Lecture Notes in Quantum Mechanics: June 2017

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/318899706
Lecture Notes in Quantum Mechanics
Book · June 2017
CITATIONS READS
0 36,582
1 author:
Salwa Alsaleh
King Saud University
24 PUBLICATIONS 85 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Geometric flows and implications to quantum gravity View project
All content following this page was uploaded by Salwa Alsaleh on 04 August 2017.
The user has requested enhancement of the downloaded file.

Lecture Notes
in
Quantum Mechanics
Salwa Alsaleh
Department of Physics and Astronomy

College of Science
King Saud University
ii
Contents
1 Review of Classical Mechanics 1

1.1 The Hamiltonian . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Hamilton’s equations . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Poisson brackets . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Phase space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.1 Example: phase-space for SHO . . . . . . . . . . . . . . . . 3
1.4 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Mathematical preliminaries 5
2.1 Abstract vector spaces . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Functions as vectors . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.3 Dual spaces and inner product . . . . . . . . . . . . . . . . . . . . . 7
2.3.1 The Bra-Ket notation . . . . . . . . . . . . . . . . . . . . . 7
2.3.2 Normed spaces . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Hilbert space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4.1 Basis of a Hilbert space . . . . . . . . . . . . . . . . . . . . 8
2.5 The space L2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.6 Outer product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.7 Linear operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.7.1 The algebra of operators . . . . . . . . . . . . . . . . . . . . 11
2.8 Spectral theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.9 Projection operators . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.9.1 The identity operator . . . . . . . . . . . . . . . . . . . . . . 13
2.9.2 Projection operators . . . . . . . . . . . . . . . . . . . . . . 13
2.10 Unitary operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.11 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.11.1 Rotation in the Euclidean 2-D space . . . . . . . . . . . . . 14
2.11.2 The differential operator . . . . . . . . . . . . . . . . . . . . 14
2.12 Commutators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.13 Function of operator . . . . . . . . . . . . . . . . . . . . . . . . . . 16
iii
iv CONTENTS
2.14 Commuting operators . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.15 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3 Postulates of Quantum Mechanics 19

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 The postulates of quantum mechanics . . . . . . . . . . . . . . . . . 19
3.2.1 The first postulate . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.2 The second postulate . . . . . . . . . . . . . . . . . . . . . 20
3.2.3 The third postulate . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.4 The forth postulate . . . . . . . . . . . . . . . . . . . . . . 21
3.3 Measurement in quantum mechanics . . . . . . . . . . . . . . . . . 22
3.4 Example: Energy states . . . . . . . . . . . . . . . . . . . . . . . . 23
3.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4 The Wavefunction 25
4.1 Position representation . . . . . . . . . . . . . . . . . . . . . . . . 25
4.2 Separation of variables in Schrödinger’s equation . . . . . . . . . . . 26
4.3 The free-particle solution . . . . . . . . . . . . . . . . . . . . . . . . 27
4.4 Probability flux and density . . . . . . . . . . . . . . . . . . . . . . 28
4.5 The Born conditions . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5 The Uncertainty Principle 33

5.1 The uncertainty principle for the free particle . . . . . . . . . . . . 33
5.2 General uncertainty principle . . . . . . . . . . . . . . . . . . . . . 35
5.3 Uncertainty in time and energy . . . . . . . . . . . . . . . . . . . . 36
5.4 Ehrenfest theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
6 Basic Scattering Theory 41

6.1 Remarks on the quantum scattering theory . . . . . . . . . . . . . . 41
6.2 Scattering/Reaction channels . . . . . . . . . . . . . . . . . . . . . 42
6.3 Scattering by potentials in 1-D . . . . . . . . . . . . . . . . . . . . 42
6.4 The optical theorem in 1-D . . . . . . . . . . . . . . . . . . . . . . 43
7 Stationary States 45
7.1 Particle in a box . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
7.1.1 Solution to Schrödinger’s equation . . . . . . . . . . . . . . . 46
7.1.2 Momentum eigenfucntions . . . . . . . . . . . . . . . . . . . 48
7.2 Simple Harmonic Oscillator . . . . . . . . . . . . . . . . . . . . . . 49
7.2.1 Quantization of the SHO Hamiltonian . . . . . . . . . . . . 49
CONTENTS v
7.2.2 The eigenfunctions . . . . . . . . . . . . . . . . . . . . . . . 51

7.2.3 Coherent states . . . . . . . . . . . . . . . . . . . . . . . . . 52
7.3 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
7.3.1 Particle in a box . . . . . . . . . . . . . . . . . . . . . . . . 52
7.3.2 Simple Harmonic Oscillator . . . . . . . . . . . . . . . . . . 53
8 Angular Momentum 55
8.1 The Classical angular momentum . . . . . . . . . . . . . . . . . . . 55
8.2 Quantisation of the angular momentum . . . . . . . . . . . . . . . 56
8.3 The spherical harmonics . . . . . . . . . . . . . . . . . . . . . . . . 57
8.4 Properties of the spherical harmonics . . . . . . . . . . . . . . . . . 57
8.5 Angular momentum eigenstates . . . . . . . . . . . . . . . . . . . . 59
8.6 The spectrum of angular momentum observable . . . . . . . . . . . 60
8.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
9 The Hydrogen Atom 63

