Stark effect in hydrogen

Masatsugu Sei Suzuki

Department of Physics, SUNY at Binghamton
(Date: March 29, 2022)

Johannes Stark (15 April 1874 – 21 June 1957) was a German physicist, and Physics
Nobel Prize laureate who was closely involved with the Deutsche Physik movement
under the Nazi regime.
http://en.wikipedia.org/wiki/Johannes_Stark

Here we apply the perturbation theory to the Stark effect in hydrogen atom. The Stark
effect can be observed as a possible shift of the energy level, when an external electric
field is applied to hydrogen atom. The hydrogen atom has an electric dipole moment
because of the positive charge of proton and a negative charge of electron separated by a
distance which is the order of Bohr radius. When an external electric field is applied to
the hydrogen atom, the perturbation Hamiltonian appears as an inner product between the
electric dipole moment and the electric field. In the absence of an external electric field,
the energy level depends on the principal quantum number n;

R
En(0)   . (R = 13.605693122994 eV)
n2

The ground state with n = 1 (l = 0, m=0) is non-degenerate. On the other hand, the first
excited state with n = 2 (l = 1, m = 1, 0, -1; l = 0, m = 0) is degenerate (22 = 4
degeneracy). The second excited state with n = 3 (l = 2, m = 2, 1, 0, -1, -2; l = 1, m = 1, 0,
-1; l = 0, m = 0) is degenerate (32 = 9 degeneracy).
In order to evaluate the shift of energy levels in the presence of electric field, we need
to apply the perturbation theory (nondegenerate case) to the ground state and the

1
perturbation theory (degenerate case) to the excited states. In these cases, it is significant
to calculate the matrix elements.
We are also interested in the change of the probability density of the wave function of
hydrogen atom, when the electric field is applied. It is expected that the ContourPlot of
the probability density appropriately changes depending on the magnitude and direction
of the electric field.

1 Hamiltonian for the hydrogen atom in the presence of an electric field

ˆ
H0 is the Hamiltonian of the hydrogen atom. We apply an external electric field 
(along the z axis) to the hydrogen atom, producing the Stark effect.

Hˆ  Hˆ 0  Hˆ 1 .

Hˆ 1   μˆ e    (erˆ)  e z  ezˆ .

where q = -e (e>0) is the electron charge and µe (=-er) is an electric dipole moment. The
vector r is the position vector of electron. The proton (charge e) is located at the origin.
The eigenstate of Hˆ 0 is given by n, l , m with the energy

R
En( 0)   .
n2

where R is the Rydberg constant. R = 13.60569193 eV.

In the presence of Hˆ 1 , the full spherical symmetry of the Hamiltonian is destroyed by
the external electric field that selects the positive z-direction, but Hˆ is still invariant
under the rotation around the z axis.

 '  R̂z  ,

 ' Hˆ  '   Hˆ  ,

or

 Rˆ z  Hˆ Rˆ z    Hˆ  ,

or

Rˆ z Hˆ Rˆ z  Hˆ ,

or

[ Hˆ , Rˆ z ]  0 .

2
Since

i i
Rˆ z  exp[ Lˆz ]  1̂  Lˆz .
ℏ ℏ

We have

[ Hˆ , Lˆz ]  0 ,

or

[ Hˆ 1, Lˆz ]  0 .

Since

Hˆ 1  ezˆ

we have

[ Lˆz , zˆ]  0 .

since

Lˆz  xˆpˆ y  yˆpˆ x

2 Selection rules
The selection rules are summarized as follows.

(i) Selection rule-1

n, l , m zˆ n' , l ' , m'  0 ,

only for m' = m.

(ii) Selection rule-2

n, l , m xˆ  iyˆ n' , l ' , m'  0 .

only for m'  m  1 .

(iii) Selection rule-3

3
 n, l , m zˆ n' , l ' , m'  0

unless l’ = l ± 1.

3. Derivation of the selection rules from commutation relations

(a) Selection rule 1
We note that the commutation relation [ Lˆz , zˆ]  0 , can be derived from the original
definition of the angular momentum.

Lˆz  xˆpˆ y  yˆ pˆ x

with

[ Lˆz , zˆ]  0 .

Using the commutation relation, we calculate the matrix element;

n, l , m [ Lˆz , zˆ] n' , l ' , m'  n, l , m Lˆz zˆ  zˆLˆz ] n' , l ' , m'
.
 (m  m' )ℏ n, l , m zˆ n' , l ' , m'  0

Thus, only for m’ = m.

n, l , m zˆ n ' , l ' , m'  0

(b) Selection rule 2

Lˆz  xˆpˆ y  yˆ pˆ x

with

[ Lˆ z , xˆ ]  iℏyˆ . [ Lˆ z , yˆ ]  iℏxˆ .

From this commutation relation, we have

n, l , m [ Lˆz , xˆ ] n ' , l ' , m'  n, l , m Lˆz xˆ  xˆLˆ z ] n' , l ' , m'

 (m  m' )ℏ n, l , m xˆ n' , l ' , m' (1)
 iℏ n, l , m yˆ n ' , l ' , m'

and

4
 n, l , m [ Lˆz , yˆ ] n ' , l ' , m'  n, l , m Lˆ z yˆ  yˆ Lˆ z ] n' , l ' , m'
 (m  m' )ℏ n, l , m yˆ n ' , l ' , m' (2)
 iℏ n, l , m xˆ n' , l ' , m'

From Eqs.(1) and (2), we get

(m  m' ) 2 n, l , m xˆ n' , l ' , m'  i( m  m' ) n, l , m yˆ n' , l ' , m'

 n, l , m xˆ n' , l ' , m'
For m'  m  1

n, l , m xˆ  iyˆ n' , l ' , m'  0 .

