Theoretical Physics

Prof. Ruiz, UNC Asheville, doctorphys on YouTube

Chapter K Notes. The Pauli Equation

K1. Measurement. Our eigenvector analysis in the previous chapter is the key to
understanding measurement in quantum mechanics. An operator stands for some
measurement you make and if the state is an eigenstate of that operator, you get the
eigenvalue. Below we measure a state in the third energy level.

Hψ 3 = E3ψ 3
The H is none other than the left side of the Schrödinger equation, called the
Hamiltonian. Sometimes we prefer to work with the abstract operator symbols, a
hallmark of Heisenberg's approach to quantum mechanics.

PK1 (Practice Problem). Show Hψ 3 = E3ψ 3 for a particle in a one-dimensional

box. Most of the solution is given below.

λ3 2π
3 =L k3 =
2 λ3
2π 3π
k3 = = ψ 3 ( x ) = A sin( k3 x ) = A sin(3π x / L )
(2 L / 3) L

ℏ2 d 2 ℏ2 d 2
H =− 2
+V = −
2m dx 2m dx 2
When you calculate Hψ 3 = E3ψ 3 you will get the energy eigenvalue. Compare your
answer to the third energy level found from the formula we found earlier.

n 2π 2ℏ 2 32 π 2ℏ 2 9π 2ℏ 2
En = E3 = =
Since we know
2mL2 , you must find
2mL 2
2mL2 .

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

Below we make a measurement and find the electron in a spin-up state.

1  1 0  1  1 
σz   =     = 
 0  0 −1 0   0 
But since the actual measured value will be h-bar over two from experiment, we write

ℏ
Sz = σz
2
as the measurement operator for the electron spin-1/2 particles. If you make two
measurements A and B where the state is an eigenstate of each operator, then

Aψ = aψ and Bψ = bψ , with

ABψ = A(bψ ) = bAψ = baψ

BAψ = B ( aψ ) = aBψ = abψ

and all is well. The measurements can be accomplished in either order. Note that when
this happens, the measurement operators commute, which you can see by subtracting
the above pair of equations.

( AB − BA)ψ = (ba − ab)ψ = 0 and [ A, B ] = 0

But remember the operator action of the Pauli matrix σx on our spin-up state?

1   0 1  1  0 
σx   =    0  = 1 
  
0 1 0    
We do not get the same state back. In fact we kicked the state into something else. The
electron now has spin down. So you can disturb the states. When this happens, the
measurement operators do not commute. Before hitting the state with σx the electron
was spin up.
1  1 0  1  1 
σz   =     =  ,
 0  0 −1 0   0 

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

But afterwards, we get
0 1 0  0  0 
σz   =   1  = − 1  .
  
1 0 −1   
Checking the commutator, you find something you already have calculated in the
previous chapter.

0 1  1 0  1 0  0 1 
[σ x , σ z ] =     −  
1 0   0 −1  0 −1 1 0 

0 −1  0 1   0 −1  0 −i 
[σ x , σ z ] =  −
  −1 0  = 2 1 0  = − 2i  i 0  = −2iσ y
 1 0       

Here is your general result from the last chapter.

σ j , σ k  = 2iε jkl σ l

When operators do not commute, the measurements disturb the states.

There is an Uncertainty Relation!

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

K2. Heisenberg Uncertainty Relation.

Werner Heisenberg (1901-1976)

Courtesy School of Mathematics and Statistics
University of St. Andrews, Scotland

Now we consider position and momentum. Recall

from our chapter introducing quantum mechanics
that
∂ 
E → iℏ p → −iℏ∇ .
∂t and

Therefore, we have the momentum operator in one

dimension as
d
p = −iℏ
dx
Does the position operator x commute with the momentum operator p? Let's check.

 d  dψ
x ( pψ ) = x  −iℏ ψ = −iℏx
 dx  dx

 d  dψ
p ( xψ ) =  −iℏ  ( xψ ) = −iℏψ − iℏx
 dx  dx
Subtracting these,

( xp )ψ − ( px)ψ = iℏψ ,

and we obtain for the commutator

[ x, p ] = iℏ
So if you make a position measurement first, then measure the momentum, you kick the
state and no longer have the position you just measured. There is an uncertainty in the
combined position and momentum measurements. The above commutator is an elegant
form for Heisenberg's Uncertainty relation.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

Think of the Schrödinger picture of quantum mechanics as the way with differential
equations. The Heisenberg picture involves operators. Both ways are equivalent
formulations. Thus we credit two physicists as architects of quantum mechanics.

