0.

1 Schrödinger Equation in 2-dimensional system

In HW problem set 5 , we introduced a simpleminded system describing the
ammonia (NH3 ) molecule, consisting of a plane spanned by the 3 hydrogen
atoms and a single N atom being either above or below the plane. Our simple
HS has only 2 dimensions, corresponding to the two possible states of the
N atom relative to the plane of 3 N atoms (above or below). This yields to
basis vectors |1i , |2i which are eigenvectors to the observable measuring the
z-position of the N atom relative to the hydrogen plane, Z:

Z |1i = +z0 |1i

Z |2i = −z0 |2i (1)

We can thus represent any state in the HS as

 
a
|Ψi = a |1i + b |2i =
b

The Hamiltonian from HW problem set 5 in matrix form is given by

 
E -V
Hi,j =
−V E

Applying this to the Schr̈odinger Equation:

      
ȧ E -V a Ea − V b
i~ = =
ḃ −V E b −V a + Eb

Adding up the two equations

i~(ȧ + ḃ) = (Ea + Eb − V a − V b) = (E − V )(a + b)

ċ = (ȧ + ḃ)
(E − V )c
i~ċ = (E − V )c; ċ =
i~
(E−V )t (E−V )t
c(t) = c0 e i~ ; (a + b)(t) = (a + b)(t0 )e i~

Subtracting the two

i~(ȧ − ḃ) = (Ea − V b + V a − Eb) = (E + V )(a − b)

1
 (E+V )t
(a − b)(t) = (a − b)(t0 )e i~

1h (E−V )t (E+V )t
i
a(t) = (a + b)(t0 )e i~ + (a − b)(t0 )e i~
2
1h (E−V )t (E+V )t
i
b(t) = (a + b)(t0 )e i~ − (a − b)(t0 )e i~
2
Et 1
h  −V t Vt
  −V t Vt
i
a(t) = e i~ a0 e i~ + e i~ + b0 e i~ − e i~
2
 
−iEt Vt Vt
a(t) = e ~ a0 cos + ib0 sin
~ ~
 
−iEt Vt Vt
b(t) = e ~ b0 cos + ia0 sin
~ ~
Over all time
a0 cos V~t + ib0 sin V~t Vt Vt
   
−iEt −iEt cos i sin
|Ψi (t) = e ~ =e ~ ~ ~ |Ψi (0)
b0 cos V~t + ia0 sin V~t i sin V~t cos V~t

Given initial conditions, you can determine the state of the system at any
time. Assume, e.g., that at
 
1
t = 0, |Ψi (0) = , a0 = 1, b0 = 0 the atom is on top
0

then we find at
 
π~ −iET 0
t=T = , |Ψi (T ) = e ~ the system will have flipped with
2V i

the nitrogen atom at the bottom.

The system moves up and down periodically. By making a measurement
of Z we get either one of the eigenvalues of this observable
  with probability
1
h1, 2| ψ(t)i. At t = 0 there is a 100% chance of getting . Even though
0
the system is perfect, after t = 0, it changes to a superposition
   of the two
1 0
eigenvectors. The outcome still has to be either or , but the
0 −1
probability of each state has to be calculated.

2
   2
 1  
P rob(+1) =    |Ψ(t)i
0 
 2
 −iEt V t 
= e ~ cos 
~ 
Vt
= cos2
~
Vt
P rob(−1) = sin2
~
The sum of the 2 probabilities always gives 1. If the probability of either is
100%, then that is the state of the system.  
E -V
We can use the Hamiltonian matrix, H = by replacing it in
−V E
iHt
the formal solution |Ψi (t) = e ~ |Ψi (0)
 
it 
E -V  
it

~
−V E E -V
e =1+ ...
~ −V E

Using e(a+b) = ea eb    
−it  E 0  it  0 -V 
=e
~
0 E e ~ −V 0

0 iV t~−1 1 iV 2 t2 ~−2 0
     
−iEt 10
=e ~ 1+ + ...
01 iV t~−1 0 2 0 iV 2 t2 ~−2
Vt Vt
 
−iEt cos i sin
yields the propagator U (t) = e ~ ~ ~
i sin V~t cos V~t
This is of course the same matrix we found earlier, confirming the equivalence
of these two approaches.
If H is Hermitian, then U is Unitary and tells what the state of the system
is at any time t
|Ψi (t) = U (t) |Ψi (0)
 
1
|Ψi (0) =
0

3
 One more way of looking at the Schrodinger equation is to look first for
the eigenvalues
H |φn i = En |φn i
Any state X
|ψi (t) = |φn i hφn | Ψi (t)
n