For m'  m  1

n, l , m xˆ n ' , l ' , m'  n, l , m yˆ n' , l ' , m'  0

(c) Selection rule 3

We can show the commutation relation

[ Lˆ2 , [ Lˆ2 , rˆ]]  2ℏ 2 ( Lˆ2 rˆ  rˆLˆ2 )

by using the two methods (i) the commutation relations and (ii) Mathematica.

Using this relation we can calculate the matrix element of the commutation relation

n, l , m [ Lˆ2 , [ Lˆ2 , rˆ]] n ' , l ' , m'  2ℏ 2 n, l , m ( Lˆ2 rˆ  rˆLˆ2 ) n ' , l ' , m'
2
.
 2ℏ 4 [l (l  1)  l ' (l ' 1)] n, l , m rˆ n' , l ' , m'

Noting that

2
2ℏ 2 n, l , m ( Lˆ2 rˆ  rˆLˆ2 ) n' , l ' , m'  2ℏ 4 [l (l  1)  l ' (l '1)] n, l , m rˆ n ' , l ' , m'

and

n, l , m [ Lˆ2 , [ Lˆ2 , rˆ]] n' , l ' , m'  ℏ 2 [l (l  1)  l ' (l ' 1)] n, l , m [ Lˆ2 , rˆ]] n' , l ' , m'
 ℏ 4 [l (l  1)  l ' (l ' 1)]2 [l (l  1) n, l , m rˆ n' , l ' , m'

Then we get

{[l (l  1)  l ' (l ' 1)]2  2l (l  1)  2l ' (l '1)} n, l , m rˆ n' , l ' , m'  0 .

5
or

(l  l '1)(l  l '1)(l  l ' )(l  l '2) n, l , m rˆ n ' , l ' , m'  0

This leads to the selection rule

n, l , m rˆ n ' , l ' , m'  0

only for l’ = l ± 1.

((Proof))

(i) Commutation relations

[ Lˆ2 , [ Lˆ2 , zˆ]  2ℏ 2 ( zˆLˆ2  Lˆ2 zˆ)

((Proof)) Griffiths, Introduction to Quantum Mechanics

We prove this by using a various kind of commutation relations.

[ Lˆ2 , zˆ ]  [ Lˆ x , zˆ ]  [ Lˆ y , zˆ ]  [ Lˆ z , zˆ]
2 2 2

 Lˆ x [ Lˆ x , zˆ]  [ Lˆ x , zˆ ]Lˆ x  Lˆ y [ Lˆ y , zˆ ]  [ Lˆ y , zˆ ]Lˆ y  Lˆ z [ Lˆ z , zˆ]  [ Lˆ z , zˆ ]Lˆ z

 iℏ(  Lˆ x yˆ  yˆ Lˆ x  Lˆ y xˆ  xˆLˆ y )

We note that

Lˆ x yˆ  [ Lˆ x , yˆ ]  yˆ Lˆ x  iℏzˆ  yˆ Lˆ x

Lˆ y xˆ  [ Lˆ y , xˆ ]  xˆLˆ y  iℏzˆ   xˆLˆ y

Then we get

[ Lˆ2 , zˆ ]  iℏ[(iℏzˆ  yˆ Lˆ x )  yˆ Lˆ x  ( iℏzˆ  xˆLˆ y )  xˆLˆ y )

 2iℏ( xˆLˆ y  yˆ Lˆ x  iℏzˆ )

Similarly,

[ Lˆ2 , xˆ ]  2iℏ ( yˆLˆz  zˆLˆ y  iℏxˆ )

and

6
 [ Lˆ2 , y ]  2iℏ( zLˆx  xLˆ z  iℏyˆ )

where

[ Lˆ x , zˆ]  [ yˆ pˆ z  zˆpˆ y , zˆ]  [ yˆ pˆ z , zˆ]  [ zˆpˆ y , zˆ]  yˆ [ pˆ z , zˆ]  iℏyˆ .

ℏ
[ Lˆ y , zˆ ]  [ zˆpˆ x  xˆpˆ z , zˆ]  [ xpˆ z , zˆ ]   xˆ[ pˆ z , zˆ ]   xˆ  iℏxˆ .
i

[ Lˆ z , zˆ]  [ xˆpˆ y  yˆ pˆ x , zˆ]  [ yˆ pˆ x , zˆ]  0 .

[ Lˆ x , xˆ ]  [ yˆ pˆ z  zˆpˆ y , xˆ ]  0 .

ℏ
[ Lˆ y , xˆ ]  [ zˆpˆ x  xˆpˆ z , xˆ ]  [ zˆpˆ x , xˆ ]  zˆ  iℏzˆ .
i

[ Lˆ z , xˆ ]  [ xˆpˆ y  yˆ pˆ x , xˆ ]  [ yˆ pˆ x , xˆ ]  iℏyˆ .

[ Lˆ x , yˆ ]  [ yˆ pˆ z  zˆpˆ y , yˆ ]  iℏzˆ .

[ Lˆ y , yˆ ]  [ zˆpˆ x  xˆpˆ z , yˆ ]  0 .