Erwin Schrödinger (1887-1961) Werner Heisenberg (1901-1976)

ℏ 2 d 2ψ Hψ = Eψ
− + Vψ = Eψ
2m dx 2 [ x, p ] = iℏ

Max Born (1882-1970)

Courtesy School of Mathematics and Statistics
University of St. Andrews, Scotland

But we should include Max Born among the architects

of quantum mechanics since he gave us the
interpretation of the wave function in terms of
probability.
P( x) dx = ψ * ( x)ψ ( x) dx
Nobel Prizes for Quantum Mechanics
Heisenberg (1932)
Schrödinger (1933)
Born (1954)

Also Grand pop to Olivia Newton-John

Schrödinger shared his prize with Dirac and Born shared the prize with another scientist
honored for another achievement in physics.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

K3. Angular Momentum. We continue with the powerful language of operators.

The angular momentum operator is found from the definition in classical physics,
  
L = r × p , but now we treat these as operators.
∧ ∧ ∧
i j k
  ∧ ∧ ∧
r× p = x y z = i ( ypz − zp y ) − j ( xp z − zpx ) + k ( xp y − ypx )
px py pz

Note that the order of position and momenta do not matter in the above equation since
the pairs have different spatial dimensions. We could not be cavalier if we had xpx. We
could not switch to pxx. We find

Lx = ypz − zp y ,

Ly = zp x − xp z ,

Lz = xp y − ypx .
Let's find a commutator.

[ Lx , L y ] =  yp z − zp y , zpx − xpz 

[ Lx , L y ] = [ ypz , zpx ] − [ ypz , xp z ] − [ zp y , zpx ] + [ zp y , xp z ]

[ Lx , L y ] = yp x [ p z , z ] − yx[ pz , p z ] − p y p x [ z , z ] + xp y [ z , pz ]

[ Lx , L y ] = yp x ( −iℏ ) − yx ⋅ 0 − p y px ⋅ 0 + xp y (iℏ )

[ Lx , L y ] = yp x ( −iℏ ) + xp y (iℏ ) = iℏ ( xp y − yp x ) = iℏLz

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

PK2 (Practice Problem). Show that along with

[ Lx , Ly ] = iℏLz , we have [ Ly , Lz ] = iℏLx and [ Lz , Lx ] = iℏLy .

Since each angular momentum component commutes with itself, we can write

 L j , Lk  = iℏε jkl Ll

But wait, this looks very similar to

σ j , σ k  = 2iε jkl σ l

This suggests that the Pauli spin matrices are proportional to a type of angular
momentum that has only two values from working with these operators on spinors! But
angular momentum has units of h-bar. So we do two things.

1. Insert the h-bar to get angular momentum units.

2. Get rid of that 2 so the commutation relation looks like the L relations.

Then,

ℏ ℏ ℏ
Sx = σx Sy = σy Sz = σz
2 2 2

 S j , S k  = iℏε jkl Sl

Summary:
  
Orbital Angular Momentum: L = r× p with  L j , Lk  = iℏε jkl Ll

 ℏ 
Intrinsic Angular Momentum for the Electron:
S= σ with  S j , S k  = iℏε jkl Sl
2
  
Total Angular Momentum: J = L+S

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

 K4. General Spin Eigenstates.

We choose our coordinate system as

shown in the figure, following the
convention used in physics. Note that in
math classes, the angle definitions are
often interchanged.

Let the unit vector

ɵ
n point along the r
direction, i.e.,
ɵ
n =ɵ
r . We want to find
the eigenvectors for a spinor along such
an arbitrary direction.

We start with our spin-up and spin-down

cases along the z-direction to be

Image Adapted from 1  0

Mauro Giachero, Wikimedia ↑=   and
↓=   ,
Creative Commons License
Attribution-Share Alike
0 1 

which are eigenvectors for σ z . We

work with Pauli matrices because the math is faster. you can always put the h-bar in
and 1/2 later to get the actual spin.

