Professional Documents
Culture Documents
LN-29 Time Evolution Operator
LN-29 Time Evolution Operator
Schrödinger picture
Heisenberg picture
Dirac picture
________________________________________________________________________
Erwin Rudolf Josef Alexander Schrödinger (12 August 1887– 4 January 1961) was an
Austrian theoretical physicist who was one of the fathers of quantum mechanics, and is
famed for a number of important contributions to physics, especially the Schrödinger
equation, for which he received the Nobel Prize in Physics in 1933. In 1935, after
extensive correspondence with personal friend Albert Einstein, he proposed the
Schrödinger's cat thought experiment.
http://en.wikipedia.org/wiki/Erwin_Schr%C3%B6dinger
________________________________________________________________________
Werner Heisenberg (5 December 1901– 1 February 1976) was a German theoretical
physicist who made foundational contributions to quantum mechanics and is best known
for asserting the uncertainty principle of quantum theory. In addition, he made important
contributions to nuclear physics, quantum field theory, and particle physics. Heisenberg,
along with Max Born and Pascual Jordan, set forth the matrix formulation of quantum
mechanics in 1925. Heisenberg was awarded the 1932 Nobel Prize in Physics for the
creation of quantum mechanics, and its application especially to the discovery of the
allotropic forms of hydrogen.
http://en.wikipedia.org/wiki/Paul_Dirac
_______________________________________________________________________
29.1 Time evolution operator
We define the Unitary operator as
(t ) Uˆ (t , t0 ) (t0 )
Normalization
(t ) (t ) (t0 ) (t0 ) 1
Then
or
t
t0 t1 t2
We note that
This should be
with
lim Uˆ (t 0 dt , t 0 ) 1̂
dt 0
Uˆ (t 0 dt , t 0 ) 1̂ i
ˆ dt
or
ˆ
ˆ (Hermitian)
We assume that
ˆ Ĥ
where Ĥ is a Hamiltonian.
t
t0 t t+dt
Hˆ
Uˆ (t dt , t 0 ) (1 i dt )Uˆ (t , t 0 )
or
Hˆ
Uˆ (t dt , t 0 ) Uˆ (t , t 0 ) i dtUˆ (t , t 0 )
Uˆ (t dt , t 0 ) Uˆ (t , t 0 ) Hˆ
lim i Uˆ (t , t 0 )
dt 0 dt
or
or
ˆ
i U (t , t 0 ) Hˆ Uˆ (t , t 0 )
t
ˆ
i U (t , t 0 ) (t 0 ) Hˆ Uˆ (t , t 0 ) (t 0 )
t
or
i (t ) Hˆ (t )
t
t
t0 Dt t
t t0
t
N
iHˆ t t0 N iHˆ
lim [1̂ ( )] exp[ (t t0 )]
N N
or
iHˆ
Uˆ (t , t0 ) exp[ (t t0 )]
i (t ) (t ) Hˆ (t ) Hˆ
t
or
i
(t ) (t ) Hˆ (t ) Hˆ
t
d Aˆ
(t ) Aˆ (t ) (t ) ( (t ) ) Aˆ (t ) (t ) (t ) (t ) Aˆ ( (t ) )
dt t t t
i Aˆ i
(t ) Hˆ Aˆ (t ) (t ) (t ) (t ) Aˆ ( ) (t )
t
i Aˆ
(t ) [ Aˆ , Hˆ ] (t ) (t ) (t )
t
or
d i Aˆ
(t ) Aˆ (t ) (t ) [ Aˆ , Hˆ ] (t ) (t ) (t )
dt t
pˆ 2
Hˆ V ( xˆ )
2m
d i
(t ) xˆ (t ) (t ) [ xˆ , Hˆ ] (t )
dt
i pˆ 2
(t ) [ xˆ , ] (t )
2m
i pˆ
i (t ) ] (t )
m
d pˆ
(t ) xˆ (t ) (t ) ] (t )
dt m
or
d 1
x p
dt m
Similarly
d i
(t ) pˆ (t ) (t ) [ pˆ , Hˆ ] (t )
dt
i
(t ) [ pˆ ,V ( xˆ )] (t )
i
(t ) V ( xˆ )] (t )
i xˆ
or
d
(t ) pˆ (t ) (t ) V ( xˆ )] (t )
dt xˆ
or
d dV
p
dt dx
The equations
d 1
x p ,
dt m
and
d dV
p
dt dx
express the Ehrenfest’s theorem. These forms recall that of classical Hamiltonian-Jacobi
equations for a particle.