[ Lˆ z , yˆ ]  [ xˆpˆ y  yˆ pˆ x , yˆ ]  iℏxˆ .
________________________________________________________________________
Then we get

7
 [ Lˆ2 ,[ Lˆ2 , zˆ]  2iℏ{[ Lˆ2 , xˆLˆ y ]  [ Lˆ2 , yˆ Lˆx ]  iℏ[ Lˆ2 , zˆ]}
 2iℏ{[ Lˆ2 , xˆ ]Lˆ y  xˆ[ Lˆ2 , Lˆ y ]  [ Lˆ2 , yˆ ]Lˆ x  yˆ[ Lˆ2 , Lˆx ]  iℏ[ Lˆ2 , zˆ]}
 2iℏ{2iℏ( yˆLˆz  zˆLˆ y  iℏxˆ ) Lˆ y  2iℏ ( zˆLˆ x  xˆLˆz  iℏyˆ ) Lˆx
 iℏ( Lˆ2 zˆ  zˆLˆ2 )}
 2ℏ 2{2( yˆLˆ z Lˆ y  zˆLˆ y  iℏxˆLˆ y )  2( zˆLˆ x  xˆLˆ z Lˆx  iℏyˆLˆ x )
2 2

 ( Lˆ2 zˆ  zˆLˆ2}
 2ℏ 2{2( yˆLˆ z  iℏxˆ ) Lˆ y  2 zˆ( Lˆx  Lˆ y )  2( xˆLˆz  iℏyˆ ) Lˆx )
2 2

 ( Lˆ2 zˆ  zˆLˆ2 )}
 2ℏ 2{2 Lˆz yˆ Lˆ y  2 zˆ( Lˆ2  Lˆ z )  2 Lˆz xˆLˆx
2

 ( Lˆ2 zˆ  zˆLˆ2 )}
 2ℏ 2{2 Lˆz ( xˆLˆ x  yˆ Lˆ y  zˆLˆz )  2 Lˆz zˆLˆz  2 zˆ ( Lˆ2  Lˆz )  ( Lˆ2 zˆ  zˆLˆ2 }
2

or

[ Lˆ2 ,[ Lˆ2 , zˆ]  2ℏ 2{2 Lˆz zˆLˆz  2 zˆLˆ2  2 zˆLˆ z  ( Lˆ2 zˆ  zˆLˆ2}
2

 2ℏ 2 ( zˆLˆ2  Lˆ2 zˆ)

where we use the relations,

xˆLˆ x  yˆ Lˆ y  zˆLˆ z  0

[ Lˆ2 , Lˆ x ]  [ Lˆ2 , Lˆ y ]  [ Lˆ2 , Lˆ z ]  0

[ Lˆz , zˆ]  0

and

[ Lˆ2 , zˆ]  2iℏ( xˆLˆ y  yˆ Lˆ x  iℏzˆ )

[ Lˆ2 , xˆ ]  2iℏ( yˆ Lˆ z  zˆLˆ y  iℏxˆ )
[ Lˆ2 , yˆ ]  2iℏ( zˆLˆ x  xˆLˆ z  iℏyˆ )

(ii) Mathematica
Using the Mathematica, we show that the above commutations are valid.

[ Lˆz , zˆ]  0 .

8
 [ Lˆ z , xˆ ]  iℏyˆ . [ Lˆ z , yˆ ]  iℏxˆ .

[ Lˆ2 , [ Lˆ2 , rˆ]]  2ℏ 2 ( Lˆ2 rˆ  rˆLˆ2 )

((Mathematica))

Clear@"Global`"D;

<< VectorAnalysis`;

SetCoordinates@Cartesian@x, y, zDD;

ux = 81, 0, 0<; uy = 8 0, 1, 0<; uz = 80, 0, 1<; r = 8x, y, z<;

Lx := Hux.H− — Cross@r, Grad @ DDL &L êê Simplify;

Ly := Huy.H− — Cross@r, Grad @ DDL &L êê Simplify;
Lz := Huz.H− — Cross@r, Grad @ DDL &L êê Simplify;

Lsq := H Nest@Lx, , 2D + Nest@Ly, , 2D + Nest@Lz, , 2DL &;

eq1 = HLz@z χ @x, y, zDD − z Lz@ χ @x, y, zDDL êê FullSimplify

0

eq2 = Lz@x χ @x, y, zDD − x Lz@ χ @x, y, zDD − — y χ @x, y, zD êê Simplify

0

eq3 = Lz@y χ @x, y, zDD − y Lz@ χ @x, y, zDD + — x χ @x, y, zD êê Simplify

0

eq4 = 2 —2 HLsq @x χ @x, y, zDD + x Lsq @ χ @x, y, zDDL êê Simplify;

eq5 = Lsq @Lsq @x χ @x, y, zDDD − Lsq @x Lsq @ χ @x, y, zDDD − Lsq @x Lsq @ χ @x, y, zDDD +
x Lsq @ Lsq @ χ @x, y, zDDD êê Simplify;

eq4 − eq5
0

4. Selection rule derived from the parity operator

 is the parity operator:
ˆ

ˆ   
 ˆ 1 ,
ˆ  

ˆz is the parity odd operator with


ˆzˆ
ˆ   zˆ ,

and

9
 ˆ n, l , m  (1)l n, l , m ,

or

n, l , m ˆ  ( 1)l n, l , m .

Then we have

n, l , m zˆ n' , l ' , m'  0 for the l-state and l'-state with the same parity.

The reason is as follows.

n, l , m ˆzˆˆ n' , l ' , m'   n, l , m zˆ n' , l ' , m' ,

or

n, l , m zˆ n' , l ' , m'  (1)l  l ' 1 n, l , m zˆ n' , l ' , m' .

When l  l' 1  2k  1(odd numbers), or l  l'  2k (even number), we have

n, l , m zˆ n' , l ' , m'  0 .

_____________________________________________________________________
5. The Stark effect on the n = 1 level
The ground state is non-degenerate.

 0  n  1, l  0, m  0

E1H0L + E1H2L

H0 L
E1
»y0 >

Fig. Shift of the energy level of the ground state in the presence of electric field.