∧ ∧ ∧ ∧ ∧
n ≡ r = sin θ cos φ i + sin θ sin φ j + cos θ k
∧  0 1   0 −i  1 0 
n⋅ σ = sin θ cos φ   + sin θ sin φ i 0  + cos θ  0 −1
1 0     
∧   cos θ sin θ ( cos φ − i sin φ ) 
n⋅ σ =  
 sin θ ( cos φ + i sin φ ) − cos θ 
∧   cos θ sin θ e − iφ 
n⋅ σ =  iφ 
 sin θ e − cos θ 

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

Eigenvalues. We now set up the eigenvalue problem.

∧   c1   cos θ sin θ e − iφ   c1   c1 
n⋅ σ   =  iφ   = λ c 
 c2  sin θ e − cos θ   c2   2
For the solution of the eigenvalues, the following determinant must vanish.

cos θ − λ sin θ e − iφ
iφ
=0
sin θ e − cos θ − λ
The result for the vanishing determinants

−(cos 2 θ − λ 2 ) − sin 2 θ = 0

− cos 2 θ + λ 2 − sin 2 θ = 0

λ 2 = cos 2 θ + sin 2 θ = 1
λ = ±1
Eigenvectors. We proceed to find the eigenvectors. We will give the eigenvector that

goes with λ = +1 the name χ

↑ and for λ = −1 we will use
χ
↓ . We use a and b
as temporary parameters for the components we use. For the first case we find

 cos θ sin θ e − iφ   a   a 
 iφ  = 
sin θ e − cos θ   b   b 

 a cos θ + b sin θ e − iφ   a 
 iφ = 
 a sin θ e − b cos θ  b 

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

For the upper component
a cos θ + b sin θ e − iφ = a

b sin θ e − iφ = a (1 − cos θ )

b (1 − cos θ ) iφ
= e
a sin θ
Do you remember an elegant identity involving a half angle and/or double angle from
trig you studied some time ago? It will unlock the secret to spinor rotations? Recall that
we derived these two identities earlier in our course.

cos(α + β ) = cos α cos β − sin α sin β

sin(α + β ) = cos α sin β + sin α cos β

From the above, we obtain.

sin 2θ = 2sin θ cos θ and cos 2θ = cos 2 θ − sin 2 θ

θ θ θ θ
sin θ = 2 sin cos and cos θ = cos 2 − sin 2
2 2 2 2
Summary:
b (1 − cos θ ) iφ
= e
a sin θ
θ θ
sin θ = 2 sin cos
2 2

θ θ
cos θ = cos 2 − sin 2
2 2
But wait!
 θ θ
1 − cos θ = 1 − cos 2  + sin 2
 2 2

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

  θ θ θ
1 − cos θ = sin 2  + sin 2 = 2 sin 2
 2 2 2
Now we are ready. The result is
θ
2 sin 2
b (1 − cos θ ) iφ 2 e iφ
= e =
a sin θ θ θ
2 sin cos
2 2
θ
sin
b 2 eiφ
=
a cos θ
2
We can pick
θ θ
a = cos and
b = sin e iφ since a * a + b*b = 1
2 2
We are normalized.

The eigenvector is

 θ 
 cos 
χ↑ =  2

θ
 e iφ sin 
 2 

PK3 (Practice Problem). Use the lower component equation in

 a cos θ + b sin θ e − iφ   a 
 iφ = 
 a sin θ e − b cos θ   b 
and normalization to obtain the same result.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

PK4 (Practice Problem). Show that the eigenvector for λ = −1 is

 − iφ θ
 e sin 
χ↓ =  2

θ
 − cos  .
 2 
Summary:

 θ   − iφ θ
 cos   e sin 
χ↑ =  2 χ↓ =  2
 
 e iφ sin θ   − cos θ 
 2   2 

†  θ θ
The dual of χ↓ is defined as
 χ ↓  =  e + iφ sin − cos 
 2 2

 θ 
cos
†  θ θ  2 
 χ ↓  χ ↑ =  e + iφ sin − cos   =0
Then  2 2   iφ θ
e sin
 2 
The eigenstates are orthogonal.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

K5. What Does It All Mean?

 θ   − iφ θ
 cos   e sin 
χ↑ =  2 2
 χ↓ =  
θ
 e iφ sin   cos θ 
 2   2 
Whenever you feel this question coming on, just
look at friendly cases. Set θ = 0°. Then,

1  0
χ ↑ z − axis =  χ ↓ z − axis = 
0 1 
We know what this means. The left spinor is an electron with spin aligned up. The right
spinor is a spin state where the spin is aligned down. Now check out θ = 90° with φ = 0°,
which is along our x-axis.
1 1
χ ↑ x − axis = 1
2  
This is an eigenstate of a spin measurement along the x-axis where the spin is up
relative to the x-axis. But it is expressed in terms of a mixture or superposition of states
relative to the z-axis.

