Introduction
State vectors
Stern Gerlach experiment
z
S
B to d e te c to r
o v e n N
c o llim a to r
In the Stern Gerlach experiment
silver atoms are heated in an oven, from which they
escape through a narrow slit,
the atoms pass through a collimator and enter an
inhomogenous magnetic field, we assume the field to
be uniform in the xy-plane and to vary in the
z-direction,
a detector measures the intensity of the electrons
emerging from the magnetic field as a function of z.
We know that
46 of the 47 electrons of a silver atom form a
spherically symmetric shell and the angular
momentum of the electron outside the shell is zero,
so the magnetic moment due to the orbital motion of
the electrons is zero,
the magnetic moment of an electron is cS, where S
is the spin of an electron,
the spins of electrons cancel pairwise,
thus the magnetic moment of an silver atom is
almost solely due to the spin of a single electron, i.e.
= cS,
the potential energy of a magnetic moment in the
magnetic field B is B, so the force acting in the
z-direction on the silver atoms is
Fz = z
So the measurement of the intensity tells how the
z-component the angular momentum of the silver atoms
passing through the magnetic field is distributed. Because complete basis (of a Hilbert space),
the atoms emerging from the oven are randomly oriented and
we would expect the intensity to behave as shown below. Postulate 2 In a measurement the system makes a
S G
c la s s ic a lly
In reality the beam is observed to split into two
components.
S G
in re a lity
Based on the measurements one can evaluate the
z-components Sz of the angular momentum of the atoms
and find out that
for the upper distribution Sz = h/2.
for the lower distribution Sz = h/2.
In quantum mechanics we say that the atoms are in the
angular momentum states h/2 and h/2.
The state vector is a mathematical tool used to represent
the states. Atoms reaching the detector can be described,
for example, by the ket-vectors |Sz ; i and |Sz ; i.
Associated with the ket-vectors there are dual bra-vectors
hSz ; | and hSz ; |. State vectors are assumed
to be a complete description of the described system,
to form a linear (Hilbert) space, so the associated
mathematics is the theory of (infinite dimensional)
linear spaces.
When the atoms leave the oven there is no reason to
expect the angular momentum of each atom to be
oriented along the z-axis. Since the state vectors form a
linear space also the superposition
c |Sz ; i + c |Sz ; i
is a state vector which obviously describes an atom with
angular momentum along both positive and negative
z-axis.
The magnet in the Stern Gerlach experiment can be
thought as an apparatus measuring the z-component of
the angular momentum. We saw that after the
measurement the atoms are in a definite angular
momentum state, i.e. in the measurement the state
c |Sz ; i + c |Sz ; i
collapses either to the state |Sz ; i or to the state |Sz ; i.
Bz
.
A generalization leads us to the measuring postulates of
z
quantum mechanics:
Postulate 1 Every measurable quantity is associated
with a Hermitean operator whose eigenvectors form a
transition to an eigenstate of the corresponding operator
and the result is the eigenvalue associated with that
eigenvector.
If A is a measurable quantity and A the corresponding
Hermitean operator then an arbitrary state |i can be
described as the superposition
X
|i = ca |a i,
a
where the vectors |a i satisfy Note The matrix representation is not unique, but
depends on the basis. In the case of our example we get
A|a i = a |a i. the 2 2-matrix representation

The measuring event A can be depicted symbolically as

 
h 1 0
Sz 7 ,
2 0 1
A
|i |a i.
when we use the set {|Sz ; i, |Sz ; i} as the basis. The
In the Stern Gerlach experiment the measurable quantity base vectors map then to the unit vectors
is the z-component of the spin. We denote the measuring  
event by SGz and the corresponding Hermitean 1
|Sz ; i 7
operator by Sz . We get 0
 
0
h |Sz ; i 7
Sz |Sz ; i = |Sz ; i 1
2
h of the two dimensional Euclidean space.
Sz |Sz ; i = |Sz ; i
2 Although the matrix representations are not unique they
|Sz ; i = c |Sz ; i + c |Sz ; i are related in a rather simple way. Namely, we know that
Theorem 1 If both of the basis {|a i} and {|b i} are
SGz
|Sz ; i |Sz ; i or orthonormalized and complete then there exists a unitary
operator U so that
SGz
|Sz ; i |Sz ; i.
|b1 i = U |a1 i, |b2 i = U |a2 i, |b3 i = U |a3 i, . . .

Because the vectors |a i in the relation
If now X is the representation of an operaor A in the
A|a i = a |a i basis {|a i} the representation X in the basis {|b i} is
obtained by the similarity transformation
are eigenvectors of an Hermitean operator they are
orthognal with each other. We also suppose that they are X = T XT,
normalized, i.e.
where T is the representation of the base transformation
ha |a i = a a .
operator U in the basis {|a i}. Due to the unitarity of the
Due to the completeness of the vector set they satisfy operator U the matrix T is a unitary matrix.
X Since
|a iha | = 1,
a
an abstract state vector, excluding an arbitrary
phase factor, uniquely describes the physical system,
where 1 stands for the identity operator. This property is
called the closure. Using the orthonormality the the states can be written as superpositions of
coefficients in the superposition different base sets, and so the abstract operators can
X take different matrix forms,
|i = ca |a i
the physics must be contained in the invariant propertices
a
of these matrices. We know that
can be written as the scalar product Theorem 2 If T is a unitary matrix, then the matrices
X and T XT have the same trace and the same
ca = ha |i. eigenvalues.
The same theorem is valid also for operators when the
An arbitrary linear operator B can in turn be written trace is defined as
with the help of a complete basis {|a i} as X
X trA = ha |A|a i.
B= |a iha |B|a iha |. a
a ,a
Since
Abstract operators can be represented as matrices:
quite obviously operators and matrices representing
|a1 i |a2 i |a3 i ... them have the same trace and the same eigenvalues,

ha1 | ha1 |B|a1 i ha1 |B|a2 i ha1 |B|a3 i ...
due to the postulates 1 and 2 corresponding to a
ha2 |
ha2 |B|a1 i ha2 |B|a2 i ha2 |B|a3 i ...
B 7 ha3 | ha3 |B|a1 i ha3 |B|a2 i

. measurable quantity there exists an Hermitean
ha3 |B|a3 i ...
operator and the measuring results are eigenvalues of
.. .. .. .. ..

. . . . . this operator,
the results of measurements are independent on the SGx
particular representation and, in addition, every |Sz ; i |Sx ; i.
measuring event corresponding to an operator reachable Again the analysis of the experiment gives Sx = h/2 and
by a similarity transformation, gives the same results. Sx = h/2 as the x-components of the angular momenta.
Which one of the possible eigenvalues will be the result of We can thus deduce that the state |Sz ; i is, in fact, the
a measurement is clarified by superposition
Postulate 3 If A is the Hermitean operator
corresponding to the measurement A, {|a i} the |Sz ; i = c |Sx ; i + c |Sx ; i.
eigenvectors of A associated with the eigenvalues {a },
For the other component we have correspondingly
then the probability for the result a is |ca |2 when the
system to be measured is in the state |Sz ; i = c |Sx ; i + c |Sx ; i.
When the intensities are equal the coeffiecients satisfy
X
|i = ca |a i.
a 1
|c | = |c | =
Only if the system already before the measurement is in a 2
definite eigenstate the result can be predicted exactly. 1
|c | = |c | =
For example, in the Stern Gerlach experiment SGz we 2
can block the emerging lower beam so that the spins of according to the postulate 3. Excluding a phase factor,
the remaining atoms are oriented along the positive our postulates determine the transformation coefficients.
z-axis. We say that the system is prepared to the state When we also take into account the orthogonality of the
|Sz ; i.

S
state vectors |Sz ; i and |Sz ; i we can write

S z 1 1
|Sz ; i = |Sx ; i + |Sx ; i

S G z
z 2 2

S G z

S
 
1 1
ei1 |Sx ; i |Sx ; i .
z
|Sz ; i =
2 2
There is nothing special in the direction x, nor for that
If we now let the polarized beam to pass through a new matter, in any other direction. We could equally well let
SGz experiment we see that the beam from the latter the beam pass through a SGy experiment, from which
experiment does not split any more. According to the
we could deduce the relations
postulate this result can be predicted exactly.
We see that 1 1
|Sz ; i = |Sy ; i + |Sy ; i
2 2
the postulate can also be interpreted so that the 
1 1

i2
quantities |ca |2 tell the probability for the system |Sz ; i = e |Sy ; i |Sy ; i ,
being in the state |a i, 2 2

the physical meaning of the matrix element h|A|i or we could first do the SGx experiment and then the
is then the expectation value (average) of the SGy experiment which would give us
measurement and
ei3 ei4
|Sx ; i = |Sy ; i + |Sy ; i
the normalization condition h|i = 1 says that the 2 2
system is in one of the states |a i.
ei3 ei4
|Sx ; i = |Sy ; i |Sy ; i.
Instead of measuring the spin z-component of the atoms 2 2
with spin polarized along the z-axis we let this polarized In other words
beam go through the SGx experiment. The result is
1
exactly like in a single SGz experiment: the beam is |hSy ; |Sx ; i| = |hSy ; |Sx ; i| =
2
again splitted into two components of equal intensity, this
1
time, however, in the x-direction. |hSy ; |Sx ; i| = |hSy ; |Sx ; i| = .
S
2

S x We can now deduce that the unknown phase factors must

z
S G x
S satisfy

S G z

S z
x 2 1 = /2 or /2.
A common choice is 1 = 0, so we get, for example,
1 1
So, we have performed the experiment |Sz ; i = |Sx ; i + |Sx ; i
2 2
SGx 1 1
|Sz ; i |Sx ; i or |Sz ; i = |Sx ; i |Sx ; i.
2 2
Thinking like in classical mechanics, we would expect
both the z- and x-components of the spin of the atoms in
the upper beam passed through the SGz and SGx
experiments to be Sx,z = h/2. On the other hand, we can
reverse the relations above and get
1 1
|Sx ; i = |Sz ; i + |Sz ; i,
2 2
so the spin state parallel to the positive x-axis is actually
a superposition of the spin states parallel to the positive
and negative z-axis. A Stern Gerlach experiment confirms
this.
S
S z

S x
S G z
S
z
S G x
S z

S G z

S z
x

After the last SGz measurement we see the beam

splitting again into two equally intensive componenents.
The experiment tells us that there are quantitities which
cannot be measured simultaneously. In this case it is
impossible to determine simultaneously both the z- and
x-components of the spin. Measuring the one causes the
atom to go to a state where both possible results of the
other are present.
We know that
Theorem 3 Commuting operators have common
eigenvectors.
When we measure the quantity associated with an
operator A the system goes to an eigenstate |a i of A. If
now B commutes with A, i.e.

[A, B] = 0,

then |a i is also an eigenstate of B. When we measure the

quantity associated with the operator B while the system
is already in an eigenstate of B we get as the result the
corresponding eigenvalue of B. So, in this case we can
measure both quantities simultaneously.
On the other hand, Sx and Sz cannot be measured
simultaneously, so we can deduce that

[Sx , Sz ] 6= 0.

So, in our example a single Stern Gerlach experiment

gives as much information as possible (as far as only the
spin is concerned), consecutive Stern Gerlach experiments
cannot reveal anything new.
In general, if we are interested in quantities associated
with commuting operators, the states must be
characterized by eigenvalues of all these operators. In
many cases quantum mechanical problems can be reduced
to the tasks to find the set of all possible commuting
operators (and their eigenvalues). Once this set is found
the states can be classified completely using the
eigenvalues of the operators.
Translations The effect of an infinitesimal translation on an arbitrary
The previous discrete spectrum state vector formalism state can be seen by expanding it using position
can be generalized also to continuos cases, in practice, by eigenstates:
replacing Z
|i T (dx )|i = T (dx ) d3 x0 |x0 ihx0 |i
00 00
summations with integrations
Z
Kroneckers -function with Diracs -function. = d3 x0 |x0 + dx00 ihx0 |i
A typical continuous case is the measurement of position:
Z
= d3 x0 |x0 ihx0 dx00 |i,
the operator x corresponding to the measurement of
the x-coordinate of the position is Hermitean, because x0 is an ordinary integration variable.
To construct T (dx0 ) explicitely we need extra constraints:
the eigenvalues {x0 } of x are real,
the eigenvectors {|x0 i} form a complete basis. 1. it is natural to require that it preserves the
normalization (i.e. the conservation of probability) of
So, we have the state vectors:

x|x0 i = x0 |x0 i h|i = h|T (dx0 )T (dx0 )|i.

Z
1 = dx0 |x0 ihx0 | This is satisfied if T (dx0 )is unitary, i.e.

Z
T (dx0 )T (dx0 ) = 1.
|i = dx0 |x0 ihx0 |i,

2. we require that two consecutive translations are
where |i is an arbitrary state vector. The quantity equivalent to a single combined transformation:
hx0 |i is called a wave function and is usually written
down using the function notation T (dx0 )T (dx00 ) = T (dx0 + dx00 ).

hx0 |i = (x0 ). 3. the translation to the opposite direction is equivalent

to the inverse of the original translation:
Obviously, looking at the expansion
Z T (dx0 ) = T 1 (dx0 ).
|i = dx0 |x0 ihx0 |i,
4. we end up with the identity operator when dx0 0:
the quantity | (x0 )|2 dx0 can be interpreted according to lim T (dx0 ) = 1.
the postulate 3 as the probability for the state being dx0 0
0 0 0 0
localized in the neighborhood (x , x + dx ) of the point x .
The position can be generalized to three dimension. We It is easy to see that the operator
0
denote by |x i the simultaneous eigenvector of the
T (dx0 ) = 1 iK dx0 ,
operators x, y and z, i.e.
where the components Kx , Ky and Kz of the vector K
|x0 i |x0 , y 0 , z 0 i are Hermitean operators, satisfies all four conditions.
x|x0 i = x0 |x0 i, y|x0 i = y 0 |x0 i, z|x0 i = z 0 |x0 i. Using the definition
The exsistence of the common eigenvector requires T (dx0 )|x0 i = |x0 + dx0 i
commutativity of the corresponding operators:
we can show that
[xi , xj ] = 0.
[x, T (dx0 )] = dx0 .
Let us suppose that the state of the system is localized at
the point x0 . We consider an operation which transforms Substituting the explicit reprersentation
this state to another state, this time localized at the point
T (dx0 ) = 1 iK dx0
x0 + dx0 , all other observables keeping their values. This
operation is called an infinitesimal translation. The it is now easy to prove the commutation relation
corresponding operator is denoted by T (dx0 ):
[xi , Kj ] = iij .
T (dx0 )|x0 i = |x0 + dx0 i.
The equations
The state vector on the right hand side is again an
eigenstate of the position operator. Quite obviously, the T (dx0 ) = 1 iK dx0
vector |x0 i is not an eigenstate of the operator T (dx0 ). T (dx0 )|x0 i = |x0 + dx0 i
can be considered as the definition of K. So, we have derived the canonical commutation relations
One would expect the operator K to have something to or fundamental commutation relations
do with the momentum. It is, however, not quite the
momentum, because its dimension is 1/length. Writing [xi , xj ] = 0, [pi , pj ] = 0, [xi , pj ] = ihij .

p = hK Recall, that the projection of the state |i along the state

vector |x0 i was called the wave function and was denoted
we get an operator p, with dimension of momentum. We like (x0 ). Since the vectors |x0 i form a complete basis
take this as the definition of the momemtum. The the scalar product between the states |i and |i can be
commutation relation written with the help of the wave functions as
Z Z
[xi , Kj ] = iij
h|i = dx0 h|x0 ihx0 |i = dx0 (x0 ) (x0 ),
can now be written in a familiar form like
i.e. h|i tells how much the wave functions overlap. If
[xi , pj ] = ihij . |a0 i is an eigenstate of A we define the corresponding
eigenfunction ua0 (x0 ) like
Finite translations ua0 (x0 ) = hx0 |a0 i.
0
Consider translation of the distance x along the x-axis:
An arbitrary wave function (x0 ) can be expanded using
T (x0 x)|x0 i = |x0 + x0 xi. eigenfunctions as
We construct this translation combining infinitesimal
X
(x0 ) = ca0 ua0 (x0 ).
translations of distance x0 /N letting N : a0
N
ipx x0

The matrix element h|A|i of an operator A can also be
T (x0 x) = lim 1
N N h expressed with the help of eigenfunctions like
ipx x0
  Z Z
= exp . h|A|i = dx0 dx00 h|x0 ihx0 |A|x00 ihx00 |i
h
Z Z
It is relatively easy to show that the translation operators = dx0
dx00 (x0 )hx0 |A|x00 i (x00 ).
satisfy
[T (y 0 y), T (x0 x)] = 0,
To apply this formula we have to evaluate the matrix
so it follows that elements hx0 |A|x00 i, which in general are functions of the
[py , px ] = 0. two variables x0 and x00 . When A depends only on the
position operator x,
Generally
[pi , pj ] = 0. A = f (x),
This commutation relation tells that it is possible to
construct a state vector which is a simultaneous the calculations are much simpler:
eigenvector of all components of the momentum operator, Z
i.e. there exists a vector h|f (x)|i = dx0 (x0 )f (x0 ) (x0 ).

|p0 i |p0x , p0y , p0z i, Note f (x) on the left hand side is an operator while f (x0 )
on the right hand side is an ordinary number.
so that

px |p0 i = p0x |p0 i, py |p0 i = p0y |p0 i, pz |p0 i = p0z |p0 i. Momentum operator p in position basis {|x0 i}
For simplicity we consider the one dimensional case.
The effect of the translation T (dx0 ) on an eigenstate of According to the equation
the momentum operator is Z
T (dx00 )|i = T (dx00 ) d3 x0 |x0 ihx0 |i
ip dx0 ip0 dx0
   
0 0
0
T (dx )|p i = 1 |p i = 1 |p0 i. Z
h h
= d3 x0 |x0 + dx00 ihx0 |i
The state |p0 i is thus an eigenstate of T (dx0 ): a result, Z
which we could have predicted because = d3 x0 |x0 ihx0 dx00 |i

[p, T (dx0 )] = 0. we can write

0
Note The eigenvalues of T (dx ) are complex because it is ip dx00
 
not Hermitean. 1 |i
h
 Z
= dx0 T (dx00 )|x0 ihx0 |i Here the left hand side is simply (x0 x00 ) and the
Z integration of the left hand side gives 2h|C|2 (x0 x00 ).
Thus the transformation function is
= dx0 |x0 ihx0 dx00 |i
 0 0
0 0 1 ip x
hx |p i =
 
exp ,
Z

= dx0 |x0 i hx0 |i dx00 0 hx0 |i . 2h h
x
and the relations
In the last step we have expanded hx0 dx00 |i as Taylor Z
series. Comparing both sides of the equation we see that 0
(x ) 0
= hx |i = dp0 hx0 |p0 ihp0 |i
Z  

p|i = dx0 |x0 i ih 0 hx0 |i ,
Z
x (p0 ) = hp0 |i = dx0 hp0 |x0 ihx0 |i.
or, taking scalar product with a position eigenstate on
both sides, can be written as familiar Fourier transforms

1
Z  0 0
ip x
hx0 |p|i = ih 0 hx0 |i. (x0 ) = dp0 exp (p0 )
x h
2h
In particular, if we choose |i = |x0 i we get 
1
Z 
ip0 x0

(p0 ) = dx0 exp (x0 ).
2h h
hx0 |p|x00 i = ih (x0 x00 ).
x0
Taking scalar product with an arbitrary state vector |i
on both sides of
Z  

p|i = dx0 |x0 i ih 0 hx0 |i
x
we get the important relation
Z  

h|p|i = dx0 (x0 ) ih 0 (x0 ).
x
Just like the position eigenvalues also the momentum
eigenvalues p0 comprise a continuum. Analogically we can
define the momentum space wave function as

hp0 |i = (p0 ).

We can move between the momentum and configuration

space representations with help of the relations
Z
(x0 ) = hx0 |i = dp0 hx0 |p0 ihp0 |i
Z
0 0
(p ) = hp |i = dx0 hp0 |x0 ihx0 |i.

The transformation function hx0 |p0 i can be evaluated by

substituting a momentum eigenvector |p0 i for |i into

hx0 |p|i = ih hx0 |i.
x0
Then

hx0 |p|p0 i = p0 hx0 |p0 i = ihhx0 |p0 i.
x0
The solution of this differential equation is
 0 0
ip x
hx0 |p0 i = C exp ,
h
where the normalization factor C can be determined from
the identity
Z
hx0 |x00 i = dp0 hx0 |p0 ihp0 |x00 i.
Time evolution operator
In quantum mechanics
unlike position, time is not an observable.
there is no Hermitean operator whose eigenvalues
were the time of the system.
time appears only as a parameter, not as a
measurable quantity.
So, contradictory to teachings of the relativity theory,
time and position are not on equal standing. In
relativistic quantum field theories the equality is restored
by degrading also the position down to the parameter
level.
We consider a system which at the moment t0 is in the
state |i. When time goes on there is no reason to expect
it to remain in this state. We suppose that at a later
moment t the system is described by the state
|, t0 ; ti, (t > t0 ),
where the parameter t0 reminds us that exactly at that
moment the system was in the state |i. Since the time is
a continuous parameter we obviously have
lim |, t0 ; ti = |i,
tt0
and can use the shorter notation
|, t0 ; t0 i = |, t0 i.
Lets see, how state vectors evolve when time goes on:
evolution
|, t0 i |, t0 ; ti.
We work like we did with translations. We define the
time evolution operator U(t, t0 ):
|, t0 ; ti = U(t, t0 )|, t0 i,
which must satisfy physically relevant conditions.
1. Conservation of probability
We expand the state at the moment t0 with the help of
the eigenstates of an observable A:
X the composition condition
|, t0 i = ca0 (t0 )|a0 i.
a0
At a later moment we get the expansion
X term O(dt).
|, t0 ; ti = ca0 (t)|a0 i.
a0
In general, we cannot expect the probability for the
system being in a specific state |a0 i to remain constant,
i.e. in most cases
|ca0 (t)| 6= |ca0 (t0 )|.
For example, when a spin 21 particle, which at the
moment t0 is in the state |Sx ; i, is subjected to an
external constant magnetic field parallel to the z-axis, it
will precess in the xy-plane: the probability for the result
h/2 in the measurement SGx oscillates between 0 and 1
as a function of time. In any case, the probability for the
result h/2 or h/2 remains always as the constant 1.
Generalizing, it is natural to require that
X X
|ca0 (t0 )|2 = |ca0 (t)|2 .
a0 a0
In other words, the normalization of the states does not
depend on time:
h, t0 |, t0 i = h, t0 ; t|, t0 ; ti
= h, t0 |U (t, t0 )U(t, t0 )|, t0 i.
This is satisfied if we require U(t, t0 ) to be unitary, i.e.
U (t, t0 )U(t, t0 ) = 1.
2. Composition property
The evolution from the time t0 to a later time t2 should
be equivalent to the evolution from the initial time t0 to
an intermediate time t1 followed by the evolution from t1
to the final time t2 , i.e.
U(t2 , t0 ) = U(t2 , t1 )U(t1 , t0 ), (t2 > t1 > t0 ).
Like in the case of the translation operator we will first
look at the infinitesimal evolution
|, t0 ; t0 + dti = U(t0 + dt, t0 )|, t0 i.
Due to the continuity condition
lim |, t0 ; ti = |i
tt0
we have
lim U(t0 + dt, t0 ) = 1.
dt0
So, we can assume the deviations of the operator
U(t0 + dt, t0 ) from the identity operator to be of the order
dt. When we now set
U(t0 + dt, t0 ) = 1 idt,
where is a Hermitean operator, we see that it satisfies
U(t2 , t0 ) = U(t2 , t1 )U(t1 , t0 ), (t2 > t1 > t0 ),
is unitary and deviates from the identity operator by the
The physical meaning of will be revealed when we
recall that in classical mechanics the Hamiltonian
generates the time evolution. From the definition
U(t0 + dt, t0 ) = 1 idt
we see that the dimension of is frequency, so it must be
multiplied by a factor before associating it with the
Hamiltonian operator H:
H = h,
or (ii) The Hamiltonain H depends on time but the operators
iH dt H corresponding to different moments of time commute.
U(t0 + dt, t0 ) = 1 .
h For example, a spin 12 particle in the magnetic field whose
The factor h here is not necessarily the same as the factor strength varies but direction remains constant as a
h in the case of translations. It turns out, however, that function of time. A formal solution of the equation
in order to recover Newtons equations of motion in the
classical limit both coefficients must be equal.
ih U(t, t0 ) = HU(t, t0 )
Applying the composition property t
U(t2 , t0 ) = U(t2 , t1 )U(t1 , t0 ), (t2 > t1 > t0 ) is now
  Z t 
we get i 0 0
U(t, t0 ) = exp dt H(t ) ,
h t0
U(t + dt, t0 ) = U(t + dt, t)U(t, t0 )
 
iH dt which, again, can be proved by expanding the exponential
= 1 U(t, t0 ),
h function as the series.
(iii) The operators H evaluated at different moments of
where the time difference t t0 does not need to be time do not commute For example, a spin 21 particle in a
infinitesimal. This can be written as magnetic field whose direction changes in the course of
 