The energy to the first order:

E1 ( 0)  R

E1   0 Hˆ 1  0  1,0,0 Hˆ 1 1,0,0  0
(1)

10
 2
1,0,0 zˆ n, l , m
e  
(2) 2 2
E1
E1  En
(0 ) ( 0)
n 1, l , m

where

R
En( 0)  
n2

Then we have

2
1 1,0,0 zˆ n, l , m
E1  E    2  e2 2 
( 2)
1
E1  En
(0) ( 0)
2 n 1, l , m

or

2
1,0,0 zˆ n, l , m
  2e 2

n 1, l , m E1  En
( 0) (0 )

The proceeding sum is certainly not zero, since there exist states n, l , m whose parity is
opposite to that of 1,0,0 . To the lowest order in , the Stark shift of the 1s ground state
is quadratic.

((Polarizability))
Electric polarizability is the relative tendency of a charge distribution, like the
electron cloud of an atom or molecule, to be distorted from its normal shape by an
external electric field, which is applied typically by inserting the molecule in a charged
parallel-plate capacitor, but may also be caused by the presence of a nearby ion or dipole.
The electronic polarizability  is defined as the ratio of the induced dipole moment P of
an atom to the electric field  that produces this dipole moment.

p  

The work done on the system as  slightly changes to  +d,

dW   pd  d

or

1
W  K    2
2

from the work-energy theorem. So that, we have

11
 1
E  K    2
2

6 Polarizability of the 1s-state

n, l , m zˆ 1,0,0
 1s  1,0,0  e  n, l , m (0) ( 0)
 ...
n 1 E1  En
l ,m

2
n, l , m zˆ 1,0,0
p   1s (ezˆ )  1s  2e   2
( 0)
 
 En
(0)
n 1 E1
l ,m

or

2
n, l , m zˆ 1,0,0
   2e 2

n 1 E1
( 0)
 En
( 0)

l ,m

where q   e . Under the perturbation, the energy shift is given by

2
n,1, m zˆ 1,0,0  2
E  e 
2 2

n 1 E1
(0)
 En
( 0)

2
l ,m

((Note-1))

*
n, l , m zˆ 1,0,0
 1s (ezˆ)  1s  ( 1,0,0  e  n, l , m  ...)(ezˆ)
 En )
( 0) (0 )
n 1 ( E1
l,m

n, l , m zˆ 1,0,0
 ( 1,0,0  e  n, l , m (0 ) (0)
 ...)
n 1 ( E1  En )
l,m

The electric field  causes an induced dipole moment to appear, proportional to .

((Note-2))
Since

1,0,0 zˆ 1,0,0  0 and En  E1  E2  E1  0

( 0) ( 0) ( 0) (0)

12
we have

2
n, l , m zˆ 1, 0, 0
  2e 2

n 1 En (0)  E1(0)
l ,m

2e 2

2
 n, l , m zˆ 1, 0, 0
E2 (0)  E1(0) n 1
l ,m

2e 2

2
 n, l , m zˆ 1, 0, 0
E2 (0)  E1(0) n,
l ,m

Here

   1,0,0 zˆ n, l , m n, l , m zˆ 1,0,0  1,0,0 zˆ 2 1,0,0

2
n, l , m zˆ 1,0,0
n, n,
l ,m l ,m

Then we have

2e 2 2e 2

2
 (0 ) (0)
n, l , m zˆ 1,0,0  (0) ( 0)
1,0,0 zˆ 2 1,0,0
E2  E1 n, E2  E1
l ,m

or

2e 2 16 3
 2
a0 2  a0  5.33a0 3
e 1 3
(1  )
2a0 4

which is consistent with the experimentally observed value:  = 4.5 a03.

1,0,0 zˆ 2 1,0,0  a0
2

((Bethe-Salpeter))

Hans Albrecht Bethe (July 2, 1906 – March 6, 2005) was a German-American physicist,
and Nobel laureate in physics for his work on the theory of stellar nucleosynthesis. A
versatile theoretical physicist, Bethe also made important contributions to quantum
electrodynamics, nuclear physics, solid-state physics and particle astrophysics. For most
of his career, Bethe was a professor at Cornell University.

13
http://en.wikipedia.org/wiki/Hans_Bethe
________________________________________________________________________
How can we calculate the exact value of ?
2 2
n, l , m zˆ 1, 0, 0 n, l , m zˆ 1, 0, 0
  2e 2

n 1 En (0)  E1(0)
 2e 2

n 1 En (0)  E1(0)
l ,m l ,m

n, l , m zˆ 1,0, 0   d 3rRnl * (r )Yl m* ( ,  )r cos  [ R10 (r )Y00 ( ,  )]

Here

1 4 0
Y00 ( ,  )  , cos   Y1 ( ,  )
4 3


1 0
n, l , m zˆ 1,0, 0   d Yl m* ( ,  ) Y1 ( ,  )  r 3 drRnl (r ) R10 (r )
3 0

1 0 1
 d Y ( ,  ) Y1 ( ,  )   l ,1 m ,0
m*
l
3 3

Then we have

14
 
1
n, l , m zˆ 1,0,0   l ,1 m, 0  r 3drRn1 (r ) R10 (r )
3 0

or


2 1
 [  r 3drRn1 (r ) R10 (r )]2  a0 f (n)
2
n,1,0 zˆ 1,0,0
3 0

where f(n) is obtained by H.A. Bethe and E.E. Salpeter [Quantum Mechanics of One- and
Two Electron Atoms, Academic Press, New York, 1957, p.262]

1 28 n 7 (n  1)2 n 5
f ( n) 
3 (n  1)2 n 5

m0 e 4 e2
En    
2n 2 ℏ 2 2n 2 a0

e2 1
En (0)
E1
(0)
 (1  2 )
2a0 n

Then we have
2
n,1, 0 zˆ 1, 0, 0 
n 2 f ( n)
  2e 2   4a03   4a0 3 0.915806  3.66326a03
n 1 En (0)  E1(0) n 2 n 2
 1

