Quantum Mechanics

Jorn Hoekstra

February 2014
Contents

1 Essential Quantum Physics 4

1.1 Bra’s and Kets . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.1 Expectation Values . . . . . . . . . . . . . . . . . . . . 9
1.2.2 Linear Operators . . . . . . . . . . . . . . . . . . . . . 9
1.2.3 Hermitian Adjoints . . . . . . . . . . . . . . . . . . . . 10
1.2.4 Commutators . . . . . . . . . . . . . . . . . . . . . . . 11
1.2.5 Heisenberg . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.2.6 Eigenvectors and -values . . . . . . . . . . . . . . . . . 16
1.2.7 Unitary Operators . . . . . . . . . . . . . . . . . . . . 19
1.2.8 Matrix or Continuous representation . . . . . . . . . . 20

2 Particles in Bound States 22

2.1 Energy Wells . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.1.1 Trapping in Potential Wells . . . . . . . . . . . . . . . 23
2.1.2 Particles in Infinite Square Potential Wells . . . . . . . 25
2.2 Particles and Potential Steps . . . . . . . . . . . . . . . . . . . 29
2.2.1 Assuming the particle has plenty of energy . . . . . . . 29
2.2.2 Assuming the particle doesn’t have enough energy . . . 33
2.3 Particles and Potential Barriers . . . . . . . . . . . . . . . . . 36
2.3.1 Through the barrier with E > V0 . . . . . . . . . . . . 36
2.3.2 Through the barrier even when E < V0 . . . . . . . . . 38
2.4 Particles Unbound . . . . . . . . . . . . . . . . . . . . . . . . 40
2.4.1 Getting particles with wave packets . . . . . . . . . . . 41
2.5 Gaussian Example . . . . . . . . . . . . . . . . . . . . . . . . 43
2.6 Harmonic Oscillators . . . . . . . . . . . . . . . . . . . . . . . 45
2.6.1 Creation and Annihilation . . . . . . . . . . . . . . . . 46
2.6.2 Finding eigenstates . . . . . . . . . . . . . . . . . . . . 47

2
 2.6.3 Harmonic Oscillators Operators as Matrices . . . . . . 54

A Proofs and Calculations 56

A.1 Expectation value Dice Example . . . . . . . . . . . . . . . . . 56
A.2 Rewriting a commutator . . . . . . . . . . . . . . . . . . . . . 59
A.3 Energy equation through Schrödinger . . . . . . . . . . . . . . 59
A.4 Normalizing a Wave Function . . . . . . . . . . . . . . . . . . 61
A.5 Solving differential equation . . . . . . . . . . . . . . . . . . . 62
A.6 Calculating the incident current density . . . . . . . . . . . . . 63
A.7 Normalizing a Gaussian . . . . . . . . . . . . . . . . . . . . . 64
A.8 Commutator of Raising- and Lowering Operator . . . . . . . . 66
A.9 First Order Derivative of a Wave Function . . . . . . . . . . . 67
A.10 Three-Dimensional Angular Momentum Commutator . . . . . 68
A.11 Lowering Operator on βmin . . . . . . . . . . . . . . . . . . . . 69
A.12 Checking Angular Momentum Operators . . . . . . . . . . . . 70

B List of Definite Integrals 71

3
Chapter 1

Essential Quantum Physics

Here the hard parts of the needed background information will be discussed
by example of two dice and their probabilities, and many more examples and
calculations.

1.1 Bra’s and Kets

Rolling to ordinary dice and calculating in how many ways the number can
be found gives the following table:

Sum of Dice Relative Probability

2 1
3 2
4 3
5 4
6 5
7 6
8 5
9 4
10 3
11 2
12 1

4
In a probability vector this becomes:

 
1
2
 
3
 
4
 
5
 
6
 
5
 
4
 
3
 
2
1

Which, controverted to probability amplitudes becomes:

√ 
√1
 2
√ 
 3
√ 
 4
√ 
√5
 
√6
 
√5
 
 
 4
√ 
 3
√ 
 2
√
1

With this te actual probability of getting any outcome becomes:

√ √ √ √ √ √ √ √ √ √ √ 2
1 + 2 + 3 + 4 + 5 + 6 + 5 + 4 + 3 + 2 + 1 = 36

5
 √
This must be 100% (or in this case 1), so everything must be divided by 36.
The probability vector thus becomes:
p  √   
1/36 1/6 1/6
√
√2/6
p
 2/36 
 √2/6
 √3/6
p   
p3/36 
 
 3/6
√     
 
 4/36  4/6 √2/6 
p   √ 
 5/36  5/6
 
5/6

p   √  √ 
 6/36 =  6/6 =  6/6
  
p  √  √ 
 5/36  5/6  5/6

p   √
 4/36  √2/6 
  
p   √ 4/6  
p3/36 
  3/6   3/6
 √  √ 
p2/36 √2/6  2/6

1/36 1/6 1/6

This can be called a ket, |ψ⟩:

 
1/6
√2/6
√ 
 3/6
 
 2/6 
√ 
√5/6
 
|ψ⟩ = √6/6
 
 5/6
 
√2/6 
 
 
 3/6
√ 
 2/6
1/6

A bra can be made with the dot-product and complex conjugate of the ket:

⟨ψ| = |ψ⟩T ∗
h √ √ √ √ √ √ √ i
⟨ψ| = 1 2 3 2 5 6 5 2 3 2 1
6 6 6 6 6 6 6 6 6 6 6

These notations have these names for the bra-ket notation (sounds as bracket).
A bra-ket can be noted as: ⟨ψ|ψ⟩. To check if the total probability is possible

6
the bra-ket can be calculated:
 
1/6
√2/6
√ 
 3/6
 
 2/6 
√ 
i √5/6
 
h √ √ √ √ √ √ √
⟨ψ|ψ⟩ = 1 2 3 2 5 6 5 2 3 2 1 · √6/6
 
6 6 6 6 6 6 6 6 6 6 6
 5/6
 
√2/6 
 
 
 3/6
√ 
 2/6
1/6
1 2 3 4 5 6 5 4 3 2 1


= 36 + 36
+ 36
+ 36
+ 36
+ 36
+ 36
+ 36
+ 36
+ 36 + 36
=1
We want working solutions, so in general ⟨ψ|ψ⟩ = 1. Furthermore:
ˆ |ψ⟩ N
P
i=1 |ϕ⟩ ⟨ϕ|ψ⟩
ˆ |⟨ψ|ϕ⟩|2 ≤ ⟨ψ|ψ⟩ ⟨ϕ|ϕ⟩ (Schwartz inequality)
2
ˆ |A · B| ≤ |A| + |B|2 2

2 kets are orthogonal if:

⟨ψ|ϕ⟩ = 0
2 kets are orthonormal if:
⟨ψ|ϕ⟩ = 0
⟨ψ|ψ⟩ = 1
⟨ϕ|ϕ⟩ = 1

1.2 Operators
An operator transforms an old bra or ket into a new one. For example,
operator A transforms them like this:
A |ψ⟩ = |ψ ′ ⟩
A ⟨ψ| = ⟨ψ ′ |

7
Five important operators are:

ˆ Hamiltonian (H), applying the Hamiltonian operator gives you E,

the enrgy of the particle; E is a scalar quantity:

H |ψ⟩ = E |ψ⟩

ˆ Unity or identity (I), the unity or identity operator leaves kets (and
bra’s) unchanged:
I |ψ⟩ = |ψ⟩

ˆ Gradient (∇), gradient operator works like this:

δ δ δ
∇ |ψ⟩ = |ψ⟩ i + |ψ⟩ j + |ψ⟩ k
δx δy δz

ˆ Linear momentum (P), the linear momentum operator looks like

this (in quantum mechanics):

P |ψ⟩ = −ih̄∇ |ψ⟩

ˆ Laplacian (∇2 ), you use the Laplacian operator to create the energy-
finding Hamiltonian operator:

δ2 δ2 δ2
∇2 |ψ⟩ = ∇ · ∇ |ψ⟩ = |ψ⟩ + |ψ⟩ + |ψ⟩
δx2 δy 2 δz 2

Operators are dependent of order, therefore for operators A and B:

AB ̸= BA

And an operator (A) is said to be linear if:

A(c1 |ψ⟩ + c2 |ψ⟩) = c1 A |ψ⟩ + c2 A |ψ⟩ (1.1)

8
1.2.1 Expectation Values
De expected value, or expectation value of A is:

expectation value = ⟨ψ| A |ψ⟩

In the dice example the ket and bra store the probabilities, so the created
operator (Roll operator, R) must store the dice values (2 through 12) for
each probability. Therefore, the operator R looks like this:
 
2 0 0 0 0 0 0 0 0 0 0
 0 3 0 0 0 0 0 0 0 0 0 
 
 0 0 4 0 0 0 0 0 0 0 0 
 
 0 0 0 5 0 0 0 0 0 0 0 
 
 0 0 0 0 6 0 0 0 0 0 0 
 
R=  0 0 0 0 0 7 0 0 0 0 0 

 0 0 0 0 0 0 8 0 0 0 0 
 
 0 0 0 0 0 0 0 9 0 0 0 
 
 0 0 0 0 0 0 0 0 10 0 0 
 
 0 0 0 0 0 0 0 0 0 11 0 
0 0 0 0 0 0 0 0 0 0 12

Doing the calculations1 we get:

⟨ψ|R|ψ⟩ = ⟨R⟩ = 7

In this answer ⟨R⟩ is called the common notation.

1.2.2 Linear Operators

Recall the condition for a operator to be linear (1.1):

A(c1 |ψ⟩ + c2 |ψ⟩) = c1 A |ψ⟩ + c2 A |ψ⟩

Then two ground rules for bra’s and kets must be called:

1. If we take the product of ⟨χ|, with the ket c |ψ⟩, where c is a complex
number:
⟨χ|c|ψ⟩ = c ⟨χ|ψ⟩
1
See proof A.1

9
 2. If we take the product of the bra ⟨χ|, with the sum of two kets:

⟨χ| (|ψ1 ⟩ + |ψ2 ⟩) = ⟨χ|ψ1 ⟩ + ⟨χ|ψ2 ⟩

To find if a operator is linear (for example the operator |ϕ⟩ ⟨χ|) one must do
the following:

1. Apply a linear combination of kets (with c1 and c2 as complex numbers)

|ϕ⟩ ⟨χ| (c1 |ψ1 ⟩ + c2 |ψ2 ⟩)

2. With the rule for the product of a bra with a sum of two kets, you can
say:

|ϕ⟩ ⟨χ| (c1 |ψ1 ⟩ + c2 |ψ2 ⟩) = |ϕ⟩ ⟨χ| c1 |ψ1 ⟩ + |ϕ⟩ ⟨χ| c2 |ψ2 ⟩

3. Which can be rewritten as:

|ϕ⟩ ⟨χ| (c1 |ψ1 ⟩ + c2 |ψ2 ⟩) = c1 |ϕ⟩ ⟨χ|ψ1 ⟩ + c2 |ϕ⟩ ⟨χ|ψ2 ⟩

When |ϕ⟩ ⟨χ| is replaced by A this is:

A(c1 |ψ1 ⟩ + c2 |ψ2 ⟩) = c1 A |ψ1 ⟩ + c2 A |ψ2 ⟩

’This is the exact rule for a linear operator, therefore |ϕ⟩ ⟨χ| is a linear
operator.