H time: H is proportional to the term S B and if now, at
U(t + dt, t0 ) U(t, t0 ) = i dt U(t, t0 ).
h the moment t = t1 the magnetic field is parallel to the
x-axis and, at the moment t = t2 parallel to the y-axis,
Expanding the left hand side as the Taylor series we end then H(t1 ) BSx and H(t2 ) BSy , or
up with [H(t1 ), H(t2 )] B 2 [Sx , Sy ] 6= 0. It can be shown that the

ih U(t, t0 ) = HU(t, t0 ). formal solution of the Schrodinger equation is now
t
This is the Schrodinger equation of the time evolution U(t, t0 ) =
operator. Multiplying both sides by the state vector  n Z t Z t1
X i
|, t0 i we get 1+ dt1 dt2
n=1
h t0 t0

ih U(t, t0 )|, t0 i = HU(t, t0 )|, t0 i. Z tn1
t dtn H(t1 )H(t2 ) H(tn ).
t0
Since the state |, t0 i is independent on the time t we can
write the Schrodinger equation of the state vectors in the This expansion is called the Dyson series. We will assume
form that our Hamiltonians are time independent until we

ih |, t0 ; ti = H|, t0 ; ti. start working with the so called interaction picture.
t Suppose that A is an Hermitean operator and
In fact, in most cases the state vector Schrodinger
equation is unnecessary because all information about the [A, H] = 0.
dynamics of the system is contained in the time evolution
operator U(t, t0 ). When this operator is known the state Then the eigenstates of A are also eigenstates of H, called
of the system at any moment is obtained by applying the energy eigenstates. Denoting corresponding eigenvalues of
definition the Hamiltonian as Ea0 we have
|, t0 ; ti = U(t, t0 )|, t0 i,
H|a0 i = Ea0 |a0 i.
We consider three cases:
(i) The Hamiltonian does not depend on time. For example,
The time evolution operator can now be written with the
a spin 12 particle in a time independent magnetic field
help of these eigenstates. Choosing t0 = 0 we get
belongs to this category. The solution of the equation
   
iHt XX
00 00 iHt
ih U(t, t0 ) = HU(t, t0 ) exp = |a iha | exp |a0 iha0 |
t h h
a0 a00
 
is X
0 iEa0 t

iH(t t0 )
 = |a i exp ha0 |.
U(t, t0 ) = exp h
a0
h
as can be shown by expanding the exponential function as Using this form for the time evolution operator we can
the Taylor series and differentiating term by term with solve every intial value problem provided that we can
respect to the time. Another way to get the solution is to expand the initial state with the set {|a0 i}. If, for
compose the finite evolution from the infinitesimal ones: example, the initial state can be expanded as
 N  
(iH/h(t t0 ) iH(t t0 )
X X
lim 1 = exp . |, t0 = 0i = |a0 iha0 |i = ca0 |a0 i,
N N h a0 a0
we get in an eigenstate |a0 i of an operator A commuting with the
 Hamiltonian H. Suppose, we are interested in the
iHt expectation value of an operator B which does not
|, t0 = 0; ti = exp |, t0 = 0i
h necessarily commute either with A or with H. At the
 
X
0 0 iEa0 t moment t the system is in the state
= |a iha |i exp .
h
0 a |a0 , t0 = 0; ti = U(t, 0)|a0 i.
In other words, the expansion coefficients evolve in the
In this special case we have
course of time as
hBi = ha0 |U (t, 0)BU(t, 0)|a0 i
 
iEa0 t
ca (t = 0) ca (t) = ca (t = 0) exp
0 0 0 .    
h 0 iEa0 t iEa0 t
= ha | exp B exp |a0 i
h h
So, the absolute values of the coefficients remain
= ha0 |B|a0 i,
constant. The relative phase between different
components will, however, change in the course of time that is, the expectation value does not depend on time.
because the oscillation frequencies of different For this reason the energy eigenstates are usually called
components differ from each other. stationary states
As a special case we consider an initial state consisting of We now look at the expectation value in a superposition
a single eigenstate: of energy eigenstates, in a non stationary state
|, t0 = 0i = |a0 i. X
|, t0 = 0i = ca0 |a0 i.
At some later moment this state has evolved to the state a0

iEa0 t
 It is easy to see, that the expectation value of B is now
0
|, t0 = 0; ti = |a i exp .
h XX 
i(Ea00 Ea0 )t

hBi = ca0 ca00 ha0 |B|a00 i exp .
Hence, if the system originally is in an eigenstate of the 0 00
h
a a
Hamiltonian H and the operator A it stays there forever.
Only the phase factor exp(iEa0 t/h) can vary. In this This time the expectation value consists of terms which
sense the observables whose corresponding operators oscillate with frequences determind by the Bohr
commute with the Hamiltonian, are constants of motion. frequency condition
Observables (or operators) associated with mutually Ea00 Ea0
commuting operators are called compatible. As mentioned a00 a0 = .
h
before, the treatment of a physical problem can in many
cases be reduced to the search for a maximal set of As an application we look at how spin 21 particles behave
compatible operators. If the operators A, B, C, . . . belong in a constant magnetic field. When we assume the
to this set, i.e. magnetic moments of the particles to be eh/2me c (like
electrons), the Hamiltonian is
[A, B] = [B, C] = [A, C] = = 0,
 
e
and if, furthermore, H= S B.
me c
[A, H] = [B, H] = [C, H] = = 0, If we choose B k z, we have
that is, also the Hamiltonian is compatible with other 
eB

operators, then the time evolution operator can be H= Sz .
me c
written as

iHt
 X 
iEK 0 t
 The operators H and Sz differ only by a constant factor,
0
exp = |K i exp hK 0 |. so they obviously commute and the eigenstates of Sz are
h 0
h
K also energy eigenstates with energies
Here K 0 stands for the collective index: ehB
E = for state |Sz ; i
2me c
A|K 0 i = a0 |K 0 i, B|K 0 i = b0 |K 0 i, C|K 0 i = c0 |K 0 i, . . .
ehB
E = + for state |Sz ; i.
Thus, the quantum dynamics is completely solved (when 2me c
H does not depend on time) if we only can find a
We define the cyclotron frequency c so that the energy
maximal set of compatible operators commuting also with
difference between the states is hc :
the Hamiltonian.
Lets now look at the expectation value of an operator. |e|B
We first assume, that at the moment t = 0 the system is c .
me c
The Hamiltonian H can now be written as Lastly we look at how the statevectors corresponding to
different times are correlated. Suppose that at the
H = c Sz , moment t = 0 the system is described by the state vector
|i, which in the course of time evolves to the state
when we assume that e < 0. |, t0 = 0; ti. We define the correlation amplitude C(t) as
All information about the evolution of the system is
contained in the operator C(t) = h|, t0 = 0; ti

ic Sz t
 = h|U(t, 0)|i.
U(t, 0) = exp .
h The absolute value of the correlation amplitude tells us
If at the moment t = 0 the system is in the state how much the states associated with different moments of
time resemble each other.
|i = c |Sz ; i + c |Sz ; i, In particular, if the initial state is an energy eigenstate
|a0 i, then  
it is easy to see that at the moment t it is in the state iEa0 t
C(t) = exp ,
  h
ic t
|, t0 = 0; ti = c exp |Sz ; i and the absolute value of the correlation amplitude is 1 at
2
  all times. When the initial state is a superposition of
ic t energy eigenstates we get
+c exp + |Sz ; i.
2  
X iEa0 t
If the initial state happens to be |Sz ; i, meaning that in C(t) = |ca0 |2 exp .
0
h
the previous equation a

When t is relatively large the terms in the sum oscillate

c = 1, c = 0,
rapidly with different frequencies and hence most
we see that the system will stay in this state at all times. probably cancel each other. Thus we expect the
This was to be expected because the state is stationary. correlation amplitude decreasing rather rapidly from its
We now assume that the initial state is |Sx ; i. From the initial value 1 at the moment t = 0.
relation We can estimate the value of the expression
1 1
|Sx ; i = |Sz ; i + |Sz ; i
 
X
2 iEa0 t
2 2 C(t) = |ca0 | exp
0
h
we see that a
1
c = c = . more concretely when we suppose that the statevectors of
2
the system comprise so many, nearly degenerate, energy
For the probabilities that at the moment t the system is eigenvectors that we can think them almost to form a
in eigenstates of Sx we get continuum. Then the summation can be replaced by the
c t integration
|hSx ; |, t0 = 0; ti|2 = cos2
2 Z 
X 
2 c t dE (E), ca0 g(E) ,

|hSx ; |, t0 = 0; ti|2 = sin .
2 a 0

EEa0

Even if the spin originally were parallel to the positive where (E) is the density of the energy eigenstates. The
x-axis a magnetic field parallel to the z-axis makes the expression
direction of the spin to rotate. There is a finite
probability for finding the system at some later moment
 
X iEa0 t
in the state |Sx ; i. The sum of probabilities C(t) = |ca0 |2 exp
0
h
a
corresponding to different orientations is 1.
It is easy to see that the expectation values of the can now be written as
operator S satisfy Z  
iEt
 
h C(t) = dE |g(E)|2 (E) exp ,
hSx i = cos c t h
2
 
h which must satisfy the normalization condition
hSy i = sin c t
2
Z
dE |g(E)|2 (E) = 1.
hSz i = 0.

Physically this means that the spin precesses in the In many realistic physical cases |g(E)|2 (E) is
xy-plane. concentrated into a small neighborhood (size E) of a
point E = E0 . Rewriting the integral representation as
 
iE0 t
C(t) = exp
h
 
i(E E0 )t
Z
dE |g(E)|2 (E) exp ,
h

we see that when t increases the integrand oscillates very

rapidly except when the energy interval |E E0 | is small
as compared with h/t. If the interval, which satisfies
|E E0 | h/t, is much shorter than E the interval
from which the integral picks up its contribution, the
correlation amplitudes practically vanishes. The
characteristic time, after which the absolute value of the
correlation amplitude deviates significantly from its initial
value 1, is
h
t .
E
Although this equation was derived for a quasi continuous
energy spectrum it is also valid for the two state system
in our spin precession example: the initial state |Sx ; i
starts to lose its identity after the time
1/c = h/(E E ) as we can see from the equation

c t
|hSx ; |, t0 = 0; ti|2 = cos2 .
2
As a summary we can say that due to the evolution the
state vector describing the initial state of the system will
not any more describe it after a time interval of order
h/E. This property is often called the time and energy
uncertainty relation. Note, however, that this relation is
of completely different character than the uncertainty
relation concerning position and momentum because time
is not a quantum mechanical observable.
Quantum statistics Thermodynamics
Density operator: We define
X = tr( ln ).
wi |i ihi |
i One can show that
is for a completely stochastic ensemble
Hermitean: = ln N,
=
when N is the number of the independent states in
normalized:
the system.
tr = 1.
for a pure ensemble
Density matrix:
X = 0.
hb00 ||b0 i = wi hb00 |i ihi |b0 i.
i
Hence measures disorder = it has something to do
Ensemble average: with the entropy.
XX The entropy is defined by
[A] = hb00 ||b0 ihb0 |A|b00 i
b0 b00
S = k.
= tr(A).
In a thermodynamical equilibrium
Dynamics
= 0,
t
|i i = |i ; t0 i |i , t0 ; ti
so
We suppose that the occupation of states is conserved, i.e. [, H] = 0
wi = constant. and the operators and H have common eigenstates |ki:
Now X H|ki = Ek |ki
(t) = wi |i , t0 ; tihi , t0 ; t|,
i
|ki = wk |ki.
so
  Using these eigenstates the density matrix can be
X represented as
ih = wi ih |i , t0 ; ti hi , t0 ; t| X
t i
t = wk |kihk|
  k
X
+ wi |i , t0 ; ti ih |i , t0 ; ti and
i
t X
= kk ln kk ,
= H H = [, H]. k

Like Heisenbergs equation of motion, but wrong sign! where the diagonal elements of the density matrix are
OK, since is not an observable.
kk = wk .
Continuum
Example: In the equilibrium the entropy is at maximum.
Z Z We maximize under conditions
3 0
[A] = d x d3 x00 hx00 ||x0 ihx0 |A|x00 i. P
U = [H] = trH = k kk Ek .

Here the density matrix is tr = 1.

!
X Hence
hx00 ||x0 i = hx00 | wi |i ihi | |x0 i X
i = kk (ln kk + 1) = 0
X
00
= wi i (x )i (x0 ). X
k

i [H] = kk Ek = 0
k
Note X X
hx0 ||x0 i = wi |i (x0 )|2 . (tr) = kk = 0.
i k
With the help of Lagrange multipliers we get
X
kk [(ln kk + 1) + Ek + ] = 0,
k
so
kk = eEk 1 .
The normalization (tr = 1) gives
eEk
kk = N
(canonical ensemble).
X
eEl
l

It turns out that

1
,
=
kB T
where T is the thermodynamical temperature and kB the
Boltzmann constant.
In statistical mechanics we define the canonical partition
function Z:
XN
Z = treH = eEk .
k
Now
eH
.
=
Z
The ensemble average can be written as
tr eH A


[A] = trA =
Z
"N #
X
Ek
hk|A|kie
k
= N
.
X
Ek
e
k

In particular we have
N
X
Ek eEk
k
U = [H] = N
X
eEk
k

= (ln Z).

Example Electrons in a magnetic field parallel to z axis.
In the basis {|Sz ; i, |Sz ; i} of the eigenstates of the
Hamiltonian
H = c Sz
we have
ehc /2
 
0
0 ehc /2
7 ,
Z
where
Z = ehc /2 + ehc /2 .
For example the ensemble averages are
[Sx ] = [Sy ] = 0,
   
h hc
[Sz ] = tanh .
2 2
Angular momentum We see that
2

0 0
O(3) Rx ()Ry () Ry ()Rx () = 2 0 0
We consider active rotations. 0 0 0
3 3 orthogonal matrix R rotation inR3 .
= Rz (2 ) 1.

Number of parameters In a Hilbert space we associate

1. RRT symmetric RRT has 6 independent R D(R),

parameters orthogonality condition RRT = 1
i.e.
gives 6 independent equations R has 9 6 = 3 free
|iR = D(R)|i.
parameters.
We define the angular momentum (J) so that (we are not
2. Rotation around n (2 angles) by the angle 3 employing properties of the classical angular momentum
parameters. x p)  
J n
3. n vector 3 parameters. D(n, d) = 1 i d
h
and require that the rotation operator D
3 3 orthogonal matrices form a group with respect to
the matrix multiplication: is unitary,
1. R1 R2 is orthogonal if R1 and R2 are orthogonnal. is decomposable,

2. R1 (R2 R3 ) = (R1 R2 )R3 , associativity. D 1, when d 0.

3. identity I = the unit matrix.
4. if R is orthogonal, then also the inverse matrix
R1 = RT is orthogonal.
The group is called O(3).
Generally rotations do not commute,
We see that J must be Hermitean, i.e.
J = J.
Moreover, we require that D satisfies the same group
properties as R, i.e.
Dx ()Dy () Dy ()Dx () = Dz (2 ) 1.

R1 R2 6= R2 R1 , Since rotations around a common axis commute a finite

rotation can be constructed as
so the group is non-Abelian.     N
J n
Rotations around a common axis commute. D(n) = lim 1 i
Rotation around z-axis:
N h N
 
iJ n

cos sin 0
= exp
h
Rz () = sin cos 0
J n (J n)2 2
0 0 1 = 1i + .
h 2h2
x x cos y sin
Rz y = x sin + y cos . We apply this up to the order O(2 ):
z z 2 2
!
iJx  Jx2 2 iJy  Jy 
 
1 1
Infinitesimal rotations up to the order O(2 ): h 2h2 h 2h2
!
2 2
iJy  Jy  iJx  Jx2 2

2
1 2

 0 1 1
2 h 2h2 h 2h2
Rz () =  ,

 1 2 0 1 1
2 3
0 0 1 = 2 Jx Jy  + 2 Jy Jx + O( )
h h
1 0 0

Jz 2
0 1 2 =1i 1.
Rx () = 2  ,
h
2
0  1 2 We see that
2 [Jx , Jy ] = ihJz .
1 2 0


0 1 0 Similarly for other components:
Ry () = .

2
 0 1 2 [Ji , Jj ] = ihijk Jk .
We consider: Now

hJx i h|Jx |i Ry0 () = Rz ()Ry ()Rz1 ()

R h|Jx |iR = h|D z ()Jx D z ()|i. Rz0 () = Ry0 ()Rz ()Ry1
0 (),

We evaluate so
   
iJz iJz R(, , ) = Ry0 ()Rz ()Ry1
0 ()Ry 0 ()Rz ()
Dz ()Jx Dz () = exp Jx exp
h h
= Ry0 ()Rz ()Rz ()
applying the Baker-Hausdorff lemma = Rz ()Ry ()Rz1 ()Rz ()Rz ()
= Rz ()Ry ()Rz ().
eiG AeiG =
 2 2
i Correspondingly
A + i[G, A] + [G, [G, A]] +
2!
 n n
i D(, , ) = Dz ()Dy ()Dz ().
+ [G, [G, [G, . . . [G, A]]] . . .] +
n!

where G is Hermitean. So we need the commutators

[Jz , Jx ] = ihJy
[Jz , [Jz , Jx ]] = ih[Jz , Jy ] = h2 Jx
[Jz , [Jz , [Jz , Jx ]]] = h2 [Jz , Jx ] = ih3 Jy
..
.

Substituting into the Baker-Hausdorff lemma we get

Dz ()Jx Dz () = Jx cos Jy sin .

Thus the expectation value is

hJx i R h|Jx |iR = hJx i cos hJy i sin .

Correspondingly we get for the other components

hJy i hJy i cos + hJx i sin

hJz i hJz i.

We see that the components of the expectation value of

the angular momentum operator transform in rotations
like a vector in R3 :
X
hJk i Rkl hJl i.
l

Euler angles

1. Rotate the system counterclockwise by the angle

around the z-axis. The y-axis of of the system
coordinates rotates then to a new position y 0 .
2. Rotate the system counterclockwise by the angle
around the y 0 -axis. The system z-axis rotates now to
a new position z 0 .
3. Rotate the system counterclockwise by the angle
around the z 0 -axis.

Using matrices:

R(, , ) Rz0 ()Ry0 ()Rz ().

SU(2) the spin returns to its original direction after time
In the two dimensional space t = 2/c .

{|Sz ; i, |Sz ; i} the wave vector returns to its original value after
time t = 4/c .
the spin operators
 
h Matrix representation
Sx = {(|Sz ; ihSz ; |) + (|Sz ; ihSz ; |)}
2 In the basis {|Sz ; i, |Sz ; i} the base vectors are
 
ih represented as
Sy = {(|Sz ; ihSz ; |) + (|Sz ; ihSz ; |)}
2    
1 0
|Sz ; i 7 |Sz ; i 7
 
h
Sz = {(|Sz ; ihSz ; |) (|Sz ; ihSz ; |)} 0 1
2
hSz ; | 7 (1, 0) hSz ; | 7 (0, 1) ,
satisfy the angular momentum commutation relations
so an arbitrary state vector is represented as
[Sx , Sy ] = ihSz + cyclic permutations.  
hSz ; |i
Thus the smallest dimension where these commutation |i 7
hSz ; |i
relations can be realized is 2.
h| 7 (h|Sz ; i, h|Sz ; i).
The state

|i = |Sz ; ihSz ; |i + |Sz ; ihSz ; |i The column vector

   
hSz ; |i c
behaves in the rotation =
  hSz ; |i c
iSz
Dz () = exp is called the two component spinor
h
like Paulis spin matrices

iSz
 Paulis spin matrices i are defined via the relations
Dz ()|i = exp |i
h  
h
(Sk )ij (k )ij ,
= ei/2 |Sz ; ihSz ; |i 2
+ei/2 |Sz ; ihSz ; |i.
where the matrix elements are evaluated in the basis
In particular: {|Sz ; i, |Sz ; i}.
Dz (2)|i = |i. For example
 
h
Spin precession S1 = Sx = {(|Sz ; ihSz ; |) + (|Sz ; ihSz ; |)},
2
When the Hamiltonian is
so
H = c Sz
(S1 )11 = (S1 )22 = 0
the time evolution operator is h
(S1 )12 = (S1 )21 = ,
2
 
iSz c t
U(t, 0) = exp = Dz (c t).
h or  
h 0 1
Looking at the equations (S1 ) = .
2 1 0
hJx i R hJx i cos hJy i sin Thus we get
hJy i R hJy i cos + hJx i sin      
0 1 0 i 1 0
hJz i R hJz i 1 = , 2 = , 3 = .
1 0 i 0 0 1
one can read that The spin matrices satisfy the anticommutation relations
hSx it = hSx it=0 cos c t hSy it=0 sin c t
{i , j } i j + j i = 2ij
hSy it = hSy it=0 cos c t + hSx it=0 sin c t
hSz it = hSz it=0 . and the commutation relations

We see that [i , j ] = 2iijk k .

Moreover, we see that Eulers angles
The spinor rotation matrices corresponding to rotations
i = i , around z and y axes are
det(i ) = 1,  i/2 
e 0
tr(i ) = 0. Dz () 7
0 ei/2
 
Often the collective vector notation cos /2 sin /2
Dy () 7 .
sin /2 cos /2
1 x + 2 y + 3 z. With the help of Eulers angles , and the rotation
matrices can be written as
is used for spin matrices. For example we get
1
X D(, , ) 7 D( 2 ) (, , ) =
a ak k

 

 
ei(+)/2 cos 2 ei()/2 sin 2
k   .

 
ei()/2 sin 2 ei(+)/2 cos 2
 
+a3 a1 ia2
= .
a1 + ia2 a3
We seek for the eigenspinor of the matrix n:
and
n = .
X
( a)( b) = j aj k bk Now
j,k sin cos
X1 n = sin sin ,
= ({j , k } + [j , k ]) aj bk
2 cos
j,k
X so
sin ei
 
= (jk + ijki i )aj bk cos
n = .
j,k sin ei cos
= a b + i (a b). The state where the spin is parallel to the unit vector n,
is obviously invariant under rotations
A special case of the latter formula is
Dn () = eiS n/h
( a)2 = |a|2 .
and thus an eigenstate of the operator S n.
Now This kind of state can be obtained by rotating the state
    |Sz ; i
iS n i n
D(n, ) = exp 7 exp = 1. angle around y axis,
h 2
   
2. angle around z axis,
1 cos i n sin =
2 2
      i.e.
cos 2 inz sin 2 (inx ny ) sin 2
      S n|S n; i = S nD(, , 0)|Sz ; i
(inx + ny ) sin 2 cos 2 + inz sin 2  
h
= D(, , 0)|Sz ; i
2
and the spinors behave in rotations like  
h
= |S n; i.

i n
 2
exp .
2 Correspondingly for spinors the vector
i/2
 
Note the notation does not mean that would behave cos e
2 
1
in rotations like a vector, k R k . Instead we have = D( 2 ) (, , 0)|Sz ; i =

sin 2 ei/2
X
k Rkl l .
l
is an eigenstate of the matrix n.

For all directions n one has SU(2)

  As a representation of rotations the 2 2-matrices
i n 
exp = 1, for any n.
D( 2 ) (n, ) = ei n/2
1
2 
=2

form obviously a group. These matrices have two

characteristic properties:
 1. unitarity
 1
  1 1
D( 2 ) = D( 2 ) ,

2. unimodularity  1 
 ( 2 )
D  = 1.

A unitary unimodular matrix can be written as

 
a b
U (a, b) = .
b a
The unimodularity condition gives
1 = |U | = |a|2 + |b|2 ,
and we are left with 3 free parameters.
The unitarity condition is automatically satisfied because
  
a b a b
U (a, b) U (a, b) =
b a b a
|a|2 + |b|2
 
0
= = 1.
0 |a|2 + |b|2
Matrices U (a, b) form a group since
the matrix
U (a1 , b1 )U (a2 , b2 ) = U (a1 a2 b1 b2 , a1 b2 + a2 b1 )
is unimodular because
|U (a1 a2 b1 b2 , a1 b2 + a2 b1 )| =
|a1 a2 b1 b2 |2 + |a1 b2 + a2 b1 |2 = 1,
and thus also unitary.
as a unitary matrix U has the inverse matrix:
U 1 (a, b) = U (a, b) = U (a , b).

the unit matrix 1 is unitary and unimodular.