((Mathematica))

15
Stark effect with n = 1

Clear@"Global`∗"D;

1 − −3 − r 2r
R @n_, _, r_D := 21+ a0 2 a0 n n − −2 r Hn − − 1L ! LaguerreLB− 1 + n − , 1 + 2 , F ;
Hn + L ! a0 n
Y@ _, m_, θ_, φ_D := SphericalHarmonicY@ , m , θ , φ D;
ψ @n_, _, m_, r_, θ_, φ_D := R @n , , r D Y@ , m , θ , φ D
f@n1_, 1_, m1_, n2_, 2_, m2_, r_, θ_, φ_D =
H− 1L m1 ψ @n1, 1, − m1, r, θ, φD r Cos@θD ψ @n2, 2, m2, r, θ, φD r2 Sin@θD êê Simplify;

Integral calculation

g@n1_, 1_, m1_, n2_, 2_, m2_D :=

f@n1, 1, m1, n2 , 2 , m2 , r, θ, φD
SimplifyBIntegrateBIntegrateBIntegrateB , 8φ, 0, 2 π <F,
a0
8θ, 0, π <F, 8r, 0, ∞<F, a0 > 0F;

Beth' s formula

1 28 n7 Hn − 1L2 n−5
h@n_D = ;
3 Hn + 1L2 n+5

TableA9n, 1, g@n, 1, 0, 1, 0, 0D2 êê N, h@nD êê N=, 8n, 2, 10<E êê TableForm

2 1 0.554929 0.554929
3 1 0.0889893 0.0889893
4 1 0.0309238 0.0309238
5 1 0.0145191 0.0145191
6 1 0.00802234 0.00802234
7 1 0.00491424 0.00491424
8 1 0.00323396 0.00323396
9 1 0.00224381 0.00224381
10 1 0.00162158 0.00162158

∞ n2 h@nD
4‚ êê N
n=2 n2 − 1

3.66326

_______________________________________________________________________
7 Stark effect on the n = 2 level
We now consider the state with n = 2.

n = 2 state (4 states-degeneracy):

l = 1 (m = ±1, 0): p-state (3 states)

l = 0 (m = 0): s-state (1 state)

Note that

R
E2( 0)  
22

is the eigenvalue of Hˆ 0 . The degenerate system with the four states:

n, l , m  2,0,0 , 2,1,1 , 2,1,0 , 2,1,1

with the same energy. For convenience we introduce the basis as

16
  1(0)  2, 0, 0 even state

 2(0)  2,1,1
 3(0)  2,1, 0 odd states
 4(0)  2,1, 1

Here we show the cross section of the probability density for n = 2 and l = 1 in the
absence of external electric field.

________________________________________________________________________
State 2,1, 0

________________________________________________________________________
State 2,1,1

17
From the selection rule, we have

2,1, m zˆ 2, 0, m '  2,1, m zˆ 2, 0, m  m ,m'

2,1, m zˆ 2,1, m '  0

2, 0, 0 zˆ n, 0, 0  0

The matrix of Hˆ 1 based on these bases is given by

 0 0 ( Hˆ 1 )13 0
 
 0 0 0 0
 
(Hˆ 1 )31 0 0 0
 
 0 0 0 0

where

( Hˆ 1 ) 31  e  3 zˆ  1  3ea0   E0
(0) ( 0)

or

E0  3ea0 (>0)

Note that

18
  3 (0) zˆ  1(0)  2,1, 0 zˆ 2, 0, 0
  dr 2,1, 0 r r zˆ 2, 0, 0
  r cos  R21 (r )*[Y10 ( ,  )]* R20 (r )Y00 ( ,  )r 2 sin  drd d
 3a0

Matrix elements of n, l ' , m' Hˆ 1 n, l , m is given by

2,1,1 2,1, 0 2,1, 1 2, 0, 0

2,1,1 0 0 0 0
2,1, 0 0 0 0  E0
2,1, 1 0 0 0 0
2, 0, 0 0  E0 0 0

where

2,1, 0 Hˆ 2, 0, 0  3e a0   E0

The reduced matrix:

2,1, 0 2, 0, 0
2,1, 0 0  E0
2, 0, 0  E0 0

We find that

Hˆ 1  1   E0  3
( 0) (0 )

Hˆ 1  2  0
(0)

Hˆ 1  3   E0  1
( 0) (0 )

Hˆ 1  4  0
(0)

2 and  4 are the eigenstates of Hˆ 1 with the energy 0.

(0) (0)

19
We now consider the matrix of Hˆ 1 in terms of the basis  1 and  3
( 0) ( 0)

 0  E0 
Hˆ 1r   
  E0 0 

1  Uˆ  1 and 3  Uˆ  3
( 0) (0)

with

 1 1 
U U12   
Uˆ   11  2 2 
 U 21 U 22   1

1 
 
 2 2

For  = -E0 (the lowest level)

 1 
 
 U11   2
1  Uˆ  1(0)    
U12   1 

 2

For  = E0, (the highest level)

 1 
 
 U12   2 
3  Uˆ  3(0 )    
U 22    1 
 2

The degenerate level of n = 2 splits into the three levels:

(i) The ground state: E  E 2 (0 )  E0

1
1  ( 1   3 )
(0 ) (0)

(ii) The first excited state with E2( 0) (double-degeneracy)

2 and  4
(0) (0)

(iii) The second excited state with E2( 0)  E0

20
 1
3  ( 1   3 )
( 0) ( 0)

»f3 >
E2H0L +E0

E2H0L »y2H0 L >,»y4H0L >

E2H0 L

H0L
E2 -E0
»f1 >

Fig. Energy splitting (Stark effect with n = 2). E0  3ea0 .