1.2.3 Hermitian Adjoints

The Hermitian adjoint (also the adjoint or Hermitian conjugate) of an
operator A is denoted A† . To find the Hermitian adjoint follow these steps:

1. Replace complex constants with their complex conjugates:

a† = a∗

2. Replace kets with their corresponding bras and visa versa.

10
 3. Replace operators with their Hermitian adjoints. In quantum mechanics
operators that are equal to their Hermitian adjoints are called Hermitian
operators, so for these operators:

A† = A

These are special as the representative matrix might be diagonalized

(all non-zero number on diagonal) and the outcome of the expectation
value is a real number (not complex).
4. Write your final equation

As an example we take the following ⟨ψ| A |ϕ⟩ (with A as a non-Hermitian

operator). With the steps above we find the Hermitian adjoint of this as
follows:

1. -
2. ⟨ψ| → |ψ⟩
|ϕ⟩ → ⟨ϕ|
3. A → A†
4. ⟨ϕ|A† |ψ⟩

Finally a few relationships concerning Hermitian adjoints are to be known:

ˆ (aA)† = a∗ A†

†
ˆ A† = A

ˆ (A + B)† = A† + B †

ˆ (AB)† = B † A†

ˆ (AB |ψ⟩)† = ⟨ψ| B † A†

1.2.4 Commutators
The measure of how different it is to apply operator A and then B, is called
the operators commutator. This is how to define the commutator of operator
A and B:
[A, B] = AB − BA

11
Two operators commuto with each other if their commutaotr is equal to zero:

[A, B] = 0

Of course every operator commutes with itself:

[A, A] = 0

And it’s easy to show that the commutator of A, B is the negative of the
commutator of B, A:
[A, B] = −[B, A]
It’s also true that commutators are linear:

[A, B + C + D + ...] = [A, B] + [A, C] + [A, D] + ...

And the Hermitian adjoint of a commutator works this way:

[A, B]† = B † , A†
 

You can also find the anticommutator, {A, B}:

{A, B} = AB + BA

Example: What can you say about the Hermitian adjoint of the commutator
of two Hermitian operators (A and B):

[A, B]† = (AB − BA)†

(AB − BA)† = B † A† − A† B †
B † A† − A† B † = BA − AB
BA − AB = −[A, B]
[A, B]† = −[A, B]

When you take the Hermitian adjoint of an expression and get the same
thing back with a negative sign in front of it, the expression is called anti-
Hermitian, so the commutator of two Hermitian operators is anti-Hermitian.
(By the way, the expectation value of an anti-Hermitian operator is guaranteed
to be completely imaginary.)

12
1.2.5 Heisenberg
Here will be given a (relatively) short calculation of the Heisenberg uncertainty
principle starting from scratch.
The uncertainty in a measurement of the Hermitian operator A is formally
given by:

1/2
∆A = A2 − ⟨A⟩2



So for hermitian operator B:

1/2
B 2 − ⟨B⟩2


∆B =

If you apply ∆A and ∆B as operators (not uncertainties) this will give you
measurement values like this:

∆A = A − ⟨A⟩
∆B = B − ⟨B⟩

Like any operator, ∆A and ∆B transform kets:

∆A |ψ⟩ = |χ⟩
∆B |ψ⟩ = |ϕ⟩

With the (earlier given) Schwartz inequality we know:

⟨χ|χ⟩ ⟨ϕ|ϕ⟩ ≥ |⟨χ|ϕ⟩|2

Because ∆A and ∆B are Hermitian ⟨χ|χ⟩ is equal to ⟨ψ|∆A2 |ψ⟩ and ⟨ϕ|ϕ⟩
equal to ⟨ψ|∆B 2 |ψ⟩. Because ∆A† = ∆A (definition of a Hermitian operator),
you can see that:

⟨χ|χ⟩ = ⟨ψ|∆A† ∆A|ψ⟩ = ⟨ψ|∆A2 |ψ⟩

So

⟨χ|χ⟩ = ∆A2



⟨ϕ|ϕ⟩ = ∆B 2



And the Schwartz inequality becomes:

∆A2 ∆B 2 ≥ |⟨∆A∆B⟩|2




13
∆A∆B can be rewritten to:
1 1
∆A∆B = [∆A, ∆B] + {∆A, ∆B}
2 2
1 1
= (∆A∆B − ∆B∆A) + (∆A∆B + ∆B∆A)
2 2
1
= 2 ∆A∆B − 2 ∆B∆A + 2 ∆A∆B + 12 ∆B∆A
1 1

= ∆A∆B
So:
1 1
∆A∆B = [∆A, ∆B] + {∆A, ∆B}
2 2
2
Which on its turn can be rewritten to :
1 1
∆A∆B = [A, B] + {∆A, ∆B}
2 2
Having in mind that the expectation value of an anti-Hermitian (anti-Hermitian
is the commutator of two Hermitian operators, [A, B]) is always imaginary,
and the expectation value of a Hermitian (as proven {∆A, ∆B} is Hermitian)
always real, we can view this as a sum of real ({∆A, ∆B}) and imaginary
([A, B]) parts:
1 1
|⟨∆A∆B⟩|2 = |⟨[A, B]⟩|2 + |{∆A, ∆B}|2
4 4
Because the right term is always positive or zero (because it must be a real
probability), we can state:
1
|⟨∆A∆B⟩|2 ≥ |⟨[A, B]⟩|2
4
Combining this with the earlier found Schwartz inequality for this case gives
you:
 1
∆A2 ∆B 2 ≥ |⟨[A, B]⟩|2



4
Next we want to remove the expectation value brackets and the fact that
∆A and ∆B are squared. So how to get ⟨∆A2 ⟩ ⟨∆B 2 ⟩ to ∆A∆B?
Both can be written as3 :
∆A2 = A2 + ⟨A⟩2 − 2A ⟨A⟩




2
See proof A.2
3
Use ∆A = A − ⟨A⟩ and ∆B = B − ⟨B⟩

14
By taking the expectation value of the last term in this equation, you get:
∆A2 = A2 + ⟨A⟩2 − 2A ⟨A⟩ = A2 − ⟨A⟩2





By squaring the earlier stated definition of ∆A 4 you get the following:

∆A2 = A2 − ⟨A⟩2



So we can conclude that:

∆A2 = ∆A2



This means that ⟨∆A2 ⟩ ⟨∆B 2 ⟩ ≥ 41 |⟨[A, B]⟩|2 becomes

1
∆A2 ∆B 2 ≥ |⟨[A, B]⟩|2
4
Which at last leads to:
1
∆A∆B ≥ |⟨[A, B]⟩|
2
To make the Heisenberg uncertainty relation we look at the momentum
operator5 , which in the x direction is:
δ
Px = −ih̄
δx
The commutator of the Px and the X operator (which just gives the position)
is:
[X, Px ] = ih̄
So from the last equation we get:
1
∆A∆B ≥ |⟨[A, B]⟩|
2
1
∆x∆px ≥ |ih̄|
2
1
∆x∆px ≥ h̄
2
h̄
∆x∆px ≥
2
Note that here ∆x and px are the uncertainties not operators.

 1/2
4 2
∆A = A2 − ⟨A⟩
5
P = −ih̄∇

15
1.2.6 Eigenvectors and -values
|ψ⟩ is the eigenvector of operator A if:
ˆ The number a is a complex constant
ˆ A |ψ⟩ = a |ψ⟩
So applying A to one of its eigenvectors returns you the eigenvector multiplied
by that eigenvectors eigenvalue (a).
The eigenvalues of Hermitian operators are real numbers, and their eigenvectors
are orthogonal.
When we look at the R operator from our dice example again:
 
2 0 0 0 0 0 0 0 0 0 0
 0 3 0 0 0 0 0 0 0 0 0 
 
 0 0 4 0 0 0 0 0 0 0 0 
 
 0 0 0 5 0 0 0 0 0 0 0 
 
 0 0 0 0 6 0 0 0 0 0 0 
 
R=  0 0 0 0 0 7 0 0 0 0 0 

 0 0 0 0 0 0 8 0 0 0 0 
 
 0 0 0 0 0 0 0 9 0 0 0 
 
 0 0 0 0 0 0 0 0 10 0 0 
 
 0 0 0 0 0 0 0 0 0 11 0 
0 0 0 0 0 0 0 0 0 0 12

The R operator works in 11-dimensional space and is Hermitian, so there

will be 11 orthogonal eigenvectors and 11 corresponding eigenvalues. The
eigenvalues of operator R would look like this:
     
1 0 0
0 1 0
     
0 0 0
     
0 0 0
     
0 0 0
     
ξ1 = 0
  ξ2 = 0
  ... 0
ξ11 =  
0 0 0
     
0 0 0
     
0 0 0
     
0 0 0
0 0 1

16
The eigenvalues are the numbers you get when you apply R to an eigenvector,
because all eigenvectors are unit vectors the eigenvalues are the numbers on
the diagonal of the R matrix, for example:

R |ξ1 ⟩ = 2 |ξ1 ⟩
R |ξ2 ⟩ = 3 |ξ2 ⟩
...
R |ξ11 ⟩ = 12 |ξ11 ⟩

When two or more eigenvalues are the same, the eigenvalue is said to be
degenerate.
Cool thing: When two Hermitian operators, A and B, commute, and if A
doesn’t have any degenerate eigenvalues, then each eigenvector of A is also
an eigenvector of B.

To find the eigenvectors and -values of an operator you have to first rewrite
the equation to a form equal to zero, basically:

A |ψ⟩ = a |ψ⟩
A |ψ⟩ = aI |ψ⟩
(A − aI) |ψ⟩ = 0

I is here the identity matrix in the form of:

 
1 0 0 0 0 0 0 0 0 0 ...
 0 1 0 0 0 0 0 0 0 0 ... 
 
 0 0 1 0 0 0 0 0 0 0 ... 
 
I= 0 0 0 1 0 0 0 0 0 0 ...
 

 0 0 0 0 1 0 0 0 0 0 ... 
 
 0 0 0 0 0 1 0 0 0 0 ... 
 
..
.