The group is called SU(2).
Comparing with the previous spinor representation
1
D( 2 ) (n, ) =
     
cos 2 inz sin 2 (inx ny ) sin 2
     
(inx + ny ) sin 2 cos 2 + inz sin 2

we see that
 

 
Re(a) = cos 2 Im(a) = nz sin
2
 

 
Re(b) = ny sin 2 Im(b) = nx sin .
2
The complex numbers a and b are known as
Cayley-Kleins parameters.
Note O(3) and SU(2) are not isomorphic.
Example
In O(3): 2- and 4-rotations 7 1
In SU(2): 2-rotation 7 1 and 4-rotation 7 1.
The operations U (a, b) and U (a, b) in SU(2)
correspond to a single matrix of O(3). The map SU(2) 7
O(3) is thus 2 to 1. The groups are, however, locally
isomorphic.
Angular momentum algebra The product of matrices belongs to the group:
It is easy to see that the operator (j)
X (j) (j)
Dm00 m (R1 R2 ) = Dm00 m0 (R1 )Dm0 m (R2 ),
J 2 = Jx Jx + Jy Jy + Jz Jz m0

commutes with the operators Jx , Jy and Jz , where R1 R2 is the combined rotation of the rotations
R1 and R2 ,
[J 2 , Ji ] = 0.
the inverse operation belongs to the group:
We choose the component Jz and denote the common (j) (j)
eigenstate of the operators J 2 and Jz by |j, mi. We know Dm0 m (R1 ) = Dmm0 (R).
(QM II) that The state vectors |j, mi transform in rotations like
2 21 3 X
J |j, mi = j(j + 1)h |j, mi, j = 0, , 1, , . . . D(R)|j, mi = |j, m0 ihj, m0 |D(R)|j, mi
2 2
m0
Jz |j, mi = mh|j, mi, m = j, j + 1, . . . , j 1, j. X (j)
= |j, m0 iDm0 m (R).
We define the ladder operators J+ and J : m0

J Jx iJy . With the help of the Euler angles

(j)
They satisfy the commutation relations Dm0 m (R) =
     
iJz iJy iJz
[J+ , J ] = 2hJz hj, m0 | exp exp exp |j, mi
h h h
[Jz , J ] = hJ 0 (j)
 2 
J , J = 0. = ei(m +m) dm0 m (),

We see that where

 
(j) 0 iJy
Jz J+ |j, mi = hJ+ Jz |j, mi = (m + 1)hJ+ |j, mi dm0 m () hj, m | exp |j, mi.
h
and (j)
Functions dm0 m can be evaluated using Wigners formula
2 2
J J+ |j, mi = J+ J |j, mi = j(j + 1)hJ+ |j, mi, (j)
dm0 m () =
so we must have X 0
(1)km+m
J+ |j, mi = c+ |j, m + 1i k
p
The factor c+ can be deduced from the normalization (j + m)!(j m)!(j + m0 )!(j m0 )!

condition (j + m k)!k!(j k m0 )!(k m + m0 )!
hj, m|j 0 , m0 i = jj 0 mm0 . 

2j2k+mm0 

2km+m0
cos sin .
We end up with 2 2
p
J |j, mi = (j m)(j m + 1)h|j, m 1i.
Orbital angular momentum
Matrix elements will be The components of the classically analogous operator
L = x p satisfy the commutation relations
hj 0 , m0 |J 2 |j, mi = j(j + 1)h2 j 0 j m0 m
[Li , Lj ] = iijk hLk .
hj 0 , m0 |Jz |j, mi = mhj 0 j m0 m
hj 0 , m0 |J |j, mi =
p
(j m)(j m + 1)hj 0 j m0 ,m1 . Using the spherical coordinates to label the position
eigenstates,
We define Wigners function: |x0 i = |r, , i,
one can show that
 
(j) 0 iJ n
Dm0 m (R) = hj, m | exp |j, mi.
h
hx0 |Lz |i = ih hx0 |i
Since
 

hx0 |Lx |i = hx0 |i
 
iJ n ih sin cot cos
[J 2 , D(R)] = [J 2 , exp ] = 0,
h  

we see that D(R) does not chance the j-quantum number, hx0 |Ly |i = ih cos cot sin hx0 |i

so it cannot have non zero matrix elements between  

states with different j values. hx0 |L |i = ihe i
i cot hx0 |i
(j)
The matrix with matrix elements Dm0 m (R) is the
1 2
  
(2j + 1)-dimensional irreducible representation of the 1
hx0 |L2 |i = h 2
+ sin
rotation operator D(R). sin2 2 sin
(j) 0
The matrices Dm0 m (R) form a group: hx |i.
We denote the common eigenstate of the operators L2 or s 
and Lz by the ket-vector |l, mi, i.e. (l) 4 m

Dm0 (, , 0) = Yl (, ) .

(2l + 1) 
Lz |l, mi = mh|l, mi ,

L2 |l, mi = l(l + 1)h2 |l, mi. As a special case

(l) (l)
Since R3 can be represented as the direct product D00 (, , 0) = d00 () = Pl (cos ).

R3 = R ,
Coupling of angular momenta
where is the surface of the unit sphere We consider two Hilbert spaces H1 and H2 . If now Ai is
(position=distance from the origin and direction) the an operator in the space Hi , the notation A1 A2 means
position eigenstates can be written correspondingly as the operator
|x0 i = |ri|ni. A1 A2 |i1 |i2 = (A1 |i1 ) (A2 |i2 )
Here the state vectors |ni form a complete basis on the in the product space. Here |ii Hi . In particular,
surface of the sphere, i.e.
Z A1 12 |i1 |i2 = (A1 |i1 ) |i2 ,
dn |nihn| = 1.
where 1i is the identity operator of the space Hi .
Correspondingly 11 A2 operates only in the subspace
We define the spherical harmonic function:
H2 of the product space. Usually the subspace of the
Ylm (, ) = Ylm (n) = hn|l, mi. identity operators, or even the identity operator itself, is
not shown, for example
The scalar product of the vector hn| with the equations
A1 12 = A1 1 = A1 .
Lz |l, mi = mh|l, mi
It is easy to verify that operators operating in different
L2 |l, mi = l(l + 1)h2 |l, mi
subspace commute, i.e.
gives
m [A1 12 , 11 A2 ] = [A1 , A2 ] = 0.
ih Y (, ) = mhYlm (, )
l In particular we consider two angular momenta J1 and J2
and operating in two different Hilbert spaces. They commute:
1 2
   
1 [J1i , J2j ] = 0.
sin + + l(l + 1) Ylm = 0.
sin sin2 2
The infinitesimal rotation affecting both Hilbert spaces is
   
Ylm and D(l) iJ 1 n iJ 2 n
1 1 =
The state h h
|ni = |, i i(J 1 1 + 1 J 2 ) n
1 .
is obtained from the state |zi rotating it first by the angle h
around y-axis and then by the angle around z-axis: The components of the total angular momentum
|ni = D(R)|zi J = J1 1 + 1 J2 = J1 + J2
= D( = , = , = 0)|zi
X obey the commutation relations
= D(, , 0)|l, mihl, m|zi.
l,m
[Ji , Jj ] = ihijk Jk ,
Furthermore
i.e. J is angular momentum.
Ylm (, )
(l) A finite rotation is constructed analogously:
X
hl, m|ni = = Dm0 m (, , 0)hl, m|zi.
m    
J 1 n J 2 n
D1 (R) D2 (R) = exp exp .
Now r h h
(2l + 1)
hl, m|zi = Ylm (0, ) = m0 , Base vectors of the whole system
4
so We seek in the product space {|j1 m1 i |j2 m2 i} for the
r
(2l + 1) (l) maximal set of commuting operators.
Ylm (, ) = Dm0 (, , = 0) (i) J 21 , J 22 , J1z and J2z .
4
Their common eigenstates are simply direct products is unitary. The elements hj1 j2 ; m1 m2 |j1 j2 ; jmi of the
transformation matrix are called Clebsch-Gordans
|j1 j2 ; m1 m2 i |j1 , m1 i |j2 , m2 i. coefficients.
Since
If j1 and j2 can be deduced from the context we often
Jz = J1z + J2z ,
denote
|m1 m2 i = |j1 j2 ; m1 m2 i. we must have
m = m1 + m2 ,
The quantum numbers are obtained from the
(eigen)equations so the Clebsch-Gordan coefficients satisfy the condition

J 21 |j1 j2 ; m1 m2 i = j1 (j1 + 1)h2 |j1 j2 ; m1 m2 i hj1 j2 ; m1 m2 |j1 j2 ; jmi = 0, if m 6= m1 + m2 .

J1z |j1 j2 ; m1 m2 i = m1 h|j1 j2 ; m1 m2 i Further, we must have (QM II)
J 22 |j1 j2 ; m1 m2 i = j2 (j2 + 1)h2 |j1 j2 ; m1 m2 i
|j1 j2 | j j1 + j2 .
J2z |j1 j2 ; m1 m2 i = m2 h|j1 j2 ; m1 m2 i.
It turns out, that the C-G coefficients can be chosen to be
(ii) J 2 , J 21 , J 22 and Jz . real, so the transformation matrix C is in fact orthogonal:
Their common eigenstate is denoted as X
hj1 j2 ; m1 m2 |j1 j2 ; jmihj1 j2 ; m01 m02 |j1 j2 ; jmi
|j1 j2 ; jmi jm

or shortly = m1 m01 m2 m02

X
|jmi = |j1 j2 ; jmi hj1 j2 ; m1 m2 |j1 j2 ; jmihj1 j2 ; m1 m2 |j1 j2 ; j 0 m0 i
if the quantum numbers j1 and j2 can be deduced from m1 m2

the context. The quantum numbers are obtained from the = jj 0 mm0 .
equations
As a special case (j 0 = j and m0 = m = m1 + m2 ) we get
J 21 |j1 j2 ; jmi = j1 (j1 + 1)h2 |j1 j2 ; jmi the normalization condition
J 22 |j1 j2 ; jmi j2 (j2 + 1)h2 |j1 j2 ; jmi
X
= |hj1 j2 ; m1 m2 |j1 j2 ; jmi|2 = 1.
J 2 |j1 j2 ; jmi = j(j + 1)h2 |j1 j2 ; jmi m1 m2

Jz |j1 j2 ; jmi = mh|j1 j2 ; jmi. Recursion formulas

Operating with the ladder operators to the state
Now |j1 j2 ; jmi we get
[J 2 , J1z ] 6= 0, [J 2 , J2z ] 6= 0,
J |j1 j2 ; jmi =
so we cannot add to the set (i) the operator J 2 , nor to X
the set (ii) the operators J1z or J2z . Both sets are thus (J1 + J2 ) |j1 j2 ; m1 m2 i
maximal and the corresponding bases complete (and m1 m2
orthonormal), i.e. hj1 j2 ; m1 m2 |j1 j2 ; jmi,
X X
|j1 j2 ; m1 m2 ihj1 j2 ; m1 m2 | = 1 or
j1 j2 m1 m2
p
XX (j m)(j m + 1)|j1 j2 ; j, m 1i
|j1 j2 ; jmihj1 j2 ; jm| = 1. X X q
j1 j2 jm
= (j1 m01 )(j1 m01 + 1)
m01 m02
In the subspace where the quantum numbers j1 and j2 |j1 j2 ; m01 1, m02 i
are fixed we have the completeness relations q
X + (j2 m02 )(j2 m02 + 1)
|j1 j2 ; m1 m2 ihj1 j2 ; m1 m2 | = 1 
m1 m2 |j1 j2 ; m01 , m02 1i
X
|j1 j2 ; jmihj1 j2 ; jm| = 1. hj1 j2 ; m01 m02 |j1 j2 ; jmi.
jm
Taking the scalar product on the both sides with the
One can go from the basis (i) to the basis (ii) via the vector hj1 j2 ; m1 m2 | we get
unitary transformation p
X (j m)(j m + 1)hj1 j2 ; m1 m2 |j1 j2 ; j, m 1i
|j1 j2 ; jmi = |j1 j2 ; m1 m2 ihj1 j2 ; m1 m2 |j1 j2 ; jmi, =
p
(j1 m1 + 1)(j1 m1 )
m1 m2
hj1 j2 ; m1 1, m2 |j1 j2 ; jmi
so also the transformation matrix p
+ (j2 m2 + 1)(j2 m2 )
(C)jm,m1 m2 = hj1 j2 ; m1 m2 |j1 j2 ; jmi hj1 j2 ; m1 , m2 1|j1 j2 ; jmi.
The Clebsch-Gordan coefficients are determined uniquely Example L + S-coupling.
by Now

1. the recursion formulas. j1 = l = 0, 1, 2, . . .

m1 = ml = l, l + 1, . . . , l 1, l
2. the normalization condition
1
X j2 = s=
|hj1 j2 ; m1 m2 |j1 j2 ; jmi|2 = 1. 2
m1 m2 1
m2 = ms =
2
l 12 , when l > 0
(
3. the sign conventions, for example
j = 1
hj1 j2 ; j m |J1z |j1 j2 ; jmi 0. 0 0 2, when l = 0.

m s
Note Due to the sign conventions the order of the
coupling is essential:
J - J - J -
|j1 j2 ; jmi = |j2 j1 ; jmi. 1 /2
m
Graphical representation of recursion formulas
l l

(m 1 - 1 ,m 2 ) (m 1 ,m 2 ) (m 1 ,m 2 + 1 )
-1 /2

J
Recursion when j1 = l and j2 = s = 1/2

J
+ Using the selection rule
- 1 1
m1 = ml = m , m2 = ms =
2 2
(m 1 ,m 2 -1 ) (m 1 ,m 2 ) (m 1 + 1 ,m 2 ) and the shorthand notation the J -recursion gives
q
Recursion formula in m1 m2 -plane (l + 21 + m + 1)(l + 12 m)hm 21 , 12 |l + 12 , mi
We fix j1 , j2 and j. Then q
= (l + m + 12 )(l m + 21 )
|m1 | j1 , |m2 | j2 , |m1 + m2 | j.
hm + 12 , 21 |l + 12 , m + 1i,
m = j2
m

2 or
1
+ m

A
2
= j
m

s
l + m + 12
1
= -j1

1 1 1
hm 2 , 2 |l + 2 , mi = hm + 12 , 12 |l + 12 , m + 1i.
l + m + 32
m

m
1
= j1

1 + m Applying the same recursion repeatedly we have

2 = -
j m 2 = -j2 hm 21 , 12 |l + 12 , mi
(a ) s s
l + m + 12 l + m + 23
D A = hm + 23 , 12 |l + 12 , m + 2i
J l + m + 32 l + m + 25
J J
+
fo rb id d e n

E
s s s
- - l + m + 12 l + m + 23 l + m + 25
J B =
J -
+ l + m + 32 l + m + 25 l + m + 27
F C hm + 52 , 12 |l + 12 , m + 3i
=
(b ) ..
. s
Using recursion formulas
l + m + 12
We see that = hl, 21 |l + 12 , l + 12 i.
2l + 1
1. every C-G coefficient depends on A,
If j = jmax = j1 + j2 and m = mmax = j1 + j2 one must
2. the normalization condition determines the absolute have
value of A,
|j1 j2 ; jmi =
3. the sign is obtained from the sign conventions. hj1 j2 ; m1 = j1 , m2 = j2 |j1 j2 ; jmi|j1 m1 i|j2 m2 i.
Now the normalization condition On the other hand we have

|hj1 j2 ; m1 = j1 , m2 = j2 |j1 j2 ; jmi|2 = 1 hj1 j2 ; m1 m2 |D(R)|j1 j2 ; m01 m02 i

XX
and the sign convention give = hj1 j2 ; m1 m2 |j1 j2 ; jmi
jm j 0 m0

hj1 j2 ; m1 = j1 , m2 = j2 |j1 j2 ; jmi = 1. hj1 j2 ; jm|D(R)|j1 j2 ; j 0 m0 i

hj1 j2 ; j 0 m0 |j1 j2 ; m01 m02 i
Thus, in the case of the spin-orbit coupling, XX (j)
= hj1 j2 ; m1 m2 |j1 j2 ; jmiDmm0 (R)jj 0
hl, 12 |l + 12 , l + 12 i = 1, jm j 0 m0

hj1 j2 ; m01 m02 |j1 j2 ; j 0 m0 i.

or s
l + m + 12 We end up with the Clebsch-Gordan series
hm 12 , 12 |l + 12 , mi = .
2l + 1 (j ) (j )
Dm11 m0 (R)Dm22 m0 (R) =
With the help of the recursion relations, normalization X1
XX 2
condition and sign convention the rest of the C-G hj1 j2 ; m1 m2 |j1 j2 ; jmi
coefficients can be evaluated, too. We get j m m0
(j)
r r hj1 j2 ; m01 m02 |j1 j2 ; jm0 iDmm0 (R).
l + m + 12 l m + 21
|j = l + 21 , mi
 
r 2l + 1 r 2l + 1

=
|j = l 12 , mi l m + 21 l + m + 21
As an application we have
2l + 1 2l + 1
1 1
 
|ml = m 2 , ms = 2 i
Z
|ml = m + 12 , ms = 12 i
. dYlm (, )Ylm
1
1
(, )Ylm
2
2
(, )
s
(2l1 + 1)(2l2 + 1)
Rotation matrices =
4(2l + 1)
If D(j1 ) (R) is a rotation matrix in the base
hl1 l2 ; 00|l1 l2 ; l0ihl1 l2 ; m1 m2 |l1 l2 ; lmi.
{|j1 m1 i|m1 = j1 , . . . , j1 } and D(j2 ) (R) a rotation matrix
in the base {|j2 m2 i|m2 = j2 , . . . , j2 }, then
D(j1 ) (R) D(j2 ) (R) is a rotation matrix in the
(2j1 + 1) (2j2 + 1)-dimensional base
{|j1 , m1 i |j2 , m2 i}. Selecting suitable superpositions of
the vectors |j1 , m1 i |j2 , m2 i the matrix takes the form
like

D(j1 ) (R) D(j2 ) (R)

D(j1 +j2 ) 0
D(j1 +j2 1)
.

..

.

0 D(|j1 j2 |)

One can thus write

D(j1 ) D(j2 ) = D(j1 +j2 ) D(j1 +j2 1) D(|j1 j2 |) .

As a check we can calculate the dimensions:

(2j1 + 1)(2j2 + 1) =
2(j1 + j2 ) + 1 + 2(j1 + j2 1) + 1
+ + 2|j1 j2 | + 1.

The matrix elements of the rotation operator satisfy

hj1 j2 ; m1 m2 |D(R)|j1 j2 ; m01 m02 i

= hj1 m1 |D(R)|j1 m01 ihj2 m2 |D(R)|j2 m02 i
(j ) (j )
= Dm11 m0 (R)Dm22 m0 (R).
1 2
3j- 6j- and 9j-symbols where
The Clebsch-Gordan coefficients obey certain symmetry 
1, when |j1 j2 | j3 j1 + j2
relations, like (j1 j2 j3 ) =
0, otherwise.
hj1 j2 ; m1 m2 |j1 j2 ; jmi
= (1)j1 +j2 j hj2 j1 ; m2 m1 |j2 j1 ; jmi 6j-symbols
Let us couple three angular momenta, j1 , j2 and j3 , to
hj1 j2 ; m1 m2 |j1 j2 ; j3 m3 i
s the angular momentum J. There are two ways:
2j3 + 1
= (1)j2 +m2 hj2 j3 ; m2 , m3 |j2 j3 ; j1 m1 i 1. first j1 , j2 j12 and then j12 , j3 J.
2j1 + 1
2. first j2 , j3 j23 and then j23 , j1 J.
hj1 j2 ; m1 m2 |j1 j2 ; j3 m3 i
Lets choose the first way. The quantum number j12 must
s
2j3 + 1
= (1)j1 m1 hj3 j1 ; m3 , m1 |j3 j1 ; j2 m2 i satisfy the selection rules
2j2 + 1
hj1 j2 ; m1 m2 |j1 j2 ; j3 m3 i |j1 j2 | j12 j1 + j2
= (1)j1 +j2 j3 hj1 j2 ; m1 , m2 |j1 j2 ; j3 , m3 i. |j12 j3 | J j12 + j3 .

Note The first relation shows that the coupling order is

The states belonging to different j12 are independent so
essential. we must specify the intermediate state j12 . We use the
We define more symmetric 3j-symbols: notation
  |(j1 j2 )j12 j3 ; JM i.
j1 j2 j3 Explicitely one has

m1 m2 m3
(1)j1 j2 m2 |(j1 j2 )j12 j3 ; JM i
hj1 j2 ; m1 m2 |j1 j2 ; j3 , m3 i. XX
= |j1 j2 ; j12 m12 i|j3 m3 i
p
2j3 + 1
m12 m3
They satisfy hj12 j3 ; m12 m3 |j12 j3 ; JM i
 
j1 j2 j3
X
= |j1 m1 i|j2 m2 i|j3 m3 i
m1 m2 m3 m1 m2 m3 m12
   
j2 j3 j1 j3 j1 j2 hj1 j2 ; m1 m2 |j1 j2 ; j12 m12 i
= =
m2 m3 m1 m3 m1 m2 hj12 j3 ; m12 m3 |j12 j3 ; JM i.
   
j1 j2 j3 j2 j1 j3
(1)j1 +j2 +j3 = Correspondingly the angular momenta coupled in way 2
m1 m2 m3 m2 m1 m3
    satisfy
j1 j3 j2 j3 j2 j1
= = |j1 (j2 j3 )j23 ; JM i
m1 m3 m2 m3 m2 m1 XX
= |j1 m1 i|j2 j3 ; j23 m23 i
 
j1 j2 j3
m1 m2 m3 m23 m1

j1 j2 j3
 hj1 j23 ; m1 m23 |j1 j23 ; JM i
= (1)j1 +j2 +j3 . X
m1 m2 m3 = |j1 m1 i|j2 m2 i|j3 m3 i
m1 m2 m3 m23
As an application, we see that the coefficients
 3 3    hj2 j3 ; m2 m3 |j2 j3 ; j23 m23 i
2 2 2 2 2 3 hj1 j23 ; m1 m23 |j1 j23 ; JM i.
1 1 , .
2 2 1 1 1 2
Both bases are complete so there is a unitary transform
vanish. between them:
On the other hand, the orthogonality properties are X
somewhat more complicated: |j1 (j2 j3 )j23 ; JM i = |(j1 j2 )j12 j3 ; JM i
   j12
XX j1 j2 j3 j1 j2 j3
(2j3 + 1) h(j1 j2 )j12 j3 ; JM |j1 (j2 j3 )j23 ; JM i.
m1 m2 m3 m01 m02 m3
j3 m3
In the transformation coefficients, recoupling coefficients it
= m1 m01 m2 m02
is not necessary to show the quantum number M , because
and Theorem 1 In the transformation
X X  j1 j30
 
j2 j3 j1 j2
X
|; jmi = |; jmih; jm|; jmi
m1 m2 m3 m1 m2 m03
m1 m2
j3 j30 m3 m03 (j1 j2 j3 ) the coefficients h; jm|; jmi do not depend on the
= p ,
2j3 + 1 quantum number m.
Proof: Let us suppose that m < j. Now
X
|; j, m + 1i = |; j, m + 1ih; j, m + 1|; j, m + 1i.

On the other hand

J+
|; j, m + 1i = p |; jmi
h (j + m + 1)(j m)
X
= |; j, m + 1ih; jm|; jmi,

so

h; j, m + 1|; j, m + 1i = h; j, m|; j, mi

The explicit expression for the recoupling coefficients will

be

h(j1 j2 )j12 j3 ; J|j1 (j2 j3 )j23 ; Ji

X
= hj12 j3 ; JM |j12 j3 ; m12 m3 i
m1 m2 m3
m12 m23
hj1 j2 ; j12 m12 |j1 j2 ; m1 m2 i
hj2 j3 ; m2 m3 |j2 j3 ; j23 m23 i
hj1 j23 ; m1 m23 |j1 j23 ; JM i.

We define the more symmetric 6j-symbols:

 
j1 j2 j12
j3 J j23
(1)j1 +j2 +j3 +J
p
(2j12 + 1)(2j23 + 1)
h(j1 j2 )j12 j3 ; J|j1 (j2 j3 )j23 ; Ji
(1)j1 +j2 +j3 +J
=p
(2j12 + 1)(2j23 + 1)
X
hj1 j2 ; m1 m2 |j1 j2 ; j12 , m1 + m2 i
m1 m2
hj12 j3 ; m1 + m2 , M m1 m2 |j12 j3 ; JM i
hj2 j3 ; m2 , M m1 m2 |j2 j3 ; j23 , M m1 i
hj1 j23 ; m1 , M m1 |j1 j23 ; JM i.