8 Charge density distribution for the Stark effect with n = 2

The charge density distribution for the 1 ,  3 ,  2 and  4
(0) (0)
is evaluated
from the CountourPlot (Mathematica) of

2 2 2 2
r 1 , r 3 , r 2 , and r  4
(0) (0)

where y = 0, in the x-z plane.

21
 10

10
10 5 0 5 10

2
ContourPlot of r 1 with y =0, in the x-z plane. When the electric field is applied
along the z axis, the average position of electrons shifts to the (-z) direction. The energy
eigenvalue is E  E 2 (0 )  E0 .

22
 10

10
10 5 0 5 10

2
ContourPlot of r 3 with y =0, in the x-z plane. When the electric field is applied
along the z axis, the average position of electrons shifts to the z direction. The energy
( 0)
eigenvalue is E  E2  E0 .

23
 10

10
10 5 0 5 10

2 2
ContourPlot of r  2  r 4
( 0) ( 0)
with y =0, in the x-z plane. When the electric field
is applied along the z axis, the average position of electrons remains unshifted in the
(0)
direction to the z axis. The energy eigenvalue is E  E2 .

Two of the four degenerate states for n = 2 (  2 and  4

(0) (0)
) are unaffected by the
electric field to the first order, and the other two linear combinations

1
1  (  1   3 ) (E = E2(0)  E0 ),
(0) ( 0)

and

1
3  (  1   3 ) (E = E2 (0 )  E0 ).
( 0) ( 0)

This means that the hydrogen atom in this unperturbed state behaves as though it has a
permanent electric-dipole moment of magnitude 3ea0, which can be oriented in three
different ways; one state parallel to the external electric field, one state anti-parallel to the
field, two states with zero component along the field (Schiff).
____________________________________________________________________
((Mathematica)) The eigenvalue problem for n = 2 is solved using the Mathematica.

24
Calculation of matrix element for the Stark effect with n = 1

R @n_, _, r_D :=
1

Hn + L !
− −3 − r
21+ a0 2 a0 n n − −2 r Hn − − 1L !

2r
LaguerreLB− 1 + n − , 1 + 2 , F
a0 n
Y@ _, m_, θ_, φ_D := SphericalHarmonicY@ , m , θ , φ D;
ψ @n_, _, m_, r_, θ_, φ_D := R @n , , r D Y@ , m , θ , φ D;
f@n1_, 1_, m1_, n2_, 2_, m2_, r_, θ_, φ_D =
H− 1L m1 ψ @n1, 1, − m1, r, θ, φD r Cos@θD ψ @n2, 2, m2, r, θ, φD
r2 Sin@θD êê Simplify;

Simplify@
Integrate@
Integrate@Integrate@f@2, 1, 0, 2, 0, 0, r, θ, φD,
8φ, 0, 2 π <D, 8θ, 0, π <D, 8r, 0, ∞<D, a0 > 0D
− 3 a0

25
E0 = 3 e a0 ¶; Eigenvalue problem for the simplified system

H22 = 880, − E0<, 8− E0, 0<<

880, − E0<, 8− E0, 0<<

eq1 = Eigensystem @H22D

88− E0, E0<, 881, 1<, 8− 1, 1<<<

ψ1 = Normalize@eq1@@2, 1DDD
1 1
9 , =
2 2

ψ2 = − Normalize@eq1@@2, 2DDD
1 1
9 , − =
2 2

UT = 8ψ1, ψ2<
1 1 1 1
99 , =, 9 ,− ==
2 2 2 2

U = Transpose@UTD
1 1 1 1
99 , =, 9 ,− ==
2 2 2 2

UH = UT
1 1 1 1
99 , =, 9 ,− ==
2 2 2 2

UH.U
881, 0<, 80, 1<<

26
________________________________________________________________________
9 n = 3 Stark effect
We consider the case of n = 3.

n = 3 state (9 states degenerate):

l = 2 (m = ±2, ±1, 0): d-state (5 states)

l = 1 (m = ±1, 0): p-state 3 states)
l = 0 (m = 0): s-state (1 state)

Note that

( 0) R
E3 
32

is the eigenvalue of Hˆ 0 .

Matrix elements of H1:

3,2,2 3,2,1 3,2,0 3,2,1 3,2,2 3,1,1 3,1,0 3,1,1 3,0,0

3,2,2 0 0 0 0 0 0 0 0 0
9
3,2,1 0 0 0 0 0  ea0 0 0 0
2
3,2,0 0 0 0 0 0 0  3 3ea0 0 0
9
3,2,1 0 0 0 0 0 0 0  ea0 0
2
3,2,2 0 0 0 0 0 0 0 0 0
9
3,1,1 0  ea0 0 0 0 0 0 0 0
2
3,1,0 0 0  3 3ea0 0 0 0 0 0  3 6ea0
9
3,1,1 0 0 0  ea0 0 0 0 0 0
2
3,0,0 0 0 0 0 0 0  3 6ea0 0 0

where

9 9
3,2,1 Hˆ 1 3,1,1   ea0 , 3,2,0 Hˆ 1 3,1,0  3 3ea0 , 3,2,1 zˆ 3,1,1   ea0 ,
2 2
3,1,0 zˆ 3,0,0  3 6ea0 .

Note that

27
 Hˆ 1 3,2,2  0 , Hˆ 1 3,2,2  0

Thus 3, 2,2 and 3, 2,2 are eigenstates of H1 with the zero energy. So we consider the
matrix under the basis { 3, 2,1 , 3,2,0 , 3, 2,1 , 3,1,1 , 3,1,0 , 3,1, 1 , 3,0,0 }.