9.1 Canonical quantisation of the H-atom Hamiltonian . . . . . . . . . 64
9.2 Quantum numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
9.3 Solution of the TISE for the Hydrogen . . . . . . . . . . . . . . . . 65
9.4 Degeneracies in the ideal H-atom . . . . . . . . . . . . . . . . . . . 66
9.5 Spin quantum number . . . . . . . . . . . . . . . . . . . . . . . . . 66
9.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
10 Spin 69
10.1 Transformation of the wavefunction . . . . . . . . . . . . . . . . . . 69
10.2 External and internal degrees of freedom for a system . . . . . . . 70
10.3 Mathematical description for internal degrees of freedom . . . . . . 71
10.4 Discovery of electron’s magnetic dipole moment . . . . . . . . . . . 72
10.5 Electron’s spin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
10.6 Infel-van der Warden Symbols . . . . . . . . . . . . . . . . . . . . . 76
10.7 Matrix representation of spin states . . . . . . . . . . . . . . . . . . 77
10.8 Geometric representation . . . . . . . . . . . . . . . . . . . . . . . . 79
10.9 Spin in constant magnetic field . . . . . . . . . . . . . . . . . . . . 80
10.9.1 Stationary states . . . . . . . . . . . . . . . . . . . . . . . . 81
10.10Electron paramagnetic resonance EPR . . . . . . . . . . . . . . . . 81
10.11Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
A The Virial Theorem 85

vi CONTENTS
B Summery of Hilbert spaces 87

B.1 Representations of Hilbert space . . . . . . . . . . . . . . . . . . . . 87
B.2 Operator properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Chapter 1
Review of Classical Mechanics
1.1 The Hamiltonian

We have seen the Hamiltonian function in classical mechanics; resembling the
energy of the dynamical system in study.1
H(~p, ~q) ≡ Energy of the system. (1.1)
Where, p~ and ~q are the generalised momentum and coordinates( configuration) for
the system. We define the generalised/ canonical momentum as:2
∂L
pi = . (1.2)
∂ q̇ i
With L being the Lagrangian function.3
1.1.1 Examples
1. The Hamiltonian for a free particle is given by:
p2
H= (1.3)
2m
2. In a central potential, the Hamiltonian takes the form :
p2 Q1 Q2
H= +g (1.4)
2m r
Where g is a constant, called coupling constant, and Q1 Q2 are charges/
masses. Depending on the type of interaction.
1
recall that H = T + V
2
Note that p~ here is not always m~v !
3
recall that L = T − V
1
2 CHAPTER 1. REVIEW OF CLASSICAL MECHANICS
3. A dipole in a magnetic field B has the following Hamiltonian:
H=µ ~
~ ·B (1.5)
4
4. The Hamiltonian for a free rotating mass is:
L2
H= (1.6)
2I
Here, L is the angular momentum of the system.
5. Lastly, a very important Hamiltonian, is the Hamiltonian for a simple har-

monic oscillator ( SHO).
P2 mω 2 q 2
H= + . (1.7)
2m 2
1.2 Hamilton’s equations

For a dynamical system with a Hamiltonian. The evolution of that system obeys
the set of equations , known as Hamilton’s equations :5
∂H
= −ṗi (1.8)
∂q i
∂H
= q̇ i (1.9)
∂pi
Some dynamical systems however, does not obey these equations. These systems
are known as Hamiltonian constraint systems.
1.2.1 Poisson brackets

Let f (~q, p~), and g(~q, p~) be functions of both ~q and p~. One may define the following
operation :
X ∂f ∂g ∂g ∂f

{f, g} ≡ i ∂p
− i (1.10)
i
∂q i ∂q ∂p i
This operation is called Poisson brackets, it shall prove importance in quanti-

sation, and also in understanding conserved quantities ( see homework).
4
recall that I = M r2
5
note that both p~ and ~q are functions of time
1.3. PHASE SPACE 3
It is interesting to Poisson bracket the Hamiltonian with components of ~q and p~,

for example :
X ∂q j ∂H ∂H ∂q j
{q j , H} = − i
i
∂q i ∂pi ∂q ∂pi
|{z}
=0
X ∂H
= δij
i
∂pi
|{z}
=q̇ i
j
= q̇ (1.11)
Same goes for:
{pj , H} = ṗj (1.12)
This actually leads us to the general result, that for any function f (~q, p~), we have:
∂f (~q, p~)
{f (~q, p~), H} = (1.13)
∂t
1.3 Phase space

Since one only needs the p’s and q’s 6 in order to fully describe the dynamical
system. The state vector is defined in 2f dimensional space, called the phase
space.
~ ≡ (~q(t), p~(t))
S(t) (1.14)
1.3.1 Example: phase-space for SHO

This phase space for the SHO in 1-D is resembled by a circle: The circle resembles
all the possible states the SHO can take, at any moment in time. In other words,
this describes the full evolution of the system in time.
Since energy is conserved for the free SHO, the system will keep being in the circle.
As for damped one, when energy is lost. The system will spiral until it comes to
a halt.
1.4 Problems
1. Show that the Hamiltonian for a particle of mass m, orbiting a mass M , and
they interact gravitationally is gien by:
L2 Mm
H= +G
2I r
6
One needs as many of them as the number of degrees of freedom f for the system
4 CHAPTER 1. REVIEW OF CLASSICAL MECHANICS
Figure 1.1: (a) The phase space for free SHO. (B) The phase space of a damped
SHO
Then derive the equations of motion for this system, comment on your re-
sults.
2. Derive and solve the equation of motion for a 2-D SHO, with m = 1 and
ω = 1.
3. Show that :
∂pj
{pj , H} =
∂t
4. Draw the shape of the phase space for a particle free-falling from altitude y0
5. Discuss conserved quantities, and Neother’s theorem in light of Hamiltonian

