n

. To this end, we define

, ,
, ,
0 1 1
0

, , , ,
n
n
m
a
n n n
C m

=
=
[

, ,
, ,
1
1
1
1
0

n
n
m
a
C m
+

=
[

, ,
, ,
0
0
0
1
0

n
n
m
a
C m
+

=
u
[
where 0, 0, u = is the vacuum. Note that for fermions, 0,1 n

= so that
, ,
1
0
1
n
m
C m

=
=
[
for all n

. The set of all

0 1
, , , , n n n

thus generated is called the

Fock space. It is straightforward to check that the exchange symmetries are
established by requiring
, a

0 a

+
( =

where the upper and lower signs are for bosons and fermions, respectively. Also,
j | , a b ab ba

Note that the anticommutator j | ,
+
is often written as { ; , . Using a

+
, ,
a

+
= ,
we have
j
, a

0 a

( =

Also,
, 1, , 1, n n

+ ,
a

+
a

+
,
a

, , , , n n


, , , ,
, , , ,
1 1 0
1 1 1
n n n n
n n n

=

=

for


=
=

j
, a

+
( =

6.1.5. Number Representation: Operators

Consider a 1-P operator , ) , A p x . Given the complete orthonormal basis

, we
can write
, ) , A A

p x A

where the matrix elements A A

= are numbers. Using a

+
= u ,
where u is the vacuum, we can write

A A a

+
= u

u
Note that the vacuum projector u u serves to confine

A to the 1-particle
subspace, i.e., 0

A + O = if the number of particles in either u or O is not one.

By removing this restriction, we obtain the desired many body version

A A

+
a

Next, we consider the 2-P potential

, )
1
1
,
2
N
i j
i j
V V
= =
=

x x
, )
1
,
N
i j
i j
V
> =
=

x x
Given the complete orthonormal 1-P basis

, a basis vector for the 2-P Hilbert
space is
1 2
1 2 1 2

where, to avoid ambiguity, we have used subscript to indicate the particle occupying
the state. Taking the hermitian conjugate, we obtain the adjoint basis vector
, )
1 2 2 1

+
=
1 2
2 1

where the order of the factors is to be noted with care. Using the completeness
condition
1 2 2 1
,
I

=

,
1 2 1 2

=

we can write the 2-P potential (1

st
quantized) operator as
1 2 2 1 1 2 2 1
1
2
V V

=

1 2 1 2 1 2 1 2
1
2
V

=

Using
1 2
1 2
= a

+
= a

+
u
1 2
2 1
=
, )
1 2

+
=
,
a

+
=
)
a

+
+
u a

= u a

we can write the 2

nd
quantized version of V as

V
1
2
a a


+ +
u =

V a

u a

where
, ) , ) , ) , ) , )
3 3 * *
1 2 1 2 1 2 2 1
, V d d V

=
} }
x x x x x x x x
As before, the vacuum projector u u serves to confine

V to the 2-particle
subspace. Removing this restriction then gives
1

2
V V

=

a

+
a

+
a

(a)
Note that , and , refer to the particle at x
1
and x
2
, respectively. Thus, in eq(a),
the first and last operators refer to particle 1.
6.1.6. Summary
For the ease of reference, we list below the salient results of the n-representation
(upper and lower signs are for bosons and fermions, respectively):
n

= a

+
a

j
, a

( =

, a

0 a

+
( =

j
, a

+
( =

continuous
, )
0 1 0 1
, , , , , , , , n n n n n n n n

=
, )
0 1 0 1
, , , , , , , ,
S
a n n n n n n n

=
1
0
S n

=
=

, )
0 1 0 1
, , , , 1 , , , ,
S
a n n n n n n n

+
=
0 0 1 1
0 1 0 1
, , , , , , , ,
n n n n n n
n n n n n n



' ' '
' ' ' =
1-P operator:

A A

=

a

+
a

2-P operator:
1

2
V V

=

a

+
a

+
a

6.2. Field Operators and Observables

Henceforth, we shall take the 1-P states to be momentum eigenstaes. For spinless
particles, we have = =
p
k

. Since k is continuous, the orthonormality and

completeness take the form
, ) ' ' = k k k k
3
1 d k =
}
k k
with
, ) ' ' = x x x x
3
1 d r =
}
x x
, ) =
k
x x k
, )
3/ 2
1
2
i
e

=
k x
Thus, the wave function for a single particle in state , ) t + can be written as
, ) , ) , t t + = + x x , )
3
d k t = +
}
x k k
, )
, )
3
3/ 2
2
i
d k
e t

= +
}
k x
k
, )
, ) , )
3
3/ 2
0
2
i
d k
e a t

= +
}
k x
k (6.12)
where weve used
, )
+
= k k , )
0 a
+
+
( =

k , )
0 a = k
The field operators are defined in the Schrodinger picture by
, )
x a

=

x , )
a

=

x
, )

+
x a

+
=

x , )
*
a

+
=

x
In the momentum basis, they become
, )
x , ) , )
3
d k a =
}
k
x k
, )
, )
3
3/ 2

2
i
d k
e a

=
}
k x
k (6.13)
, )

+
x , ) , )
3 *
d k a
+
=
}
k
x k
, )
, )
3
3/ 2

2
i
d k
e a

+
=
}
k x
k (6.14)
The corresponding commutation relations are
, )
,

x , )
' = (

'

x x j
, a

0 a

( =

, )
,
+

x , )

+
= ' (

'

x x , a

0 a

+
( =

(6.15)
, )
,

x , )

+
= ' (

'

x x j
, a

+
(

= '

x x
, ) ' ' = = x x x x (6.16)
Consider the operator
, )
= x , )

+
x , )
x
we have
, )
3 3
d x d x =
} }
x , )

+
x , )
x
, ) , )
, )
3 3
3
3/ 2 3/ 2

2 2
i i
d k d k
d x e a e

' +
'
=
} } }
k x k x
k , )
a ' k
, ) , )
, )
3 3
3/ 2 3/ 2

2 2
d k d k
a

+
'
=
} }
k , ) , ) , )
3
2 a ' ' k k k
, )
3
d k a
+
=
}
k , )
a k , )
3
d k n N = =
}
k
where N is the total number of particles. Hence, , )
x is the number density
operator at x. The 2
nd
quantized forms discussed in 6.1.5 can be rewritten in terms
of the field operators.
1-P operator:
, )
3 3

, A d k d k A ' ' =
} }
k x p k , )
a
+
k , )
a ' k
3
d x =
}
, )

+
x , A
i
V
| |

|
\ .
x

, )
x
2-P operator:
, )
3 3 3 3
1 2 3 4 1 2 3 4
1

,
2
V d k d k d k d k V ' =
} } } }
k k x x k k , )
1
a
+
k , )
2
a
+
k , )
4
a k , )
3
a k
3 3
1
2
d x d x' =
} }
, )

+
x , )

+
' x , ) , V ' x x , )
' x , )
x (6.21)
6.2.a. Field Operators and Observables
Note: This section is the same as 6.2 but with different normalization.
Henceforth, we shall take the 1-P states to be momentum eigenstaes. For spinless
particles, we have = =
p
k

. Since k is continuous, the orthonormality and

completeness take the form
, ) , )
3
2 ' ' = k k k k
, )
3
3
1
2
d k

=
}
k k
with
, ) ' ' = x x x x
3
1 d x =
}
x x
, ) =
k
x x k
i
e

=
k x
Thus, the wave function for a single particle in state , ) t + can be written as
, ) , ) , t t + = + x x
, )
, )
3
3
2
d k
t