We can handle analogously the coupling of 4 angular

momenta. Transformations from a coupling scheme to
another are mediated by the 9j-symbols:

j1 j2 j12
j3 j4 j34
j13 j24 j

h(j1 j2 )j12 (j3 j4 )j34 ; j|(j1 j3 )j13 (j2 j4 )j24 ; ji

p .
(2j12 + 1)(2j34 + 1)(2j13 + 1)(2j24 + 1
Tensor operators Applying the Baker-Hausdorff lemma
We have used the vector notation for three component
operators for example to express the scalar product, like eiG AeiG =
i2 2
 
0 0 0 0 A + i[G, A] + [G, [G, A]] +
p x = p x x + p y y + pz z .
2!
 n n
Classically a vector is a quantity that under rotations i
+ [G, [G, [G, . . . [G, A]]] . . .] +
transforms like V R3 (or C 3 ), i.e. if R O(3), then n!
3
X we end up with the commutators
Vi0 = Rij Vj .
j=1
[Jj , [Jj , [ [Jj , Vi ] ]]].
In quantum mechanics V is a vector operator provided
that hV i C 3 is a vector: These will be evaluated in turn into operators Vi and Vk
(k 6= i, j).
R h|Vi |iR = h|D (R)Vi D(R)|i A vector operator (V ) is defined so that it satisfies the
3
X commutation relation
= Rij h|Vj |i,
j=1 [Vi , Jj ] = ihijk Vk .
|i H, R O(3).
We can easily see that for example p, x and J are vector
Thus we must have operators.
X In classical mechanics a quantity which under rotations
D (R)Vi D(R) = Rij Vj .
transforms like
j
XXX
Thus the infinitesimal rotations T ijk Rii0 Rjj 0 Rkk0 Ti0 j 0 k0 ,
| {z } i0 j 0 k0
iJ n n indeces
D(n) = 1
h
is called a Cartesian tensor of the rank n.
satisfy Example The dyad product of the vectors U and V
   
iJ n iJ n
1+ Vi 1 + Tij = Ui Vj
h h
i is a tensor of rank 2.
= Vi + (J nVi Vi J n) + O(2 )
X h Cartesian tensors are reducible, for example the dyad
= Rij Vj product can be written as
j
U V (Ui Vj Uj Vi )
or Ui Vj = ij +
 X 3 2
Vi + [Vi , J n] = Rij (n)Vj .  
ih Ui Vj + Uj Vi U V
j + ij .
2 3
Substituting the explicit expressions for infinitesimal
rotations, like We see that the terms transform under rotations
2 differently:
1 2

 0
R(z) = 1 2
2
0 , U 3 V ij is invariant. There is 1 term.


0 0 1
(Ui Vj Uj Vi )
2 retains its antisymmetry. There are 3
we get
 terms.
Vx + [Vx , Jz ] = Vx Vy + O(3 ).
ih
 
Ui Vj + Uj Vi U V
Handling similarly the other components we end up with 2 3 ij retains its symmetry and
tracelessness. There are 5 terms.
[Vi , Jj ] = ihijk Vk .
We recognize that the number of terms checks and that
the partition might have something to do with the
Finite rotation
angular momentum since the multiplicities correspond to
A finite rotation specified by Euler angles is accomplished
the multiplicities of the angular momenta l = 0, 1, 2.
by rotating around coordinate axises, so we have to (k)
consider such expressions as We define the spherical tensor Tq of rank k so that the
    argument n of the spherical function
iJj iJj
exp Vi exp . Ylm (n) = hn|lmi
h h
is replaced by the vector V : Under the infinitesimal rotations
 
m=q
Tq(k) = Yl=k (V ). iJ n
D(n) = 1
h
Example The spherical function Y1 :
r r r a spherical tensor behaves thus like
3 3 z (1) 3
Y10 = cos = 7 T0 = Vz 
iJ n
 
iJ n

4 4 r 4 1+ Tq(k)
1
r r   h h
3 x iy (1) 3 Vx iVy
Y11 = 7 T1 = . k  
4 2r 4 2 X (k)
0 iJ n
= Tq0 hkq | 1 + |kqi
0
h
Similarly we could construct for example a spherical q =k

tensor of rank 2: k k
(k) (k)
X X
r r = Tq0 hkq 0 |kqi + iTq0 hkq 0 |J n|kqi,
2 15 (x iy)2 (2) 15 q 0 =k q 0 =k
Y2 = 7 T2 = (Vx iVy )2 .
32 r2 32
or
(k)
X
The tensors
(k)
Tq
are irreducible, i.e. there does not exist [J n, Tq(k) ] = Tq0 hkq 0 |J n|kqi.
any proper subset q0

Choosing n = z and x iy we get

{Tp(k)
1
, Tp(k)
2
, . . .} {Tq(k) |q = k, . . . , +k},
(k) (k)
which would remain invariant under rotations. [Jz , Tq ] = hqTq

Transformation of spherical tensors and

Under the rotation R an eigenstate of the direction (k)
[J , Tq ] = h
p (k)
(k q)(k q + 1)Tq1 .
transforms like
Example Decomposition of the dyad product.
|ni |n0 i = D(R)|ni. We form spherical tensors of rank 1 from the vector
operators U and V :
The state vectors |lmi, on the other hand, transform
under the rotation R1 like
U0 = Uz , V0 = Vz ,
(l) Ux iUy Vx iVy
X
D(R1 )|l, mi = |l, m0 iDm0 m (R1 ). U1 = , V1 = .
m0 2 2
So we get Now

Ylm (n0 ) = hn0 |lmi = hn|D (R)|lmi (0) U V U+1 V1 + U1 V+1 U0 V0

T0 = = ,
X (l) 3 3
= hn|D(R1 )|lmi = hn||lm0 iDm0 m (R1 )
(U V )q
m0 Tq(1) = ,
X 0 (l) i 2
= Ylm (n)Dm0 m (R1 ) (2)
m0
T2 = U1 V1 ,
X 0 (l) U1 V0 + U0 V1
= Ylm (n)Dmm0 (R). (2)
T1 = ,
m0 2
(2) U+1 V1 + 2U0 V0 + U1 V+1
We define a tensor operator Ylm (V ) so that T0 = .
6
0 (l)
X
D (R)Ylm (V )D(R) = Ylm (V )Dmm0 (R). In general we have
m0 (k ) (k )
Theorem 1 Let Xq1 1 and Zq2 2 be irreducible spherical
(k) tensors of rank k1 and k2 . Then
Generalizing we define: Tq is a (2k + 1)-component
spherical tensor of rank k if and only if
XX
Tq(k) = hk1 k2 ; q1 q2 |k1 k2 ; kqiXq(k1 1 ) Zq(kq 2 )
q1 q2
k
(k) (k)
X

D (R)Tq(k) D(R) = Dqq0 (R)Tq0
is a (irreducible) spherical tensor of rank k.
q 0 =k (k)
Proof: We show that Tq transforms like
or equivalently
k
(k) (k)
X

(k) Pk (k) (k) D (R)Tq(k) D(R) = Dqq0 (R)Tq0 .
D(R)Tq D (R) = q 0 =k D q 0 q (R)Tq 0 . q 0 =k
Now Proof: Due to the property

D (R)Tq(k) D(R) [Jz , Tq(k) ] = hqTq(k)

XX
= hk1 k2 ; q1 q2 |k1 k2 ; kqi we have
q1 q2
h0 , j 0 m0 |[Jz , Tq(k) ] hqTq(k) |, jmi
D (R)Xq(k1 1 ) D(R)D (R)Zq(k2 2 ) D(R)
XXXX = [(m0 m)h qh] h0 , j 0 m0 |Tq(k) |, jmi = 0,
= hk1 k2 ; q1 q2 |k1 k2 ; kqi
q1 q2 q10 q20 so
(k ) (k ) (k ) (k )
h0 , j 0 m0 |Tq(k) |, jmi = 0,
Xq0 1 Dq0 1q1 (R1 )Zq0 2 Dq0 2q2 (R1 )
X1X X 1
XXX 2
X2 if m0 6= q + m
= hk1 k2 ; q1 q2 |k1 k2 ; kqi Theorem 3 (Wigner-Eckardts theorem) The matrix
k00 q1 q2 q10 q20 q 00 q0 elements of a tensor operator between eigenstates of the
hk1 k2 ; q10 q20 |k1 k2 ; k 00 q 0 i angular momentum satisfy the relation
(k00 ) (k ) (k )
hk1 k2 ; q1 q2 |k1 k2 ; k 00 q 00 iDq0 q00 (R1 )Xq0 1 Zq0 2 , h0 j 0 kT (k) kji
1 2 h0 , j 0 m0 |Tq(k) |, jmi = hjk; mq|jk; j 0 m0 i p ,
2j + 1
where we have substituted the Clebsch-Gordan series
expansion where the reduced matrix element h0 j 0 kT (k) kji depends
neither on the quantum numbers m, m0 nor on q.
(j ) (j )
Dm11 m0 (R)Dm22 m0 (R) = (k)
Proof: Since Tq is a tensor operator it satisfies the
1
X XX 2 condition
hj1 j2 ; m1 m2 |j1 j2 ; jmi
(k)
p
j m m0 [J , Tq(k) ] = h (k q)(k q + 1)Tq1 ,
(j)
hj1 j2 ; m01 m02 |j1 j2 ; jm0 iDmm0 (R) so
h0 , j 0 m0 |[J , Tq(k) ]|, jmi
Taking into account the orthogonality of the p (k)
= h (k q)(k q + 1)h0 , j 0 m0 |Tq1 |, jmi.
Clebsch-Gordan coefficients
X Substituting the matrix elements of the ladder operators
hj1 j2 ; m1 m2 |j1 j2 ; jmihj1 j2 ; m1 m2 |j1 j2 ; j 0 m0 i we get
m1 m2 p
= jj 0 mm0 (j 0 m0 )(j 0 m0 + 1)h0 , j 0 , m0 1|Tq(k) |, jmi
p
= (j m)(j m + 1)h0 , j 0 , m0 |Tq(k) |, j, m 1i
we get
(k)
p
+ (k q)(k q + 1)h0 , j 0 , m0 |Tq1 |, jmi.
D (R)Tq(k) D(R)
XXXXX If we now substituted j 0 j, m0 m, j j1 , m m1 ,
= kk00 qq00 hk1 k2 ; q10 q20 |k1 k2 ; k 00 q 0 i k j2 and q m2 , we would note that the recursion
k00 q10 q20 q 00 q0
formula above is exactly like the recursion formula for the
00
(k )
Dq0 q00 (R1 )Xq0 1 Zq0 2 ,
(k ) (k ) Clebsch-Gordan coefficients,
1 2 p
(j m)(j m + 1)hj1 j2 ; m1 m2 |j1 j2 ; j, m 1i
which can be rewritten as p
= (j1 m1 + 1)(j1 m1 )
D (R)Tq(k) D(R) hj1 j2 ; m1 1, m2 |j1 j2 ; jmi
p
X X (k ) (k )
+ (j2 m2 + 1)(j2 m2 )
= hk1 k2 ; q10 q20 |k1 k2 ; kq 0 iXq0 1 Zq0 2
1 2 hj1 j2 ; m1 , m2 1|j1 j2 ; jmi.
q0 q10 q20
P
(k) Both recursions are of the form j aij xj = 0, or sets of
Dq0 q (R1 )
X (k) (k) X (k) linear homogenous simultaneous equations with the same
(k)
= Tq0 Dq0 q (R1 ) = Dqq0 (R)Tq0 coefficients aij . So we have two sets of equations
q0 q0 X X
aij xj = 0, aij yj = 0,
j j
Matrix elements of tensor operators
one for the matrix elements (xi ) of the tensor operator
Theorem 2 The matrix elements of the tensor operator
(k) and the other for the Clebsch-Gordan coefficients (yi ).
Tq satisfy These sets of equations tell that
h0 , j 0 m0 |Tq(k) |, jmi = 0,
xj yj
unless m0 = q + m. = j and k fixed,
xk yk
so xj = cyj while c is a proportionality coefficient The coefficient cjm does not depend either on the
independent of the indeces j. Thus we see that quantum number m, because J V is a scalar operator,
so we can write it briefly as cj . Because cj does not
h0 , j 0 m0 |Tq(k) |, jmi depend on the operator V the above equation is valid
= ( constant independent on m, q and m0 ) also when V J and 0 , or
hjk; mq|jk; j 0 m0 i.
h, jm|J 2 |, jmi = h2 j(j + 1) = cj hjkJ kji.
If we write the proportionality coefficient like
If we now apply the Wigner-Eckart theorem to the
h0 j 0 kT (k) kji operators Vq and Jq we get
p
2j + 1
h0 , jm0 |Vq |, jmi h0 jkV kji
0 = .
we are through. h, jm |Jq |, jmi hjkJ kji
According to the Wigner-Eckart theorem a matrix
element of a tensor operator is a product of two factors, for the ratios of the matrix elements. On the other hand,
of which the right hand side of this equation is

hjk; mq|jk; j 0 m0 i depends only on the geometry, i.e. h0 , jm|J V |, jmi

,
on the orientation of the system with respect to the h, jm|J 2 |, jmi
z-axis.
so
h0 jp
0
kT (k) kji
depends on the dynamics of the h0 , jm|J V |, jmi
2j + 1 h0 , jm0 |Vq |, jmi = hjm0 |Jq |jmi
system. h2 j(j + 1)
As a special case we have the projection theorem: Generalizing one can show that the reduced matrix
Theorem 4 Let (k)
elements of the irreducible product Tq of two tensor
1 1 (k1 ) (k2 )
J1 = (Jx iJy ) = J , J0 = J z operators, Xq1 and Zq2 , satisfy
2 2
h0 j 0 ||T (k) kji
be the components of the tensor operator corresponding to
 
0 XX k1 k2 k
the angular momentum. Then = 2k + 1(1)k+j+j
00 00
j j0 j 00
j
0 0 h0 , jm|J V |, jmi
h , jm |Vq |, jmi = hjm0 |Jq |jmi. 0 0
h j kX (k1 ) 00 00 00 00
k j ih j |Z (k2 ) kji.
h2 j(j + 1)
Proof: Due to the expansions
(0) U V U+1 V1 + U1 V+1 U0 V0
T0 = = ,
3 3
(U V )q
Tq(1) = ,
i 2
(2)
T2 = U1 V1 ,
(2) U1 V0 + U0 V1
T1 = ,
2
(2) U+1 V1 + 2U0 V0 + U1 V+1
T0 =
6
we can write

h0 , jm|J V |, jmi
= h0 , jm|(J0 V0 J+1 V1 J1 V+1 ), jmi
= mhh0 , jm|V0 |, jmi
h p
+ (j + m)(j m + 1)h0 , j, m 1|V1 |, jmi
2
h p
(j m)(j + m + 1)h0 , j, m + 1|V+1 |, jmi
2
= cjm h0 jkV kji,

where, according to the Wigner-Eckart theorem the

coefficient cjm does not depend on , 0 or V .
Symmetry it follows now
[G, H] = 0,
Symmetries, constants of motion and so according to the Heisenberg equation of motion
degeneracies
Looking at the Lagrange equation of motion dA 1
= [A, H]
  dt ih
d L L
=0 we have
dt qi qi dG
= 0.
of classical mechanics one can see that if the Lagrangian dt
L(qi , qi ) is invariant under translations, i.e. In the Heisenberg formalism the observable G is thus a
constant of motion. if H is invariant for example under
L(qi , qi ) L(qi + qi , qi ) = L(qi , qi ),
translations then the momentum is constant of
the momentum motion.
L
pi = rotations then the angular momentum is a constant
qi
motion.
is a conserved quantity, i.e.
  Let us suppose now that the Hamiltonian is symmetric
dpi d L under the operations S generated by G:
= = 0.
dt dt qi
S HS = H
Formulating classical mechanics using the Hamiltonian
[S, H] = 0
function H(qi , pi ) the equations of motion take the forms
[G, H] = 0.
H
pi = Let |g 0 i be the eigenstates of G, i.e.
qi
H G|g 0 i = g 0 |g 0 i
qi = .
pi
and let the system at the moment t0 be in the eigenstate
Also looking at these one can see that if H is symmetric |g 0 i of G. Since the time evolution operator is a
under the operation functional of the Hamiltonian only,
qi qi + qi U = U [H],
there exists a conserved quantity: so
[G, U ] = 0.
pi = 0.
At the moment t we then have
In quantum mechanics operations of that kind
(translations, rotations, . . .) are associated with a unitary G|g 0 , t0 ; ti = GU (t0 , t)|g 0 i = U (t0 , t)G|g 0 i
symmetry operator. = g 0 |g 0 , t0 ; ti,
Let S be an arbitrary symmetry operator. We say that
the Hamiltonian H is symmetric, if or an eigenstate associated with a particular eigenvalue of
G remains always an eigenstate belonging to the same
[S, H] = 0, eigenvalue.
Let us consider now the energy eigenstates |ni, i.e.
or due to the unitarity of the operator S equivalently
H|ni = En |ni.
S HS = H.
When the Hamiltonian is symmetric under the operations
The matrix elements of the Hamiltonian are then S we have
invariant under that operation. H(S|ni = SH|ni = En S|ni.
In the case of a continuum symmetry we can look at
infinitesimal operations If now
|ni =
6 S|ni,
i
S = 1 G, then the energy states En are degenerate. Thus a
h
symmetry is also usually associated with a degeneracy.
where the Hermitean operator G is the generator of that Let us suppose now that the symmetry operation S can
symmetry. From the condition be parametrized with a continuous quantity, say :

S HS = H S = S().
When the Hamiltonian is symmetric under these
operations all states S()|ni have the same energy.
Example Rotations D(R).
If
[D(R), H] = 0,
then
[J , H] = 0, [J 2 , H] = 0.
So there exist simultaneous eigenvectors |n; jmi of the
operators H, J 2 ja Jz . Now all rotated states

D(R)|n; jmi

belong to the same energy eigenvalue. We know that

(j)
X
D(R)|n; jmi = |n; jm0 iDm0 m (R),
m0

that is, every rotated state is a superposition of (2j + 1)

linearly independent states. The degeneracy is thus
(2j + 1)-fold.
Example Atomic electron.
The potential acting on an electron is of form

U = V (r) + VLS L S.

Now
[J , H] = 0, [J 2 , H] = 0,
where
J = L + S.
The energy levels are thus (2j + 1)-foldly degenerated.
Lets set the atom in magnetic field parallel to the z-axis.
The Hamiltonian is then appended by the term

Z = cSz .

Now
[J 2 , Sz ] 6= 0,
so the rotation symmetry is broken and the (2j + 1)-fold
degeneracy lifted.
Parity are equivalent:
The parity or space inversion operation converts a right
handed coordinate system to left handed: T (dx0 ) = T (dx0 ).

x x, y y, z z. Substituting
i 0
T (dx0 ) = 1 dx p,
This is a case of a non continuous operation, i.e. the h
operation cannot be composed of infinitesimal operations. we get the condition
Thus the non continuous operations have no generator.
We consider the parity operation, i.e. we let the parity {, p} = 0 or p = p,
operator to act on vectors of a Hilbert space and keep
the coordinate system fixed: or the momentum changes its sign under the parity
operation.
|i |i.
Angular momentum and parity
Like in all symmetry operations we require that is In the case of the orbital angular momentum
unitary, i.e.
= 1. L=xp
Furthermore we require: one can easily evaluate

h| x|i = h|x|i |i. L = x p = x p = (x) (p)
So we must have = L,
x = x,
so the parity and the angular momentum commute:
or
x = x. [, L] = 0.
The operators x ja anti commute. In R3 the parity operator is the matrix
Let |x0 i be a position eigenstate, i.e.

1 0 0
x|x0 i = x0 |x0 i. P = 0 1 0 ,
0 0 1
Then
x|x0 i = x|x0 i = (x0 )|x0 i, so quite obviously
and we must have
P R = RP, R O(3).
|x0 i = ei | x0 i.
We require that the corresponding operators of the
The phase is usually taken to be = 0, so Hilbert space satisfy the same condition, i.e.

|x0 i = | x0 i. D(R) = D(R).

Applying the parity operator again we get Looking at the infinitesimal rotation

2 |x0 i = |x0 i D(n) = 1 iJ n/h,

or we see that
2
= 1. [, J ] = 0 or J = J ,
We see that which is equivalent to the transformation of the orbital
angular momentum.
the eigenvalues of the operator can be only 1, We see that under
1 = = . rotations x and J transform similarly, that is, like
vectors or tensors of rank 1.
Momentum and parity space inversions x is odd and J even.
We require that operations
We say that under the parity operation
translation followed by space inversion
odd vectors are polar,
space inversion followed by translation to the
opposite direction even vectors are axial or pseudovectors.
Let us consider such scalar products as p x and S x. The explicit expression for spherical functions is
One can easily see that under rotation these are invariant, s
scalars. Under the parity operation they transform like m m (2l + 1)(l m)! m
Yl (, ) = (1) Pl ()eim ,
4(l + m)!
p x = (p) (x) = p x
S x = S (x) = S x. from which as a special case, m = 0, we obtain
r
We say that quantities behaving under rotations like 0 2l + 1
scalars, spherical tensors of rank 0, which under the Yl (, ) = Pl (cos ).
4
parity operation are
Depending on the degree l of the Legendre polynomial it
even, are (ordinary) scalars, is either even or odd:
odd, are pseudoscalars. Pl (z) = (1)l Pl (z).

We see that
Wave functions and parity
Let be the wave function of a spinles particle in the hx0 ||, l0i = (1)l hx0 |, l0i,
state |i, i.e.
(x0 ) = hx0 |i. so the state vectors obey
Since the position eigenstates satisfy |, l0i = (1)l |, l0i.
|x0 i = | x0 i, Now
[, L ] = 0
the wave function of the space inverted state is
and
hx0 ||i = hx0 |i = (x0 ). Lr |, l0i |, l, ri,
Suppose that |i is a parity eigenstate, i.e. so the orbital angular momentum states satisfy the
relation
|i = |i. |, lmi = (1)l |, lmi.

The corresponding wave function obeys the the relation Theorem 1 If

[H, ] = 0,
(x0 ) = hx0 ||i = hx0 |i = (x0 ),
and |ni is an eigenstate of the Hamiltonian H belonging
i.e. it is an even or odd function of its argument. to the nondegenerate eigenvalue En , i.e.
Note Not all physically relevant wave function have
H|ni = En |ni,
parity. For example,
then |ni is also an eigenstate of the parity.
[p, ] 6= 0,
Proof: Using the property 2 = 1 one can easily see that
so a momentum eigenstate is not an eigenstate of the the state
1
parity. The wave function corresponding to an eigenstate (1 )|ni
2
of the momentum is the plane wave
is a parity eigenstate belonging to the eigenvalue 1. On
p0 (x0 ) = eip x /h ,
0 0
the other hand, this is also an eigenstate of the
Hamiltonia H with the energy En :
which is neither even nor odd.
Because 1 1
H( (1 )|ni) = En (1 )|ni.
[, L] = 0, 2 2

the eigenstate |, lmi of the orbital angular momentum Since we supposed the state |ni to be non degenerate the
(L2 , Lz ) is also an eigenstate of the parity. Now states |ni and 21 (1 )|ni must be the same excluding a
phase factor,
R (r)Ylm (, ) = hx0 |, lmi. 1
(1 )|ni = ei |ni,
2
In spherical coordinates the transformation x0 x0 so the state |ni is a parity eigen state belonging to the
maps to eigenvalue 1
Example The energy states of a one dimensional
r r
harmonic oscillator are non degenerate and the
(cos cos ) Hamiltonian even, so the wave functions are either even
+ (eim (1)m eim ). or odd.
Note The nondegeneracy condition is essential. For Let us suppose that at the moment t0 = 0 the state of the
p2 system is |Li. At a later moment, t, the system is
example, the Hamiltonian of a free particle, H = 2m , is
even but the energy states descibed by the state vector

2 |L, t0 = 0; ti
p0
H|p0 i = |p0 i 1
2m = (eiES t/h |Si + eiEA t/h |Ai)
2
are not eigenstates of the parity because 1 iES t/h
= e (|Si + ei(EA ES )t/h |Ai),
2
|p0 i = | p0 i.
because now the time evolution operator is simply
The condition of the theorem is not valid because the
states |p0 i and | p0 i are degenerate. We can form parity U(t, t0 = 0) = eiHt/h .
eigenstates
1/ 2(|p0 i | p0 i), At the moment t = T /2 = 2h/2(EA ES ) the system is
in the pure |Ri state and at the moment t = T again in
which are also degenerate energy (but not momentum) its pure initial state |Li. The system oscillates between
eigenstates. The corresponding wave functions the states |Li and |Ri at the angular velocity

p0 (x0 ) = hx0 | p0 i = eip x /h

0 0
are neither even nor odd, whereas
0 0 0
hx |(|p i + | p i) cos p x /h 0
hx0 |(|p0 i | p0 i) sin p0 x0 /h
are.
Example One dimensional symmetric double well
E A ES
= .
h
When V , then EA ES . Then the states |Li and
0 |Ri are degenerate energy eigenstates but not parity
eigenstates. A particle which is localized in one of the
wells will remain there forever. Its wave function does
not, however, obey the same symmetry as the
Hamiltonian: we are dealing with a broken symmetry.

Selection rules
Suppose that the states |i and |i are parity eigenstates:

|i =  |i
|i =  |i,

where  and  are the parities (1) of the states. Now

S y m m e tric (S ) A n tis y m m e tric (A ) h|x|i = h| x |i =   h|x|i,

The ground state is the symmetric state |Si and the first
so
excited state the antisymmetric state |Ai:
h|x|i = 0 unless  =  .
H|Si = ES |Si Example The intensity of the dipole transition is
|Si = |Si proportional to the matrix element of the operator x
H|Ai = EA |Si between the initial and final states. Dipole transitions are
thus possible between states which have opposite parity.
|Ai = |Ai, Example Dipole moment.
If
where ES < EA . When the potential barrier V between
[H, ] = 0,
the wells increases the energy difference between the
states decreases: then no non degenerate state has dipole moment:

lim (EA ES ) 0. hn|x|ni = 0.