dynamics.
Chapter 2
Mathematical preliminaries
2.1 Abstract vector spaces

In previous courses, the notion of a vector was introduced as being an n-tuple of
ordered numbers ( either real or complex). However, one can have a more general
definition for a vector as being an element of a vector space. A vector space is a
set that satisfies the following property
Let V be a vector space, and |ψi and |φi are elements of it . Then :
α|ψi + β|φi, (2.1)
is also an element of that vector space, where α and β are complex or real numbers.
The expression (2.1) is called the superposition of the vectors |ψi and |φi.
The dimension of V could either be finite , countably infinite or uncountably infinite
( see next section). For finite dimensional- or countably infinite- vector spaces. It
is possible to represent a vector |ψi as a column matrix :
 
f1
 f2 
|f i ⇔  f  (2.2)
 
 3 
..
.
We can represent the components of the above vector in what-so-called a spike

diagram representing the magnitude of each component of the vector |ψi.
2.2 Functions as vectors

It may seem unfamiliar to most readers that functions could be considered as
vectors of an uncountably infinite dimensional vector space. In order to see this,
5
6 CHAPTER 2. MATHEMATICAL PRELIMINARIES
Figure 2.1: Spike digram of a vector in 3 dimensional vector space, and 30 dimen-
sional one.
we use the spike diagrams discussed above.This time using a continuous parameter
x taking real-number values instead of the discrete index i. The spike digram
for such vector would look like: In fact, this spike digram looks familiar to the
Figure 2.2: Spike digram of a vector having infinite components
classical graph of a real-valued function f (x): in this way, a function is just a
Figure 2.3: Classical graph of a function f (x)
single vector in an infinite dimensional space. It should be noted that to make the
transition to infinite dimensions mathematically meaningful, you need to impose
some smoothness constraints on the function. Typically, it is required that the
function is continuous, or at least integrable in some sense. These details are not
important for our purpose, thus we shall not discuss them further.
2.3. DUAL SPACES AND INNER PRODUCT 7
2.3 Dual spaces and inner product

One may define a map hf | that sends a vector |φi in V to the real or complex
numbers. Defined as - for finite dimensional vector spaces or countably infinite
ones: X
hf |φi = fi φi (2.3)
i
1
or for functions : Z
hf |φi = dxf (x)∗ φ(x) (2.4)
We can easily show that the set of such maps form a vector space themselves, sometimes they
which we call the dual vector space and denoted by V ∗ . Moreover, the operation are called linear
between a vector and a (dual) vector is called inner product . functionals
2.3.1 The Bra-Ket notation

Every vector space has its dual. In the notation adopted in quantum mechanics,
and invented by Paul Dirac. A member of a vector space is called a Ket and
denoted by |ψi, as we have seen, for discrete components it is represented as a
column matrix .On the other hand, the elements of the dual space are denoted by
a Bra hf | in Dirac notation. And represented as a row matrix: 2
hf |⇔ f1∗ f2∗ f3∗ · · ·

(2.5)
The notation adopted is known as the Bra-Ket notation . Since for every vector
space there is a dual space. One may turn a Ket vector into a Bra vector, by a
one-to-one map. Hence defining the inner product in the vector space itself, calling
it an inner-product space.
Two vectors |ψi and |φi are called orthogonal if and only if :
hφ|ψi = 0
2.3.2 Normed spaces

In a similar sense to the magnitude of a vector in the space ( like velocity), we can
extend this notion to any vector with the norm function. k · k . There are many
ways one can define the norm of a vector. However, we are only interested in the
norm defined by the inner product:
p
kψk ≡ hψ|ψi (2.6)
1
The star resembles the complex conjugation f (x)∗
2
The inner product is carried like matrix multiplication between a row and a column
Note that the norm is always a real number . Vectors with a unit norm is called
normal vectors .
2.4 Hilbert space

An inner product vector space H, with a norm defined in (2.6), in addition to
another property called completeness 3 is called a Hilbert space. Hilbert spaces
are extremely important in quantum mechanics. They replace the phase spaces of
classical mechanics. Hilbert spaces can be finite dimensional or infinite dimensional
( both cases we call it separable Hilbert spaces), the latter infinite dimensional
space might have either continuous and discrete representations.
2.4.1 Basis of a Hilbert space

We can express any vector in the Hilbert space as a linear superposition of dim H
other vectors. Hence, it is possible to construct orthonormal basis for a Hilbert
space. You need as many of them as the dimension of the vector space itself. Given
a vector |ψi in the Hilbert space having a set of orthonormal basis S = {|ei i}i , |ψi
is uniquely expressed in terms of the basis S :
X
|ψi = hei |ψi |ei i (2.7)
i
The coefficients cj = hej |ψi is sometimes called Fourier coefficients. We can

have the same argument for the Bra-vectors:
X
hψ| = hψ|ei i hei | (2.8)
i
The basis satisfy4 :

hei |ej i = δij (2.9)
For function spaces, the basis have continuous parameter x instead of a discrete
index the basis therefore are denoted by |xi, for all x a real number. A function φ
is written as :
φ(x) = hx|φi
3
complete vector space is a space that all Cauchy sequences converge
4
There is a fundamental difference by what we mean by basis here and the conventional basis
tough in Classical mechanics The basis we use are called Schauder basis, whilst the latter is
known as Hamel basis
2.5. THE SPACE L2 9
Instead of using the Bra-Ket notation for functions, we shall only denote them by
φ(x).
Note that we need some-sort of structure in the Hilbert space to insure that (2.7)
converges if the sum is infinite. We call the Hilbert space in with the series of
this type the `2 space, or the space of square-summable sequences. Moreover, the
integrals used in this course - and in quantum mechanics in general- are known
as the Lebesgue integrals , they are different from Riemann integrals defined in
calculus courses.
Here, we have illustrated the most relevant properties of Hilbert spaces that con-
cern us in the study of introductory quantum mechanics. A lot of mathematical
details and rigour has been spared in this lecture. It is urged from the reader to
conduct a further reading in the theory of Hilbert space; please consult: Functional
Analysis by M. Reed and B. Simon.
2.5 The space L2

