Perturbation Theory Examples

A. Charged particle in a one dimensional box

Consider a particle of mass m and charge q moving in the potential
0 V for x & x = < > a a and V f . The wavefunction is zero where the
potential is infinite and so we need only consider the motion between 0 & a. The
Schrodinger equation is with
0 0 or x =
0 0 0 0
N N N
H E =

2 2
0
2
2
d
m dx
=

H ,
0
2
N
N x
sin
a a

= and
2
2
0
2
N
N
E
m a

=

with . If we introduce a constant electric field, 1 N , = 2 3 , F Fx =

=

along the x axis the charged particle experiences an additional potential energy V q
and so the perturbed problem is
Fx
N

( )
0
N N

H qFx E = . The first order correction to the energy of the N

th
state is

( ) 1 0 0 0 0
2
N N N N N
qFa
E qFx qF x = = =

where we have used the fact that the average value of x is a/2 for all states. Note that each
state is shifted by the same amount and whether the energy increases or decrease depends
on the signs of q & F. The second order correction to the energy is given by

( )
( )
2 2
0 0 0 2
2
0 0 0 2 2
1
N K N
N
K N K N
N K
qFx x
qF
E
E E E N K


= =


0
K


where
2
2
0
1
2
E
m a

=

. The required matrix element is

0 0
2
0 0
2 2
a
N K
N x K x a
x x sin sin dx sin N sin K d
a a a

= =



Exercise: Evaluate
0
sin N sin K d

J. F. Harrison Michigan State University 9/19/2006 1
B. Harmonic Oscillator with a cubic perturbation

Suppose we are interested in estimating the eigenvalues and eigenvectors associated with
the Hamiltonian

2 2
2 3 0 3
2
1

2 2
d
H kx x H
m dx
x = + + = +

The first order correction to the energy of the state
0
n
is then zero by parity arguments.

(1) 0 3 0
0
n n n
E x = =

The second order correction is

2 2
2
0 0 0 3 0 3
(2)
0 0

( ) ( )
n p n p
n
n p
p n p n p n
V x n x p
E
h n p h n p
E E



= = =

So we need to evaluate matrix elements of
3
x between various harmonic oscillator states.
Using recursion relationships between the Hermite polynomials many texts show that

, 1 , 1
1
( )
2 2
j i j i ij
j j
i x j x

+
+
= + =

and this result allows us to evaluate matrix elements of
l
x for any positive integer value
of using the resolution of the identity. Since the set of eigenfunctions is complete
we may write
l
0
n

0 0
0 0

1
n n
n n
n n

= =
= =



So

2 2
0
( )
ij
n
i x j x i xx j i x n n x j

=
= = =

or in a more explicit matrix form

2
0
( ) ( ) ( )
ij in nj
n
x x x

=
=

Using the above result for ( )
ij
x the sum can be evaluated and we obtain

J. F. Harrison Michigan State University 9/19/2006 2
2
, 2 , 2 2 2
( 1) 2 1 ( 1)( 2)
( )
4 2 4
ij j i ij j i
i i i i i
x


+
+ + + +
= + +

Note the systematics. The matrix element ( )
ij
x is zero unless i & j differ by 1, say j is
equal to i . The matrix element 1& 1 i +
2
( )
ij
x is zero unless j equals i i . The
matrix element (
2, , 2 i +
3
)
ij
x will vanish unless 3, 1, 1& 3 i i i j i = + + whereas
4
( )
ij
x will
vanish except for , and so on. Returning to the perturbation sum
for
4, 2, , 2, i i i i = + + 4 j i
(2)
n
E we see

3 2 3 2 3 2 3 2 3 2
2 2
, 3 , 1 , 1 , 3 (2)
( ) ( ) ( ) ( ) ( )
( ) 3 1 1 3
np n n n n n n n n
n
p n
x x x x x
E
h n p h


+ +

= = + +

x


Now
3 2
, 3 , , 3
0
( ) ( ) ( )
n n n k k n
k
x x

+ +
=
=

and since ( ) vanishes unless
, 3 k n
x
+
2 k n or n+4 = + we have

3 2 2
, 3 , 4 4, 3 , 2 2, 3
( ) ( ) ( ) ( ) ( )
n n n n n n n n n n
x x x x x
+ + + + + +
= +
+


and since ( ) , we have
2
, 4
0
n n
x
+
=

3 2
, 3 , 2 2, 3 3
( 2)( 1)( 3
( ) ( ) ( )
(2 )
n n n n n n
n n n
x x x

+ + + +
) + + +
= =

In a similar fashion we find

3
3
, 1 3
( 1)
( ) 3
(2 )
n n
n
x

+
+
=

3
3
, 1 3
( ) 3
(2 )
n n
n
x

=

3
, 3 3
( 1) 2)
( )
(2 )
n n
n n n
x

=

Assembling these components we find

J. F. Harrison Michigan State University 9/19/2006 3
2
(2) 2
3
30
( 11/ 30)
(2 )
n
E n n
h


= + +

The first order correction to the wavefunction depends on the same matrix elements so