________________________________________________________________________
29.7 Example: Spin precession
2 B Sˆz
ˆz B
ˆz .
2 Sˆ
Hˆ
ˆ z B ( B z )B B
ˆ zB
e
Since the Bohr magneton µB is given by B ,
2mc
eB eB
B B 0
2mc 2 mc 2
or
eB
0 (angular frequency of the Larmor precession)
mc
ˆ
Hˆ 0z
2
i i
(t ) exp[ Hˆ t ] (t 0) exp[ 0ˆ z t ] (t 0)
2
Note that the time evolution operator coincides with the rotation operator
i
Rˆ z ( 0 t) exp[ 0
ˆ z t]
2
We assume that
i2
e cos( )
i i
(t 0) exp[ ˆ z ] exp[ ˆ y ] i 2
n
2 2 2
e sin( )
2
The average
i i
Sx t
(t ) Sˆ x (t ) n exp[ 0ˆ z t ] x exp[ 0ˆ z t ] n
2 2 2
i i
Sy (t ) Sˆ y (t ) n exp[ 0ˆ z t ] y exp[ 0ˆ z t ] n
t 2 2 2
i i
Sz t
(t ) Sˆ z (t ) n exp[ 0ˆ z t ] z exp[ 0ˆ z t ] n
2 2 2
Here we have
it2 0 0 1 it2 0
i i e 0 e 0
exp[ 0ˆ z t ] x exp[ 0ˆ z t ] it 0
0 0
it
2 2 0
2 1
e 2
e 0
0 eit 0
it
e
0
0
it2 0 0 i it2 0
i i e 0 e 0
exp[ 0ˆ z t ] y exp[ 0ˆ z t ] it 0
0 0
it
2 2 0
2 i
e 2
e 0
0 ieit 0
it
ie
0
0
i i 1 0
exp[ 0ˆ z t ] z exp[ 0ˆ z t ]
2 2 0 1
Thus we have
S x t sin cos(0t )
2
sin [cos(0t ) cos sin(0t ) cos ]
2
Sy sin [sin(0t ) cos cos(0t ) sin ]
t 2
At t = 0,
Sx 0
sin cos
2
Sy sin sin
0 2
Sz 0
cos
2
Sx t
S x 0 cos(0t ) S y sin(0t )
0
Sy S x 0 sin(0t ) S y cos(0t )
t 0
Sz t
Sz 0
cos
2
Sx t i S y e i 0 t ( S x 0
i Sy )
t 0
i i
Sx t
(t ) Sˆ x (t ) n exp[ 0ˆ z t ]ˆ x exp[ 0ˆ z t ] n
2 2 2
n ˆ x cos(0t ) ˆ y sin(0t ) n
2
where
Sx ̂ x
0
2n n
We also get
i i
Sy (t ) Sˆ y (t ) n
exp[ 0ˆ z t ]ˆ y exp[ 0ˆ z t ] n
t 2 2 2
ˆ x sin(0t ) ˆ y cos(0t )
2n n
and
Sz t
(t ) Sˆ z (t )
i i
n
exp[ 0ˆ z t ]ˆ z exp[ 0ˆ z t ] n
2 2 2
n ˆ z n S z 0
2
Then
Sx t
S x 0 cos(0t ) S y sin(0t )
0
Sy S x 0 sin(0t ) S y cos(0t )
t 0
and
Sz t
Sz 0
________________________________________________________________________
29.8 Baker-Hausdorf lemma
In the commutation relations, [ Jˆ z , Jˆ x ] iJˆ y , we put Jˆ z ̂ z and Jˆ x ̂ x
2 2
Then we have
[ ˆ z , ˆ x ] i ˆ y or [ˆ z , ˆ x ] 2iˆ y .
2 2 2
Similarly, we have
We notice the following relations which can be derived from the Baker-Hausdorf
lemma:
x x2 x3
exp( Aˆ x) Bˆ exp( Aˆ x) Bˆ [ Aˆ , Bˆ ] [ Aˆ ,[ Aˆ , Bˆ ]] [ Aˆ ,[ Aˆ ,[ Aˆ , Bˆ ]]] ...