= +
}
x k k
, )
, )
3
3
2
i
d k
e t

= +
}
k x
k
, )
, ) , )
3
3
0
2
i
k
e t a
d

= +
}
k x
k (6.12)
where weve used
, )
+
= k k , )
0 a
+
+
( =

k , )
0 a = k
The field operators are defined in the Schrodinger picture by
, )
x a

=

x , )
a

=

x
, )

+
x a

+
=

x , )
*
a

+
=

x
In the momentum basis, they become
, )
x
, )
, ) , )
3
3

2
d k
a

=
}
k
x k
, )
, )
3
3

2
i
d k
e a

=
}
k x
k (6.13)
, )

+
x
, )
, ) , )
3
*
3

2
d k
a

+
=
}
k
x k
, )
, )
3
3

2
i
d k
e a

+
=
}
k x
k (6.14)
The corresponding commutation relations are
, )
,

x , )
' = (

'

x x j
, a

0 a

( =

, )
,
+

x , )

+
= ' (

'

x x , a

0 a

+
( =

(6.15)
, )
,

x , )

+
= ' (

'

x x j
, a

+
(

= '

x x
, ) ' ' = = x x x x (6.16)
Consider the operator
, )
= x , )

+
x , )
x
we have
, )
3 3
d x d x =
} }
x , )

+
x , )
x
, ) , )
, )
3 3
3
3 3

2 2
i i
d k d k
d x e a e

' +
'
=
} } }
k x k x
k , )
a ' k
, ) , )
, )
3 3
3 3

2 2
d k d
a
k

+
'
=
} }
k , ) , ) , )
3
2 a ' ' k k k
, )
, )
3
3

2
d k
a

+
=
}
k , )
a k
, )
, )
3
3

2
d k
n N

= =
}
k
where N is the total number of particles. Hence, , )
x is the number density
operator at x. The 2
nd
quantized forms discussed in 6.1.5 can be rewritten in terms
of the field operators.
1-P operator:
, ) , )
, )
3 3
3 3
,
2 2

A A
d k d k

'
' =
} }
k x p k , )
a
+
k , )
a ' k
3
d x =
}
, )

+
x , A
i
|
|
V
|

\ .
x

, )
x
2-P operator:
, ) , ) , ) , )
, )
3 3 3 3
1
1 2
2 3 4
3 3 3 3 3 4
2 2 2
1

,
2
2
V V
d k d k d k d k

' =
} } } }
k k x x k k
, )
1
a
+
k , )
2
a
+
k , )
4
a k , )
3
a k
3 3
1
2
d x d x' =
} }
, )

+
x , )

+
' x , ) , V ' x x , )
' x , )
x (6.21)
6.3. Equation of Motion and Lagrangian Formalism for Field
Operators
6.3.1. Equation of Motion
6.3.2. Lagrangian
6.3.1. Equation of Motion
Field operators in the Heisenberg picture are defined by eq(5.35) as
, ) , )

, exp exp
i i
t H t H t
| | | |
=
| |
\ . \ .
x x

(6.22)
, ) , )

, exp exp
i i
t H t H t
+ +
| | | |
=
| |
\ . \ .
x x

Similarly,
, ) , )

, , exp , exp
i i
H t H t H H t
| | | |
=
| |
\ . \ .
p x p x

, )

, H = p x
In general, commutation relations in the Heisenberg picture take on simple forms only
when the operators are at the same time. For example, it is easy to verify the
following equal time commutation relations:
, ) , ) , ) , )
, , , , , , 0 t t t t
+ +
' ' ( = = (

x x x x

, ) , ) , )
, , , t t
+
' ' ( =

x x x x

For a system of particles subject to an external potential U and mutual 2-body

interaction potential V, the hamiltonian can be written as
, ) , ) , )
, ) , ) , ) , ) , )
2
3
3 3
2

, ,
2
1
, , , , ,
2
H d x t t
m
d x d x t
U
t V t t


+
+ +
| |
=
|
\ .
' ' ' ' +
V +
}
} }
x x
x x x x
x
x x

The commutators in the equation of motion [see (5.36)]