Functions form a vector space, as we have seen previously, nevertheless, we need
to imply an additional restriction on functions in order to form a Hilbert space.
Let φ(x) be a function defined on the interval [a, b], the norm of this function is
defined to be - in accordance to the formal definition of the norm- :
s
Z b
kφ(x)k = φ(x)∗ φ(x) dx (2.10)
a
For the norm to exist, the function φ(x) needs to be square integrable on the
interval [a, b]. It is not hard to show that the set of square-integrable functions
on the same interval form a Hilbert space. This Hilbert space is known as the
L2 (R; dµ) space. It reads; the space of square-integrable function on the inter-
val/Region R 5 , with respect to the measure dµ. By measure we mean the volume
element that we integrate over. In 1-D case dµ = dx. Sometimes, one wishes to
define a weight w(x) for the space, but this is out of the scope of this course.
The space L2 can have basis of orthogonal, and normalised functions (u1 (x), u2 (x), . . . )
depending on the interval of interest. For example the classical orthogonal poly-
nomials including:
• Hermite polynomials:
2 dn −x2
Hn (x) = Kn−1 ex e . (2.11)
dxn
5
Functions of L2 could be of several variables, real or complex-valued
They could form basis for the space L2 (−∞, +∞; dx).
• Legendre polynomials:
dn
Lνn = Kn−1 x− νex xν+1 ex .

n
(2.12)
dx
2
They could form basis for the space L (0, +∞; dx).
• Legendre polynomials ( of the first kind):
dn
Pn = Kn−1 1 − x2

n
(2.13)
dx
They could form basis for the space L2 (0, 1; dx).
The second example ofL2 spaces, are the ones used in Fourier analysis. Where
cos(nkx) and sin(nkx) form an orthonormal basis for a given L2 space with an
interval L. Recall that we can expand any function in a Fourier series :
∞
X nπx nπx
f (x) = An cos( ) + Bn sin( ) (2.14)
n=0
L L
Where :
Z L
nπx
An = f (x) cos(
)dx (2.15)
0 L
Z L
nπx
Bn = f (x) sin( )dx (2.16)
0 L
Or expanding the function in a continuous basis ( Fourier integral).
Note Advanced readers might not find the discussion in this lecture formal nei-
ther accurate enough, as discussing mathematical rigour of Hilbert spaces is very
distant from the course aims.
2.6 Outer product

For a finite-dimensional vector space, the outer product can be understood as
simple matrix multiplication:
   
φ1 φ1 ψ1∗ φ1 ψ2∗ · · · φ1 ψN∗
 φ2 
.   ∗ ∗ ∗
 φ2 ψ1∗ φ2 ψ2∗ · · ·
 φ2 ψN∗ 
|φi hψ|=  ..  ψ1 ψ2 · · · ψN =  ..

.. .. .. 
 .   . . . . 
∗ ∗ ∗
φN φN ψ1 φN ψ2 · · · φN ψN
The outer product is an N×N matrix.
2.7. LINEAR OPERATORS 11
2.7 Linear operators

The term ’linear operator’ is used in many contexts, in quantum mechanics how-
ever, we are interested -mainly- in linear operators, acting on a Hilbert space.
Linear operators are maps from a Hilbert space to itself ( known mathematically
as Endomorphisms ). In simple words, they send ’kets’ to ’kets’. Operators 6 are
represented as square matrices ( for finite dimensional or countably infinite dimen-
sional Hilbert spaces). Hence, all the algebra of matrices will apply to operators.
Such as :
2.7.1 The algebra of operators

1. Linearity:
Let Â and B̂ be operators acting on the Hilbert space H,α and β are scalars,
and |ψi and |φi be vectors in H. Then the following properties hold:
(αÂ + β B̂)|ψi = α(Â|ψi) + β(B̂|ψi) (2.17)
Moreover:
Â(α|ψi + β|φi) = α(Â|ψi) + β(Â|φi) (2.18)
A result from above : X
Â|ψi = hi|ψi Â|ii (2.19)
i
2. Eigenvalue:
A scalar λ is called an eigenvalue if it satisfied the equation:
Â|ψi = λ|ψi (2.20)
Called the eigenvalue equation, and the vector |ψi is called an eigenket/
eigenvector. This equation is equivalent to :

det Â − λIˆ = 0 (2.21)
For Iˆ or just I being the identity operator. An important result from this
property is the spectral decomposition, that we shall discuss later in this
lecture.
3. Self-adjointness ( Hermitian operators)

Let Â be an operator, then Â† is the hermitian conjugate of this oper-
ator It is simply the hermitian matrix in the matrix representation of Â .
6
we shall always mean linear operator when we use the term operator from now on
The hermitian conjugate acts on the dual space of H ( acts on the Bras).
The following properties for the hermitian conjugation are listed below ( for
reminding)
†
• Â† = Â – involutiveness
• If an inverse for the operator exists then:

† −1
Â−1 = Â†
• Antilinearity:
†
αÂ + β B̂ = α∗ Â† + β ∗ B̂ †
• (ÂB̂)† = B̂ † Â† .
An operator is called self-adjoint , if it is equal to its hermitian conjugate:
Â† = Â
In other words, it acts both on the Kets and on the Bras. An important
theorem for self-adjoint operators is stated below:
All the eigenvalues for a self-adjoint operator are real
Other properties of the matrices can be revised from a linear algebra book.
2.8 Spectral theorem

An important result from linear algebra is the spectral decomposition of an opera-
tor in terms of its eigenvalues and eigenvectors. If the set eigenvectors {|u1 i, |u2 i, |u3 i · · · }
form a basis for the Hilbert space, and the set of eigenvalues {λi } satisfying:
Â|uj i = λj |uj i (2.22)
Then the operator Â is decomposed as follows:

X
Â = λi |ui ihui | (2.23)
i
Note that the outer product |ui ihui | is gives the identity matrix / operator I. 7
Hence, we may diagonalise the operator Â if we have found all of its eigenvalues.
7
If two eigenkets have different eigenvalues they ought to be orthogonal
2.9. PROJECTION OPERATORS 13
As for uncountably-infinite dimensional Hilbert spaces , which we refer to by:

inseparable Hilbert spaces, the spectral theorem reads :
Z
Â = dµ(λ)λ (2.24)
where, dµ(λ) is the integration measure that depends on the nature of spectrum
for the measurement outcomes in the theory. We are not going to go further in
the details, as they are beyond the scope of our course.
2.9 Projection operators

2.9.1 The identity operator
From the commutativity of Kets with (complex) scalars now follows that
X
|ei ihei | = Iˆ (2.25)
i∈N
must be the identity operator, which sends each vector to itself. This can be
inserted in any expression without affecting its value, for example
X X X
hv|wi = hv| |ei ihei |wi = hv| |ei ihei | |ej ihej |wi = hv|ei ihei |ej ihej |wi
i∈N i∈N j∈N
(2.26)
2.9.2 Projection operators

We define the projection operator P̂α for a normalised vector |αi,as :
P̂α ≡ |αihα| (2.27)
. Observe that the projection operator is self-adjoint. and satisfies the identity:
P̂α2 = |αihα||αihα| = |αihα| = P̂α (2.28)
.
2.10 Unitary operators

An operator Û is called unitary if it preserves the inner product for two vectors,
and thereby the norm . This also can be stated as:
ˆ
Û † Û = Û Û † = I. (2.29)
Therefore, a unitary, self-adjoint operator is its own inverse .
2.11 Examples
2.11.1 Rotation in the Euclidean 2-D space
A vector in the 2-D plane is represented by :

x
|ri = (2.30)
y
The basis vectors are:

1 0
|ex i = , |ey i = (2.31)
0 1
We can define a rotation operator R̂(ϑ), that acts on the vector |ri by rotating it
with an angle ϑ. This operator has a matrix representation:

cos ϑ − sin ϑ
R̂ = (2.32)
sin ϑ cos ϑ
Figure 2.4: The rotation in 2-D space, carried put by the rotation operator.
2.11.2 The differential operator

There are operators also acting on function space, they have the same properties
as the operators discussed above, but with slight modifications. For example, the
operators acting on function space cannot have an explicit matrix representation.
2.12. COMMUTATORS 15
The most famous operators which act on function space are the differential oper-
ators; denoted by L̂. There are a variety of differential operators. They play an
important rôle in the theory of differential equations. In fact, most of the prob-
lems in quantum mechanics are related to the analysis of the differential operators
related to dynamical observables; as we shall see.
Take the function f (x) = eλx . It is the eigenfunction of the differential opera-
d
tor , with an eigenvalue λ. Hence, we conclude that eλx , is a solution to the
dx
differential equation:
d
(f (x) = λf (x) (2.33)
dx
d2
The operator 2
has two eigenfunctions e+λx and e−λx they resemble solutions
dx
for the differential equation :
d2
(f (x) = λf (x) (2.34)
dx2
And by the superposition principle, a general solution would be:
f (x) = Ae+λx + Be−λx (2.35)
2.12 Commutators
Just like ordinary matrix multiplication, the product between operators is generally
non-commutative. In fact, this particular property of operator multiplication is
behind the unfamiliar phenomena observed in quantum mechanics, thus generally
we have :
ÂB̂ 6= B̂ Â (2.36)
We define the commutator between two operators as:
[Â, B̂] ≡ ÂB̂ − B̂ Â (2.37)
The commutator satisfies the following properties:

[αA + βB, C] = α[A, C] + β[B, C] linearity in both slots.
[A, [B, C]] + [B, [C, A]] + [C, [A, B]] = 0 Jacobi Identity For the last prop-
[AB, C] = [A, C]B + A[B, C] Product rule
erty, it is a rule of thumb to think of the commutator as a kind of derivative
DC = [·, C]:
DC (AB) = DC (A)B + ADC (B)
2.13 Function of operator