3,2,1 3,2,0 3,2,1 3,1,1 3,1,0 3,1,1 3,0,0

9
3,2,1 0 0 0  ea0 0 0 0
2
3,2,0 0 0 0 0  3 3ea0 0 0
9
3,2,1 0 0 0 0 0  ea0 0
2

9
3,1,1  ea0 0 0 0 0 0 0
2
3,1,0 0  3 3ea0 0 0 0 0  3 6ea0
9
3,1,1 0 0  ea0 0 0 0 0
2
3,0,0 0 0 0 0  3 6ea0 0 0

This matrix consists of three submnatrices.

(i)
3,2,0 3,1,0 3,0,0
3,2,0 0  3 3ea0 0
3,1,0  3 3ea0 0  3 6ea0
3,0,0 0  3 6ea0 0

or

 0  3 3ea0 0 
 
M 1    3 3ea0 0  3 6ea0 
 
 0  3 6ea0 0 

Eigensystem[M1] (Mathematica is used for the calculation)

1 1 3
E1 = 9ea0 = 3E0 1  [ 3,2,0  3,1,0  3,0,0 ]
3 2 2
1
E2= 0, 2  [ 2 3,2,0  3,0,0 ]
3

28
 1 1 3
E3 = -9ea0 = -3E0 3  [ 3,2,0  3,1,0  3,0,0 ]
3 2 2

(ii)
3,2,1 3,1,1
9
3,2,1 0  ea0
2
9
3,1,1  ea0 0
2

 9 
 0  ea0 
M2   2 
  9 ea 0 
 0
 2 

Eigensystem[M1] (Mathematica is used for the calculation)

9 3 1
E4 = ea0 = E0 4  [ 3,2,1  3,1,1 ]
2 2 2
9 3 1
E5 = - ea0= - E0 5  [[ 3,2,1  3,1,1 ]]
2 2 2

(iii)
3,2,1 3,1,1
9
3,2,1 0  ea0
2
9
3,1,1  ea0 0
2

 9 
 0  ea0 
M3   2 
  9 ea 0 
 0
 2 

Eigensystem[M3] (Mathematica is used for the calculation)

9 3 1
E4 = ea0 = E0 6  [ 3,2,1  3,1,1 ]
2 2 2
9 3 1
E5 = - ea0 = - E0 7  [[ 3,2,1  3,1,1 ]]
2 2 2

29
 »y1 > H0 L
E3 +3.0E0

»y4 >, »y6 > H0 L

E3 +1.5E0

H0L
E3 »»y2 > H0 L
E
»»3,2,2>, »3,2,-2> 3

H0 L
E3 -1.5E0
»y5 >, »y7 >

H0 L
E3 -3.0E0
»y3 >

Fig. Energy splitting (Stark effect with n = 3). E0  3ea0 .

10 Charge density distribution for the Stark effect with n = 3

First we show the cross section of the probability density of n = 3 and l = 1 in the absence
of electric field.

State 3,1, 0

30
________________________________________________________________________
State 3,1,1

31
Next we show the cross section of the probability density of n = 3 and l = 1 in the
presence of an electric field.

____________________________________________________________________

32
 20

10

10

20
20 10 0 10 20

2
ContourPlot of r  1 with y =0, in the x-z plane. The energy eigenvalue is E3( 0)  3E0 .

33
 20

10

10

20
20 10 0 10 20

2
ContourPlot of r  3 with y =0, in the x-z plane. The energy eigenvalue is E3( 0)  3E0 .

34
 20

10

10

20
20 10 0 10 20

2
ContourPlot of r  2 with y =0, in the x-z plane. The energy eigenvalue is E3( 0)  3E0 .

35
 20

10

10

20
20 10 0 10 20

2 2
ContourPlot of r  7 and r  5 with y =0, in the x-z plane. The energy eigenvalue
( 0)
is E
3  1.5E0 .

36
 20

10

10

20
20 10 0 10 20

2 2
ContourPlot of r  6 and r  4 with y =0, in the x-z plane. The energy eigenvalue
( 0)
is E
3  1.5 E0 .

37
 20

10

10

20
20 10 0 10 20

2 2
ContourPlot of r 3,2,2 and r 3,2,2 with y =0, in the x-z plane. The energy
( 0)
eigenvalue is E 3 .

REFERENCES
L.I. Schiff, Quantum Mechanics (McGraw-Hill, New York, 1995).
H.A. Bethe and E.E. Salpeter, Quantum Mechanics of One- and Two Electron Atoms,
Academic Press, New York, 1957, p.262]
Stephen Gasiorowicz, Quantum Physics, 3rd edition (John Wiley & Sons, 2003).
____________________________________________________________________

APPENDIX-I

The wavefunction of hydrogen atom:

38
 nlm (r , ,  )  r n, l , m
(n  l  1)! 1l  l 3/ 2 r 2r m
 2 a0 exp( )n l  2 r l L2nll11 ( )Yl ( ,  )
(n  l )! na 0 na0

The matrix element:

n' , l ' , m' zˆ n, l , m   r 2 sin drdd n'l 'm ' (r , ,  )r cos  nlm (r , ,  )
*

Calculation of matrix elements for n = 3

R @n_, _, r_D :=
1 − −3 − r 2r
21+ a0 2 a0 n n − −2 r Hn − − 1L ! LaguerreLB− 1 + n − , 1 + 2 , F
Hn + L ! a0 n

Y@ _, m_, θ_, φ_D := SphericalHarmonicY@ , m , θ , φ D;