1! 2! 3!
exp[i ˆ z ]ˆ x exp[i ˆ z ] ˆ x cos ˆ y sin
2 2
((Proof))
We note that
i ˆ
x , A ˆ z , and Bˆ ̂ x .
2
Then we have
x x2 x3
I exp[ xˆ z ]ˆ x exp[ xˆ z ] ˆ x [ˆ z , ˆ x ] [ˆ z , [ˆ z , ˆ x ]] [ˆ z , [ˆ z , [ˆ z , ˆ x ]]]
1! 2! 3!
x4
[ˆ z ,[ˆ z ,[ˆ z ,[ˆ z ,ˆ x ]]]] .....
4!
1 i 1 i 1 i 1 i
2 3 4
I ˆ x 2iˆ y [ˆ z ,2iˆ y ] [ˆ z ,[ˆ z ,2iˆ y ]] [ˆ z ,[ˆ z ,[ˆ z ,2iˆ y ]]] .
1! 2 2! 2 3! 2 4! 2
or
2 i 3 1 4
I ˆ x ˆ y i 2
[ˆ y , ˆ z ]
( 2 i )[ˆ z , [ ˆ y , ˆ z ]] (2i )[ˆ z , [ˆ z , [ˆ y , ˆ z ]]].....
2 3! 23 4! 2 4
2 i 3 1 4
ˆ x ˆ y i 2 2iˆ x ( 2 i )( 2 i )[ ˆ z , ˆ x ] (2i )(2i )[ˆ z , [ˆ z , ˆ x ]] ...
2 3! 23 4! 2 4
or
2 i 3 1 4
I ˆ x ˆ y ˆ x ( 2i )( 2i ) 2
ˆ y (2i )(2i )(2i )(2i )ˆ x ...
2 3! 23 4! 24
2 3 4
ˆ x ˆ y ˆ x ˆ y ˆ x ...
2 3! 4!
2 4 3
ˆ x (1 ...) ˆ y (1 ...)
2 4! 3!
ˆ x cos ˆ y cos
(t ) s (t )
s (t ) Uˆ (t , t0 ) s (t0 )
Uˆ (t,t 0 ) Uˆ 1 (t,t0 ) .
In the Schrodinger picture, the average of the operator Âs in the state s (t ) is defined
by
s (t ) Aˆ s s (t ) .
H (t ) s (t0 ) .
H Aˆ H (t ) H s Aˆ s (t ) s ,
or
i s (t ) Hˆ s (t ) s (t )
t
or
or
d ˆ
i U (t , t0 ) Hˆ s (t )Uˆ (t , t0 )
dt
or
d ˆ i
U(t,t0 ) Hˆ s (t)Uˆ (t,t 0 )
dt
and
d ˆ i
U (t,t0 ) Uˆ (t,t0 ) Hˆ s (t)
dt
where
Hˆ H (t ) Uˆ (t , t0 ) Hˆ s (t )Uˆ (t , t0 )
d dAˆ (t)
i Aˆ H (t) [ Aˆ H (t), Hˆ H (t)] i( s ) H
dt dt
Hˆ s(t) Hˆ , Aˆ s (t) Aˆ
i
Hˆ t
Uˆ e
i ˆ i ˆ
Ht Ht
Aˆ H Uˆ Aˆ sUˆ e Aˆ s e
Hˆ H Hˆ s
d ˆ
i AH [ Aˆ H , Hˆ H ]
dt
We get an analogy between the classical equations of motion in the Hamiltonian form
and the quantum equations of motion in the Heisenberg’s form. ÂH is called a constant
of the motion, when [ Aˆ , Hˆ ] =0 at all times.
H H
Therefore [ Aˆ H , Hˆ H ] means [ Aˆ S , Hˆ S ] 0
1
HS pˆ S V ( xˆS ) ,
2
2m
1
HH pˆ V( xˆ H ),
2
2m H
[ xˆ H , pˆ H ] xˆ H pˆ H pˆ H xˆ H
Uˆ xˆUˆUˆ pˆ Uˆ Uˆ pˆ UˆUˆ xˆUˆ
Uˆ [ xˆ , pˆ ]Uˆ iUˆ Uˆ i1̂
[ xˆ H , pˆ H ] Uˆ [ xˆ , pˆ 2 ]Uˆ
2
2iUˆ pˆ Uˆ
2ipˆ H
or
d 1
xˆ H [ xˆ H , Hˆ H ] pˆ H .