Important to know is that the solution to (A − aI) |ψ⟩ = 0 only exist if the
determinant of the matrix A − aI is 0:

det(A − aI)

17
 Example: Find the eigenvalues and eigenvectors of operator A:
 
−1 −1
A=
2 −4
First we want to convert it to the form A − aI:
 
−1 − a −1
A − aI =
2 −4 − a
Next find the determinant:
det(A − aI) = (−1 − a)(−4 − a) + 2
det(A − aI) = a2 + 5a + 6
det(A − aI) = (a + 2)(a + 3)
Given is that det(A − aI) = 0 so we have a1 = −2 and a2 = −3. First fill a1
in:  
1 −1
A − a1 I =
2 −2
So you have:     
1 −1 ψ1 0
=
2 −2 ψ2 0
Because every row of the matrix must be true, you know ψ1 = ψ2 . And that
means that, up to an arbitrary constant, the eigenvector corresponding to a1
is:  
1
ξ1 = c
1
Drop the arbitrary constant and just write it as a matrix:
 
1
ξ1 =
1
Then for a2 = −3 you get:
 
2 −1
A − a2 I =
2 −1
Then:     
2 −1 ψ1 0
=
2 −1 ψ2 0
So 2ψ1 − ψ2 = 0, thus ψ1 = ψ2 ÷ 2. Which means that the eigenvector is:
   
1 1
ξ2 = c =
2 2

18
1.2.7 Unitary Operators
Applying the inverse of an operator undoes the work the operators did:

A−1 A = AA−1 = I

Qunatum physics calculations are sometimes limited to working with unitary

operators, U , where the operator’s inverse is equal to its adjoint, U −1 = U †
6
. This gives you the following equation:

U †U = U U † = I

The product of two unitary operators, U and V , is also unitary because:

(U V )(U V )† = (U V )(V † U † ) = U (V V † )U † = U U † = I

When you use unitary operators, kets and bras transform this way:
ˆ |ψ ′ ⟩ = U |ψ⟩
ˆ ⟨ψ ′ | = ⟨ψ| U

You can transform other operators using unitary operators:

A′ = U AU †

The preceding equation means:

ˆ |ψ⟩ = U † |ψ ′ ⟩
ˆ ⟨ψ| = ⟨ψ ′ | U
ˆ A = U † A′ U

Remember the following properties of unitary operators:

ˆ If an operator is Hermitian, then its unitary transformed version, A =
U AU † , is also Hermitian.
ˆ The eigenvalues of A and its unitary transformed version, A′ = U AU † ,
are the same.
ˆ Commutators that are equal to complex numbers are unchanged by
unitary transformations: [A′ , B ′ ] = [A, B].
6
The adjoint can be found with the dot-product and the complex conjugate, AT ∗ = A†

19
1.2.8 Matrix or Continuous representation
Setting the determinant equal to zero is a good way to find the eigenvalues
of an operator. One commonly used one is the Hamiltonian operator that
finds the energy of a system7 . The E is an eigenvalue of the H operator so
the earlier given solution is valid, this would however lead to:
 
H11 − E H12 H13 H14 ...
 H21
 H22 − E H23 H24 ... 

 H31
det  H 32 H 33 − E H 34 ... =0

 H41
 H 42 H 43 H44 − E ... 

..
.

Which is solvable, but not when the number of eigenstates is infinite. Then
you can no longer use a discrete basis for your operators, bras and kets - you
use a continuous basis.
In the continuous basis summations become integrals, for example:
∞
X Z
|ϕn ⟩ ⟨ϕn | = I → dϕ |ϕ⟩ ⟨ϕ| = I
n=1

And of course with the idea of the identity operator, every ket (|ψ⟩) can be
expanded in a basis of other kets (|ϕ⟩):
Z
|ψ⟩ = dϕ |ϕ⟩ ⟨ϕ|ψ⟩

If you look at the position operator R, in a continuous basis. Applying this

operator gives you r, the position vector:

R |ψ⟩ = r |ψ⟩

In this equation, applying the position operator to a state vector returns the
locations, r, that a particle may be found at. You can expand any ket in the
position basis like this:
Z
|ψ⟩ = d3 r |r⟩ ⟨r|ψ⟩

7
H |ψ⟩ = E |ψ⟩

20
And this becomes: Z
|ψ⟩ = d3 r ψ(r) |r⟩

Important - ψ(r) = ⟨r|ψ⟩ is the wave function for the state vector |ψ⟩, it’s
the ket’s representation in the position basis. Or in common terms, it’s just
a function where the quantity |ψ(r)|2 d3 r represents the probability that the
particle will be found in the region d3 r at r.

In wave mechanics you put bras and kets in the position basis. Therefore
you go from talking about |ψ⟩ to ⟨r|ψ⟩, which equals ψ(r). So H |ψ⟩ = E |ψ⟩
becomes the following:
⟨r| H |ψ⟩ = Eψ(r)
The H operator is the total energy of the system, kinetic (p2 /2m) plus
potential (V (r))8 so you get the following:

p2
H= + V (r)
2m
We know that the momentum operator is:
δ δ δ
H = −ih̄ |ψ⟩ i − ih̄ |ψ⟩ j − ih̄ |ψ⟩ k
δx δy δz
Substituting this operator for p gives:

−h̄2 δ 2 δ2 δ2
 
H= + + + V (r)
2m δx2 δy 2 δz 2

With the Laplacian operator:

δ2 δ2 δ2
∇2 |ψ⟩ = ∇ · ∇ |ψ⟩ = |ψ⟩ + |ψ⟩ + |ψ⟩
δx2 δy 2 δz 2

You can rewrite the equation to the following (called the Schrödinger equation):

−h̄2 2
H ψ(r) = ∇ ψ(r) + V (r) ψ(r) = E ψ(r) (1.2)
2m

8
Etotal = Ekinetic + Epotential = Ek + Ep is the basis of Newtonian physics

21
Chapter 2

Particles in Bound States

Here the best part of quantum physics (to some) will be discussed, the finding
of energy levels and wave functions for trapped particles.

2.1 Energy Wells

A square well is a potential (that is, a potential energy well) that forms a
square shape. An example of this is Figure 2.1.

a x
Figure 2.1: A square well

22
 This well is called, an infinite square well as where x < 0 and x > a V
goes to infinity. Or:

ˆ V (x) = ∞, where x < 0

ˆ V (x) = 0, where 0 ≤ x ≤ a
ˆ V (x) = ∞, where x > a

The particle is in a bound state (it must remain within 0 and a), the wave
function for a particle in a bound state is:
r
2 nπ
ψ(x) = sin x n = 1, 2, 3... (2.1)
a a
This is the wave function for the states n = 1, 2, 3....
With the Schrödinger equation (1.2) the equation for the energy can be
postulated1 :
h̄2 π 2 n2
E= n = 1, 2, 3... (2.2)
2ma2

2.1.1 Trapping in Potential Wells

Consider the following potential.

V
V2

V1

x
x1 x2

Figure 2.2: A potential well

1
See proof ...

23
We know that that a particle’s kinetic energy summed with its potential
energy is a constant, equal to this equation:

p2
E= +V (2.3)
2m

If the particles total energy is less than V1 , the particle will be trapped in
the potential well, as seen in Figure 2.2; tot get ot of the well, the particles
kinetic energy would have to become negative to satisfy the equation, which
is impossible.

If a particles energy, E, is greater than the potential V1 in Figure 2.2 the

praticle can escape from the potential well. There are two possible cases:
V1 < E < V2 and E > V2 . Here we look at them separately.

Case 1: V1 < E < V2

If V1 < E < V2 , the particle in the potetnial well has enough energy to
overcome the left-barrier but not the one on the right. The particle is thus
free to move to negative infinity, making its allowed x-region between −∞
and x1 .
Here the allowed energy values are continuous, not discrete, because the
particle isn’t completely bound. The energy eigenvalues are not degenerate
- that is, no two energy eigenvalues are the same.
The Schrödinger equation is a second order differential equation, so it has
two linearly independent solutions: however, in this case, only one of those
solutions is physical and doesn’t diverge.
The wave function in this case turns out to oscillate for x < x2 and to decay
rapidly for x > x2 .

Case 2: E > V2
If E > V2 , the particle isn’t bound at all and is free to travel from negative
infinity to positive infinity.
The energy spectrum is continuous and the wave function turns out to be
a sum of a function moving to the right and one moving to the left. The
energy levels of the allowed spectrum are therefore doubly degenerate.
That’s all the overview you need - time to start solving the Schrödinger
equation for various different potentials, starting with the easiest.

24
2.1.2 Particles in Infinite Square Potential Wells
Here we are going to take a closer look a the beautiful infinite square well.

Finding the wave function

Looking at the square well in Figure 2.1 we know:

ˆ V (x) = ∞, where x < 0

ˆ V (x) = 0, where 0 ≤ x ≤ a
ˆ V (x) = ∞, where x > a

Then we take the three-dimensional Schrödinger equation (1.2), and convert

it to a one dimensional form:

−h̄2 δ 2
E ψ(x) = ψ(x) + V (x)ψ(x) (2.4)
2m δx2
And V (x) = 0 inside the well so:

−h̄2 δ 2
E ψ(x) = ψ(x)
2m δx2
Which is usually written as:

δ2
ψ(x) + k 2 ψ(x) = 0
δx2

where k 2 = 2mE
h̄2
(k is called the wave number).
Because this is a second-order differential equation you get two independent
solutions:

ψ1 (x) = A sin(kx)
ψ2 (x) = B sin(kx)

A and B are yet to be determined constants.

δ2 2
The general solution of δx 2 ψ(x) + k ψ(x) = 0 is the sum of ψ1 (x) and ψ2 (x):

ψ(x) = A sin(kx) + B cos(kx) (2.5)

25
Determining the Energy Levels
To find A and B the boundary conditions need to be taken into account, the
wave function must disappear at the boundaries of the square well, so
ˆ ψ(0) = 0
ˆ ψ(a) = 0
ψ(0) = 0 tells you that B must be zero because cos(0) = 1. And the fact
that ψ(a) = 0 tells you that ψ(a) = A sin(ka) = 0. Because sine is zero when
its argument is a multiple of π, this means that

ka = nπ n = 1, 2, 3...