V

The same holds for any quantity if the corresponding

We form the superpositions
operator o is odd:
1 o = o.
|Li = (|Si + |Ai)
2
1
|Ri = (|Si |Ai),
2
which are neither energy nor parity eigenstates.
Lattice translations Obviously we have
We consider a particle in the one dimensional periodic
potential H|i = E0 |i.
V (x a) = V (x).
Furthermore we get

X
X
a a a (a)|i = ein |n + 1i = ei(n1) |ni
n= n=
(a ) =e i
|i.

Thus every state corresponding to a value of the

continuous parameter has the same energy, i.e. the
a a a ground state of the system infinitely degenerate.
(b ) Let us suppose further that
The Hamiltonian of the system is not in general invariant |ni is a state localized at the point n so that
under translations
(a)|ni = |n + 1i,
(l)x (l) = x + l, (l)|x0 i = |x0 + li.
hx0 |ni =
6 0 (but small), when |x0 na| > a.
However, when l is exactly equal to the period of the
lattice a we have Due to the translation symmetry the diagonal elements of
the Hamiltonian H in the base {|ni} are all equal to
(a)V (x) (a) = V (x + a) = V (x). eachother:
hn|H|ni = E0 .
Because the operator corresponding to the kinetic energy Let us suppose now that
in the Hamiltonian is translationally invariant the whole
Hamiltonian H satisfies the condition hn0 |H|ni =
6 0 only if n0 = n or n0 = n 1.

(a)H (a) = H, We are dealing with the so called tight binding

approximation.
which, due to the unitarity of the translation operator When we define
can be written as
[H, (a)] = 0. = hn 1|H|ni,

The operators H and (a) have thus common eigenstates. we can write
Note The operator (a) is unitary and hence its
eigenvalues need not be real. H|ni = E0 |ni |n + 1i |n 1i,
Let us suppose that the potential barrier between the
where we have exploited the orthonormality of the basis
lattice points is infinitely high. Let |ni be the state
{|ni}. Thus the state |ni is not an energy eigen state.
localized in the lattice cell n, i.e.
Let us look again at the trial
hx0 |ni =
6 0 only if x0 na.
X
|i = ein |ni.
Obviously |ni is a stationary state. Because all lattice n=
cells are exactly alike we must have
Like before we have
H|ni = E0 |ni, n.
(a)|i = ei |i.
Thus the system has countably infinite number of ground Furthermore
states |ni, n = , . . . , . X
Now H ein |ni
(a)|ni = |n + 1i, X X
= E0 ein |ni ein |n + 1i
so the state |ni is not an eigenstate of the translation X
(a). ein |n 1i
Lets try X X
X = E0 ein |ni (eini + ein+i )|ni
|i ein |ni, X

= (E0 2 cos ) ein |ni.

where is a real parameter and The earlier degeneracy will be lifted if 6= 0 and

. E0 2 E E0 + 2.
Blochs theorem
Let us consider the wave function hx0 |i. In the
translated state (a)|i the wave function is

hx0 | (a)|i = hx0 a|i

when the operator (a) acts on left. When it acts on

right we get
hx0 | (a)|i = ei hx0 |i,
so we have
hx0 a|i = hx0 |iei .
This equation can be solved by substituting
0
hx0 |i = eikx uk (x0 ),

when = ka and uk (x0 ) is a periodic function with the

period a.
We have derived a theorem known as the Bloch theorem:
Theorem 1 The wave function of the eigenstate |i of
the translation operator (a) can be written as the procuct
0
of the plane wave eikx and a function with the period a.
Note When deriving the theorem we exploited only the
fact that |i an eigenstate of the operator (a) belonging
to the eigenvalue ei . Thus it is valid for all periodic
systems (whether the tight binding approximation holds
or not)
With the help of the Bloch theorem the dispersion
relation of the energy in the tight binding model can be
written as

E(k) = E0 2 cos ka, k .
a a
This continuum of the energies is known as the Brillouin
zone.
Time reversal (reversal of motion) The fact that the operator K does not change the base
The Newton equations of motion are invariant under the states can be justified like:
transformation t t: if x(t) is a solution of the The state |a0 i represented in the base {|a0 i} maps to the
equation column vector
mx = V (x) 0
0
then also x(t) is a solution. ..

At the moment t = 0 let there be a particle at the point .

x(t = 0) with the momentum p(t = 0). Then a particle at 0
0
|a i 7
1 ,

the same point but with the momentum p(t = 0)
follows the trajectory x(t). 0

In the quantum mechanical Schrodinger equation .
..


h2 2
 0
ih = + V , which is unaffected by the complex conjugation.
t 2m
Note The effect of the operator K depends thus on the
due to the first derivative with respect to the time, choice of the basis states.
(x, t) is not a solution eventhough (x, t) were, but If U is a unitary operator then the operator = U K is
(x, t) is. In quantum mechanics the time reversal has antiunitary.
obviously something to do with the complex conjugation. Proof: Firstly
Let us consider the symmetry operation
(c1 |i + c2 |i) = U K(c1 |i + c2 |i)
|i |i, |i |i. = (c1 U K|i + c2 U K|i)
We require that the absolute value of the scalar product = (c1 |i + c2 |i),
is invariant under that operation:
so is antiliniear. Secondly, expanding the states |i and
|h|i| = |h|i|. |i in a complete basis {|a0 i} we get

X
There are two possibilities to satisfy this condition: |i |i = ha0 |i U K|a0 i
a0
1. h|i = h|i, so the corresponding symmetry X
operator is unitary, that is = ha |i U |a0 i
0

a0

X
h|i h|U U |i = h|i. = h|a0 iU |a0 i
a0
The symmetries treated earlier have obeyed this
condition. and
X X
2. h|i = h|i = h|i, so the symmetry operator |i = ha0 |i U |a0 i h| = ha0 |iha0 |U .
cannot be unitary. a0 a0

We define the antiunitary operator so that Thus the scalar product is

XX
h|i = h|i h|i = ha00 |iha00 |U U |a0 ih|a0 i
a00 a0
(c1 |i + c2 |i) = c1 |i + c2 |i, X
= h|a0 iha0 |i = h|i
where a0
= h|i .
|i |i = |i, |i |i = |i
The operator is thus indeed antiunitary.
If the operator satisfies only the last condition it is called Let be the time reversal operator. We consider the
antilinear. transformation
We define the complex conjugation operator K so that |i |i,
Kc|i = c K|i. where |i is the time reversed (motion reversed) state.
If |i is the momentum eigenstate |p0 i, we should have
We present the state |i in the base {|a0 i}. The effect of
the operator K is then |p0 i = ei | p0 i.
X K
X Let the system be at the moment t = 0 in the state |i.
|i = |a0 iha0 |i |i = ha0 |i K|a0 i
At a slightly later moment t = t it is in the state
a0 a0
 
iH
X
0 0
= ha |i |a i. |, t0 = 0; t = ti = 1 t |i.
a0 h
We apply now, at the moment t = 0, the time reversal and
operator and let the system evolve under the
Hamiltonian H. Then at the moment t the system is in h| |i = h|i = h|i
the state   = h| |i = h| 1 |i
iH
1 t |i. = h| 1 |i.
h
If the motion of the system is invariant under time In partcular, for a Hermitean observable A we have
reversal this state should be the same as
h|A|i = h|A1 |i.
|, t0 = 0; ti,
We say that the observable A is even or odd under time
i.e. we first look at the state at the earlier moment t reversal depending on wheter in the equation
and then reverse the direction of the momentum p.
A1 = A
Mathematically this condition can be expressed as
    the upper or the lower sign holds. This together with the
iH iH
1 t |i = 1 (t) |i. equation
h h h|A|i = h|A1 |i
Thus we must have imposes certain conditions on the phases of the matrix
elements of the operator A between the time reversed
iH|i = iH|i, states. Namely, they has to satisfy
where |i stands for an arbitrary state vector. h|A|i = h|A|i .
If were linear we would obtain the anticommutator
relation In particular, the expectation value satisfies the condition
H = H.
h|A|i = h|A|i.
If now |ni is an energy eigenstate corresponding to the
eigenvalue En then, according to the anticommutation Example The expectation value of the momentum
rule operator p.
H|ni = H|ni = (En )|ni, We require that
and the state |ni is an energy eigenstate corresponding h|p|i = h|p|i,
to the eigenvalue En . Thus most systems (those, whose
energy spectrum is not bounded) would not have any so p is odd, or
ground state. p1 = p.
Thus the operator must be antilinear, and, in order to
The momentum eigenstates satisfy
be a symmetry operator, it must be antiunitary. Using
the antilinearity for the right hand side of the condition p|p0 i = p1 |p0 i
iH|i = iH|i = (p0 )|p0 i,

i.e. |p0 i is the momentum eigenstates correponding to

we can write it as
the eigenvalue p0 :
iH|i = iH|i.
|p0 i = ei | p0 i.
So, we see that the operators commute:
Similarly we can derive for the position operator x the
H = H. expressions

x1 = x
Note We have not defined the Hermitean conjugate of
the antiunitary operator nor have we defined the |x0 i = |x0 i
meaning of the expression h|. That being, we let the
when we impose the physically sensible condition
time reversal operator to operate always on the right
and with the matrix element h||i we mean the h|x|i = h|x|i.
expression (h|) (|i).
Let be an arbitrary linear operator. We define We consider the basic commutation relations

|i |i, [xi , pj ]|i = ihij |i.

so that Now
h| = h| [xi , pj ]1 |i = ihij |i,
from which, using the antilinearity and the time reversal
properties of the operators x and p we get
[xi , (pj )]|i = ihij |i. the state |ni was supposed to be nondegenerate they
We see thus that the commutation rule
[xi , pj ]|i = ihij |i
remains invariant under the time reversal.
Correspondingly, the requirement of the invariance of the
commutation rule
H|ni = H|ni = En |ni,
so the states |ni and |ni have the same energy. Because
must represent the same state. The wave function of the
state |ni is hx0 |ni and the one of the state |ni
correspondingly hx0 |ni . These must be same (or more
accurately, they can differ only by a phase factor which
does not depend on the coordinate x0 ), i.e.
hx0 |ni = hx0 |ni

[Ji , Jj ] = ihijk Jk For example the wave function of a nondegenerate

groundstate is always real.
leads to the condition For a spinles particle in the state |i we get

J 1 = J .
Z
|i = dx0 hx0 |i|x0 i
This agrees with transformation properties of the orbital Z
angular momentum x p. = dx0 hx0 |i |x0 i = K|i,

Wave functions i.e. the time reversal is equivalent to the complex

We expand the state |i with the help of position conjugation.
eigenstates: Z On the other hand, in the momentum space we have
|i = d3 x0 |x0 ihx0 |i. Z
|i = d3 p0 | p0 ihp0 |i
Now Z
Z
|i = d3 x0 |x0 ihx0 |i = d3 p0 |p0 ihp0 |i ,
Z
= d3 x0 |x0 ihx0 |i , because
|p0 i = | p0 i.
so under the time reversal the wave function The momentum space wave function transform thus
under time reversal like
(x0 ) = hx0 |i
(p0 ) (p0 ).
transforms like
(x0 ) (x0 ). We consider a spin 21 particle the spin of which is oriented
If in particular we have along n. The corresponding state is obtained by rotating
the state |Sz ; i:
(x0 ) = R(r)Ylm (, ),
|n; i = eiSz /h eiSy /h |Sz ; i,
we see that
where and are the direction angles of the vector n.
Ylm (, ) Ylm (, ) = (1)m Ylm (, ). Because
Because Ylm is the wave function belonging to the state J 1 = J .
|lmi we must have we see that
m
|lmi = (1) |l, mi. |n; i = eiSz /h eiSy /h |Sz ; i.
The probability current corresponding to the wave
Furthermore, due to the oddity of the angular
function R(r)Ylm seems to turn clockwise when looked at
momentum, it follows that
from the direction of the positive z-axis and m > 0. The
probability current of the corresponding time reversed h
state on the other hand turns counterclockwise because m Jz |Sz ; i = |Sz ; i,
2
changes its sign under the operation.
The spinles particles obey so we must have
Theorem 1 If the Hamiltonian H is invariant under the |Sz ; i = |Sz ; i,
time reversal and the energy eigenstate |ni nondegenerate
then the corresponding energy eigenfunction is real (or where is an arbitrary phase factor. So we get
more generally a real function times a phase factor
independent on the coordinate x0 ). |n; i = |n; i.
On the other hand we have Thus we must have

|n; i = eiSz /h ei(+)Sy /h |Sz ; i, 2 = (1)2j .

so Often one chooses

iSz /h iSy /h
|n; i = |n; i = e e |Sz ; i |jmi = i2m |j, mi.
iSz /h i(+)Sy /h
= e e |Sz ; i.

Writing Spherical tensors

= U K, U unitary Let us suppose that the operator A is either even or odd,
and recalling that the complex conjugation K has no i.e.
effect on the base states we see that A(1) = A.
 
2Sy We saw that then we have
= eiSy /h K = i K.
h
h|A|i = h|A|i.
Now
In an eigenstate of the angular momentum we have thus
eiSy /h |Sz ; i = +|Sz ; i
eiSy /h |Sz ; i = |Sz ; i, h, jm|A|, jmi = h, j, m|A|, j, mi.
1 Let now A be a component of a Hermitian spherical
so the effect of the time reversal on a general spin 2 state
is tensor:
A = Tq(k) .
(c |Sz ; i + c |Sz ; i) = +c |Sz ; i c |Sz ; i.
According to the Wigner-Eckart theorem it is sufficient to
Applying the operator once again we get consider only the component q = 0.
We define T (k) to be even/odd under the time reversal if
2 (c |Sz ; i + c |Sz ; i)
(k) (k)
= ||2 c |Sz ; i ||2 c |Sz ; i Tq=0 1 = Tq=0 .
= (c |Sz ; i + c |Sz ; i),
Then we have
i.e. for an arbitrary spin orientation we have (k) (k)
h, jm|T0 |, jmi = h, j, m|T0 |, j, mi.
2
= 1.
The state |, j, mi is obtained by rotating the state
From the relation |, jmi:
D(0, , 0)|, jmi = ei |, j, mi.
|lmi = (1)m |l, mi
On the other hand, due to the definition of the spherical
we see that for spinles particles we have tensor
2 = 1. k
(k) (k)
X

D (R)Tq(k) D(R) = Dqq0 (R)Tq0 ,
In general, one can show that q 0 =k

2 |j half integeri = |j half integeri we get

2 |j integeri = +|j integeri. (k)
X (k)
D (0, , 0)T0 D(0, , 0) = D0q (0, , 0)Tq(k) .
Generally we can write q

= eiJy /h K. Now
(k)
D00 (0, , 0) = Pk (cos ) = (1)k ,
Now
e2iJy /h |jmi = (1)2j |jmi, so we have
so (k)
D (0, , 0)T0 D(0, , 0)
  (k)
2 |jmi = eiJy /h |jmi = (1)k T0 + (q 6= 0 components).

= ||2 e2iJy /h |jmi Furthermore

(k)
= (1)2j |jmi. h, jm|Tq6=0 |, jmi = 0,
since the m selection rule would require m = m + q. So the Hamiltonian of an electron contains such terms as
we get
S B, p A + A p.
(k)
h, jm|T0 |, jmi
(k)
The magnetic field B is external, independent on the
= h, jm|D (0, , 0)T0 D(0, , 0)|, jmi system, so
(k)
= (1)k h, jm|T0 |, jmi. [, B] = 0 ja [, A] = 0.
On the other hand, S and p are odd, or
Note Unlike under other symmetries the invariance of
the Hamiltonian under the time reversal S1 = S ja p1 = p,
[, H] = 0, so
[, H] 6= 0.
does not lead to any conservation laws. This is due to the
fact that the time evolution operator is not invariant: We say that magnetic field breaks the time reversal
symmetry and lifts the Kramers degeneracy.
U (t, t0 ) 6= U (t, t0 ).

Time reversal and degeneracy

Let us suppose that

[, H] = 0.

Then the energy eigenstates obey

H|ni = En |ni
H|ni = En |ni.

If we now had
|ni = ei |ni,
then, reapplying the time reversal we would obtain

2 |ni = ei |ni = |ni,

or
2 = 1.
This is, however, impossible if the system j is half integer,
because then 2 = 1. In systems of this kind |ni and
|ni are degenerate.
Example Electon in electromagnetic field
If a particle is influenced by an external static electric
field
V (x) = e(x),
then clearly the Hamiltonian

p2
H= + V (x)
2m
is invariant under the time reversal:

[, H] = 0.

If now there are odd number of electrons in the system

the total j is half integer. Thus, in a system of this kind
there is at least twofold degeneracy, so called Kramers
degeneracy.
In the magnetic field

B =A
Perturbation theory We write

Stationary perturbation methods |ni = |n(0) i + |n(1) i + 2 |n(2) i +

Let us suppose that n = (1) 2 (2)
n + n + .

we have solved completely the problem Because

H0 |n (0)
i= En(0) |n(0) i. hn(0) |V n |ni = 0,

The basis {|n(0) i} is now complete. we have, on the other hand

the states |n(0) i are non degenerate. n = hn(0) |V |ni.

we want to solve the problem Thus we get

(H0 + V )|ni = En() |ni . (1) 2 (2)
n + n +
Usualy the index is dropped off. = hn(0) |V |n(0) i + 2 hn(0) |V |n(1) i + .

When we denote Equalizing the coefficients of the powers of the parameter

we get
n En En(0) ,
(1)
the eigenvalue equation to be solved takes the form O(1 ) : n = hn(0) |V |n(0) i
(2)
O(2 ) : n = hn(0) |V |n(1) i
(En(0) H0 )|ni = (V n )|ni. .. ..
. . .
(0) 1 (0) N (N ) (0) (N 1)
Note Because the expression (En H0 ) |n i is O( ) : n = hn |V |n i
(0) .. ..
undefined the operator (En H0 )1 is not well defined. . .
So, in the equation above we cannot invert the operator
(0)
(En H0 ). We substitute into the expression
Now
hn(0) |V n |ni = hn(0) |En(0) H0 |ni, n
|ni = |n(0) i + (V n )|ni
so in the state (V n )|ni there is no component along En(0) H0
the state |n(0) i.
for the state vector the power series of the state vector
We define a projection operator as
and the energy correction and we get
X
n = 1 |n(0) ihn(0) | = |k (0) ihk (0) |.
|n(0) i + |n(1) i + 2 |n(2) i +
k6=n
n
Now = |n(0) i + (0) (V (1) 2 (2)
n n )
En H0
1 X 1 (|n(0) i + |n(1) i + ).
n = |k (0) ihk (0) |
En(0) H0 (0)
k6=n En
(0)
Ek
Equalizing the coefficients of the linear -terms we get in
and the first order
(V n )|ni = n (V n )|ni.
O() : |n(1) i
Since in the limit 0 we must have
n (1)
n
|ni |n(0) i, = V |n(0) i n |n(0) i
En(0) H0 En(0) H0
the formal solution is of the form n
= V |n(0) i,
1 En(0) H0
|ni = cn ()|n(0) i + n (V n )|ni,
En(0) H0 because
where n (1)
n |n
(0)
i = 0.
lim cn () = 1 We substitute |n(1) i into the expression
0

and
(2)
n = hn
(0)
|V |n(1) i,
cn () = hn(0) |ni.
Diverting from the normal procedure we normalize so
n
(2)
n = hn
(0)
|V V |n(0) i.
hn(0) |ni = cn () = 1. En(0) H0
 (0) (0)
We substitute this further into the power series of the Perturbation expansions converge if |Vij /(Ei Ej )| is
state vectors and we get for the coeffients of 2 the small. In general, no exact convergence criterion is
condition known.
n n The state |ni is not normalized. We define the normalized
O(2 ) : |n(2) i = V V |n(0) i state
En(0) H0 En(0) H0
|niN = Zn1/2 |ni,
n n
hn(0) |V |n(0) i V |n(0) i. so that
En(0) H0 En(0) H0
hn(0) |niN = Zn1/2 hn(0) |ni = Zn1/2 .
Likewise we could continue to higher powers of the
parameter . This method is known as the Thus the normalization factor Zn is the probablility for
Rayleigh-Schrodinger perturbation theory. the perturbed state to be in the unperturbed state.
The explicit expression for the second order energy The normalization condition
correction will be
N hn|niN = Zn hn|ni = 1
n
(2)
n = hn (0)
|V V |n (0)
i
En(0) H0 gives us
n
Zn1 = hn|ni = (hn(0) | + hn(1) | + 2 hn(2) | + )
X
= hn(0) |V |k (0) ihk (0) | |l(0) ihl(0) |V |n(0) i
En(0) H0
k,l (|n(0) i + |n(1) i + 2 |n(2) i + )
X hk (0) |l(0) i = 1 + 2 hn(1) |n(1) i + O(3 )
= Vnk (0)
Vln
k,l6=n En(0) El X |Vkn |2
= 1 + 2 (0) 2
+ O(3 ).
(0)
X |Vnk |2 k6=n (En Ek )
= (0)
.
k6=n En(0) Ek
Up to the order 2 the probability for the perturbed state
Thus, up to the second order we have to lie in the unperturbed state is thus

n En En(0) X |Vkn |2
Zn 1 2 .
X |Vnk |2 (0)
(En(0) Ek )2
= Vnn + 2 (0)
+ . k6=n

k6=n En(0) Ek
The latter term can be interpreted as the probability for
Correspondingly, up to the second order the state vector the leakage of the system from the state |n(0) i to other
is states.
X Vkn Example The quadratic Stark effect.
|ni = |n(0) i + |k (0) i (0) We consider hydrogen like atoms, i.e. atoms with one
(0)
k6=n En Ek
electron outside a closed shell, under external uniform
electric field parallel to the positive z-axis. Now
X X Vkl Vln
+2 |k (0) i (0) (0)
k6=n
(0) (0)
l6=n (En Ek )(En El ) p2
H0 = + V0 (r) and V = e|E|z.
! 2m
Vnn Vkn
(0) We suppose that the eigenstates of H0 are non degenerate
(En(0) Ek )2
(not valid for hydrogen). The energy shift due to the
+ . external field is
We see that the perturbation mixes in also other states X |zkj |2
(than |n(0) i). k = e|E|zkk + e2 |E|2 (0) (0)
+ ,
We see that j6=k Ek Ej

in the 1st order we need only the matrix element Vnn . where
in the 2nd order the energy levels i and j repel each zkj = hk (0) |z|j (0) i.
(0) (0)
other. Namely, if Ei < Ej , then the contributions Since we assumed the states |k (0) i to be non degenerate
of one of these states to the energy corrections of the they are eigenstates of the parity. So, according to the
other are parity selection rule the matrix element of zkk vanishes.
|Vij |2 Indeces k and j are collective indeces standing for the
(2)
i = (0) (0)
<0 quantum number triplet (n, l, m). According to the
Ei Ej Wigner-Eckart theorem we have
(2) |Vij |2
j = (0) (0)
>0 zkj = hn0 , l0 m0 |z|n, lmi
Ej Ei
hn0 l0 kT (1) knli
and the energy levels move apart from each other. = hl1; m0|l1; l0 m0 i ,
2l + 1
where we have written the operator z as the spherical Degeneracy
(0)
tensor Lets suppose that the energy state ED is g-foldly
(1)
T0 = z. degenerated:
In order to satisfy zkj 6= 0 we must have (0)
H0 |m(0) i = ED |m(0) i, |m(0) i D, dimD = g.
m0 = m and We want to solve the problem
l0 = l 1, l, l + 1. From these l0 = l is not suitable (H0 + V )|li = El |li
due to the parity selection rule.
with the boundary condition
So we get X
lim |li hm(0) |l(0) i|m(0) i,
0
l0 = l 1

0 0 0 mD
hn , l m |z|n, lmi = 0 unless .
m0 = m
i.e. we are looking for corrections to the degenerated
states. With the help of the energy correction we have to
We define the polarizability so that
solve
(0)
1 (ED H0 )|li = (V l )|li.
= |E|2 .
2 We write again
As a special case we consider the ground state |li = |l(0) i + |l(1) i + 2 |l(2) i +
|0(0) i = |1, 00i of hydrogen atom which is non degenerate (1) (2)
when we ignore the spin. The perturbation expansion l = l + 2 l + ,
gives so we get

X |hk (0) |z|1, 00i|2
= 2e2 , (0)
(0) (0) (ED H0 )(|l(0) i + |l(1) i + 2 |l(2) i + )
k6=0 E0 Ek
(1) (2)
= (V l 2 l )
where the summing must be extended also over the
(0) (1) 2 (2)
continuum states. (|l i + |l i + |l i + ).
Let us suppose that
Equalizing the coefficients of the powers of we get in
(0) (0) the first order
E0 Ek constant,
(0)
(ED H0 )|l(1) i
so that (1)
X X = (V l )|l(0) i
(0) 2 (0) 2
|hk |z|1, 00i| |hk |z|1, 00i|
" #
=
(1)
X
(0) (0) (0)
k6=0 all ks = (V l ) |m ihm |l i .
mD
= h1, 00|z 2 |1, 00i.
(0)
Taking the scalar product with the vector hm0 | and
In the spherically symmetric ground state we obviously
recalling that
have
1 (0) (0)
hz 2 i = hx2 i = hy 2 i = hr2 i. hm0 |(ED H0 ) = 0,
3
Using the explicit wave functions we get we end up with the simultaneous eigenvalue equations
(1) (0)
X
hz 2 i = a20 . Vm0 m hm(0) |l(0) i = l hm0 |l(0) i.
m
Now (1)
The energy corrections l are obtained as eigenvalues.
 2 
From the equation
(0) (0) (0) (0) e 1
E0 + Ek E0 + E1 = 1 ,
2a0 4 (0)
(ED H0 )|l(1) i = (V l )|l(0) i
so
8a0 16a30 we also see that
< 2e2 a20
2 = 5.3a30 .
3e 3 (1)
hm(0) |V l |l(0) i = 0 |m(0) i D.
The exact summation gives
(1)
Thus the vector (V l )|l(0) i has no components in the
9a30 subspace D. Defining a projection operator as
= = 4.5a30 .
2 g
X X
D = 1 |m(0) ihm(0) | = |k (0) ihk (0) |
mD k6D
we can write We put the atom in external electric field parallel to the
(1) (1) z-axis:
(V l )|l(0) i = D (V l )|l(0) i = D V |l(0) i. V = ez|E|.
We get the equation Now z is the q = 0 component of a spherical tensor:
(0) (1)
(ED H0 )|l(1) i = D (V l )|l(0) i, z = T0 .
(0) According to the parity selection rule the operator V now
where now the operator (ED H0 ) can be inverted:
has nonzero matrix elements only between states with
D l = 0 and l = 1, and due to the m-selection rule all states
|l(1) i = (0)
V |l(0) i
ED H0 must have the same m:
X |k (0) iVkl 2s 2p, 0 2p, 1 2p, 1
= (0) (0)
.
k6D ED Ek 2s 0 0 0
2p, 0 0 0 0
V = .
When we again normalize 2p, 1 0 0 0 0
2p, 1 0 0 0 0
hl(0) |li = 1,
The nonzero matrix elements are
we get from the equation
(0)
h2s|V |2p, m = 0i = h2p, m = 0|V |2si = 3ea0 |E|.
(ED H0 )|li = (V l )|li
The eigenvalues of the perturbation matrix are
for the energy shift
(1)
= 3ea0 |E|
l = hl(0) |V |li.
and the eigenvectors
We substitute the power series and get
1
hl(0) |V (|l(0) i + |l(1) i + 2 |l(2) i + ) |i = (|2s, m = 0i |2p, m = 0i).
2
(1) (2)
= l + 2 l + .
Note The energy shift is a linear function of the electric
The second order energy correction is now field. The states |i are not parity eigenstates so it is
perfectly possible that they have permanent electric
(2) D dipole moment hzi = 6 0.
l = hl(0) |V |l(1) i = hl(0) |V (0)
V |l(0) i
ED H0
Nearly degenerated states
X |Vkl |2
= (0) (0)
. Let the states m D to be almost degenerate. We write
k6D ED Ek
V = V1 + V 2 ,
Thus the perturbation calculation in a degenerate system
proceeds as follows: where
X X
1 Identify the degenerated eigenstates. We suppose V1 = |m(0) ihm(0) |V |m0(0) ihm0(0) |
that their count is g. Construct the mD m0 D
g g-perturbation matrix V . V2 = V V1 .