dt m
Similarly
d
i pˆ H [ pˆ H , Hˆ H ] Uˆ [ pˆ , Hˆ ]Uˆ Uˆ [ pˆ ,Vˆ ( xˆ )]Uˆ (i) V ( xˆ H )
dt xˆ H
or
d V ( xˆ H )
pˆ H ()
dt xˆ H
1
V ( xˆ H ) m 2 xˆ H
2
d
pˆ H m 2 xˆ H
dt
d i i
( xˆ H pˆ H ) i ( xˆ H pˆ H )
dt m m
i
( xˆ H pˆ H ) Aˆ H e it
m
or
d i i
( xˆ H pˆ H ) i ( xˆ H )
dt m m
i
( xˆ H pˆ H ) Bˆ H eit
m
i
Aˆ H xˆ H (0) pˆ H (0)
m
i
Bˆ H xˆ H (0) pˆ H (0)
m
Note that xˆ H (0) and pˆ H (0) correspond to the operators in the Schrödinger picture. From
these equations, we get final results
1
xˆ H xˆ H (0) cos t pˆ H (0) sin t `
m
These look to the same as the classical equation of motion. We see that x̂H and p̂ H
operators oscillate just like their classical analogue.
An advantage of the Heisenberg picture is therefore that it leads to equations which are
formally similar to those of classical mechanics.
((Note))
d2 dxˆ H ˆ pˆ H ˆ 1 m 2 2 2 2
2
i ˆ
x H [ , H H ] [ , H H ] [ ˆ
p H , ˆ
x H ] [ ˆ
p H , ˆ
x H ] 2 xˆ H
dt 2 dt m m 2 2 2 i
or
d2
xˆ H 2 xˆH
dt 2
d 1
xˆ H |t 0 pˆ H (0) , xˆ H |t 0 xˆ H (0)
dt m
The solution is
dxˆ H pˆ (0)
|t 0 [Cˆ1 sin(t ) Cˆ 2 cos n(t )]t 0 Cˆ 2 H
dt m
Thus we have
pˆ (0)
Cˆ 2 H .
m
and
pˆ H (0)
xˆ H xˆ H (0) cos(t ) sin(t )
m
________________________________________________________________________
Paul Ehrenfest (January 18, 1880 – September 25, 1933) was an Austrian and Dutch
physicist and mathematician, who made major contributions to the field of statistical
mechanics and its relations with quantum mechanics, including the theory of phase
transition and the Ehrenfest theorem.
http://en.wikipedia.org/wiki/Paul_Ehrenfest
________________________________________________________________________
29.15 Analogy with classical mechanics
In the classical mechanics, dynamical variables vary with time according to the
Hamilton’s equations of motion,
dq j H
,
dt p j
where qj and pj are a set of canonical co-ordinate and momentum, and H is the
Hamiltonian expressed as a function of them,
dA A dq j A dp j
dt
j q j dt p j dt
A H A H
j q j p j p j q j
[ A, H ]classical
Hˆ Hˆ 0 Vˆs (t)
where Hˆ 0 is independent of t.
i
Hˆ 0t
s (t ) e
I (t )
or
i ˆ
H 0t
I (t ) e s (t )
We assume that
I (t ) Aˆ I (t ) I (t ) s (t ) Aˆ s s (t )
or
i i ˆ
Hˆ 0t H 0t
s (t ) e
Aˆ I (t )e s (t ) s (t ) Aˆ s s (t )
or
i i ˆ
Hˆ 0t H 0t
e
Aˆ I (t )e Aˆ s
i ˆ i ˆ
H 0t H 0t
Aˆ I (t ) e Aˆ s e
or
i i i i
d i Hˆ 0 t Hˆ 0 t Hˆ 0 t Hˆ 0 t
i Aˆ I (t ) i [ Hˆ 0e Aˆ s e e Aˆ s e Hˆ 0 ]
dt
[ Aˆ I (t ), Hˆ 0 ]
Thus every operator behaves as if it would in the Heisenberg representation for a non-
interacting system.