Of course n = 0 is technically a solution, it yields ψ(x) = 0 for all x, which

is not normalizable, so not a physical solution.
The equation can be rewritten to:
nπ
k= n = 1, 2, 3...
a
Substituting this into the original equation:
2mE n2 π 2
= 2
h̄2 a
n2 h̄2 π 2
E= (2.6)
2ma2
The first physical state (n = 1) gives:

h̄2 π 2
E=
2ma2

Normalizing the wave function

To normalize the wave function we need the total probability of finding the
particle between x = 0 and x = a to be 1:
Z a
|ψ(x)|2 dx = 1
0

And we know:  nπx 

ψ(a) = A sin
a
26
So substituting this:
Z a   nπx 2
2
|A| sin  dx = 1
 
0 a

Which results in the following normalized wave function2 :

r
2  nπx 
ψ(x) = sin n = 1, 2, 3...
a a

Time dependence to wave functions

We know the Hamiltonian operator:

Hψ(r) = Eψ(r)

This is actually called the time-independent Schrödinger equation. Its time-

dependent form looks like this:

δ
ih̄ Ψ(r, t) = HΨ(r, t) (2.7)
δt
Combining this with equation 1.2 gives the following form:

δ h̄2 2
ih̄ Ψ(r, t) = − ∇ Ψ(r, t) + V (r, t)Ψ(r, t) (2.8)
δt 2m
But dealing with only one dimension, x, this becomes:

δ h̄2 δ 2
ih̄ Ψ(x, t) = − Ψ(x, t) + V (x, t)Ψ(x, t)
δt 2m δx2
When the potential doesn’t vary in time this becomes:

δ
ih̄ Ψ(x, t) = E Ψ(x, t)
δt
Which is quite easily solvable, in this case the solution is:

Ψ(x, t) = ψ(x)e−iEt/h̄
2
See proof ...

27
This is a very useful and neat solution, because the time-dependent wave
function for a particle in a square well just becomes:
r
2  nπx  −iEt/h̄
Ψ(x, t) = sin e n = 1, 2, 3...
a a
With the proved energy equation (2.6) this the overall time-dependent wave
function becomes:
r
2  nπx  − in2 h̄2 π22 t
Ψ(x, t) = sin e 2ma n = 1, 2, 3...
a a

Shifting square well potentials

We know now that the standard infinite square well looks like this:
ˆ V (x) = ∞, where x < 0
ˆ V (x) = 0, where 0 ≤ x ≤ a
ˆ V (x) = ∞, where x > a
However, we can shift this potential ever so slightly until the origin (x = 0)
is the middle of the well, making it symmetric around the origin. Keeping
the same distances this would mean:
ˆ V (x) = ∞, where x < − a2
ˆ V (x) = 0, where − a2 ≤ x ≤ a
2
ˆ V (x) = ∞, where x > a
2
a
In the wave function we only need to shift the x a little bit by adding 2
to
the equation. Resulting in:
r
2  nπ  a 
ψ(x) = sin x+ n = 1, 2, 3...
a a 2
A little mathematics3 gives you:
r
2  nπ  a 
ψ(x) = cos x+ n = 1, 3, 5...
a a 2
r
2  nπ  a 
ψ(x) = sin x+ n = 2, 4, 6...
a a 2
3
See proof ...

28
2.2 Particles and Potential Steps
In reality one rarely comes across a truly infinite potential, here we will look
at more real-world examples. Here potential is not infinity but a certain
finite value, V0 .

V0

0
Figure 2.3: A potential step, E > V0

There are two possible cases depending on the value of E:

1. E > V0 : Classically, when E > V0 , you expect the particle to be able

to continue on to the region x > 0.
2. E < V0 : When E < V0 , you would expect the particle to bounce back
and not be able to get to the region x > 0 at all.

The first is the case in our example (Figure 2.3) so E > V0 .

2.2.1 Assuming the particle has plenty of energy

When the particle has more energy (E) than the potential (V0 ), the Schrödinger
equation looks like this:
δ 2 ψ1
ˆ For the region x < 00: δx2
(x) + k1 2 ψ1 (x) = 0
Here, k1 2 = 2mE
h̄2
δ 2 ψ2
ˆ For the region x > 00: δx2
(x) + k2 2 ψ2 (x) = 0
Here, k2 2 = 2m(E−V
h̄2
0)

29
This means that one two different k’s and therefore two different wave functions:

k1 k2

0
Figure 2.4: The value of k by region, where E > V0

The first equation can be quite easily solved when treated as a second order
differential equation4 , the solution that follows is:

ψ1 (x) = Aeik1 x + Be−ik1 x , where x < 0

And consequently solving the second equation:

ψ2 (x) = Ceik2 x + De−ik2 x , where x > 0

In these equations eikx represents wave travelling in the +x direction, and

e−ikx represents plane waves travelling in the −x direction.
However the wave can only be reflected by going to the right, not the left, so
D must be zero. The wave function becomes:

ˆ Where x < 00: ψ1 (x) = Aeik1 x + Be−ik1 x

ˆ Where x > 00: ψ2 (x) = Ceik2 x

Here Aeik1 x is the incident wave, Be−ik1 x the reflected wave and Ceik2 x is the
transmitted wave.

4
See proof A.5

30
 With the reflecting and transmission coefficients one can calculate the
probability the particle will be reflected or transmitted through. These are
defined in terms of the current density J(x); this is given in terms of the
wave function by:
dψ ∗ (x)
 
ih̄ ∗ dψ(x)
J(x) = ψ(x) − ψ (x) (2.9)
2m dx dx
If Jr is the reflected current density, and Ji is the incident current density,
then R, the reflection coefficient is
Jr
R=
Ji
T, the transmission coefficient, is5
Jt
T =
Ji
To start with Ji we only need to know the wave function, in this case Aeik1 x ,
thus the incident current density is6
dψi∗ (x)
 
ih̄ ∗ dψi (x) h̄k1 2
Ji = ψi (x) − ψi (x) = |A|
2m dx dx m
Following this Jr and Jt become:
h̄k1 2
Jr = |B|
m
h̄k2 2
Jt = |C|
m

All of these combined give us:

Jr |B|2
R= =
Ji |A|2
Jt k2 |C|2
T = =
Ji k1 |A|2
5
With Jt as the transmitted current density
6
See proof A.6

31
The only boundary conditions known in this case, are that ψ(x) and dψ(x)/dx
are continuous across the potential step’s boundary. In other words, the two
wave functions must touch at x = 0:

ψ1 (0) = ψ2 (0)
dψ1 dψ2
(0) = (0)
dx dx
Starting with the first one, plugging both wave functions (ψ1 (x) and ψ2 (x))
in gives you A + B = C. Substituting them both in the second one gives:

k1 A − k1 B = k2 C

Solving for B in terms of A and subsequently solving for C in terms of A,

gives you this result:
k1 − k2
B= A
k1 + k2
2k1
C= A
k1 + k2
Filling these in in the original equations for the reflection and transmission
coefficients, R and T , we get:
(k1 −k2 )2
|B|2 (k1 +k2 )2
· |A|2 (k1 − k2 )2
R= = =
|A|2 |A|2 (k1 + k2 )2
2
k2 |C|2 k2 (k14k+k1 2 )2 |A|2 k2 4k1 2 4k1 k2
T = 2
= 2
= · 2
=
k1 |A| k1 |A| k1 (k1 + k2 ) (k1 + k2 )2

According to classical physics there should be no reflection at all, but when

k1 ̸= k2 there will indeed be a particle reflection.
Also not that as k1 goes to k2 , R goes to 0 and T goes to 1, which is what
you would expect7 .

7
As k1 ≡ E and k2 ≡ E − V0 , when k1 becomes smaller, E becomes smaller (compared
to V0 ) and thus the ’leap’ is smaller

32
 V

V0
E

0
Figure 2.5: A potential step, E < V0

2.2.2 Assuming the particle doesn’t have enough energy

With the earlier example (as seen in Figure 2.3), the particle had enough
energy to make it into te region x > 0.
We will first start looking at the region x < 0, from the previous section we
know:
d2 ψ1
ˆ dx2
(x) + k1 2 ψ1 (x) = 0
ˆ k1 2 = 2mE
h̄2
ˆ ψ1 (x) = Aeik1 x + Be−ik1 x x<0

And for x > 0 we know:

d2 ψ2
ˆ dx2
(x) + k2 2 ψ2 (x) =0
ˆ k2 2 = 2m(E−V
h̄2
0)

However this would make k2 imaginary (E − V0 is less then zero) which is

impossible. When we change the sign in the Schrödinger equation this is
solved, and we know:
d2 ψ2
ˆ dx2
(x) − k2 2 ψ2 (x) =0
ˆ k2 2 = 2m(Vh̄02−E)

33
 And this Schrödinger equation we have two solutions8 :

ˆ ψ(x) = Ce−k2 x
ˆ ψ(x) = Dek2 x

And so the general solution is:

ψ2 (x) = Ce−k2 x + Dek2 x x>0

Because the wave function must be finite D must be zero. This term goes
to infinity, which will not be a problem with the first term as it is limited to
x > 0). Therefore the solution is:

ψ2 (x) = Ce−k2 x x>0 (2.10)

These two wave functions can be put in terms of the incident, reflected and
transmitted wave functions (ψi (x), ψr (x) and ψt (x)), you have the following:

ˆ ψi (x) = Aeik1 x
ˆ ψr (x) = Be−ik1 x
ˆ ψt (x) = Ce−k2 x

As for the coefficients:

Jr
R=
Ji
Jt
T =
Ji
These are must easier to solve now, because Jt of course is:

dψt∗ (x)
 
ih̄ ∗ dψt (x)
Jt = ψt (x) − ψt (x)
2m dx dx

However ψt (x) is completely real, therefore this becomes:

 
ih̄ dψt (x) dψt (x)
Jt = ψt (x) − ψt (x) =0
2m dx dx
8
See proof A.5, and flip the indicated sign

34
When Jt = 0, T = 0 which means that nothing gets through (or R = 1) and
you get complete reflection, as the classical solution says.

There is, however, a non-zero change of finding the particle in the region
x > 0. Look at the probability density:
P (x) = |ψt (x)|2
Plugging in equation 2.10 gives you:
P (x) = |ψt (x)|2 = |C|2 e−2k2 x
The continuity conditions can be used to solve for C in terms of A:
ˆ ψ(0) = ψ1 (0)
ˆ dψ1
dx
(0) = dψ2
dx
(0)
Using the conditions gives you:
4k1 2 |A|e−2k2 x
P (x) = |C|2 e−2k2 x =
k1 2 + k2 2
This does fall very quickly to zero as x gets larger, but near x = 0, it has a
non-zero value:

Figure 2.6: The value of k by region, E < V0

Now we have taken a look at infinite square wells and potential steps its time
to see what happens if the steps do not stretch out into infinity, this brings
us to potential barriers.