2 Diagonalize the perturbation matrix. We proceed so that

3 The resulting eigenvalues are first order corrections 1. we diagonalize the Hamiltonian H0 + V1 exactly in
for the energy shifts. The corresponding eigenvectors the basis {|m(0) i} and
are those zeroth order eigenvectors to which the
corrected eigenvectors approach when 0. 2. handle the term V2 like in an ordinary non
degenerate perturbation theory. This is possible since
4 Evaluate higher order corrections using non
degenerate perturbation methods but omit in the hm0(0) |V2 |m(0) i = 0 m, m0 D.
summations all contributions coming from the
degenerated state vectors of the space D. Example Weak periodic potential.
Now
Example The Stark efect in the hydrogen atom. p2
The hydrogen 2s (n = 2, l = 0, m = 0) and 2p H0 =
2m
(n = 2, l = 1, m = 1, 0, 1) states are degenerate. Their and for the perturbation
energy is
(0)
ED = e2 /8a0 . V (x) = V (x + a).
We denote the unperturbed eigenstates by the wave Correspondingly the energy up to the second order is
vector:
h2 k 2 (2) (0)
X |Vn |2
H0 |ki = |ki, Ek = Ek + V 0 + (0) (0)
.
2m n6=0 Ek Ek+nK
so that
(0) h2 k 2 Let us suppose now that
Ek = .
2m
nK
We impose the periodic boundary conditions k .
2
hx0 |ki = k (x0 ) = hx0 + L|ki = k (x0 + L),
We diagonalize the Hamiltonian in the basis
for the wave function and get
{|ki, |k + nKi},
0 1 0 2
k (x ) = eikx , k = n, n I. i.e. we diagonalize the matrix
L L
Because the potential V is periodic it can be represented |ki |k + nKi
!
as the Fourier series |ki
(0)
Ek + V0 Vn
(0) .
X
inKx
|k + nKi Vn Ek + V 0
V (x) = e Vn ,
n= Its eigenvalues are
where (0) (0)
K = 2/a E + Ek+nK
Ek = V0 + k
2
is the reciprocal lattice vector. The only nonzero matrix v
!2
u E (0) E (0)
u
elements are now k k+nK
t + |Vn |2 .
2
hk + nK|V |ki = Vn ,
(0) (0)
because When |Ek Ek+nK |  |Vn |, it reduces to two solutions
Z
1X 0 0 0 0
hk 0 |V |ki = Vn dx0 eik x einKx einkx Ek+ = Ek + V 0
(0)
L n
(0)
X Ek = Ek+nK + V0 ,
= Vn k+nK,k0 .
n
which are first order corrected energies. In the limiting
So the potential couples states case we get
(0)
|ki, |k + Ki, . . . , |k + nKi, . . . . lim Ek = EnK/2 + V0 |Vn |.
knK/2
The corresponding energy denominators are
(0) (0) Brillouin-Wigner perturbation theory
Ek Ek+nK .
We start with the Schrodinger equation
In general
(0)
Ek 6= Ek+nK
(0) (En H0 )|ni = V |ni,

except when and take on both sides the scalar product with the state
nK |m(0) i, and get
k
.
2
(0)
We suppose that the condition (En Em )hm(0) |ni = hm(0) |V |ni.
nK We correct the state |n(0) i. We write the corrected state
k 6=
2 |ni in the form
holds safely. The first order state vectors are then X
|ni = |m(0) ihm(0) |ni = |n(0) ihn(0) |ni + n |ni
X Vn m
|k (1) i = |ki + |k + nKi (0) (0)
, X
n6=0 Ek Ek+nK = |n(0) i + |m(0) ihm(0) |ni,
m6=n
and the wave functions
which has been normalized, like before,
1 0 X 1 0 V
(1)
k (x0 ) = eikx + ei(k+nK)x (0) n (0) .
L n6=0 L Ek Ek+nK hn(0) |ni = 1.
We substitute into this the scalar products

hm(0) |V |ni
hm(0) |ni = (0)
,
En E m

and end up with the fundamental equation of the

Brillouin-Wigner method
X
|ni = |n(0) i + |m(0) i (0)
hm(0) |V |ni.
m6=n
En E m

Iteration gives us the series

X 1
|ni = |n(0) i + |m(0) i (0)
hm(0) |V |n(0) i
m6=n
E n Em
XX 1
+ 2 |l(0) i (0)
hl(0) |V |m(0) i
m6=n l6=n En El
1
(0)
hm(0) |V |n(0) i
En Em
X XX 1
+ 3 |k (0) i (0)
hk (0) |V |l(0) i
m6=n l6=n k6=n E n E k
1 1
(0)
hl(0) |V |m(0) i (0)
hm(0) |V |n(0) i
En El E n Em
+ .

Note This is not a power series of the parameter

because the energy denominators
(0)
En Em = En(0) Em
(0)
+ n

depend also on the parameter according to the equation

n = (1) 2 (2)
n + n +
Time dependent potentials The interaction picture observables AI are defined so that
We have solved the problem
AI eiH0 t/h AS eiH0 t/h .
H0 |ni = En |ni
In particular we have
completely and want to solve the eigenstates of the
Hamiltonian VI = eiH0 t/h V eiH0 t/h .
H = H0 + V (t).
We see that the equation governing the time dpendence
Since the Hamiltonian depends on time we have of the interaction picture state vectors is

6 eiHt/h ,
U=
ih |, t0 ; tiI
t
so a system in a stationary state |ii can in the course of  iH0 t/h 
time get components also in other stationary states. = ih e |, t0 ; tiS
t
= H0 eiH0 t/h |, t0 ; tiS
Pictures of the time evolution
+eiH0 t/h (H0 + V )|, t0 ; tiS
At the moment t = 0 let the system be in the state
X = eiH0 t/h V eiH0 t/h eiH0 t/h |, t0 ; tiS ,
|i = cn (0)|ni
n so

|, t0 ; tiI = VI |, t0 ; tiI .
ih
and at the moment t in the state t
X If now AS does not depend on time we can derive
|, t0 = 0; ti = cn (t)eiEn t/h |ni.
n dAI 1
= [AI , H0 ],
dt ih
Note The time dependence of the coefficients cn (t) is due
only to the potential V (t). The effect of the Hamiltonian which in turn resembles the Heisenberg equation of
H0 is in the phase factors eiEn t/h . motion provided that in the latter we substitute for the
Schrodingers picture total Hamiltonian H the stationary operator H0 .
The evolution of the state vectors is governed by the time We expand state vectors in the base {|ni}:
evolution operator: X
|, t0 ; tiI = cn (t)|ni.
|, t0 = 0; tiS = U(t)|, t0 = 0i. n

Heisenbergs picture If now t0 = 0 so multiplying the previous expansion

The state vectors remain constant, i.e. X
|, t0 = 0; ti = cn (t)eiEn t/h |ni
|, t0 = 0, tiH = |, t0 = 0i. n

On the other hand, the operators depend on time. We on both sides by the operator eiH0 t/h we get
can go from the time independent operators of the X
Schrodinger picture to the operators of the Heisenberg |, t0 = 0; tiI = cn (t)|ni,
picture via the transformation n

i.e. the coefficients cn are equal. We just derived the

AH (t) = U (t)AS U(t).
equation

If the Hamiltonian does not depend on time then ih |, t0 ; tiI = VI |, t0 ; tiI .
t
HH (t) = U (t)HU(t) = H From this we get

and
dAH 1 1 ih hn|, t0 ; tiI = hn|VI |, t0 ; tiI
= [AH , HH ] = [AH , H]. t
dt ih ih
X
= hn|VI |mihm|, t0 ; tiI .
Interaction picture m
The state vectors depend on time as
The matrix elements of the operator VI are
iH0 t/h
|, t0 ; tiI e |, t0 ; tiS .
hn|VI |mi = hn|eiH0 t/h V eiH0 t/h |mi
At the moment t = 0 we have obviously = Vnm (t)ei(En Em )t/h

|iS = |iI = |iH . .

Because we furthermore have The system oscillates between the states |1i and |2i with
the angular velocity
cn (t) = hn|, t0 ; tiI , s 
2 ( 21 )2
we can write the equation governing the time dependence = + .
h2 4
of the superposition coefficients as

d X The amplitude of the oscillations is at its maximum, i.e.

ih cn (t) = Vnm einm t cm (t), we are in a resonance, when
dt m
E2 E 1
where 21 = .
h
En Em
nm = mn . Example Spin magnetic resonance.
h
1
This system of differential equations can be written We put a spin 2 particle into
explicitely in the matrix form time independent magnetic field parallel to the z
axis,
c1
c2 time dependent magnetic field rotating in the xy
ih c3

plane,
..

. i.e.
V12 ei12 t B = B0 z + B1 (x cos t + y sin t)

V11 c1
V21 ei21 t V c2
22 when the fields B0 and B1 are constant. Since the

= .

V33
c3

magnetic moment of the electron is
.. .. ..

..
. . . . e
= S,
me c
Example Two state systems.
Suppose that the Hamiltonian is the sum of the terms
 
H0 = E1 |1ih1| + E2 |2ih2| (E1 < E2 ) ehB0
H0 = (|Sz ; ihSz ; | |Sz ; ihSz ; |)
2me c
and that the time dependent potential is like
 
ehB1
V (t) =
2me c
V (t) = eit |1ih2| + eit |2ih1|. 
cos t(|Sz ; ihSz ; | + |Sz ; ihSz ; |)

The matrix elements Vnm are now + sin t(i|Sz ; ihSz ; | + i|Sz ; ihSz ; |) .

V12
= V21 = eit If e < 0, then
V11 = V22 = 0,
|e|hB0 |e|hB0
E2 = E = > E 1 = E = .
so transitions between the states |1i and |2i are possible. 2me c 2me c
The system of differential equations to be solved is We can thus identify in the above treated two stated
system the quantities:
ihc1 = eit ei12 t c2
ihc2 = eit ei21 t c1 , |Sz ; i 7 |2i (higher state)
|Sz ; i 7 |1i (lower state)
where
E2 E1 |e|B0
21 = 12 = . 7 21
h me c
We can see that the solution satisfying the initial ehB1
7 , 7 .
conditions 2me c
c1 (0) = 1, c2 (0) = 0 Comparing with our earlier discussion on the spin
is precession we see that

2 /h2 if B1 = 0 and B0 6= 0, the the expectation value

|c2 (t)|2 = 2 hSx,y i rotates in the course of time counterclockwise
2 /h + ( 21 )2 /4
but the probabilities |c1 |2 and |c2 |2 remain still
2 1/2
( )
2
( 21 ) constant.
sin2 + t
h2 4
if B1 6= 0, the the coefficients |c1 |2 and |c2 |2 are
2 2
|c1 (t)| = 1 |c2 (t)| . functions of time.
When the resonance condition

21

holds the probability for the spin flips

|Sz ; i |Sz ; i

is very high.
Because experimental production of rotating magnetic
fields is difficult it is common to use a field oscillating for
example along the x axis. This can be divided into
components rotating counterclockwise and clockwise:

2B1 x cos t
= B1 (x cos t + y sin t)
+B1 (x cos t y sin t).

In experiments one usually has

B1
 1,
B0
so
|e|B1 |e|B0 B1 B1
= = = 21  21 .
h 4me c me c 4B0 4B0
If now the component rotating counterclockwise triggers
the resonance condition

21 ,

the the transition probability due to this component is

2 /h2
|cR (t)|2 = 2
/h + ( 21 )2 /4
2

2 1/2
( )
2
( 21 )
sin2 + t
h2 4
 
sin2 t .
h
The clockwise rotating component,

= 21 ,

contributes
2 /h2
|cL (t)|2
/h2 + 21
2 2
( 1/2 )
2 2 2
sin + 21 t
h2
 |cR (t)|2 .
Time dependent perturbation theory Here
In the interaction picture the time evolution operator is hn|UI (t, 0)|ii = cn (t)
determined by the equation
is the same as the superposition coefficient we used before.
|, t0 ; tiI = UI (t, t0 )|, t0 ; t0 iI . From the relation binding the interaction and Schrodinger
picture state vectors we get
Since the time evolution of the state vectors is governed
by the equation |, t0 ; tiI = eiH0 t/h |, t0 ; tiS
= eiH0 t/h U(t, t0 )|, t0 ; t0 iS

ih |, t0 ; tiI = VI |, t0 ; tiI = eiH0 t/h U(t, t0 )eiH0 t0 /h |, t0 ; t0 iI ,
t
= VI UI (t, t0 )|, t0 ; t0 iI ,
so the time evolution operators of these pictures are
we see that obtained with the help of the formula

UI (t, t0 ) UI (t, t0 ) = eiH0 t/h U(t, t0 )eiH0 t0 /h .

ih |, t0 ; t0 iI = VI UI (t, t0 )|, t0 ; t0 iI .
t
The matrix elements of the operator UI (t, t0 ) can now be
The interaction picture time evolution operator satisfies calculated from the relation
thus the equation
hn|UI (t, t0 )|ii = ei(En tEi t0 )/h hn|U(t, t0 )|ii.
d
ih UI (t, t0 ) = VI (t)UI (t, t0 ).
dt We see that
As the initial condition we have obviously
the matrix element hn|UI (t, t0 )|ii is not quite the

UI (t, t0 )t=t0 = 1. transition amplitude hn|U(t, t0 )|ii,

Integration gives the transition probabilities satisfy

i
Z t |hn|UI (t, t0 )|ii|2 = |hn|U(t, t0 )|ii|2 .
UI (t, t0 ) = 1 VI (t0 )UI (t0 , t0 ) dt0 .
h t0
Note If the states |a0 i and |b0 i are not eigenstates of H0
By iteration we end up with Dysons series then
|hb0 |UI (t, t0 )|a0 i|2 6= |hb0 |U(t, t0 )|a0 i|2 .
UI (t, t0 )
Z t
" Z t0
# In this case the matrix elements are evaluated by
i 0 i 00 00 00 0 expanding the states |a0 i and |b0 i in the base {|ni}
=1 VI (t ) 1 VI (t )UI (t ) dt dt
h t0 h t0 formed by the eigenstates of H0 .
i
Z t Let us suppose now that at the moment t = t0 the system
=1 dt0 VI (t0 ) is in the eigenstate |ii of H0 . This state vector can always
h t0
be multiplied by an arbitrary phase factor, so the
 2 Z t Z t0
i 0 Schrodinger picture state vector |i, t0 ; t0 iS can be chosen
+ dt dt00 VI (t0 )VI (t00 )
h t0 t0 as
n Z t
 Z t0 |i, t0 ; t0 iS = eiEi t0 /h |ii.
i 0
+ + dt dt00 Then in the interaction picture we have
h t0 t0
Z t(n1)
|i, t0 ; t0 iI = |ii.
dt(n) VI (t0 )VI (t00 ) VI (t(n) )
t0
+. At the moment t this has evolved to the state
X
Let us suppose again that we have solved the problem |i, t0 ; tiI = UI (t, t0 )|ii = cn (t)|ni,
n
H0 |ni = En |ni so
cn (t) = hn|UI (t, t0 )|ii,
completely. Let the initial state of the system be |ii at
the moment t = t0 = 0, i.e. as we already noted.
Now
|, t0 = 0; t = 0iI = |ii.
1. substitute the Dyson series into this
At the moment t this has evolved to the state
2. expand the coefficient as a power series of the
|i, t0 = 0; tiI = UI (t, 0)|ii perturbation
X
= |nihn|UI (t, 0)|ii. cn (t) = c(0) (1) (2)
n (t) + cn (t) + cn (t) + ,
n
 (k) 1600
3. equalize the terms cn with the perturbation terms
of the order k, 1400

1200
4. denote
ei(En Ei )t/h = eini t . 1000

800 t=40
We get
600

c(0)
n (t) = ni 400
t=20

Z t
i
c(1)
n (t) = hn|VI (t0 )|ii dt0 200
t=10
h t0
Z t 0

i 0
-1 -0.5 0 0.5 1

= eini t Vni (t0 ) dt0 f

h t0
 2 X Z t Z t0
i 0 t2
c(2)
n (t) = dt0 dt00 einm t Vnm (t0 )
h m t0 t0
imi t00 00
e Vmi (t ).
1/t

The probability for the transition from the state |ii to the
state |ni can be written as

Pr(i n) = |cn (t)|2 = |c(1) (2) 2

n (t) + cn (t) + | .

When t is large then |cn (t)|2 6= 0 only if

Fermis golden rule
2 2h
Consider the constant perturbation t = .
|En Ei |

0, when t < 0
V (t) = If now t is the time the perturbation has been on then
V (time independent) when t 0.
transitions are possible only if
switched on at the moment t = 0. At the moment t = 0
let the system be in the pure state |ii. Now tE h.

c(0) = c(0) Note If the energy is conserved exactly, i.e.

n n (0) = in
Z t
i 0
E n = Ei ,
c(1)
n = Vni eini t dt0
h 0
Vni then
= (1 eini t ). 1
En Ei |c(1) 2
|Vni |2 t2 .
n (t)| =
h2
The transition probability to the state |ni is thus The transition probability is proportional to the square of
the on-time of the perturbation (and not linearly
|Vni |2
|c(1)
n |
2
= (2 2 cos ni t) proportional to the time).
|En Ei |2 In general we are interested in transitions in which the
4|Vni |2
 
2 (En Ei )t initial state |ii is fixed but the final state |ni can be any
= sin .
|En Ei |2 2h state satisfying the energy conservation rule

The quantity En Ei
En E i
The total probability for such a transition is now
h
is almost continuous because usually the En states form
Pr(i f )
almost a continuum. The transition probability is now X
= |c(1)
n (t)|
2

|Vni |2 n
|c(1) 2
n | = f (), En Ei

h2 Z
= dEn (En )|c(1)
n |
2
where
4 sin2 t/2  
|Vni |2
2 (En Ei )t
Z
f () = . = 4 sin (En ) dEn .
2 2h |En Ei |2
 0 (1)
Here (E) is the density of states, i.e. the term eini t is same as in the coefficient cn , so it
contributes only if En Ei when t .
(E)dE = the number of states between(E, E + dE). 0
if Em in the term einm t differs from En and at the
Because same time from Ei it oscillates rapidly and
1 sin2 xt contributes nothing.
lim = (x),
t tx2
So we can write
we get
 2 

2  Vnm Vmi 
 
2 En Ei En Ei
 
1 t wif = V +
X
(E )

.
lim 2 sin t = 2
 ni
h  Ei Em 
 n 

t (En Ei ) 2h 4h 2h m 
En Ei
t
= (En Ei ).
2h In the second order the term Vnm Vmi can be thougth to
describe virtual transitions
The transition probability is thus
  |ii |mi |ni.
2 
lim Pr(i f ) = |Vni |2 (En )t ,

t h En Ei
Harmonic perturbations
where |Vni |2 is the average of the term |Vni |2 . Consider the potential
Note The total transition probability depends linearly
on time t. V (t) = Veit + V eit ,
The transition rate w is defined to be the transition
probability per unit time. We end up with the Fermi which is again assumed to be switched on at the moment
golden rule t = 0. When t < 0, the system is supposed to be in the
! state |ii. The first order term is now
d X (1) 2
wif = |cn (t)| i
Z t
0 it 0
 0
dt n c(1)
n = Vni eit + Vni e eini t dt0
  h 0
2  " #
= |Vni |2 (En ) . 1 1 ei(+ni )t 1 ei(ni )t

h En Ei = Vni + Vni .
h + ni + ni
Quite often this is also written as
  This is of the same form as in the case of our earlier step
2 potential, provided that we substitute
win = |Vni |2 (En Ei ),
h
En Ei
ni = ni .
but then one implicitely
R assumes that it will be integrated h
in the expression dEn (En )win . (1)
When t , |cn |2 is thus non zero only if
Second order corrections
In the second order we got ni + 0 or En Ei h
ni 0 or En Ei + h.
 2 X Z t Z t0
i 0
c(2)
n (t) = dt0 dt00 einm t Vnm (t0 ) Obviously, if the first term is important the second one
h m t0 t0
does not contribute and vice versa. The energy of a
00
eimi t Vmi (t00 ), quantum mechanical system is not conserved in these
transitions but the external potential either gives
so in the case of the potential (absorption) or takes (stimulated emission) energy
 to/from the system. Analogically to the constant
0, when t < 0 potential the transition rate will be
V (t) =
V (time independent) when t 0.
2
we have win = |Vni |2 (En Ei h).
h
 2 X Z t Z t0
i 0 00
c(2)
n = Vnm Vmi dt0 einm t dt00 eimi t
h m 0 0
Z t
i X Vnm Vmi 0 0
= (eini t einm t )dt0 .
h m Em Ei 0

Above
Energgy shifts and line widths Thus, up to the second order the coefficient ci is
2
Evolution of the initial state e2t

i t i
We consider the case where the initial and final states are ci (t) 1 Vii e + |Vii |2 2
h h 2
the same. We switch the interaction on slowly:
|Vmi |2 e2t
 X
i
V (t) = et V. + .
h 2(Ei Em + ih)
m6=i
Here 0 at the end.
We suppose that in the far past, t = , the system has For the logarithmic time derivative of the coefficient ci up
been in the state |ii. to the second order in the perturbation V we get
Check
If n 6= i, then the perturbation theory gives i |Vii |2 X |Vmi |2
Vii +
ci i h (Ei Em + ih)
m6=i
c(0)
n (t) = 0
t ci h i Vii
1
Z
i t0 ini t0 0
c(1)
n (t) = Vni lim e e dt h
h t0 t0
|Vmi |2
 X
i i
i et+ini t Vii + ,
= Vni . h h Ei Em + ih
m6=i
h + ini
Up to the lowest non vanishing order the transition where we have already set et 1.
probability is Note We cannot set in the denominator = 0, because
|Vni |2

e2t
 the states Em can form nearly a continuum in the vicinity
2
|cn (t)| of Ei .
h2 2 + ni
2
The logarithmic derivative is thus time independent, i.e.
and the transition rate correspondingly of form
ci (t) i
2|Vni |2 e2t = i .
 
d
|cn (t)|2 . ci (t) h
dt h2 2 + ni
2
The solution satisfying the initial condition ci (0) = 1 is
Now
ci (t) = eii t/h .
lim 2 2 = (ni ) = h(En Ei ),
0 + ni
Note i is not necessarily real.
so in the limit 0 we get the Fermi golden rule
  We interprete this so that the state |ii evolves gradually
2 like
win |Vni |2 (En Ei ).
h |ii ci (t)|ii = eii t/h |ii.
This is equivalent with our previous result. In the Schrodinger picture the latter contains also a phase
Let now n = i. We get factor, or
(0)
ci = 1 eii t/h |ii 7 eii t/hiEi t/h |ii.
Z t
(1) i 0 i
ci = Vii lim et dt0 = Vii et Due to the perturbation the energy levels shift like
h t0 t0 h
Ei Ei + i .
 2 X We expand now i as the power series in the
(2) i
ci = |Vmi |2 perturbation:
h m
t t0 (1) (2)
i = i + i + .
Z Z
0 iim t0 +t0 00
+t00
lim dt e dt00 eimi t
t0 t0 t0
 2 X Comparing with our previous expression
i 2
= |Vmi |
h X |Vmi |2
m i = Vii + ,
t 0 0 Ei Em + ih
eimi t +t
Z
m6=i
0 iim t0 +t0
lim dt e
t0 t0 i(mi i)
2 we see that in the first order we have
e2t

i
= |Vii |2 2 (1)
i = Vii .
h 2
|Vmi |2 e2t
 X
i This is equivalent with the time independent perturbation
+ .
h 2(Ei Em + ih) theory.
m6=i
Because the energies Em for almost a continuum we can so i or excluding the factor -2, the imaginary part of
in the second order term the energy shift is the width of the decay line and the
real part of the energy shift what is usually called the
X |Vmi |2 energy shift.
Ei Em + ih In the case of harmonic perturbations we can repeat the
m6=i
same derivation provided that we substitute
replace the summation with the integration. To handle
the limit 0 we recall from the function theory that Em Ei 7 Em Ei h.
Z Z
f (z) f (z)
lim dz = dz if (0),
0+ z + i z

where stands for the principal value integral. A

common shorthand notation for this is
1 1
lim = i(x).
0 x + i x
Thus we get

(2)
X |Vmi |2
Re(i ) =
E i Em
m6=i
(2)
X
Im(i ) = |Vmi |2 (Ei Em ).
m6=i

The right hand side of the latter equation is familiar from

the Fermi golden rule, so we can write
X 2 X 2 h
(2)
i
wim = |Vmi |2 (Ei Em ) = Im i .
h h
m6=i m6=i

The coefficient ci (t) can be written with the help of the

energy shift as

ci (t) = e(i/h)[Re(i )t]+(1/h)[Im(i )t] .