Just like scalars, one can a function of an operator: f (Â) This is justified because,
one can expand the function f (Â) as a series:
f2
f (Â) ≈ f0 I + f1 Â + ÂÂ + . . . (2.38)
2!
Since operator product is defined, the function itself is well-defined as well
Commutator of a function is given by :
!
df (B̂)
[f (B̂), Â] = [B̂, Â] (2.39)
dB̂
Provided that
[[B̂, Â], Â] = 0
Another important formula to learn is the Hadamard Lemma:
1
eÂ B̂e−Â = B̂ + [Â, B̂] + [Â, [Â, B̂]] + . . . (2.40)
2!
2.14 Commuting operators

For two commuting operators,
[Â, B̂] = 0 (2.41)
one can find a common set of eigenbasis :
Â|ii = ai |ii (2.42)

B̂|ii = bi |ii (2.43)
The eigenkets form a mutual eigenbasis for the states Â|ψi and B̂|ψi. Hence one
can simultaneously diagonalise both operators.
2.15 Problems
1. Find the eigenvalues of the operator :
 
1 i 0
Â = −i 2 −i
0 i 1
Can we diagonalise it ?.
2.15. PROBLEMS 17
2. Is the operator d/dx acting on the L2 Hilbert space hermitian ? How about
−id/dx ?
3. Show that for a Hermitian operator T̂ the operator Û ≡ eiT̂ is unitary.
4. Is the function cos kx an eigenfunction for the operator d/dx ? What about
the operator d2 /dx2 ?
5. Show that for an operator Â commuting with the Hamiltonian operator

Ĥ we can write its time evolution as:
eiĤt Âe−iĤt
Hint: use Hadmard lemma This is known as Heisenberg equation
6. We define the expected value - average value- of an operator with respect to

a state vector |ψi as:
hÂi ≡ hψ|Â|ψi
Show that if |ψi is expanded in terms of the eigenbasis of Â, then the expected
value is the sum of the eigenvalues of the operator an , i.e.
X
hÂi = c n an
n
for some constants cn .
7. Recall that we defined the identity operator as the outer product of the basis
: X
I= |iihi|
i
Using this definition, show that this operator sends the ket |ψi to itself (
does not change the ket), then show this for the Bra vector hψ|.
8. Discuss why if [Â, B̂]|ψi 6= 0 one cannot find a mutual eigenbasis to expand
|ψi with ?
Chapter 3
Postulates of Quantum Mechanics
3.1 Introduction
In modern physics course and in modern physics lab, we have seen the clear motiva-
tion for using quantum mechanics in place of classical physics for the fundamental
description of nature. Classical physics is merely an approximation for the real
picture of the quantum world .
One of most important lessons learnt from quantum mechanics is the fact that mea-
surement affects the system, no matter how hard ones tries to avoid such effect, it
shall remain present ( there are exceptions known as weak measurements). That
implies that the order of measuring dynamical quantities of the system matters,
hence non-commutativity is the heart and soul of quantum mechanics. Max
Born tried to make an interpretation for these observable facts. His interpreta-
tion states that the quantum system is non-deterministic and each outcome of the
measurement has its own probability. Moreover, the system before measurement
takes a superposition of all of its possible states. The experimental motivation
for these statements are assumed to be known by the reader.
3.2 The postulates of quantum mechanics

An axiomatic approach for quantum mechanics is followed in these lectures. The
fundamental properties of quantum mechanics are derived from these postulates,
as we shall see for the uncertainty principle for example.This approach is more
modern and provides a deeper understanding of quantum mechanics. We start
by having a classical system , that owns a Hamiltonian H and described by a
phase space M. We also can have a Poisson brackets {·, ·}. In order to make the
quantum leap we have to make the following changes:
19
20 CHAPTER 3. POSTULATES OF QUANTUM MECHANICS
3.2.1 The first postulate

The phase space for the system is changed into a Hilbert space :
M −→ H
The state of the system becomes a ‘Ket’|ψi in H instead of a vector in the phase
space as we have seen in lecture (1). This abstract vector is unlike the state vector
for the classical system has no direct physical meaning . As one can multiply this
vector by any complex/real number and get the same state for physical system
. 1 . Hence if |ψi describes the system , then a|ψ for a a complex/real number
describes the same state for the physics system, This however has rare exceptions
( known as Berry phase). Therefore, one can normalise the state vector. such that
it satisfies:
hψ|ψi = 1 (3.1)
In the probabilistic picture, this reads out as:

X X
hψ|ψi = |ψi |2 = P (i)
i i
X
⇔ P (i) = 1 (3.2)
i
Surely the sum of all the probabilities of the possible measurable quantities for the
system is ought to equal one. This postulates implies what Max Born suggested for
the statistical nature of quantum mechanics, or more generally the superposition
principle ( the quantum system takes all of its possible configurations when not
Recall measured ). .
Schrödinger’s
cat!
3.2.2 The second postulate
Any dynamical observable for a classical system ω(p, q) is defined to be a func-
tion on the phase space. Upon quantisation, these observables will be resembled
by a linear self-adjoint (hermitian) operators acting on the Hilbert space
Ω̂. Measurement is expressed mathematically by acting the operator correspond-
ing to the physical observable on the state vector. The possible values (outcomes)
for a measurement is the set of eigenvalues for that operator. This a direct physical
result from the spectral theorem.
1
In fact, the physical state is described by a ray in H, not a one vector.
3.2. THE POSTULATES OF QUANTUM MECHANICS 21
3.2.3 The third postulate