35
2.3 Particles and Potential Barriers
When we take a finite potential step, this becomes a potential barrier, which
is thus set up something like this:
ˆ V (x) = 0, where x < 0
ˆ V (x) = V0 , where 0 ≤ x ≤ a
ˆ V (x) = 0, where x > a

In Figure 2.7 you can see what this looks like:

V0

x
a
0
Figure 2.7: A potential barrier, E > V0

2.3.1 Through the barrier with E > V0

Following the principles as in earlier examples, we get:
δ 2 ψ1
ˆ For the region x < 00: δx2
(x) + k1 2 ψ1 (x) = 0
where, k1 2 = 2mE
h̄2
δ 2 ψ2
ˆ For the region 0 ≤ x ≤ aa: δx2
(x) + k2 2 ψ2 (x) = 0
where, k2 2 = 2m(E−V
h̄2
0)

δ 2 ψ3
ˆ For the region x > 00: δx2
(x) + k1 2 ψ3 (x) = 0
where, k1 2 = 2mE
h̄2

36
Giving you the (standard) solutions:
ˆ ψ1 (x) = Aeik1 x + Be−ik1 x
ˆ ψ2 (x) = Ceik2 x + De−ik2 x
ˆ ψ3 (x) = Eeik1 x + F e−ik1 x = Eeik1 x
With the continuity conditions, we know:
ψ1 (0) = ψ2 (0)
dψ1 dψ2
(0) = (0)
dx dx

ψ2 (a) = ψ3 (a)
dψ2 dψ3
(a) = (a)
dx dx
From which follows:
ˆ A+B =C +D
ˆ ik1 (A − B) = ik2 (C − D)
ˆ Ceik2 a + De−ik2 a = Eeik1 a
ˆ ik2 Ceik2 a − ik2 De−ik2 a = ik1 Eeik1 a
Putting all of these together we get E in terms of A:
−1
E = 4k1 k2 Ae−ik1 a 4k1 k2 cos(k2 a) − 2i(k1 2 + k2 2 ) sin(k2 a)


Now for the transmission coefficient, T :

|E|2
T =
|A|2
Which works out to be
" 2 #−1
k1 2 − k2 2

1
T = 1+ sin2 (k2 a)
4 k1 k2

Note that as k1 goes to k2 , T goes to 1, which is what you should expect.

R in this case equals:
(k2 2 − k1 2 ) sin2 (k2 a)
R=
4k1 2 k2 2 + (k1 2 − k2 2 )2 sin2 (k2 a)

37
The probability density, |ψ(x)|2 , for this potential barrier looks like Figure
2.8.

Figure 2.8: |ψ(x)|2 for a potential barrier, E > V0

2.3.2 Through the barrier even when E < V0

What happens when the situation is like Figure 2.9.

V0

x
a
0
Figure 2.9: A potential barrier, E < V0

38
The Schrödinger looks like this:
δ 2 ψ1
ˆ For the region x < 00: δx2
(x) + k1 2 ψ1 (x) = 0
where, k1 2 = 2mE
h̄2
δ 2 ψ2
ˆ For the region 0 ≤ x ≤ aa: δx2
(x) − k2 2 ψ2 (x) = 0 (otherwise k is
imaginary)
where, k2 2 = 2m(Vh̄02−E)
δ 2 ψ3
ˆ For the region x > 00: δx2
(x) + k1 2 ψ3 (x) = 0
where, k1 2 = 2mE
h̄2

This means that the the wave functions looks like this:

ˆ ψ1 (x) = Aeik1 x + Be−ik1 x

ˆ ψ2 (x) = Cek2 x + De−k2 x
ˆ ψ3 (x) = Eeik1 x + F e−ik1 x = Eeik1 x

This looks like:

Figure 2.10: |ψ(x)|2 for a potential barrier, E < V0

And the reflection and transmission coefficients, R and T ? Here is what they
equal:

|B|2
R=
|A|2
|E|2
T =
|A|2

39
And again the continuity conditions to determine A, B and E:
ψ1 (0) = ψ2 (0)
dψ1 dψ2
(0) = (0)
dx dx

ψ2 (a) = ψ3 (a)
dψ2 dψ3
(a) = (a)
dx dx
With a fair bit of algebra we get:
4k1 2 k2 2
R=
(k1 2 + k2 2 ) sinh(k2 a)
"  2 2 #−1
2 k1 + k2 2 2
T = cosh (k2 a) + sinh (k2 a)
2k1 2 k2 2
Despite the equation’s complexity it is amazing that T can be non-zero.

This phenomenon where particles can get to classically penetrable barriers

is called tunneling. In general the probability of the particle going through
can be determined with the transmission coefficient.
With the WKB (Wentzel-Kramers-Brillouin) approximation, you can calculate
(relatively easy) how large the probability of the particle going through
is. This is done by approaching said potential as a sum of many square
potentials:    Z x2 
2 p
T exp − 2m(V (x) − E) dx
h̄ x1

2.4 Particles Unbound

Here we are going to take a look at free particles. Apart from potential wells
and barriers there are plenty free particles in the universe, lets what quantum
physics has t say about them.

Firstly recall the one-dimensional Schrödinger equation:

−h̄2 d2
ψ(x) + V (x) ψ(x) = E ψ(x)
2m dx2
40
For a free particle (V (x) = 0) we have:

−h̄2 d2
ψ(x) − E ψ(x) = 0
2m dx2
As seen a lot of times before this leads to:

d2
ψ(x) + k 2 ψ(x) = 0
dx2
ψ(x) = Aeikx + Be−ikx

Which with time-dependence because:

+ Be−ikx−
iEt/h̄ iEt/h̄
Ψ(x, t) = Aeikx−

This is a perfectly good solution to the Schrödinger equation, it is however

very unphysical. The probability density (|ψ9(x)|2 ) is not normalizable:

|ψ(x)|2 = |A|2 or |B|2

The problem for this particles is that the probability density is uniform
through all x’s. This is because you know k precisely - and p = h̄k and
E = h̄k 2 /2m. So when you know p and E exactly, this causes an enormous
uncertainty in x and t.
For the earlier example, the wave function ψ(x) can not be normalized. When
we do this separately for the first and second term we get:
Z +∞ Z +∞
∗ 2
ψ1 (x)ψ1 (x)dx = |A| dx → ∞
−∞ −∞
Z +∞ Z +∞
ψ2 (x)ψ2 ∗ (x)dx = |B|2 dx → ∞
−∞ −∞

2.4.1 Getting particles with wave packets

Important: Any linear combination of solutions of the Schrödinger equations
is a solution on itself. You want to add various wave functions together
so you get a wave packet, which is a collection of wave functions such that

41
they interfere constructively at one location and interfere destructively (go
to zero) at all other location:
∞
X kx − Et/h̄)
Ψ(x, t) = ϕn ei(
n=1

Which usually written as a continuous integral:

Z +∞
1 kx − Et/h̄)
Ψ(x, t) = √ ϕ(k, t)ei( dk
2π −∞

ϕ(x, t) is the amplitude of each component wave function, and you can find
it from the Fourier transform of the equation:
Z +∞
1
Ψ(x, t)e−i(
kx − Et/h̄)
ϕ(p, t) = √ dx
2πh̄ −∞
Because k = p/h̄, you can write the wave packet equations like this:
Z +∞
1 px − Et/h̄)
Ψ(x, t) = √ ϕ(p, t)ei( dp
2πh̄ −∞
Z +∞
1
Ψ(x, t)e−i(
kx − Et/h̄)
ϕ(p, t) = √ dx
2πh̄ −∞
This look pretty confusing, as ϕ(p, t) is defined in terms Ψ(x, t), and Ψ(x, t)
is defined in terms ϕ(p, t).
Just remember that the to are not definitions of ϕ(p, t) and Ψ(x, t); just
equations relating the two.

42
2.5 Gaussian Example
For a concrete overview of the discussed theory, we select an actual wave
packet shape, the so-called Gaussian, as seen in Figure 2.11.

|Ψ(x, t)|2

Figure 2.11: A Gaussian wave packet

The amplitude ϕ(k) you may choose is

−a2 (k − k0 )2
 
ϕ(k) = Aexp
4

Firstly we want to normalize this to determine A:

Z +∞
1= |ϕ(k)|2 dk
−∞

Substituting ϕ(k) gives you this equation:

+∞
−a2
Z  
2
1 = |A| exp (k − k0 )2 dk
−∞ 2

43
Calculating this integral gives9 :
r
2π 2
1 = |A|
a2
 2 1/4
a
A=
2π

So here is you wave function:

 2 1/4 Z +∞  2
−a (k − k0 )2 ikx

1 a
ψ(x) = √ exp e dk
2π 2π −∞ 4

Which can be evaluated to give you:

1/4
−x2 ik0 x
 

2
ψ(x) = exp e
πa2 a2

Now you can use this to determine the probability that the particle will be
in, say, the region 0 ≤ x ≤ a/2. The probability:
Z a/2

|ψ(x)|2 dx
0

So doing the math:

1/2 Z a/2
−2x2
  
2 1
exp dx =
πa2 0 a2 3

9
See proof A.7

44
2.6 Harmonic Oscillators
To start looking at harmonic oscillation we firstly have to return to classical
physics. We know the following (from Hooke’s law):

F = −Cx

In which C is the spring constant and x is displacement. Which Newton’s

second law (F = ma) this becomes:

ma + Cx = 0

We know that acceleration is given by:

dv d2 x
a= = 2
dt dt
Substituting for a:
d2 x
ma + Cx = m + Cx = 0
dt2
Dividing by mass:
d2 x Cx
+ =0
dt2 m
If we take k/m = ω 2 (where ω is the angular frequency), this becomes:

d2 x
+ ω2x = 0
dt2
Solving this equation we get:

x = A sin(ωt) + B cos(ωt)

Here the sine and cosine prove that the solution is an oscillating one.

45
 Because the Hamiltonian (H) gives the total energy it is the sum of kinetic
and potential energies:
H = Ek + Ep
We know the equations for these energies, the potential energy is here rewritten
to fit a harmonic oscillator:
p2
ˆ Ek = 2m
ˆ Ep = 1
2
kx2 = 21 mω 2 x2

So,
P2 1
+ mω 2 X 2
H=
2m 2
Where X and P are the position and momentum operator respectively.
So using the Hamiltonian on one of its eigenvectors we get the total energy:

P2 1
H |ψ⟩ = |ψ⟩ + mω 2 X 2 |ψ⟩ = E |ψ⟩
2m 2
To solve this rather difficult equation we are going to need new operators.

2.6.1 Creation and Annihilation

We make two new operators, the creation and annihilation operator. Here is
what they do:

ˆ Creation operator: The creation operator raises the energy level of

an eigenstate by one level, so if the harmonic oscillator is in the fourth
energy level, the creation operator raises it to the fifth level.
ˆ Annihilation operator: The annihilation does the reverse, lowering
eigenstate one level.