We define
i 2
Im(i ).
h h
Then
|ci (t)|2 = e2Im(i )t/h = ei t/h .
Thus the quantity i tells us at which rate the state |ii
disappears. The quantity
h
i =
i
is thus the average life time of the state |ii.
In the Schrodinger picture the time evolution is
Z
iEi t/h h
ci (t)e = f (E)eiEt/h dE,
2
where the energy spectrum
Z
f (E) = ei[Ei +Re(i )]t/hi t/2h eiEt/h dt

is the Fourier transform of the coefficient ci (t)eiEi t/h .

Now
1
|f (E)|2 ,
{E [Ei + Re(i )]}2 + 2i /4
Radiation and matter Earlier we saw that in the case of the harmonic potential
We handle the interaction of radiation and matter
semiclassically: V (t) = Veit + V eit

the radiation field classically, transitions are possible if

the matter quantum mechanically, ni + 0 or En Ei h

OK, if there is large number of photons in the ni 0 or En Ei + h,
volume 3 , or
in the case of the spontaneous emission we impose a eit stimulated emission
fictive field equivalent with the quantum theory. it
e absorption.
The vector potential A of the classical radiation field can
always be chosen to satisfy the transverse condition:
A = 0. The electric and magnetic field are obtained Absorption
from the vector potential as In the case of the radiation field,
1 eA0  i(/c)nx 
E = A Vni = e  p
c t me c ni
B = A. is the matrix element corresponding to the absorption.
The energy flux energy/unit area/unit time is The transition rate is then
 2
B2 2 e2
|A0 |2 |hn|ei(/c)nx  p|ii|2

c Emax win =
cU = + max . h m2e c2
2 8 8
(En Ei h).
For a monochromatic plane wave we have
h i We should note that
A = A0  ei(/c)nxit + ei(/c)nx+it ,
if the final states |ni form a continuum we integrate
where n and  are the directions of the propagation and weighing with the state density (En ).
polarization of the plane wave. Due to the transverse
condition if the final states |ni are discrete they, nevertheless,
A=0 are not ground states so that their energy cannot be
extremely accurate.
we have n. The energy flux is then
collisions can broaden the energy levels.
1 2
cU = |A0 |2 .
2 c the incoming radiation is not usually completely
A particle in the radiation field has the mechanical monochromatic.
momentum So we write the -function as
 e 2 e e e2   1
p A = p2 p A A p + 2 A2 ( ni ) = lim .
( ni )2 + 2 /4

c c c c 0 2
2
e e
= p2 2 A p + 2 A2 , We define the absorption cross section:
c c
(energy/unit time) absorbed by the atom (i n)
since due to the transvers condition abs = .
energy flux of the radiation field
p A = A p.
Since in every absorption process the atom absorbs the
The Hamiltonian of an electron in the field is now energy h, we have
1  e 2 hwin
H = p A + e(x) abs =
2me c 1 2 2
p2 e A
+ e(x) A p, 2 c 0
2me me c 2 e2
h |A0 |2 |hn|ei(/c)nx  p|ii|2
when we drop off the term |A|2 . Now h m2e c2
=
1 2
|A0 |2
 
e
Ap 2 c
me c (En Ei h)
 
e 4 2 h e2
 
= A0  p = |hn|ei(/c)nx  p|ii|2
me c
h i m2e hc
ei(/c)nxit + ei(/c)nx+it . (En Ei h).
Here e2 /hc is the fine structure constant 1/137. Integration gives
In order the absorption to be possible the energy Z X
quantum h of the radiation must be of the order of the abs () d = 4 2 ni |hn|x|ii|2 .
energy level spacing: n

Ze2 Ze2 The oscillator strength is defined as follows:

h ,
(a0 /Z) Ratom 2me ni
fni |hn|x|ii|2 .
h
when Z is the atomic number. Now
One can show that it satisfies so called
c chRatom 137Ratom Thomas-Reiche-Kuhns sum rule:
=
2 Ze2 Z X
fni = 1.
or
Ratom Z n
 1.
2137 We see that
We expand the exponential function in the expression for Z
4 2 h
 2 
e
the cross section as the power series abs () d = = 2 2 c .
2me me c2

ei(/c)nx = 1 + i n x + . This is known as the oscillator sum rule of classical
c
electrodynamics.
Now
Photoelectric effect
Z Z The initial state |ii is atomic but the final state |ni is in
hn xi Ratom Ratom = ,
c c 137Ratom 137 the continuum formed by the plane waves |kf i. In the
absorption cross section we have now to weigh the
so it is usually enough if we keep only the term 1. We
function (ni ) with the final state density (En ):
have then the so called electric dipole approximation.
Thus in the electric dipole approximation 4 2 h
abs = |hn|ei(/c)nx  p|ii|2
m2e
hn|ei(/c)nx  p|ii  hn|p|ii.
(En )(En Ei h).
We choose
Under the periodic boundary conditions in the L-sided
 k x and n k z.
cube we have
eikf x
0
Let the states |ni be the solutions of the problem
hx0 |kf i = ,
L3/2
p2 where
H0 |ni = En |ni, H0 = + e(x)
2me 2ni
ki = , ni = 0, 1, 2, . . . .
L
Because
ihpx When L , the variable n, defined via the relation
[x, H0 ] = ,
me n2 = n2x + n2y + n2z ,
we have
can be considered continuous. Then the volume in the
me
hn|px |ii = hn|[x, H0 ]|ii solid angle d bounded by the surfaces n0 = n and
ih n0 = n + dn is n2 dn d.
= ime ni hn|x|ii. The final state energy is
(1)
Since x is a superposition of the spherical tensors T1 we h2 kf2 h2 n2 (2)2
get the selection rules E= = .
2me 2me L2
m0 m = 1 The number of states with the wave vector kf in the
interval (E, E + dE) and in the solid angle is
|j 0 j| = 0, 1.
 3
dn L
dkf
If we had 2
n d dE = k2f d dE
dE 2 dE
 k y, the same selection rules were valid.  3
L me
= kf dE d.
 k z, we should have m0 = m, because z = T0 .
(1) 2 h2
The differential cross section is now
In the dipole approximation the absorption cross section
is d 4 2 h me kf L3
= |hkf |ei(/c)nx  p|ii|2 2 .
abs = 4 2 ni |hn|x|ii|2 ( ni ). d 2
me h (2)3
Example Emission of an electron from the innermost
shell.
The wave function of the initial state is approximately
like the one of the hydrogen ground state provided we
substitute a0 a0 /Z:
 3/2
0 Z
hx |ii eiZr/a0 .
a0

The matrix element is now

hkf |ei(/c)nx  p|ii

eikf x i(/c)nx0
Z 0

=  d3 x0 3/2
e
" L
 3/2 #
Zr/a0 Z
(ih) e .
a0

Integrating by parts and noting that due to the

transversal condition  n = 0 we have
h i
 ei(/c)nx = 0.
0

We get

hkf |ei(/c)nx  p|ii

h kf
Z
d3 x0 ei(kf c n)x atom (x0 ).
0
= 3/2
L
Thus the matrix element is proportional to the Fourier
transform of the atomic wave function with the respect of
the variable  
q = kf n.
c
As the final result we can write the differential cross
section as
d ( kf )2 Z 5 1
= 32e2 kf  .
d me c a50 (Z/a0 )2 + q 2 4


If now  k x and n k z, the differential cross section can

be written using the polar angle , the azimuthal angle
and the relations

kf = kf (sin cos , sin sin , cos )

2
( kf ) = kf2 sin2 cos2
 2
q2 = kf2 2kf cos + .
c c
Relativistic quantum mechanics Here Li is the linear Lagrangian density.
We go to continuum by substituting the limits
Classical fields
We suppose that the Lagrange function a dx
m
= linear mass density
L = L(qi , qi ) = T V a
i+1 i
of classical mechanics does not depend explicitely on
a x
time. From the Hamilton variation principle ka Y = Youngs modulus.
Z t2

L(qi , qi ) dt = 0, qi (t)t=t = 0 Now Z
1,2
t1
L= L dx,
one can derive the equations of motion
  where "  2 #
d L L 1
= 0. L= 2 Y .
dt qi qi 2 x
The Hamiltonian function of the Hamiltonian mechanics In the continuous case the Hamiltonian variation
is X principle takes the form
H= pi qi L, Z t2
i
L dt
where the canonically conjugated momentum pi of qi is t1
Z t2 Z  
L
pi = . = dt dx L , ,
qi t x
Z 1Z   
The equations of motion are now L L
= dt dx +
(/x) x
H  
qi = L
pi +
(/t) t
H
pi = .
Z Z 
L L
qi = dt dx + ()
(/x) x

Many body system
L
+ ()
- k (h - h ) - k (h - h )
(/t) t
i i-1 i i+ 1 Z Z   
L L
(i-1 )a ia (i+ 1 )a = dt dx


x (/x)
 
L
.
t (/t)
To get the variation to vanish for all one must satisfy
h i-1
h i
h i+ 1
the Euler-Lagrange equation

We consider N identical particles coupled to eachother by L L L

+ = 0.
identical parallel springs. The Lagrangian of the system is x (/x) t (/t)

L = T V When "  2 #
N 1 2
1 X 2 L= Y ,
mi k(i+1 i )2 ,

= 2 x
2 i
then
when i is the deviation of the particle i from its
equilibrium position ia. L
= 0
We write this as
N L 2
1 X 2 = Y = Y
mi k(i+1 i )2

L = x (/x) x x x2
2 i 2
L
N
"  2 # = 2.
X 1 m 2 i+1 i t (/t) t
= a ka
2 a i a
i Substituting into the Euler-Lagrange equation we get
N
2 2
X
= aLi .
Y 2 = 0,
i x t2
which describes
p a wave progating in one dimension with is invariant under Lorentz transformations.
the velocity Y /. Using the four-vector notation the Euler-Lagrange
We define the canonically conjugated momentum equation can be written into the compact form
L
 
= L L
= 0.
x (/x )
and the Hamiltonian density We see that the field equation derived from the
Lagrangian density L is covariant provided that the
H = L.
Lagrangian density L itself is relativistically scalar
Now (invariant under Lorentz transformations).
= ,
so
"  2 #
1
H = 2 2 Y
2 x
 2
1 2 1
= + Y .
2 2 x
The Lagrangian formalism generalizes easily to three
dimensions. The Euler-Lagrange equation takes then the
form
3
X L L L
+ = 0.
xk (/xk ) t (/t)
k=1

Covariant formulation
We employ the metrics where the components of a
four-vector b are

b = (b1 , b2 , b3 , b4 ) = (b, ib0 ).

In particular the coordinate four-vector is

x = (x1 , x2 , x3 , x4 ) = (x, ict).

Under Lorentz transformations the coordinate vector

transforms according to the equation

x0 = a x .

The coefficients of the Lorentz transformation satisfy the

orthogonality condition

a a = , (a1 ) = a ,

so that
x = (a1 ) x0 = a x0 .
We define the four-vector so that under Lorentz
transformations it transforms like x .
Now
x
0 = 0 = a ,
x x x x
so the four-gradient /x is a four-vector.
The scalar product of the four-vectors b and c,
3
X
b c = b c = bj cj + b4 c4
j=1
= b c b0 c0 ,
Klein-Gordons equation so the solution describes a free field. We include the term
We consider the scalar field (x) which, according to its
definition, behaves under Lorentz transformation like Lint = ,

0 (x0 ) = (x). where is the (usually position dependent) density of the

source. The field equation is now
Now
L = L(, /x ). 2 = .

Since we want When we choose

= G(x)
the Lagrangian density to be invariant under Lorentz
transformations and seek for a stationary solution we end up with the
equation
a linear wave equation, (2 2 ) = G(x).
the Lagrangian density can contain only the terms We substitute using its Fourier transform
Z
1
2 and . (x) = d3 keikx (k),
x x (2)2/3
One possible form for the Lagrangian density is where Z
1
  (k) = d3 x eikx (x).
1 2 2 (2)3/2
L= + .
2 x x We end up with the algebraic equation
Substituting this into the Euler-Lagrange equation G
(k2 2 )(k) =
(2)3/2
 
L L
= 0,
x (/x ) of the Fourier components. Its solution is
we get   G 1
(k) = .
+ 2 = 0. (2)2/3 k 2 + 2
x x
If we employ the notation Taking the Fourier transform we get the solution

1 2 G er
2
= 2 2, (x) = ,
c t 4 r

we end up with the Klein-Gordon equation known as the Yukawa potential. Lets suppose that the
meson field of a nucleon at the point x1 satisfies the
2 = 0. equations
(22 2 ) = G(x1 x2 ).
Its solution is thus the Yukawa potential.
Heuristic derivation
We substitute into the relativistic energy-momentum G e|x2 x1 |
relation (x2 ) = .
4 |x2 x1 |
E 2 |p|2 c2 = m2 c4
the operators Because the Hamiltonian density was

H = L,
E 7 ih , pk 7 ih ,
t xk the Hamiltonian density of the interaction is
and get 
2 m2 c2
 Hint = Lint
2
2 2 + = 0.
c t h2 and the total interaction Hamiltonian
When we set Z Z
mc 1 Hint = Hint d x = d3 x.
3
= , [] = ,
h length
we end up with the Klein-Gordon equation. We see that the interaction energy of nucleons located at
There are no sources in the Lagrangian density the points x1 and x2 is

G e|x2 x1 |
 
1 2 2 (1,2)
L= + Hint = .
2 x x 4 |x2 x1 |
Note Unlike in the Coulomb case, this interaction is In a small neighborhood of the solutions and the
atractive and short ranged. Lagrangian density is invarinat so we must have
In the reality there are 3 mesons ( + , 0 , ), with
different charges but with (almost) equal masses, L = 0.
consistent with the thory. We expand our theory so that
we consider two real fields, 1 and 2 , for two particles Thus we get
s
with equal masses. From these we construct the complex = 0,
fields x
1 + i2 where

 
=
2 s = i .
x x
1 i2
= . We see that
2
The Lagrangian density for the free fields can be written a complex field is associated with a conserved
using either the complex or real fields: four-vector density s ,
if we exchange , then s s .
   
1 1 1 2 2 1 2 2 2 2
L = + 1 + 2
2 x x 2 x x
  We interpret this so that

= + 2 .
x x s is the charge current density,
Considering the fields and independent we get two carries the charge e,
Euler-Lagrange equations
carries the charge e,
L L
= 0 the previous real field corresponds to neutral mesons.
x (/x )
L L
= 0,
x ( /x )
which can be further written as two Klein-Gordon
equations
2 = 0
2 = 0.
We define the first order gauge transformation so that the
fields transform under it like
0 = ei
0 = ei ,
when is a real parameter. Let be now an arbitrary,
infinitesimally small, number. Then
= i
= i .
The Lagrangian density transforms then as
  
L L
L = +
(/x ) x
  
L L
+ +
( /x ) x
  
L L
=
x (/x )
  
L L
+
x ( /x )
 
L L
+ +
x (/x ) ( /x )
 

= i .
x x x
Photons where
We consider a radiation field whose vector potential A = |k|c.
satisfies the transversality condition
Now
A = 0.
A(x, t)
1 XX
Because the electric and magnetic fields = (ck, (t)() eikx + ck, (t)() eikx )
V
1 k
E = A 1 XX
c t = (ck, (0)() eikx + ck, (0)() eikx ),
B = A V
k
satisfy the free space Maxwell equations where we have employed the four-vector notation
E = 0 k x = k x t = k x |k|ct.
B
E =
t The Hamiltonian function of the classical radiation field is
B = 0 Z
1 E 1
B = , H = (|B|2 + |E|2 ) d3 x
c t 2
Z
1 
| A|2 + |(1/c)(A/t)|2 d3 x.

the vector potential satisfies the wave equation =
2
1 2A A straightforward calculation shows that
2 A = 0.
c2 t2 XX
H= 2(/c)2 ck, ck, .
We write the vector potential at the moment t = 0 as a
superposition of the periodically normalized plane waves k
in an L-sided cube, Because the coefficients

uk, (x) = () eikx , ck, (t) = ck, (0)eit

like: satisfy the equation of motion

1 X X
A(x, t)|t=0 = (ck, (0)uk, (x) ck, = 2 ck, ,
V
k =1,2
it would look like the classical radiation field were
+ck, (0)uk, (x)).
composed of independent harmonic oscillators.
We define the variables
Here V = L3 and () , = 1, 2 are real polarization
vectors. 1
Qk, = (c + ck, )
Due to the transversality condition we have c k,
i
Pk, = (ck, ck, ).
() k = 0, c

so the polarization can chosen so that the vectors With the help of these the Hamiltonian function can be
((1) , (2) , k/|k|) form a righthanded rectangular written as
coordinate system. The Fourier components uk, satisfy XX 1
the orthogonality conditions H= (Pk2 , + 2 Q2k, ).
2
Z k
1
d3 x uk, uk0 ,0 = kk0 0 . Since
V
H
Due to the periodicity conditions the wave vectors can = Pk,
take the values Qk,
H
kx , ky , kz = 2n/L, n = 1, 2, . . . . = +Qk, ,
Pk,
At the moment t 6= 0 the vector potential is obtained the variables Pk, and Qk, are canonically conjugated
simply by setting and the Hamiltonian function the sum of the total
energies of the corresponding harmonic oscillators.
ck, (t) = ck, (0)eit
Thus the classical radiation field can be thought to be a
ck, (t) = ck, (0)eit , collection of independent harmonic oscillators. There
 every oscillator is characterized by the wave vector k Thus we can write
and the polarization () ,
a |nk , i = c+ |nk , + 1i
the dynamic variables of every oscillator are k,
combinations of Fourier coefficients. ak, |nk , i = c |nk , 1i.

We quantize these oscillators by postulating that Pk, Because the states |nk , i are normalized we can calculate
and Qk, are not any more pure numbers but operators the coefficients as
which satisfy the canonical commutation rules
|c+ |2 = |c+ |2 hnk , + 1|nk , + 1i
[Qk, , Pk ,0 ] = ihkk
0 0 0
= hnk , |ak, a |nk , i
[Qk, , Qk0 ,0 ] = 0 k,
= hnk , |Nk, + [ak, , a ]|nk , i
[Pk, , Pk0 ,0 ] = 0. k,
= nk , + 1,
We define dimensionless combinations ak, and a of |c |2 = hnk , |a ak, |nk , i = nk , .
k, k,
the operators Pk, and Qk, as
We choose the phase of the coefficients so that at the
1 moment t = 0 we have
ak, = (Qk, + iPk, )
2h q
1 a |nk , i = nk , + 1|nk , + 1i
a = (Qk, iPk, ). k,
k, 2h ak, |nk , i = nk , |nk , 1i.
p

It is easy to see that they satisfy the commutation

relations Because

[ak, , a 0 0 ] = kk0 0 nk , = hnk , |Nk, |nk , i = hnk , |a ak, |nk , i

k , k,
[ak, , ak0 ,0 ] = [a , a 0 0 ] = 0. and because the norm of a vectors is always non-negative
k, k ,
we must have
Note In these relations the operators must be evaluated nk , 0.
at the same moment, i.e. [ak, , a 0 0 ] stands in fact for
k , From this we can deduce that the only possible
the commutator [ak, (t), a 0 0 (t)]. eigenvalues are
k ,
We further define the Hermitean operator nk , = 0, 1, 2, . . . .

Nk, = a ak, . We interprete

k,
the state |nk , i to contain exactly nk , photons,
It is easy to see that each of which is characterized by a wave vector k and
a polarization () .
[ak, , Nk0 ,0 ] = kk0 0 ak,

[a , Nk0 ,0 ] = kk0 0 a . the operator a to create a photon with the wave
k, k, k,
vector k and the polarization () .
Due to the Hermiticity the eigenvalues nk , of the
operator Nk, are real and the eigenvectors the operator ak, to destroy a photon with the wave
vector k and the polarizartion () .
Nk, |nk , i = nk , |nk , i
the operator Nk, to count the number of photons
form an orthonormal complete basis. with the wave vector k and the polarization () in
With the help of the commutation rule the state
[a , Nk0 ,0 ] = kk0 0 a The state composed of various kind of photons is a direct
k, k,
product of individual vectors |nki ,i i:
we see that
|nk1 ,1 , nk2 ,2 , . . . , nki ,i , . . .i
Nk, a |nk , i = (a Nk, + a )|nk , i
k, k, k, = |nk1 ,1 i |nk2 ,2 i |nki ,i i .
= (nk , + 1)a |nk , i.
k,
The vector |0i stands for the state that has no kind of
Similarly we can show that photons, i.e.

Nk, ak, |nk , i = (nk , 1)ak, |nk , i. |0i = |0k1 ,1 i |0k2 ,2 i .

Application of any operator ak, onto this results always where
zero. We say that |0i represents the vacuum. = |k|c.
It is easy to see that a general normalized photon state
can be constructed applying operations a When we choose the energy scale so that
k,
consecutively: H|0i = 0,
n
Y (ak , ) ki ,i the Hamiltonian takes the form
|nk1 ,1 , nk2 ,2 , . . .i = qi i |0i.
ki ,i n ki ,i ! XX
H= hNk, .
k

Note Since the operators a and a 0 0 commute the
k, k ,
order of operators does not matter. The many photon When it acts on a many photon state the result is
states are symmetric with respect to the exchange of
photons. We say that the photons obey the Bose-Einstein H|nk1 ,1 , nk2 ,2 , . . .i
statistics or that they are bosons.
X
= nki ,i hi |nk1 ,1 , nk2 ,2 , . . .i.
Since the numbers nk , tell us the number of photons of i
type (k, ) in the volume under consideration we call
them the occupation numbers of the state. The quantum mechanical momentum operator is exactly
Correspondingly the space spanned by the state vectors is of the same form as the classical function (the Poynting
called the occupation number space. vector):
In the quantum theory the Fourier coefficients of a Z
1
classical radiation field must be replaced by the P = (E B) d3 x
corresponding non-commuting creation and annihilation c
XX 1
operators. Substituting = hk(a ak, + ak, a )
2 k, k,
p
ck, 7 c h/2 ak, (t) k
XX 1
= hk(Nk, + ).
ck, 7 c h/2 a (t)
p
k, 2
k
we get Since here the summation goes over all wave vectors the
r term associated with the factor 1/2 will not appear in the
1 X h h
A(x, t) = c ak, (t)() eikx final result the terms hk and hk cancelling each other.
V 2 For the momentum operator we get thus
k,
i
+a (t)() eikx .
XX
k, P = hkNk, .
k
Note Here A is an operator defined at every point of the
space whereas A of the classical theory is a three For one photon states we have
component field defined at every point. The variables x
and t are both in classical and quantum mechanical cases Ha |0i = ha |0i
k, k,
variables parametrizing the fields. Fields like the operator
A are called field operators or quantized fields. P a |0i = hka |0i,
k, k,
Also in the quantum theory the Hamiltonian is of the
so
form Z
1
H= (B B + E E) d3 x. h = h|k|c = photon energy
2
Substituting the field operator A into the equations hk = photon momentum.