, ) , )

, , , i t t H
t

c
(
=

c
x x
can be evaluated using
j , a bc abc bca = abc bac bac bca =
j j , , a b c b a c =

j , ab c abc cab = abc acb acb cab =
j j , , a b c a c b =

Thus,
, ) , ) , )
, ) , ) , ) , ) , ) , )
, , , ,
, , , , , , , ,
t t t
t t t t t t


+
+ +
' ' (

' ' ' ' ( = (

x x x
x x x x x x

, ) , )
, t ' ' = x x x
, ) , ) , )
, ) , ) , ) , ) , ) , )
, , , ,
, , , , , , , ,
t t t
t t t t t t


+ +
+ + + +
' '' (

' '' ' '' ( ( =

x x x
x x x x x x

, ) , ) , ) , )
, , t t
+ +
' '' '' ' = x x x x x x
so that, with a little arrangement, we have
, ) , ) , )
, ) , ) , ) , )
2
2
3
, ,
2
, , , ,
i t t
t m
d x t V t
U
t


+
| | c
=
|
c
V
\ .
' ' ' ' +
+
}
x x
x x x
x
x x

(6.24)
6.3.2. Lagrangian
The equation of motion (6.24) can of course be derived from a suitable Lagrangian
using the principle of least action. Now, in a variational calculation, operators and
functions behave identically except that operators need not commute. Hence, the
technique can be demonstrated by the derivation of the Schrodinger equation for
wavefunctions. By carefully maintaining the order of all quantities, we obtain
results that are valid for field operators.
An action that gives rise to the time-dependent 1-P Schrodinger equation is
, ) , )
3
2
2
* , ,
2
S dt d x t U i t
t m

c
+
| |
=
|
\ .
V
c
} }
x x

(6.25)
3
dt d x =
} }

Since is a complex function, each , ) x represents 2 degrees of freedom. The
corresponding independent variables may be chosen to be Re and Im .
However, a more manageable choice is and *. [ Note, * can be calculated
from only after the real axis is chosen. Until then, and * are independent. ]
Variation on * gives
, )
0
*
t

c
=
c c

, )
0
*
j

c
=
c c

2
2
* 2
U
t m
i

c c
= + V
c
| |
|
\ .
c

Thus, the Euler-Lagrange equation

, )
, )
0
* * *
t j
t j

c c c
c c =
c c c c c

becomes
2
2
2
U
t m
i


c
= V +
| |

\
c
|
.

which is the desired Schrodinger equation.
Variation on gives
, )
*
t
i

c
=
c c

, )
2
*
2
j
j
m

c
= c
c c

[integration by part performed]
* U

c
=
c

2
* 0 *
2
*
j j t
m
U i + c c = c

or
2
2
*
*
2
U
t m
i


c
= V +
c
| |
|
\ .

The generalized momentum conjugate to is
, )
*
t
i
c
H
c
=
c
=

(6.26)
But the momentum conjugate to * vanishes identically. The Hamiltonian density
is
t
= H c
2
2
*
2
U
m

| |
= V +
|
\ .

so that the Hamiltonian becomes

3
2
2
*
2
U
m
H d x
| |
V + =
|
\ .
}

(6.27)
In the 2
nd
quantization, the wavefunctions and * are replaced by the field
operators and
+
. On the other hand, if we treat (6.25) as the action for a
classical field and apply the quantization rule to the conjugate variables , ) , t x and
, ) , t H x so that
, ) , ) , )

, , , t t i
1 ' ' H =
]
x x x x

we recover the commutation relation (6.16). Thus, 2

nd
quantization is equivalent to
quantizing a classical field; hence the name quantum field theory. Obviously,
nothing new was gained in the derivation of the Schrodinger equation. However,
new physics may be expected when the procedure is applied to other classical fields
such as the electromagnetic fields.
6.4. Second Quantization for Fermions