Hˆ 0 t
i
i I (t ) i e s (t )
t t
i ˆ i ˆ
H 0t H 0t
Hˆ 0e s (t ) e i s (t )
t
Since
i s (t ) [ Hˆ 0 Vˆs (t )] s (t )
t
Hˆ 0 t
i i i
Hˆ 0 t Hˆ 0 t
i I (t ) i e s (t ) Hˆ 0e s (t ) e [ Hˆ 0 Vˆ (t )] s (t )
t t
or
i i
Hˆ 0t Hˆ 0t
i I (t ) e Vˆ (t )e I (t ) VˆI (t ) I (t )
t
or
i I (t ) VˆI (t ) I (t )
t
where
i ˆ i ˆ
H 0t H 0t
VˆI (t ) Vˆs (t )e (Schrödinger-like)
We assume that
I (t ) Uˆ I (t , t0 ) I (t0 )
i I (t ) VˆI (t ) I (t )
t
ˆ
i U I (t , t0 ) VˆI (t )Uˆ I (t , t0 )
t
t
i
Uˆ I (t , t0 ) 1 VˆI (t ' )Uˆ I (t ' , t0 )dt '
t0
t t'
i i
Uˆ I (t , t0 ) 1 VˆI (t ' )[1 VˆI (t ' ' )Uˆ I (t", t0 )dt"]dt '
t0 t0
t t t'
i i
1 ( ) VˆI (t ' )dt '( ) 2 dt ' dt"VˆI (t ' )VˆI (t ' ' ) ...
t0 t0 t0
I (t ) Uˆ I (t , t0 ) I (t0 )
where
i i ˆ
Hˆ 0 t H 0t
s (t ) e
I (t ) , or I (t ) e s (t )
and
i ˆ
H 0t
I (t ) e Uˆ s (t , t0 ) s (t0 )
i ˆ i
H 0t Hˆ 0t0
e
Uˆ s (t , t0 )e
I (t0 )
Then we have
i ˆ i
H 0t Hˆ 0 t 0
Uˆ I (t,t 0 ) e
Uˆ s (t,t 0 )e
Hˆ 0 n En n
i
( Ent Emt0 )
n Uˆ I (t , t0 ) m e n Uˆ s (t , t0 ) m
2 2
n Uˆ I (t , t0 ) m n Uˆ s (t , t0 ) m
((Remark))
When
[ Hˆ 0 , Aˆ ] 0 and [ Hˆ 0 , Bˆ ] 0
in general,
2 2
b' Uˆ I (t , t0 ) a ' b' Uˆ s (t , t0 ) a'
Because
_______________________________________________________________________
29.18 Application of Schrödinger and Heisenberg pictures
Simple harmonics:
i
Hˆ t
Uˆ e
i ˆ i ˆ
Ht Ht
Aˆ H Uˆ Aˆ sUˆ e Aˆ s e
1 2 m02 2
H
ˆ pˆ xˆ
2m 2
1
xˆ H xˆ cos t pˆ sin t `
m
and
pˆ H pˆ cos t m xˆ sin t
0 1 0 0 0
1 0 2 0 0
1 0 2 0 3 0
xˆ
2 0 0 3 0 4
0 0 0 4 0
and
((Discussion))
(t ) xˆ (t ) (0) xˆ H (0)
1
(0) xˆ cos t pˆ sin t (0)
m
1
(0) xˆ (0) cos t (0) pˆ (0) sin t
m
(t ) pˆ (t ) (0) pˆ H (0)
(0) pˆ cos t mxˆ sin t (0)
(0) pˆ (0) cos t m (0) xˆ (0) sin t
Suppose that
(1) (0)
1
0 21 2
6
we can calculate the matrix elements (0) xˆ (0) and (0) pˆ (0) as follows.
1
0 1 0 6
1 1 1
2
2 2
(0) xˆ (0) 1 0
6 6 2 6
0
6
0 2 1
6
1 2
(1 2 )
2 3
(2) ( 0)
1
0 1
2
1
1 1 1 0 1 2 1
(0) xˆ (0)
2 2 2 1 0 1 2
2
1
1 1 1 0 1 2
(0) pˆ (0) 0
2 2 2 1 0 1
2
1
(t ) xˆ (t ) cos t
2
and
m
(t ) pˆ (t ) sin t
2
(t ) e iHt /
ˆ 1
0 1
2
e
1 iE 0 t /
0 e iE1t / 1
2
1 e 0
iE t /
iE t /
2 e 1
(t ) e
1 iE 0 t / iE1t /
2
e
iE1t /
1 1
2 2
eiE0 t / iE1t /
iE 0 t /
e
1 1 1
( e i 0 t e i 0 t ) cos 0t
2 2 2
________________________________________________________________________