To make this we start by denoting two new dimensionless operators , p and

q, which relate to P and X this way:
P
p= √ (2.11)
mh̄ω
r
mω
q=X (2.12)
h̄

46
With these we can make the annihilation operator, a, and the creation
operator, a† :
1
a = √ (q + ip) (2.13)
2
1
a† = √ (q − ip) (2.14)
2
With these equations we can rewrite the harmonic oscillator Hamiltonian in
terms of a and a† :  
† 1
H = h̄ω a a +
2
Then a† a became another operator: the N or number operator :
 
1
H = h̄ω N + (2.15)
2
The N operator returns the number of the eigenstate, n is here the nth
eigenstate:
N |n⟩ = n |n⟩
Keeping in mind that H |n⟩ = En |n⟩, we can rewrite equation 2.15 to:
 
1
En = n + h̄ω n = 0, 1, 2...
2

So for the ground state (n = 0), first (n = 1) and second (n = 2) excited

state we get:
1
E0 = h̄ω
2
3
E1 = h̄ω
2
5
E2 = h̄ω
2

2.6.2 Finding eigenstates

Firstly take a look at the commutator of a and a† :
 † 1
a, a = [q + ip, q − ip]
2
47
Which equals10 :
 † 1
a, a = [q + ip, q − ip] = −i[q, p]
2
 †
This break down to a, a = 1. And  putting† this together with equation
†
2.15, lead to [a, H] = h̄ωa and a , H = −h̄ωa .
All this boils down to:

H(a |n⟩)
= (aH−h̄ωa) |n⟩
= (En −h̄ω)(a |n⟩)

This shows that a |n⟩ is also an eigenstate of the harmonic oscillator, with
energy En − h̄ω (one level beneath En ).
The same applies to a† :

H(a† |n⟩)
= (a† H+h̄ωa† ) |n⟩
= (En +h̄ω)(a† |n⟩)

Which raises (+h̄ω) the energy En by one level.

From these equations we can derive:

a |n⟩ = C |n − 1⟩ (2.16)
†
a |n⟩ = D |n + 1⟩ (2.17)

To find the constants C and D we first write down a few things:

1. ⟨n − 1|n − 1⟩ = ⟨n + 1|n + 1⟩ = ⟨n|n⟩ = 1

2. a† a = N
3. N |n⟩ = n |n⟩
4. a, a† = aa† − a† a = 1 aa† = a† a + 1
 
→

10
See proof A.8

48
 We are going to start with finding C, to begin we square the formula.
After which we use the given rules to find C:

(⟨n| a† )(a |n⟩) = C 2 ⟨n − 1|n − 1⟩ (1)

(⟨n| a† )(a |n⟩) = C 2
⟨n|a† |n⟩ = C 2 (2)
⟨n|N |n⟩ = C 2 (3)
n ⟨n|n⟩ = C 2 (1)
C 2 =√n
C= n
Then for a† , again we start by squaring the whole thing and then simplify it
until we find D:

(⟨n| a)(a† |n⟩) = D2 ⟨n + 1|n + 1⟩ (1)

(⟨n| a)(a† |n⟩) = D2
⟨n|aa† |n⟩ = D2 (4)
⟨n|a† a|n⟩ + ⟨n|n⟩ = D2 (1)
⟨n|a† a|n⟩ + 1 = D2 (2)
⟨n|N |n⟩ + 1 = D2 (3)
n ⟨n|n⟩ + 1 = D2 (1)
D2 =√n + 1
D = n+1
These results give us:
√
a |n⟩ = n |n − 1⟩
√
a† |n⟩ = n + 1 |n + 1⟩
Now that we know this, we can derive all successive energy states from only
the ground state:
ˆ |1⟩ = a† |0⟩ = |1⟩

2
ˆ |2⟩ = √1 a†
2
|1⟩ = √1
2!
a† |0⟩

3
ˆ |3⟩ = √1 a†
3
|2⟩ = √1
3!
a† |0⟩

4
ˆ |4⟩ = √1 a†
4
|3⟩ = √1
4!
a† |0⟩
So in general:
1
n
|n⟩ = √ a† |0⟩
n!

49
Working in position space
We still don’t know what |0⟩ actually is, we want a spacial eigenstate (like
ψ0 (x)) of this eigenvector. In other words, we are looking for ⟨x|0⟩ = ψ0 (x).
And to find this we need the representations of a and a† in position space.

The p operator is defined as:

P
p= √
mh̄ω

From section 1.2 we know P for one-dimension (x):

 2
−ih̄ d h̄ d
p= √ = −i
mh̄ω dx mω dx

And writing x0 = (h̄/mω)1/2 , this becomes:

d
p = −ix0
dx
The definition of the a operator is:
1
a = √ (q + ip)
2
And q is: r
mω X
q=X =
h̄ x0
Thus,
 
1 X d
a= √ + x0
2 x0 dx
 
1 2 d
= √ X + x0
x0 2 dx

And a† turns out to be:

 
† 1 2 d
a = √ X − x0
x0 2 dx

50
Now to be clever; if we use the lowering operator, a, on |0⟩, we must 0 because
there is no lower state. Therefore a |0⟩ = 0 must be true, with the ⟨x| bra
this means ⟨x| a |0⟩ = 0.
Substituting this in the earlier equation:

⟨x| a |0⟩ = 0

1√ d


x0 2
⟨x| X + x0 2 dx |0⟩ = 0 ⟨x|0⟩ = ψ0 (x)

1√ d


x0 2
X + x0 2 dx ψ0 (x) = 0
  √
1√
x0 2
xψ0 (x) + x0 2 dψdx
0 (x)
=0 ·x0 2

xψ0 (x) + x0 2 dψdx

0 (x)
=0

dψ0 (x) −xψ0 (x)

dx
= x0 2

This differential equation give the following solution11 :

−x2
 
ψ0 (x) = A exp
2x0 2

Which means that the ground state of a quantum mechanical harmonic

oscillator is a Gaussian curve (like Figure 2.11).

Wave functions and more levels

As always to make a fully functional wave function we need to find A, and

how? As always we just need to normalize the earlier function:

Z ∞
1= |ψ0 (x)|2 dx
−∞

11
See proof A.9

51
Which becomes:
  2 2
∞
−x
Z
2
1=A exp dx
−∞ 2x0 2
√
1 = A2 π x0
1
A2 = √
x0 π
1
A = 1/4 1/2
π x0

Which gives us the wave function:

−x2
 
1
ψ0 (x) = 1/4 1/2 exp
π x0 2x0 2

Now to make it even more interesting we are going to find the wave function
of the first exited state. We know ψ1 (x) = ⟨x|1⟩ and |1⟩ = a† |0⟩, so

ψ1 (x) = ⟨x|a† |0⟩

The definition of a† :  
† 1 2 d
a = √ X − x0
x0 2 dx
Therefore, ψ1 (x) = ⟨x|a† |0⟩ becomes

⟨x|a† |0⟩ = x 1√2 ⟨x| X − x0 2 dx

d


|0⟩ = ψ1 (x) ⟨x|0⟩ = ψ0 (x)
0

1√ d


ψ1 (x) = x0 2
X − x0 2 dx ψ0 (x)
 
ψ1 (x) = x0 2
1√
xψ0 (x) − x0 2 dψdx
0 (x) dψ0 (x)
dx
= (−x)
ψ (x)
x0 2 0

 
ψ1 (x) = x0 2
1√
xψ0 (x) − x0 2 (−x)
x0 2
ψ0 (x)
√
2
ψ1 (x) = x0
x ψ0 (x) definition of ψ0 (x)
√  
2 −x2
ψ1 (x) = π 1/4 x0 3/2
x exp 2x0 2

52
This is a very nice result, but what about the second excited level, or even
further? We can find ψ2 (x) with:
1
2
ψ2 (x) = √ ⟨x| a† |0⟩
2!

Substituting for a† , this becomes:

 
1 1 2 d
ψ2 (x) = √ x − x0 ψ0 (x)
2! 2x0 2 dx

This can be generalized to the following monster of an equation:

 n  2
1 1 2 d −x
ψn (x) = 1/4 n 1/2 n+1/2 x − x0 exp
π (2 n!) x0 dx 2x0 2

To make this a little easier physicists came up with the nth hermite polynomial,
Hn (x):  n  2  2
2 d −x x −x
x − x0 exp 2
= Hn ( ) exp
dx 2x0 x0 2x0 2
Hn (x) is defined as:

dn
Hn (x) = (−1)n exp(x2 ) exp(−x2 )
dxn
Some solutions to this are:

H0 (x) = 1 H0 ( xx0 ) = 1
H1 (x) = 2x H1 ( xx0 ) = x2x
0
2
4x2
H2 (x) = 4x2 − 2 H2 ( x0 ) = x0 4 − x20 2
x
3
H3 (x) = 8x3 − 12x H3 ( xx0 ) = 8x
x0 6
− 12x
x0 4
4 2
H4 (x) = 16x4 − 48x2 + 12 H4 ( x0 ) = x0 8 − 48x
x 16x
x0 6
+ x12
0
4
x 32x 5 160x 3 120x
H5 (x) = 32x5 − 160x3 + 120x H5 ( x0 ) = x0 16 − x0 8 + x0 6
With this the general quantum harmonic oscillators wave function looks like:
   2
1 1 x −x
ψn (x) = 1/4 n 1/2 n+1/2 Hn exp
π (2 n!) x0 x0 2x0 2
1
where x0 = (h̄/(mω)) /2

53
2.6.3 Harmonic Oscillators Operators as Matrices
Because the energy levels of an harmonic oscillator a evenly spaced it sometimes
is easier to view it in terms of matrices. In the case the ground state and
first excited energy level are:
   
1 0
0 1
   
0 0
   
0 0
   
|0⟩ = 
0 |1⟩ = 0
  
0 0
   
. .
   
. .
. .
The N operator (which just returns the energy level) is then fairly simple.
As an example we can look at N |2⟩:
      
0 0 0 0 ... 0 0 0
0 1 0 0 . . . 0 0 0
      
0 0 2 0 . . . 1 2 1
      
0 0 0 3 . . . 0 0 0
      
N |2⟩ = 
 0 0 0 0 . . .  0 = 0 = 2 0 = 2 |2⟩
     
0 0 0 0 . . . 0 0 0
      
 . . . . . . .  .   .  .
      
 . . . . . . .  .   .  .
. . . . ... . . .
For the lowering operator a, is a little bit more difficult, as an example we
take a |1⟩ which should result in |0⟩:
 √    
0 1 √0 0 ... 0 1
0 0
 2 √0 . . .   0
 1  

 0 0
0 0 0 3 . . .    

 0 0
0 0 0 0 . . .    

a |1⟩ = 
0 0 0 0  0 = 0 = |0⟩
. . .    
0 0
0 0 0 0 . . . 
   

 . .
. . . . . . .    

. . . . . . . . .

. . . . ... . .

54
a† looks very much like a (as
√ one would expect), in√the example we will take
† †
a |1⟩ which should equal 2 |2⟩ (because a |n⟩ = n + 1 |n + 1⟩):

√0 0 0 0 ...
    
0 0
 1 0 0 0 . . . 1 √0 
√
   
 
 0 2 0 0 . . .  0  2
 √    
 0  0  √
 0 0 3 √0 . . .    

a |1⟩ =  0
 0 0 4 . . . 