Because observables are linked to operators, this inherits non-commutativity in
quantum mechanics. The third postulate of quantum mechanics is known as the
canonical quantisation postulate. We replace the Poisson brackets with com-
mutators.
[·, ·] ←→ i~{·, ·} (3.3)
2
In particular the for Q̂ and P̂ , the configuration space/ position and momentum
operators will satisfy the canonical commutation relation:
[Q̂i , P̂j ] = i~δji Iˆ (3.4)
We have learnt that if two operators do not commute, then one cannot have a
mutual set of complete eigenbasis to simultaneously diagonalise them . This phys-
ically means that one cannot measure both observables with absolute accuracy.
Leading to the uncertainty principal.
3.2.4 The forth postulate

The fourth postulate discusses the time evolution of a quantum system. The time
evolution of a state is dictated by Schrödinger’s equation.
∂
Ĥ|ψ(t)i = i~ (|ψ(t)i) (3.5)
∂t
Note here we have the states being time-dependent, whilst operators are not.
Sometimes the operators themselves are time-dependent not the states. The first
’picture’ is known as Schrödinger’s picture, and the second is Heisenberg picture.
Both pictures are physically equivalent. The equation of motion for the opera-
tors - in Heisenberg picture- is known as Heisenberg equation. We have seen in
the tutorials, that for an operator Â in Schrödinger’s picture, we can write it in
Heisenberg picture by Hadmard formula:
i i
Â(t)Hesenberg = e ~ Ĥt ÂSchrödinger e− ~ Ĥt (3.6)
Using the correspondence between Poisson brackets, and quantum commutators

(3.3). One can directly arrive to the Heisenberg equation by quantising Hamil-
ton’s equations:
!
d i ∂ Â
Â(t) = [Ĥ, Â(t)] + eiHt/~ e−iHt/~ (3.7)
dt ~ ∂t
2
note that every two operators would commute when ~ −→ 0
3.3 Measurement in quantum mechanics
We have used the word ’measurement’ a lot in this lecture, it may be confusing for
the reader what is really meant by it within the context of quantum mechanics.
One may picture a ’physicist’ in a lab who intend to measure a quantum system
when thinking of the word measurement. However, measurement may not involve
any experiment of actual detectors, rather simply it is any interaction between the
quantum system and a classical object, which we call the apparatus.
In pure quantum mechanical view, there is no meaning for a path of a particle, as
Figure 3.1: Measurement in quantum mechanics means any interaction between a

quantum system and a classical object
defining a path requires an absolute knowledge of the momentum and position of

that particle simultaneously. Surely this is impossible by the uncertainty principle.
Moreover, there is no meaning for the speed in quantum mechanics, because - by
definition- speed needs knowing the position of a particle exactly at each moment
in time. This has no meaning in quantum mechanics, because of the measurement
problem. Yet, one may define a velocity in a different perspective.
These reflections into the postulates of quantum mechanics lead us into the conclu-
sion that in pure quantum mechanical world, there is no meaning for any dynamics
we are familiar with from the classical physics. They appear in quantum mechan-
ics merely due to measurement. Hence, we don not only need classical mechanics
are the limit of quantum mechanics for macroscopic systems. Moreover, for the
construction of quantum theory itself !
3.4. EXAMPLE: ENERGY STATES 23
3.4 Example: Energy states

A bound electron is found to have discrete energies. The state for the electron is
expanded in terms of the Hamiltonian eigen-energy states:
∞
X
|ψi = αn |En i (3.8)
n=1
Such that αn is the probability amplitude for the nth 3

energy state and |En i
satisfies the eigenvalue problem:
Ĥ|En i = En |En i (3.9)
which is -in fact- schrödinger’s equation. We can write the Hamiltonian in matrix
form :  
E1 0 0 ... 0 ...
 0 E2 0 . . . 0 . . . 
 
 0 0 E3 0 0 ... 
 
 0 0 · · · En 0 . . . 
Ĥ =  (3.10)
 .. .. .. .. . . 
 . . . . . ... 

0 0 0 0 ... 
 
.. .. .. .. .. . .
. . . . . .
If the measurement resulted the particle having the energy state Ej . The state
vector |ψi is then projected into the eigenstate |Ej i casing of what-so-called the
wavefunction collapse :
hEj |ψi = αj (3.11)
We may also calculate the expected-value for the energy:
hEi = hψ|Ĥ|ψi (3.12)
expanding this as we learnt:

∞
X
hψ|Ĥ|ψi = |αn |2 En (3.13)
n=1
3.5 Problems
1. Wilson’s chamber is a sealed environment containing a supersaturated vapour,
when a charged particle - like an electron or alpha particle- passes through
the chamber. It leaves a track of cloud behind it. Discuss why we can ’see’
Figure 3.2: Alpha particles from a Radium source in a cloud chamber, notice the
path of the particle
the quantum particle’s path in this case although we have stated there is no
path defined for quantum particles ?
2. If we let the position operator be a multiplicative one; i.e Q̂i = q i ; show
that -in order to satisfy the commutation relation discussed in the lecture
∂
the momentum operator needs to be P̂j = ~i j .
∂q
∂
3. From the previous problem, why we can’t use ~ . As a definition for the
∂qj
momentum operator, and −iq i as the position operator ?
4. Which is ‘bigger’ the phase space or the Hilbert space ? Provide a supporting
argument for your answer .
5. In a thought experiment , imagine having a quantum coin. A coin which
obeys the laws of quantum mechanics. Describe it mathematically
6. An electron can take 3 possible energy states E1 = 0.5eV , E2 = 1.2eV and
E3 = 1.6eV . With probabilitie: P1 = 0.8, P2 = 0.13 and P3 = 0.07.
(a) Write the Hamiltonian in matrix form.
(b) Write the normalised eigenbasis
(c) Find hEi and σ(E).
(d) What is the state ket after measuring the system and finding it taking
the second energy state ?
(e) Show that hψ|ψi = 1
3
Remember that |αn |2 = P (En )
Chapter 4
The Wavefunction
4.1 Position representation