ψ @n_, _, m_, r_, θ_, φ_D := R @n , , r D Y@ , m , θ , φ D

f@n1_, 1_, m1_, n2_, 2_, m2_, r_, θ_, φ_D =

H− 1L m1 ψ @n1, 1, − m1, r, θ, φD r Cos@θD ψ @n2, 2, m2, r, θ, φD r2 Sin@θD êê Simplify;
g@n1_, 1_, m1_, n2_, 2_, m2_D :=
Integrate@Integrate@Integrate@f@n1, 1, m1, n2 , 2 , m2 , r, θ, φD, 8φ, 0, 2 π <D, 8θ, 0, π <D,
8r, 0, ∞<D êê Simplify;

Matrix element calculation

α = Simplify@g@3, 2, 1, 3, 1, 1D, a0 > 0D; β = Simplify@g@3, 2, 0, 3, 1, 0D, a0 > 0D;

γ = Simplify@g@3, 2, − 1, 3, 1, − 1D, a0 > 0D;
δ = Simplify@g@3, 1, 0, 3, 0, 0D, a0 > 0D;

M1 = 880, β, 0<, 8β, 0, δ<, 80, δ, 0<<

990, − 3 3 a0, 0=, 9− 3 3 a0, 0, − 3 6 a0=, 90, − 3 6 a0, 0==

eq1 = Eigensystem @ M1D

1 3 1 3
980, − 9 a0, 9 a0<, 99− 2 , 0, 1=, 9 , , 1=, 9 , − , 1===
2 2 2 2

M2 = 880, α<, 8α, 0<<

9 a0 9 a0
990, − =, 9− , 0==
2 2

eq2 = Eigensystem @ M2D êê Simplify

9 a0 9 a0
99− , =, 881, 1<, 8− 1, 1<<=
2 2

M3 = 880, α<, 8α, 0<<

9 a0 9 a0
990, − =, 9− , 0==
2 2

Eigensystem @ M3D
9 a0 9 a0
99− , =, 881, 1<, 8− 1, 1<<=
2 2

39
________________________________________________________________________
APPENDIX-II Polarizability

F  eE [dyne = erg/cm],

p  E  er .

Then we get

er e 2r erg  cm 2
  [ ]  [cm 3 ] .
E eE erg / cm

We have two spheres, each of radius a, one of which has volume charge density + and
the other of which has density -. The vector from the center of the positive sphere to the
center of the negative sphere is d. The two spheres have a region of overlap and we want
the electric field within this region.

Es
Er

We see that the electric field inside a uniformly positively charged sphere is (restoring the
vector notation)

4 4r 3  4
Er   r,
4r 2 3 3

or

40
 4
Er  r
3

where r is the vector from the center of the sphere to the point in question. Now suppose
that s is the vector from the center of the negative sphere to the same point. Because the
charge is negative, we get

4
Es   s
3

So the total electric field is, using the superposition

4 4
E  Er  Es  (r  s )  d
3 3

For the system (sphere, radius a) with the total charge q, the charge density  is obtained
as

4R3
q 
3

Then we have

4 3q qd p
E d  3  3
3 4R 3
R R

or

p  R3 E  E

The polarizability is given by

  R3 .

For the hydrogen, if we take R  aB = 0.52917721092 A, we get

 = aB3=1.48185 x 10-25 cm3.

Another method to calculate the value  is shown as follows. Here we use the formula

a2 2 2
r2  n [5n  1  3l (l  1)] .
2Z 2

41
For n = 1, Z = 1, and l = 0, the radius R can be evaluated as

R r 2  3aB

for a = aB. Then  is calculated as

  R3  3 3aB 3  5.196aB 3

Experimentally,  for hydrogen is  = 6.67 x 10-25 cm3 = 4.50 aB3.

______________________________________________________________________
APPENDIX-III Selection rule

((Spherical tensor of rank 1))

1 1
T0(1)  zˆ , T1(1)   ( xˆ  iyˆ ) , 1 
T(1) ( xˆ  iyˆ )
2 2

((Wigner-Eckart theorem)) selection rule

j ', m ' Tˆq( k ) j , m  0

unless

m'  m  q ,

j '  j  k , j  k  1, ..., j  k ( D j '  D j  Dk  D j  k  D j  k 1  ...  D| j  k | )

((Parity))

ˆ l , m  (1)l l , m

and

ˆ xˆˆ   xˆ , ˆ yˆˆ   yˆ , ˆ zˆˆ   zˆ (odd parity)

((Example))
Using the Wigner-Eckart theorem and the property of the parity operator, we have

42
 l ', m ' zˆ l , m  l ', m ' Tˆ0(1) l , m  0

for m  m ' m  0 , l  l ' l  1

l ', m ' xˆ  iyˆ l , m   2 l ', m ' Tˆ1(1) l , m  0

for m  m ' m  1 , l  l ' l  1

l ', m ' xˆ  iyˆ l , m  2 l ', m ' Tˆ(1)

1 l, m  0

for m  m ' m  1 , l  l ' l  1

APPENDIX-IV Selection rule for polarizaed light waves

(a) Linearly polarized light wave

Vˆz  p  E
  qE0 cos t (rˆ  e z )
  qE0 zˆ cos  t
ˆ it ]
  qE0 Re[ ze

The transition matrix element;

l ', m ' Vˆz l , m   qE0 Re[ l ', m ' zˆ l , m eit ]

This element is not zero when

l  1 , and m  0

(b) Right-hand and left-hand polarized light wave

Vˆx , y  p  E
  qE0 [cos  t (rˆ  e x ) ∓ sin t (rˆ  e y )
  qE0 ( xˆ cos t ∓ yˆ sin t )
  qE0 Re[( xˆ  iyˆ )eit ]

The transition matrix element;

l ', m ' Vˆx , y l , m   qE0 Re[ l ', m ' x  iy l , m eit ]

43
This element is not zero when

l  1 , and m  1

Note that the transition of m  0 .

44