∂ X
i~ |Ψi = H |φn i hφn | Ψi
∂t n

∂ X
i~ |Ψi = En |φn i hφn | Ψi
∂t n

If Ψ is an eigenfunction, |Ψi = |φn i

∂
i~ |Ψi = En |φn i
∂t
En becomes a number so we can write
iEn t
|Ψi (t) = e ~ |Ψi (0)

This means that the initial state vector at t = 0 simply gets multiplied
with a complex phase at all t and doesn’t change the underlying physical
state (stationary = time-independent solutions to the Schrödinger equation).
The states can be considered stationary since they don’t change over time.
They also tend to be stable since the expectation value of any observable is
constant:
iEn t −iEn t
hOi(t) = hφn (0)| e ~ Oe ~ |φn (0)i = hOi(0)
This means that, for instance, a stationary state cannot radiate electro-
magnetic waves (which requires an oscillating dipole moment, i.e. a time-
dependent expectation value for X).
As shown above, any wavefunction Ψ(t = 0) can be broken down as a
linear superposition of the eigenstates |φn i of H:
X
|ψ(0)i = an (0) |φn i

with
hφn | |ψ(0)i = an (0)

4
 −iEn t
an (t) = e ~ an (0)
The coefficients follow the rule
X −iEn t
|Ψi (t) = an (0)e ~ |φn i

Now for any state we have

X −iEn t
|Ψ(t)i = hφn (0)| Ψ(0)i e ~ |φn (0)i
n

(φn is always taken at t = 0.)

It is easy to show that this gives once again the same result when using
the eigenstates and eigenvalues for H found in HW problem set 5.

0.2 Uncertainty Principle

Recall that when you measure an observable Ω you get ωi , some eigenvalue
of the observable. When you predict a measurement of Ω, you must predict
a range of probabilities associated with each eigenstate of that observable.

Expectation Value
X
hΩi = ωi P (ωi ) = hΨ | Ω | Ψi
i

RMS a.k.a. Standard Deviation a.k.a. σ

X
h(Ω − hΩi)2 i = (ωi − hΩi)2 P (ωi ) = (∆Ω)2
i

Suppose we have another operator defined by

Ω̂ = Ω − hΩiΨ 1

Then  2
(∆Ω)2 = hΨ | Ω̂2 | Ψi = hΩ̂Ψ | Ω̂Ψi = | Ω̂Ψi
 

Let’s conjure up another operator Λ, then

 2
(∆Λ)2 = | Λ̂Ψi
 

5
We can derive a relationship between any two operators by using the following
relation as well as the Schwartz inequality

Ω̂Λ̂ + Λ̂Ω̂ Ω̂Λ̂ − Λ̂Ω̂ 1h i 1h i

Ω̂Λ̂ = + = Ω̂, Λ̂ + Ω̂, Λ̂
2 2 2 + 2
Then,

(∆Ω)2 (∆Λ)2 =?
 2
2 2
(∆Ω) (∆Λ) ≥ hΨ | Ω̂Λ̂ | Ψi
 
 2
2 2
1 1 
(∆Ω) (∆Λ) ≥  hΨ | [Ω̂, Λ̂]+ | Ψi + hΨ | [Ω̂, Λ̂] | Ψi
2 2
1
(∆Ω)2 (∆Λ)2 ≥ |hΨ | [Ω, Λ] | Ψi|2
4

(The last equation follows because the first scalar product in the previous
equation must be real - assuming both operators are Hermitian, you showed
their anti-commutator is Hermitian - while the second term must be purely
imaginary, again as shown before. This in turn means that the absolute
square of the whole impression is the sum of the squares of the two individual
terms, which is of course larger than just one of these terms squared).
Thus we obtain a generalized uncertainty principle
1
∆Ω∆Λ ≥ |hΨ | [Ω, Λ] | Ψi|
2
h ∂
As an example, recall that [x, p] = i~1, p = i ∂x
then...
~
∆p∆x ≥
2
It is important to understand that the actual result for the product ∆Ωψ ∆Λψ
depends not only the operators themselves (and, in particular, their commu-
tator), but also on the state |ψi. In particular, if |ψi is an eigenvector to
either one of the two operators, the product of uncertainties will vanish. On
the other hand, it is also important to realize that, very often, the actual
product of uncertainties can be (much) larger than predicted by the r.h.s.,
since we dropped the anticommutator term and it is only an inequality. Fi-
nally, make sure you understand that the two uncertainties ∆Ωψ and ∆Λψ

6
are the intrinsic uncertainties encoded in the wave functions; they do not
tell us what the outcome is if we measure first Ω and then Λ or vice versa,
but only what the uncertainty of the predicted outcome is if we decide to
measure either of them.