0 =  0  = 2 |2⟩
  
 0  0 
 0 0 0 0 . . .    

.  . 
 . . . . . . . 
   

 . . . . . . .  .   . 
. . . . ... . .

Lastly the Hamiltonian, we know the possible outcomes are always in the
form, En = (n + 21 )h̄ω, therefore the Hamiltonian in matrix form is:
 
1 0 0 0 ...
0 3 0 0 . . .
 
0 0 5 0 . . .
 
0 0 0 7 . . .
h̄ω 
0

H= 0 0 0 . . .
2 0

 0 0 0 . . .

. . . . . . .
 
. . . . . . .
. . . . ...

So these are the matrices for if you are interested in looking at it this way.

55
Appendix A

Proofs and Calculations

Here any and all proofs and calculations skipped over before will be written
out as a guide through the book.

A.1 Expectation value Dice Example

The expectation value ⟨ψ|R|ψ⟩ was given by:

 
1/6

2 0 0 0 0 0 0 0 0 0 0 √ 
0 3 0 0 0 0 0 0 0 0 0  2/6
 √ 
 

 0 0 4 0 0 0 0 0 0 0 0   3/6
 
0 0 0 5 0 0 0 0 0 0 0   2/6 
 √ 


0 0 0 0 6 0 0 0 0 0 0   5/6
h
1
√
2
√
3 2
√
5
√ √
6 5 2
√ √
3 2
i

1 
 √ 
6 6 6 6 6 6 6 6 6 6 6 
0 0 0 0 0 7 0 0 0 0 0   6/6
 √ 

 0 0 0 0 0 0 8 0 0 0 0   5/6
 

 0 0 0 0 0 0 0 9 0 0 0   2/6 
 √ 

 0 0 0 0 0 0 0 0 10 0 0   3/6
 √ 
 0 0 0 0 0 0 0 0 0 11 0   2/6
0 0 0 0 0 0 0 0 0 0 12 1/6

56
Breaking this we start with the last two terms:

 
1/6

2 0 0 0 0 0 0 0 0 0 0 √ 
0 3 0 0 0 0 0 0 0 0 0  2/6
 √ 
 

 0 0 4 0 0 0 0 0 0 0 0   3/6
 
0 0 0 5 0 0 0 0 0 0 0   2/6 
 √ 


0 0 0 0 6 0 0 0 0 0 0   5/6
 √ 



 0 0 0 0 0 7 0 0 0 0 0   6/6
 √  =

 0 0 0 0 0 0 8 0 0 0 0   5/6
 

 0 0 0 0 0 0 0 9 0 0 0   2/6 
 √ 

 0 0 0 0 0 0 0 0 10 0 0   3/6
 √ 
 0 0 0 0 0 0 0 0 0 11 0   2/6
0 0 0 0 0 0 0 0 0 0 12 1/6

 
2/6 √ 0 0 0 0 0 0 0 0 0 0

 0 2/2 √0 0 0 0 0 0 0 0 0 


 0 0 2 3/3 0 0 0 0 0 0 0 0 


 0 0 0 5/3 √0 0 0 0 0 0 0 

0 0 0 0 5 √0 0 0 0 0 0
 
 
0 0 0 0 0 7 6/6 √0 0 0 0 0
 
 
0 0 0 0 0 0 4 5/3 0 0 0 0
 
 
0 0 0 0 0 0 0 3 √0 0 0
 
 
 

 0 0 0 0 0 0 0 0 5 3/3 √0 0 

 0 0 0 0 0 0 0 0 0 11 2/6 0 
0 0 0 0 0 0 0 0 0 0 2

57
Multiplying this by the first term, ⟨ψ|:
 
2/6 0 0 0 0 0 0 0 0 0 0
 0 √2/2 0 0 0 0 0 0 0 0 0 

 0 √ 
 0 2 3/3 0 0 0 0 0 0 0 0 

 0 0 0 5/3 √0 0 0 0 0 0 0 
 
 0 0 0 0 5 √0 0 0 0 0 0
 

 0 0 0 0 0 7 6/6 √0 0 0 0 0
 

 0 0 0 0 0 0 4 5/3 0 0 0 0
 

 0 0 0 0 0 0 0 3 √0 0 0
 

 
 0 0 0 0 0 0 0 0 5 3/3 √0 0 
 
 0 0 0 0 0 0 0 0 0 11 2/6 0 
0 0 0 0 0 0 0 0 0 0 2
h √ √ √ √ √ √ √ i
· 1 2 3 2 5 6 5 2 3 2 1
6 6 6 6 6 6 6 6 6 6 6

1 1 1 5 5 7 10 5 11 1
 
= 18 6 3 9 6 6 9
1 6 18 3
1 1 1 5 5 7 10 5 11 1
= + + + + + + +1+ + +
18 6 3 9 6 6 9 6 18 3
=7

58
A.2 Rewriting a commutator
In the example [∆A, ∆B] needed to be rewritten, this is done as follows:

[∆A, ∆B] = ∆A∆B − ∆B∆A

Using the established facts that:

∆A = A − ⟨A⟩
∆B = B − ⟨B⟩

This can be said:

[∆A, ∆B] = ∆A∆B − ∆B∆A

= (A − ⟨A⟩)(B − ⟨B⟩) − (B − ⟨B⟩)(A − ⟨A⟩)
= (AB − B ⟨A⟩ − A ⟨B⟩ + ⟨A⟩ ⟨B⟩) − (BA − B ⟨A⟩ − A ⟨B⟩ + ⟨A⟩ ⟨B⟩)
= AB − BA − B ⟨A⟩ + B ⟨A⟩ − A ⟨B⟩ + A ⟨B⟩ + ⟨A⟩ ⟨B⟩ − ⟨A⟩ ⟨B⟩
= AB − BA
= [A, B]

A.3 Energy equation through Schrödinger

Firstly recall the wave function for a particle in a bound state and the
Schrödinger equation (equation 2.1 and 1.2 respectively):
r
2 nπ
ψ(x) = sin x
a a
−h̄2 2
E ψ(x) = ∇ ψ(x) + V (x) ψ(x)
2m

Because the square in one dimensional (only x), the Laplacian operator
shrinks to only the x-dimension. And, because the particle cannot move
past 0 or a, the potential is always 0 for the particle. All of this gives us a
much easier equation:
−h̄2 δ 2
E ψ(x) = ψ(x)
2m δx2
59
Firstly the partial derivatives:
r
2 nπ
ψ(x) = sin x
a a
r  
δ 2 nπ nπ
ψ(x) = · · cos x
δx a a a
r 
δ2 n2 π 2

2 nπ
2
ψ(x) = · 2
· − sin x
δx a a a
r
n2 π 2 2 nπ
=− 2 sin x
a a a
n2 π 2
= − 2 ψ(x)
a

Substituting this in the Schrödinger equation:

h̄2
 2 2 
nπ
E ψ(x) = − · − 2 ψ(x)
2m a
2 2 2
h̄ n π
E ψ(x) = ψ(x)
2ma2
h̄2 n2 π 2
E=
2ma2

This is the final version for the example, for those who want an even more
simplified version, this can be achieved with the definition of h̄:

h
h̄ =
2π

Then the equation becomes:

 
h2
4π 2
n2 π 2
E= 2
 22ma
2

h n
4
=
2ma2
h2 n2
E=
8ma2
60
A.4 Normalizing a Wave Function
Given was: Z a
  nπx 2
2
|A| sin  dx = 1
 
0 a
First we are going to look at the integral part of this:
Z a 
nπx 2
sin  dx

0 a
This can be calculated like so:
Z a  Z a
nπx 2 nπx
sin  dx = sin2 ( ) dx

0 a 0 a
Z a  
1 1 2nπ
= − cos x dx
0 2 2 a
   a
1 1 2nπ a
= x − sin x ·
2 2 a 2nπ 0
 
1 a
= a− sin(2nπ) − (0)
2 4nπ
 
1 a
= a− sin(2nπ)
2 4π · n
 
1 a
= a =
2 2
Substituting this into the original normalization condition:
a
|A|2 · =1
2
1
|A|2 = a


2
2
|A|2 =
a
r
2
A=
a
Which brings the normalized wave function to the form:
r
2  nπx 
ψ(x) = sin n = 1, 2, 3...
a a

61
A.5 Solving differential equation
The given equation was:
d2 ψ1
2
(x) + k1 2 ψ1 (x) = 0
dx
This can easily be solved by stating:
ψ1 (x) = ep
Which makes the equation:
p2 ep + k1 2 ep = 0
0
p2 + k1 2 = p
e
p = −k1 2
2

p2 = −1 · k1 2
√ √
p = −1 · k1 ∨ p = −1 · −k1
p = ik1 ∨ p = −ik1
Substituted this gives:
ψ1 (x) = Aeik1 x + Be−ik1 x
Here we add the constants A and B, which are undetermined real constants.

The same principal can be used to solve the very similar equation:
d2 ψ
2
(x) − k2 2 ψ2 (x) = 0
dx
Again we state ψ(x):
ψ(x) = eq
And then:
q 2 eq − k2 2 eq = 0
0
q 2 − k2 2 = q
e
q = k2 2
2

q = k2 ∨ q = −k2
ψ(x) = ek2 x ∨ ψ(x) = e−k2 x

62
A.6 Calculating the incident current density
Given is the following equation:

dψ ∗ (x)
 
ih̄ ∗ dψ(x)
J(x) = ψ(x) − ψ (x)
2m dx dx

The current we are interested in has the following wave function:

ψi (x) = Aeik1 x

Plugging this wave function in the earlier equation we get:

dψi∗ (x)
 
ih̄ ∗ dψi (x)
Ji = ψi (x) − ψi (x)
2m dx dx

Then we have to postulate ψi∗ (x) and both derivatives:

ψi∗ (x) = A∗ e−ik1 x

dψi
(x) = ik1 · Aeik1 x
dx
dψi∗
(x) = −ik1 · A∗ e−ik1 x
dx
Then we just simply substitute these into the equation:

ih̄ 
Aeik1 x · −ik1 · A∗ e−ik1 x − A∗ e−ik1 x · ik1 · Aeik1 x





Ji =
2m
ih̄ 
−ik1 · |A|2 e0 − ik1 · |A|2 e0



=
2m
ih̄ 
−2ik1 |A|2

=
2m
−2i2 h̄k1 2
= |A|
2m
h̄k1 2
Ji = |A|
m
The Jr and Jt coefficients can be retrieved from some minor adaptions to
this calculation.