1 The photon mass will be

E = A
c t
1 2
B = A (mass)2 = (E |p|2 c2 )
c4
and noting that this time the Fourier coefficients do not 1
= [(h)2 (h|k|c)2 ]
commute we get c4
= 0.
1 XX
H = h(a ak, + ak, a )
2 k, k, The photon state is also characterized by its polarization
k () . Since () transforms under rotations like a vector
XX 1 the photon is associated with one unit of angular
= (Nk, + )h,
2 momentum, i.e. the spin angular momentum of the
k
photon is one. We define the circularly polarized x and t in the expression for the field operator A are
combinations not quantum mechanical variables but simply
parameters which the operator A depends on. For
1
() = ((1) i(2) ). example, it is not allowed to interprete the variables
2 x and t as the space-time coordinates of a photon.
We rotate these vectors by an infinitesimal angle the quantized field A operates at every point (x, t) of
around the progation direction k. Their changes are the space where it with a certain probability creates
and annihilates excitation states called photons.

() = ((2) i(1) ) Thus photons can be interpreted as the quantum
2 mechanical excitations of the radiation field.
= i () .
We consider photon emission and and absorption of non
We select the propagation direction k as the quantization relativistic atomic electrons. The relevant interaction
axis and compare this expression with the transformation Hamiltonian is of the form
properties of angular momentum eigenstates X e
Hint = A(xi , t) pi

i

i
m ec
|jmiR = 1 Jz |jmi = (1 im )|jmi.
e2

h
+ 2 2 A(xi , t) A(xi , t) ,
2me c
We see that
where the transversality condition is taken into account
the spin components of the polarizations () are by replacing the operator pi A with the operator A pi .
m = 1. The summation goes over all electrons participating in
the state m = 0 is missing due to the transversality the process. The symbols xi stand for their position
condition. coordinates.
Note If we had to consider the interaction of spin and
our original linear polarization states are 50/50 radiation we should also include the term
mixtures of m = 1 and m = 1 states. (spin)
X eh
Hint = i [ A(x, t)]|x=xi .
Hence the photon spin is always either parallel or i
2me c
antiparallel to the direction of the propagation.
The operators ak, and a are time dependent and so This time the Hamiltonian operator Hint operates not
k, only on the atomic states but also on the photon states.
they satisfy the Heisenberg equations of motion
In the quantum theory of radiation
i the vector describing the initial state |ii is the direct
ak, = [H, ak, ]
h product of an atomic state A and a (many) photon
i XX
= [h 0 Nk0 ,0 , ak, ] state characterized by the occupation numbers nk , :
h 0 0
k
= iak, |ii = |Ai |nk , i = |A; nk , i.

like also the vector describing the final state |f i is the direct
product of an atomic state B and a (many) photon
a = ia .
k, k, state characterized by the occupation numbers nk0 ,0 :
These equations have solutions
|f i = |Ai |nk0 ,0 i = |A; nk0 ,0 i.
ak, = ak, (0)eit
a = a (0)eit .
k, k, Absorption
Now
The final form of the field operator is then
r |ii = |A; nk , i
1 X h h
A(x, t) = c ak, (0)() eikxit |f i = |B; nk , 1i.
V 2
k,
i In the first order perturbation theory the amplitude of
+ a (0)() eikx+it .

the process
k,
|ii |f i
We should note that
is the matrix element of the interaction operator HI
the operator A is Hermitean. between the states |ii and |f i. Up to this order
 only ak, leads to a nonzero matrix element, is a which adds one photon to the final state. The
k,
eventhough the field operator A is a linear relevant matrix element is now
superposition of creation and annihilation operators
a and ak, , respectively. hB; nk , + 1|Hint |A; nk , i
k, s
the term A A is out of the question in this process e (nk , + 1)h X
= hB|eikxi pi () |Aieit .
because it either changes the number of photons by me 2V i
two or does not change it at all.
If nk , is very large then
The first order transition matrix element is now
q
nk , + 1 nk , ,
p
hB; nk , 1|Hint |A; nk , i
e and the semiclassical and quantum mechanical treatment
= hB; nk , 1|
me c coincide.
r
h If nk , is small the semiclassical method fails completely.
ak, (0)eikxi it pi () |A; nk , i
X
c
2V In particular, the semicalssical treatment of the
i
s spontaneous emission, nk , = 0, is impossible. The
e nk , h X semiclassical method can be applied if we insert the atom
= hB|eikxi pi () |Aieit . into the fictitious radiation field
me 2V i
(emis) ikx+it
A(emis) = A0 e ,
Comparing this with the matrix element of the
semiclassical perturbation potential where s
eA0  i(/c)nx  (emis)
(nk , + 1)h

Vni = e  p A0 =c () .
me c ni 2V
we see that they both give exactly the same result The field A(emis) is not
provided we use in the semiclassical theory the equivalent directly proportional to the number of photons nk , ,
radiation field
(abs) ikxit the complex conjugate of the field A(abs) .
A(abs) = A0 e ,
Example Spontaneous emission from the state A to the
where the amplitude is state B.
s In the first order the transition rate is
(abs)
nk , h
A0 =c () . wAB
2V
2
Because the transition probability is = |hB; 1k, |Hint |A; 0i|2 (EB EA + h)
h
 2
according to the semiclassical theory directly 2 e2 h X ikxi ()

=  hB|e  pi |Ai

proportional to the intensity of the radiation, 2
h 2me V  
i
|A0 |2 nk , , (EB EA + h).

according to the quantum theory directly Like in the photoelectric efect we can deduce that the
proportional to the occupation number nk , , number of the allowed photon states (E, d) in the
energy interval (h, h + d(h)) and in the solid angle d
both the semiclassical and quantum mechanical results is
give equivalent results also at low intensities, i.e. when V 2
(E, d) = n2 dn d = d(h) d.
nk , is small. (2)3 hc3
The transition rate of photons emitting into a certain
Emission
solid angle is thus
Now
 2
|ii = |A; nk , i 2 e2 h X ikxi ()
 V 2 d
wd = hB|e
 p |Ai ,

i
h 2m2e V  i  (2)3 hc3
 
|f i = |B; nk , + 1i
where h = EA EB .
and in the first order the only potential term of the field
We consider only hydrogen like atoms so that only one
r
1 X h h electron participates in the process and we restrict to the
A(x, t) = c ak, (0)() eikxit dipole approximation. Then
V 2
k,
e2
|hB|p|Ai () |2 d.
i
+ a (0)() eikx+it
wd =
8 2 m2e hc3
k,
Earlier we saw that in the term A A creation and annihilation operators
appear as quadratic,
ime (EB EA )
hB|p|Ai = hB|x|Ai
h only the quadratic term A A contributes in the first
= ime xBA . order perturbation theory.
Only two of the terms of the form
We let the symbol () stand for the angle between the
vector xBA and the polarization direction () , i.e. aa , a a, aa, a a

cos (1) = sin cos in the operator A A have non zero matrix elements
cos (2) = sin sin , provided that
0

when and are the direction angles of the vector x. a creates a photon of the type (k0 , ( ) ),
Then a annihilates a photon of the type (k, ( ),
e2 3
wd = |xBA |2 cos2 () d.
8 2 hc3 and then
The total transition rate is obtained by integrating over h1k0 ,0 |ak, a 0 0 |1k, i = 1.
k ,
all propagation directions k/|k| and summing over both
Now
polarizations:
e2 3 hB; 1k0 ,0 |Hint |A; 1k, i
w= |xBA |2 .
3hc3  2 
 e 
The life time of a state was obtained from the formula = hB; 1k0 ,0  
2 A(x, t) A(x, t)A; 1k, i

2me c
1 X  2
 e
= wABi ,
A = hB; 1k0 ,0  (a a 0 + a 0 0 ak, )
i 2me c2 k, k ,0 k ,
c2 h

0 0
() ( ) ei(kk )xi( )t A; 1k, i
0 
where we have to sum also over the magnetic quantum
numbers m. For example the life time of the hydrogen 2p 2V 0
state is e2 c2 h 0 0

(2p 1s) = 1.6 109 s. = 2

2() ( ) ei( )t hB|Ai,
2me c 2V 0

Electron photon scattering where again the exponential functions ekx are replaced
We consider the process by the constant 1 (the long wave length approximation).
In the first order we have thus
|1k, i |1k0 ,0 i,
i t if i t0
Z
c(1) (t) = e Vf i (t0 ) dt0
i.e. h t0
1 e2 c2 h 0
before the scattering the atom is in the state A, and = 2
2AB () ( )
ih 2me c 2V 0
k and () are the wave vector and polarization of Z t
the incoming photon.
0 0
ei(h +EB hEA )t /h dt0 ,
0
after the scattering the atom is in the state B, k0 is
0
the wave vector and ( ) the polarization vector of where = |k|c and 0 = |k0 |c. Now
the outgoing photon.
in the transition amplitude c(1) (t) the interaction is
The Hamiltonian of the interaction is in fact of second order: A A.

e e2 in the second order correction c(2) (t) the term A p

Hint = A(x, t) p + A(x, t) A(x, t). is also of second order.
me c 2m2e c2

Because To collect all contributions up to the second order in the

interaction we have to consider also the correction c(2) (t),
the number of photons does not change in the into which we take all double actions of the operator
scattering, A p. Now

in order to be non zero the matrix element of the

 2 X Z t Z t0
i 0 0

interaction must contain products of photon creation c(2)

(t) = dt dt00 eif m t Vf m (t0 )
h m t0 t0
and annihilation operators, 00
eimi t Vmi (t00 ).
in the term A p creation and annihilation operators
appear as linear, Thus there are two possibilities: the interaction A p can
 at the moment t1 annihilate the incoming photon photons c/V , so the differential cross section is
(k, () ) and at some later time t2 create the  0
0 d  0
outgoing photon (k0 , ( ) ) or = r0 2
AB () ( )
d 
at the moment t1 create the outgoing photon 0
1 X (p ( ) )BI (p () )IA

0 (0 )
(k ,  ) and at some later time t2 annihilate the me EI EA h
incoming photon (k, () ). I
0 2
(p () )BI (p ( ) )IA 
Between the moments t1 and t2 the atom is in the state I, +  ,
EI EA + h 0
which normally is neither of the states A and B.
In the intermediate state I there are thus two where r0 2.82 1013 cm is the classical radius of
possibilities: either there are no photons present or both electron. This expression is known as the
incoming and outgoing photons are present. We get thus Kramers-Heisenberg formula.
(in the dipole approximation) Example Elastic scattering.
2 Z t Z t2 Now A = B ja h = h 0 . Using the commutation
2 
1 c h e relations of the operators x and p, the completeness of
c(2) (t) = 2
dt2 dt1
(ih) 2V 0 m e c 0 0 the intermediate states and the relation
X
0 0
hB|p ( ) |Iiei(EB EI +h )t2 /h pAB = ime AB xAB
I
we can write
hI|p () |Aiei(EI EA h)t1 /h
0 1 Xh 0
() ( ) = (x () )AI (p ( ) )IA
X () i(EB EI +h)t2 /h
+ hB|p  |Iie ih
I
I i
0
(p () )AI (x ( ) )IA

0 0
hI|p ( ) |Aiei(EI EA h )t1 /h
1 X 2 0
2 = (p () )AI (p ( ) )IA ,
c2 h

e me h IA
= I
ih2V 0 me c
where IA = (EI EA )/h.
X (p (0 ) )BI (p () )IA
We see that
i
EI EA h 0

0
! AA () ( )
(p () )BI (p ( ) )IA 0
"
+ 1 X (p ( ) )AI (p () )IA
EI EA + h 0
me h IA
Z t I
0 0
#
dt2 ei(EB EA +h h)t2 /h . (p () )AI (p ( ) )IA
0 +
IA +
For the transition rate we get combining the terms c(1) (t) " 0
1 X (p ( ) )AI (p () )IA
and c(2) (t) and taking into account the relation =
me h IA (IA )
I
Z t 2 0
#
ixt0 0 (p () )AI (p ( ) )IA
 
lim 
 e dt  = 2t(x) .
t 0 IA (IA + )
the expression
If is small then
Z
wd = (|c(1) + c(2) |2 /t)(E, d) dE 1 1 (/IA )
.
IA IA
2  2 2
c2 h e2 V 0

2 Then
= d
h 2V 0 me c2 (2)3 hc3 X 1 h 0

0 2 (p ( ) )AI (p () )IA
AB () ( ) IA

I
0
i
0 (p () )AI (p ( ) )IA
1 X (p ( ) )BI (p () )IA

Xh 0
me EI EA h = m2e (x ( ) )AI (x () )IA
I
0 2 I
(p () )BI (p ( ) )IA  0
i
+  . (x () )AI (x ( ) )IA
EI EA + h 0
0

Because in the initial state there was exactly one photon = m2e ([x ( ) , x () ])AA
in the volume V and the flux density of the incoming = 0.
The differential cross section is now
 2  X  3
d r0 1
4 

=
d me h IA
I
(0 )
[(p  )AI (p () )IA
2
() (0 )

+ (p  )AI (p  )IA ]
 r m 2  X 1
0 e
= 4 
h IA
I
0
[(x ( ) )AI (x () )IA
2
0
+(x () )AI (x ( ) )IA ] .


At long wave lengths the differential cross section obeys

the Rayleigh law or
d 1
4.
d
Now

for ordinary colourless gases IA corresponds to wave

lengths in the ultraviolet,
for the visible light we have then  IA ,
so our approximations are valid in the atmossphere. The
theory explains why the sky is blue and the sunset red.
Diracs equation Here is a two component wave function (spinor).
We construct relativistically covariant equation that takes We define new two component wave functions
into account also the spin.  
(R) 1
The kinetic energy operator is = ih ih
mc x0
p2 (L) = .
H (KE) = .
2m
It is easy to see that these satisfy the set of simultaneous
Previously we derived for Pauli spin matrices the relation equations
( a)2 = |a|2 ,
 

ih ih (L) = mc(R)
x0
so we can also write  

( p)( p) ih ih (R) = mc(L) .
H (KE) = . x0
2m
We define yet new two component wave functions
However, when the particle moves under the influence of
a vector potential these expressions differ. Substituting A = (R) + (L)
B = (R) (L) .
p 7 p eA/c
These in turn satisfy the matrix equation
the latter takes the form
ih ih
   
x0 A A
   
1 eA eA = mc .
p p

2m c c ih ih B B
x0
 2
1 eA We now define the four component wave function
= p
2m c   (R)
+ (L)
 
i

eA
 
eA
 A
+ p p = =
2m c c
B (R) (L)
2
and the 4 4-matrices

1 eA eh
= p B,
2m c 2mc 
ik
0

k =
where we have used the identities ik
0
 
1 0
( a)( b) = a b + i (a b) 4 = .
0 1

and We end up with the Diracs equation

p A = ih( A) A p.  
mc
Let us suppose that for the relativistically invariant + 4 + =0
(ix0 ) h
expression
(E 2 /c2 ) p2 = (mc)2 for free spin- 21 particles. Employing the four vector
notation the equation takes the form
the operator analogy  
mc
1 (op) 2 + = 0.
E p2 = (mc)2 x h
c2
Note The Dirac equation is in fact a set of four coupled
holds. Here linear differential equations. The wave function is the

E (op) = ih = ihc four component vector
t x0

and 1
p = ih. 2
= 3 .

We write the operator equation into the form
4
! !
E (op) E (op) This kind of a four component object is called bispinor or
p + p = (mc)2
c c Diracs spinor. Explicitely written down the Dirac
equation is
or
4 X4  




 X  mc 
ih + ih ih ih = (mc)2 . ( ) + = 0.
x0 x0 =1
x h
=1
Note The fact that the Dirac spinor happens to have Forming the Hermitean conjugate of the Dirac equation
four components has nothing to do with our four  
dimensional space-time; does not transform like a four mc
+ =0
vector under Lorentz transformations. x h
It is easy to verify that the gamma-matrices (Dirac
matrices) satisfy the anticommutation rule we get
mc
k + 4 + = 0.
{ , } = + = 2 . xk x4 h
We multiply this from right by the matrix 4 and end up
Furthermore, every is Hermitian,
with the adjungated equation
= , mc
+ = 0.
and traceless, i.e. x h
Tr = 0.
Here we have used the relations
Lets multiply the equation

= =
x4 (ict) x4
 
mc
+ 4 + =0 k 4 = 4 k .
(ix0 ) h

on both sides by the matrix 4 and we get Lets multiply the original Dirac equation
   
mc
ch4 ih + 4 mc2 = 0. + =0
t x h

Denote from left with the adjungated spinor and the


1 0
 adjungated equation
= 4 =
0 1 mc
  + = 0
0 k x h
k = i4 k = ,
k 0
from right with the spinor and subtract the resulting
which satisfy the relations equations. We then get

{k , } = 0
( ) = 0.
2 = 1 x
{k , l } = 2kl . The quantity
When we now write
s = ic = (c , ic )
H = ich + mc2 ,
thus satisfies a continuity equation. According to Greens
the Dirac equation takes the familiar form theorem we have
Z Z
4 d x = d3 x = constant,
3
H = ih .
t
We define the adjungated spinor like: where the constant can be taken to be 1 with a suitable
normalization of . Because 4 = is positively
= 4 . definite it can be interpreted as a probability density.
Then
Explicitely, if is a column vector sk = ick = c k
can be identified as the density of the probability current.

1
2 Note It can be shown that s transforms like a four
= 3 ,

vector, so the continuity equation is relativistically
4 covariant.
It can be proved that any sets of four matrices and 0
then and are row vectors satisfying the anticommutation relations

= (1 , 2 , 3 , 4 ) { , } = 2
= (1 , 2 , 3 , 4 ). {0 , 0 } = 2 ,
are related to eachother through a similarity Suppose now that
transformation with a non-singular 4 4-matrix S:
E mc2 , |eA0 |  mc2
S S 1 = 0 . and measure the energy starting from the rest energy:
With the help of the matrices 0 the original Dirac E (NR) = E mc2 .
equation can be written as
We expand
 
mc c2 1

2mc2

S 1 0 S + S 1 S = 0. =
x h E eA0 + mc2 2m 2mc2 + E (NR) eA0
" #
Multiplying this from left with the matrix S we get 1 E (NR) eA0
= 1 + .
  2m 2mc2
mc
0 + 0 = 0,
x h This can be taken to be the power series in (v/c)2 since

where E (NR) eA0 [p (eA/c)]2 /2m mv 2 /2.

0 = S. Taking into account only the leading term we get
Thus Diracs equation is independent on the explicit form
   
1 eA eA
of the matrices ; only the anticommutation of the p p A = (E (NR) eA0 )A .
2m c c
matrices is relevant. If the matrices 0 are Hermitean the
transformation matrix S can be taken to be unitary. It is This can be written as
" 2 #
easy to show that then the probability density and 1

eA eh
current, for example, are independent on the p B + eA0 A = E (NR) A .
2m c 2mc
representation:
0 0 0 = . Up to the zeroth order of (v/c)2 the component A is
thus the two component Schrodinger-Pauli wave function
2
(multiplied with the factor eimc t ) familiar from the
Vector potential
non-relativistic quantum mechanics. The equation
When the system is subjected to a vector potential   
eA 1
A = (A, iA0 ), p A = (E eA0 + mc2 )B
c c

we make the ordinary substitutions tells us that the component B is roughly by the factor
|p (eA/c)|/2mc v/2c
ih(/x ) 7 ih(/x ) eA /c.
less than A . Due to this, provided that E mc2 , A
The Dirac equation takes now the form is known as the big and B as the small component of the
  Dirac wave function .
ie mc We obtain relativistic corrections only when we consider
A + = 0.
x hc h the second or higher order terms of the expansion
c2 2mc2
 
Assuming that A does not depend on time the time 1
=
dependence of the spinor can be written as E eA0 + mc2 2m 2mc2 + E (NR) eA0
" #
= (x, t)|t=0 eiEt/h . 1 E (NR) eA0
= 1 + .
2m 2mc2
Let us write now the Dirac equation for the components
A and B : Let us suppose now that
   A = 0.
eA 1
p B = (E eA0 mc2 )A
c c The wave equation is then
  
eA 1 HA A = E (NR) A ,
p A = (E eA0 + mc2 )B .
c c
where
With the help of the latter equation we eliminate B
!
1 E (NR) eA0
from the upper equation and get HA = ( p) 1 ( p) + eA0 .
2m 2mc2
c2
      
eA eA This wave equation looks like a time independent
p p A
c E eA0 + mc2 c Schrodinger equation for the wave function A .
= (E eA0 mc2 )A .
However,
 evaluating corrections up to the order (v/c)2 the Writing E (NR) p2 in the form 12 {E (NR) , p2 } we get
wave function A is not normalized because the
p2 p4

probability interpretation of Diracs theory requires
+ eA0
that Z 2m 8m3 c2

(A A + B B ) d3 x = 1, 1  2
+ {p , (E (NR) eA0 )}
8m2 c2
where B already of the order v/c. (NR)

2( p)(E eA0 )( p)
explicitely writing down the expression for the
operator HA we see that it contains the = E (NR) .
non-Hermitian term ihE p.
Because for arbitrary operators A and B
the equation is not an eigenvalue equation since HA
itself contains the term E (NR) . {A2 , B} 2ABA = [A, [A, B]]

Up to the order (v/c)2 the normalization condition can holds we can, by setting
now be written as
p = A
p2
Z  
(NR)
A 1+ A d3 x 1, E eA0 = B,
4m2 c2
reduce the equation into the form
because according to the equation
 2
   p p4
eA 1 + eA0
p A = (E eA0 + mc2 )B 2m 8m3 c2
c c
eh2

eh (E p)
E
we have 4m2 c2 8m2 c2
p
B A . = E (NR) .
2mc
It is worthwhile to define the new two component wave
In the derivation of the equation we have employed the
function :
relations
= A ,
where [ p, (E (NR) eA0 )] = ieh E
2
p [ p, ieh E] = eh2 E
=1+ .
8m2 c2 2eh (E p),
Now is up to the order (v/c)2 normalized correctly
because the validity of which can be verified by noting that

p2 A0 = E
Z Z  
3
d x A 1 + A d3 x.
4m2 c2 E = 0.

We multiply the equation The resulting equation is a proper Schrodinger equation

for a two component wave function.
HA A = E (NR) A , Physical interpretation
We look at the meaning of the terms in the equation
on both sides with the operator
 2
p p4
1 = 1 (p2 /8m2 c2 ), + eA0
2m 8m3 c2
eh2

and get eh (E p)
E
1 HA 1 = E (NR) 2 . 4m2 c2 8m2 c2
Explicitely, up to the order (v/c)2 this can be written as = E (NR) .

p2
p2 p2 p2
   
1. The term 2m + eA0 gives the non-realtivistic
+ eA0 , + eA0 Schrodinger equation.
2m 8m2 c2 2m
! p4
( p) E (NR)
eA0
 2. The term 3 2 is a relativistic correction to the
( p) 8m c
2m 2mc2 kinetic energy as can be seen from the expansion

p2
 
(NR)
1
p |p|2 |p|4
=E . (mc2 )2 + |p|2 c2 mc2 = + .
4m2 c2 2m 8m3 c2
 eh (E p)
3. The term describes the interaction
4m2 c2
between the spin of a moving electron and electric field.
Intuitively this, so called Thomas term, is due to the fact
that the moving electron experiences an apparent
magnetic field E (v/c). If the electric field is a central
field,
eA0 = V (r),
it can be written in the form
 
eh h 1 dV
(E p) = (x p)
4m2 c2 4m2 c2 r dr
1 1 dV
= S L,
2m2 c2 r dr
where we have substituted

S = h/2.

So we actually have a spin orbit interaction.

2
4. The term eh2 2 E is known as the Darwin term.
8m c
Its meaning can be deduced when we note that E is
the charge density. For example, in the hydrogen atom
where E = e(x) it causes the energy shift

e2 h2 e2 h2
Z 
(Schro) 2 3 (Schro) 2 

2 2 (x)| | d x = 2 2 | | ,
8m c 8m c 
x=0
which differs from zero only in the s-state.