1 1  1  0 
χ ↑ x − axis = 0  +  
2   2 1 

1 1
χ ↑ x − axis = χ ↑ z − axis + χ ↓ z − axis
2 2
So for the spin up along the x-axis, when measured with respect to the z-axis you can
get either up or down. In fact the probability for each is 1/2.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

What about when θ = 180°?

 θ 
 cos 
χ↑ =  2

 e iφ sin θ 
 2 
becomes

0 1  iφ 0
χ ↑ (θ = 180°) =  iφ 
= 0   +e 1 
e  0   
χ ↑− z − axis = c1 χ ↑ z − axis + c2 χ ↓ z − axis

The state for you, standing on your head, is up since it is your eigenstate. But for the
person oriented along the positive z-axis, the spin is flipped. We get this result from the
probabilities. The probability for regular z-axis spin up is c1*c1 = 0..For spin down we
find

c2 * c2 = e − iφ e iφ = 1 .
What about θ = 360°?
 −1
χ ↑ (θ = 360°) =  
0
Well, we get 100% probability that we are spin up again since

c1 * c1 = ( −1)( −1) = 1
But we DO NOT get the exact same thing back due to a phase factor. You have to go
two complete 360° rotations for that due to the half angles.

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

 Courtesy Council on Science and Technology, Princeton University

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

K6. The Pauli Equation

We will assume time-independent potentials in this section and thus work with the time-
independent form of the Schrödinger equation. We also consider spin-1/2 particles such
as electrons.

To include spin we promote our wave function ψ to

ψ 1 
ψ = 
ψ 2 
Replace the Schrödinger equation

ℏ2 2
− ∇ ψ + Vψ = Eψ
2m
with

ℏ 2 2 ψ 1  ψ 1  ψ 1 
− ∇   +V ψ  = E ψ  ,
2m ψ 2   2  2
where the potential is a matrix potential

V11 V12 
V = .
V
 21 V22 

The H operator

ℏ2 2
H =− ∇ +V becomes
2m

ℏ 2 1 0  2 V11 V12 
H =−   ∇ + 
2m  0 1  V21 V22  .

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

 Wolfgang Pauli (1900-1958)
Courtesy School of Mathematics and Statistics
University of St. Andrews, Scotland

The Pauli Effect. A joke, but taken seriously by

Pauli and some others. Background: Physicists
are either theorists or experimentalists for many
years now. You either work with the math and
calculations to explain observed phenomena or
spend your career making the measurements and
gathering the data. However, there are rare
exceptions like the great Enrico Fermi.

Theorists have an image of being inept in lab.

Instead, they feel at home with mathematical
physics. So they might break things in lab, ruin
experiments, or do something not too intelligent,
even putting themselves and others in danger.

So the Pauli effect is this - if a theorist walks into a lab where an experimentalist is
working, the experimental equipment will break or malfunction by just the mere
presence of the theorist. It is said the greater the theorist, the greater the destruction.

A famous story is the mysterious failure of an experiment in a lab in Germany. But Pauli
was not even present. However, later they discovered Pauli was at the train station in
that town at the time of the malfunction.

Pauli and the famous psychiatrist Carl Jung, a student who broke away from Freud,
analyzed this in terms of Jung's concept of synchronicity, where coincidences are
believed to happen for a reason with some underlying connection.

A Famous Pauli Put Down. "Not only is it not right, it's not even wrong!"

Pauli Tough on a Student in Class. One day in lecture Pauli said something was
trivial. The student did not see it, so the student asked Pauli for an explanation. Pauli
left the room and came back a few minutes later. On his return, Pauli said "It is trivial!"
Apparently Pauli left the lecture hall to check his comment out in his office and satisfied
himself that it was indeed an easy calculation. This was not quite the answer the
student was hoping for.

Max Born Comments on His Assistant Pauli. "Since the time when he was assistant
in Göttingen, I knew he was a genius, comparable only to Einstein himself." (from
Quips, Quotes and Quanta: An Anecdotal History of Physics by Anton Z. Capri).

Michael J. Ruiz, Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License