63
A.7 Normalizing a Gaussian
Given was the following integral:

+∞
−a2
Z  
2
1 = |A| exp (k − k0 )2 dk
−∞ 2

Firstly we want to make the exponent easier to work with:

+∞
−a2 2
Z  
2 2
1 = |A| exp (k − 2k0 k + k0 ) dk
−∞ 2
Z +∞  2
a2 2

2 −a 2 2
1 = |A| exp k + a k0 k − k0 dk
−∞ 2 2

Which has the form:

Z +∞
2 +bx+c
eax dx
−∞

This exponential has a standard integral B.9, which gives the solution:

r
π − b2 −4ac
 
·e 4a
a

In this case:

x=k
a2
a=
2
b = −a2 k0
a2 2
c = k0
2

64
Therefore the solution becomes:
 2

2 2
(
−a2 k0 ) −4· a2 · a2 k0 2
+∞
−a2 2 a2
Z  
π
s 
2
4· a2
exp k + a2 k0 k − k0 2 dk =  2 · e
−∞ 2 2 a
2
4a4 k0 2
r
2π a4 k0 2 − 4
= ·e 2a2
a2
r
2π a4 k0 2 −a4 k0 2
= ·e 2a2
a2
r
2π 0
= · e 2a2
a2
r
2π
= · e0
a2
r
2π
=
a2
So the solution of the equation:
r
2 2π
1 = |A|
a2
1
|A|2 = q
2π
a2
r
a2
=
2π
sr
a2
A=
2π
 2 1/4
a
A=
2π

65
A.8 Commutator of Raising- and Lowering
Operator
We take the commutator of the raising (a† ) and lowering (a) operators:
 †
a, a = aa† − a† a

The definition of these are:

1
a = √ (q + ip)
2
1
a† = √ (q − ip)
2

Then with some mathematics:

 
 † 1 1
a, a = √ (q + ip), √ (q − ip)
2 2
1 1 1 1
= √ (q + ip) √ (q − ip) − √ (q − ip) √ (q + ip)
2 2 2 2
1 1
= (q + ip)(q − ip) − (q − ip)(q + ip)
2 2
1
= ((q + ip)(q − ip) − (q − ip)(q + ip))
2
 † 1
a, a = [q + ip, q − ip]
2
1
q 2 − qip + ipq − ipip − q 2 + qip − ipq − ipuo




=
2
1
= (−2qip + 2ipq)
2
= (−qip + ipq)
= i(−qp + pq)
= −i(qp − pq)
 †
a, a = −i[q, p]

66
A.9 First Order Derivative of a Wave Function
We begin with the given equation:
dψ0 −x
(x) = 2 ψ0 (x)
dx x0
Then we state:
ψ0 (x) = eθ
We use this because we want an x (or θ) in leftover equation before ψ,
substitute this gives:
d θ −x θ
e = 2·
dx x0
−x
θ′ · eθ = 2 · eθ
x0
−x
θ′ = 2
x0
To find the original θ we simply take the integral of the θ′ (the derivative of
θ):

− 12 x2


−x
Z
dx =
x0 2 x0 2
−x2
=
2x0 2
Which gives the following wave function:
−x2
ψ0 (x) = e 2x0 2

67
A.10 Three-Dimensional Angular Momentum
Commutator
To find the commutator of two components of the angular momentum, for
example in x and y direction (with Lx and Ly respectively), we have to find
[Lx , Ly ], firstly we defined these:
ˆ Lx = Y Pz − ZPy
ˆ Ly = ZPx − XPz
ˆ Lz = XPy − Y Px

Then this becomes:

[Lx , Ly ] = [Y Pz − ZPy , ZPx − XPz ]

= ((Y Pz − ZPy ) (ZPx − XPz ) − (ZPx − XPz ) (Y Pz − ZPy ))
= (Y Pz ZPx − Y Pz XPz − ZPy ZPx + ZPy XPz ) −
(ZPx Y Pz − ZPx ZPy − XPz Y Pz + XPz ZPy )
= (Y Pz ZPx − ZPx Y Pz ) + (XPz Y Pz − Y Pz XPz ) +
(ZPx ZPy − ZPy ZPx ) + (ZPy XPz − XPz ZPy )
= [Y Pz , ZPx ] + [XPz , Y Pz ] + [ZPx , ZPy ] + [ZPy , XPz ]
= ([Y Pz , ZPx ] − [Y Pz , XPz ]) + ([ZPx , ZPy ] + [ZPy , XPz ])

This is very nice but this can be simplified, to make this clear we do this
separate:

[Y Pz , ZPx ] − [Y Pz , XPz ] [ZPx , ZPy ] + [ZPy , XPz ]

Y Pz ZPx − ZPx Y Pz − Y Pz XPz + XPz Y Pz ZPx ZPy − ZPy ZPx + ZPy XPz − XPz ZPy
Y (Pz ZPx − ZPx Pz − Pz XPz + XPz Pz ) (ZPx Z − ZZPx + ZXPz − XPz Z) Py
Commutator of one operator [XP Pz , Pz ] = 0 Commutator of one operator [Z Z Px , Z ] = 0
Y (Pz ZPx − ZPx Pz ) (ZXPz − XPz Z) Py
Y (Pz Z − ZPz ) Px X (ZPz − Pz Z) Py
Y [Pz , Z]Px X[Pz , Z]Py
Reassembling this, that becomes:

[Lx , Ly ] = Y [Pz , Z]Px + X[Pz , Z]Py

= X[Pz , Z]Py − Y [Z, Pz ]Px

68
We know that the commutator of a Position Operator (example; Z) and the
matching Momentum Operator (example; Pz ) is equal to ih̄:

[Z, Pz ] = ih̄ (A.1)

Which turns the earlier equation into:

[Lx , Ly ] = X[Pz , Z]Py − Y [Z, Pz ]Px

= X(ih̄)Py − Y (ih̄)Px
= ih̄ (XPy − Y Px )

This we defined to be:

XPy − Y Px = Lz
Therefore:
[Lx , Ly ] = ih̄Lz (A.2)

A.11 Lowering Operator on βmin

We start by lowering the lowest value of β (βmin ):

L− |l, βmin ⟩ = 0

Applying the raising operator to this does not make any difference:

L+ L− |l, βmin ⟩ = 0

Now we simply use the definitions of L2 and Lz (and replace l for α):

(L2 − Lz 2 + h̄Lz ) |α, βmin ⟩ = 0

(αh̄2 − βmin 2 h̄2 + βmin h̄2 ) |α, βmin ⟩ = 0
h̄2 (α − βmin 2 + βmin ) = 0
α = βmin 2 − βmin
α = βmin (βmin − 1)

69
A.12 Checking Angular Momentum Operators
In this section we will see if the found operators are indeed correct by
attempting to prove through matrix-calculations that the following equation
still holds:
[Lx , Ly ] = Lx Ly − Ly Lx = ih̄Lz (A.3)
The first step is to find a matrix representation of the needed Lx Ly and Ly Lx
operators. We begin with the first:
    
2 0 1 0 0 −i 0 2 i 0 −i
h̄  h̄ 
Lx Ly = 1 0 1  i 0 −i = 0 0 0
2 2
0 1 0 0 i 0 i 0 −i

And in the same manner,

    
2 0 −i 0 0 1 0 2 −i 0 i
h̄  h̄ 
Ly Lx = i 0 −i 1 0 1 = 0 0 0
2 2
0 i 0 0 1 0 −i 0 i

And so:
 
2i 0 0
h̄2 
[Lx , Ly ] = 0 0 0 
2
0 0 −2i
 
1 0 0
= ih̄2 0 0 0
0 0 −1
= ih̄Lz

70
Appendix B

List of Definite Integrals

∞ √ 1√
Z
xe−x dx = π (see also Gamma function) (B.1)
0 2
Z ∞
a
e−ax cos bxdx = 2 (B.2)
a ⊣ b2
Z0 ∞
b
e−ax sin bxdx = 2 (B.3)
0 a + b2
Z ∞ −ax
e sin bx b
dx = tan−1 (B.4)
0 x a
Z ∞ −ax
e − e−bx b
dx = ln (B.5)
0 x a
Z ∞ r
−ax2 1 π
e dx = for a > 0 (the Gaussian integral) (B.6)
0 2 a
Z ∞ r  2
−ax2 1 π −b
e cos bxdx = e 4a (B.7)
0 2 a
Z ∞ r  Z ∞
1 π b2 −4ac b 2

−(ax2 +bx+c) 2
e dx = e 4n
· erfc √ ,where erfc(p) = √ e−x dx
0 2 a 2 a π p
(B.8)
Z ∞ r 2
π b −4ac

e−(ax +bx+c) dx =
2
e 4a (B.9)
−∞ a
Z ∞
Γ(n + 1)
xn e−ax dx = (B.10)
0 an+1
(B.11)

71
 ∞
Z r
2 −ax2 1 π
xe dx = for a > 0 (B.12)
0 4 a3
Z ∞
2n − 1 ∞ 2(n−1) −ax2
Z r
2n −ax2 (2n)! π
x e dx = x e dx 2n+1 2n+1
for a > 0, n = 1, 2, 3 . . .
0 2a 0 n!2 a
(B.13)
Z ∞
2 1
x3 e−ax dx = 2 for a > 0 (B.14)
2a
Z0 ∞ Z ∞
2 n 2 n!
x2n+1 e−ax dx = x2n−1 e−ax dx = n+1 for a > 0, n = 0, 1, 2 . . .
0 a 0 2a
(B.15)
Z ∞ m+1


2 Γ 2
xm e−ax dx = m+1
(B.16)
0 2a( r2 )
Z ∞
6 1 π −2√ab
e(−ax − x2 ) dx =
2
e (B.17)
0 2 a
Z ∞
x π2
dx = ζ(2) = (B.18)
0 ex − 1 6
Z ∞ n−1
x
dx = Γ(n)ζ(n) (B.19)
0 ex − 1
Z ∞
x 1 1 1 1 π2
dx = 2 − 2 + 2 − 2 + · · · = (B.20)
ex + 1 1 2 3 4 12
Z0 ∞
sin mx 1 m 1
dx = coth − (B.21)
0 e2πx − 1 4 2 2m
Z ∞ 
1 −x dx
−e = γ (where γ is Euler-Mascheroni constant)
0 1+x x
(B.22)
Z ∞ −x2
e − e−x γ
dx = (B.23)
0 x 2
Z ∞
e−x

1
− dx = γ (B.24)
0 ex − 1 x
Z ∞ −ax
e − e−bx 1 b2 + p 2
dx = ln 2 (B.25)
0 x sec px 2 a ↓ p2
Z ∞ −ax
e − e−bx b a
dx = tan−1 − tan−1 (B.26)
0 x csc px p p
(B.27)

72
 ∞
e−ax (1 − cos x) a  a2 + 1 
Z  
−1
dx = cot a − ln  2 
x2 2 a
Z0 ∞
2 √
e−x dx = π
Z−∞
∞
1 2 (2n + 1)! √
x2(n+1) e− 2 x dx = 2π for n = 0, 1, 2, . . .
−∞ 2n n!

73