3 0 3 0 3 0 3 0 3 0
, 3 3 , 1 1 , 1 1 , 3 3 (1)
( ) ( ) ( ) ( ) ( )
( ) 3 1 1 3
np p n n n n n n n n n n n n
n
p n
x x x x x
h n p h


+ + + +

= = + + +

C. Harmonic Oscillator in a Constant Electric Field

Consider a one dimensional harmonic oscillator in a constant electric field , and let the
charge on the oscillator be q. If the oscillator is on the x axis, the Hamiltonian is
F

2 2
2
2
1

( )
2 2
d
H kx
m dx
= + +

q x

In one dimension
d
Fx
dx
x

= and since the field is constant this integrates to

( ) (0) x Fx Fx = where we will neglect the constant (0) which simply shifts the
zero of energy. We then seek the eigenvalues and eigenfunctions of the Hamiltonian

2 2
2
2
1

2 2
d
H kx
m dx
= +

qFx

We can solve this problem exactly and then compare the perturbation and exact results.
Lets solve it exactly.

First note that we can complete the square on the potential

2 2
2 2
1 2
2 2 2
k qF k qF qF
kx qFx x x x
k k


= =





k

and then define the new variable
qF
x
k
= , resulting in the Schrodinger equation

( )
2
2 2
2
2
1

( ) ( ) ( )
2 2 2
qF
d
H k
m d k
E

= + + =

or

J. F. Harrison Michigan State University 9/19/2006 4
( )
2
2 2
2
2
1
( ) ( )
2 2 2
qF
d
k E
m d k

+ = +

This is the harmonic oscillator equation, so, as we have seen above

2
1/ 2
( ) ( )
n n n
N H e



= and ( )
2
1
( )
2
n
2 E h n qF k = +

Note that all of the levels have been lowered by the same amount and the wavefunctions
have all been shifted along the x axis so that they are centered at
qF
x
k
= . If q and F are
both positive the equilibrium point is shifted in the x + direction as expected.
The explicit dependence of the first two wavefunctions on the electric field is

( )
( )
( ) ( )
( )
1/ 4
0
1/ 4
3
1
2
2
/ 2
/ 2
( )
4
( )
qF
x
k
qF
x
k
x
qF
x x
k
e
e

=
=



Now lets use perturbation theory to solve the problem.

The perturbation is and the first order correction to the energy is zero by parity.
The second order correction is then
qFx

( )
2 2
2
2 pn ( )
n
p n p n
x
p qFx n
( qF )
E
hv( n p ) h ( n p )

= =




From above the only terms that will appear in the summation are 1 p n = so we have

2 2 2
2 1 1
1 1
( ) n,n n.n
n
( x ) ( x ) ( qF )
E
h
+

= +

2 2
1 1
1
2 2
n,n n,n
n n
( x ) & ( x )

+
+
= =

so

( )
2
2
2
( )
n
qF
E
k
=

J. F. Harrison Michigan State University 9/19/2006 5
Note that because the first order correction to the energy vanishes, the third order
correction has the form

(3)
0 0
np pt tn
n
np nt
p n t n
x x x
E
E E

=

which is identically zero. Indeed all higher corrections must be

zero and the exact energy of the oscillator in the field is

( )
2
1
2
2
n
qF
E h ( n )
k
= + as found in the exact solution. In this problem the second order
correction is the total correction. The correction to the wavefunction is given by

( ) ( ) ( )
( )
0 0 0
1 1
, 1 , 1 (1) 0 0
1 1
1
1 1 2
p n n
np n n n n
n n n
p n
x x x
qF qF qF
n n
h n p h h
+
+
+

= = + =

+

Exercise:
We have seen that the exact solution for the ground state of the harmonic oscillator in a
constant electric field is

( )
( )
1/ 4
0
2
/ 2
( )
qF
x
k
x e

= while perturbation theory tells us that

(1) 0
1 0
2
qF
h

= Show that these two results are consistent. Hint: expand the exact
solution in a power series in the electric field.



J. F. Harrison Michigan State University 9/19/2006 6