We start naturally from the postulates of quantum mechanics, having the state ket
|ψ(t)i ( in Schrödinger’s picture) that evolves in time by Schrödinger’s equation:
∂
Ĥ|ψ(t)i = i~ |ψ(t)i (4.1)
∂t
What we are interested in knowing for a free particle is its position, we therefore
project the state ket into the position space , and get the wavefunction:
ψ(x, t) = hx|ψ(t)i (4.2)

def
The Hilbert space is therefore : H = L2 ( ] − ∞, +∞[ ; dx ) 1 The position
operator is simply a multiplicative operator :
X̂ψ(x, t) = xψ(x, t) (4.3)
It has a continuous spectrum of eigenvalues being the position(s) of the quantum

particle in the 1-dimensional space. Generalisation to 3-D space is straightforward
- see homework- . The task now is to find the momentum operator in the position
representation, this can be done from investigating the canonical commutation
ˆ as following:
relation [X̂, p̂x ] = i~I,
[X̂, p̂x ]ψ(x, t) =i~ψ(x, t)

xp̂x ψ(x, t) − p̂x (xψ(x, t)) =i~ψ(x, t) (4.4)
1
This Hilbert space implies that the particle could be anywhere in the universe until measured;
but the propability vanishes at infinity points
25
26 CHAPTER 4. THE WAVEFUNCTION
Rearranging the above equation :
p̂x (xψ(x, t)) = −i~ψ(x, t) + xp̂x ψ(x, t)

~ ∂
⇔ p̂x (xψ(x, t)) = (xψ(x, t)) (4.5)
i ∂x
Hence we conclude that the momentum operator in the position representation is
given by :
def ~ ∂
p̂x = (4.6)
i ∂x
4.2 Separation of variables in Schrödinger’s equa-

tion
Now, we attempt to quantise the free particle system. This is done first by defin-
ing the wavefunction and the Hilbert space. Now what is left is to quantise the
Hamiltonian; recall that the Hamiltonian for the free particle is given by:
p2
H(p) = (4.7)
2m
p̂2
The Hamiltonian operator that acts on the free-particle Hilbert space is Ĥ = ,
2m
since we have the position representation for the momentum operator p̂. We have
the Hamiltonian operator:
def ~2 ∂ 2
Ĥ = − (4.8)
2m ∂x2
Plunging it in the Schrödinger’s equation; to obtain:
~2 ∂ 2 ψ(x, t) ∂ψ(x, t)
− 2
= i~ (4.9)
2m ∂x ∂t
Observe the similarity between Schrödinger’s equation and the classical wave equa-
tion. However, we have derived this equation from an axiomatic approach. In order
to solve (4.9) we need to use a mathematical trick known as the separation of
variables. Assume that we can write the wavefunction as the product of two
functions:
ψ(x, t) = ϕ(x)h(t) (4.10)
Substituting in (4.9), and rearranging, we obtain:
~2 1 d2 ϕ(x) 1 dh(t)
− 2
= i~ (4.11)
2m ϕ(x) dx h(t) dt
4.3. THE FREE-PARTICLE SOLUTION 27
Each side of the equation (4.11) depends only on one variable, and since they equal
each other. This implies :
~2 1 d2 ϕ(x)
− = Const. (4.12a)
2m ϕ(x) dx2
1 dh(t)
i~ = Const. (4.12b)
h(t) dt
Evidently, the ’constant’ is indeed the eigen-energy of the particle E. We start
by solving the second equation (4.12b), the solution yields, what-so-called the
stationary states time evolution:
h(t) = e−iωt . (4.13)
with ω = E~ . Observe this result can be obtained directly from the Heisenberg
picture ( Show how !) All systems of which we can separate their time dependence
in this way is called stationary states. They shall be the main focus in these notes.
4.3 The free-particle solution

We now turn to the spacial part of Schrödinger’s equation
~2 d2 ϕ(x)
− = Eϕ(x) (4.14)
2m dx2
This has a particular solution of the form
u(x) = Ceikx (4.15)
With k 2 = 2mE
~2
, having the dimension of inverse length; we recognise k being the
wavenumber. The solution (4.15) can be written in terms of the momentum - by
the relation p = ~j.-:
i
u(x) = Ce ~ px (4.16)
This solution is known as the plane wave solution. It represents a wave propagating
in the +x direction. This solution can be used to find ϕ(x) by the superposition
principle; the ’constant’ of integration is not constant in fact. Rather, it is a
function of p . In order to see this, recall that the eigen-energy of the free particle
2
p2
E = 2m = (~k)
2m
putting this in (4.9) we shall have a continuous spectrum of
eigen-energies. Then use the spectral theorem we shall have therefore :
Z +∞
1 i
ϕ(x) = √ ϕ̃(p)e ~ px dp (4.17)
2π~ −∞
Which is simply the Fourier transform of the momentum wavefunctions ϕ̃(p) =

i
hp|ϕi . We then identify the plane-wave solution as hx|pi = e ~ px . We can use
the wavefunctions to calculate the probability of finding the particle as a given
position x0 , |ϕ(x0 )|2 or momentum p0 , |ϕ̃(p0 )|2 .
Since momentum cannot be determined with absolute certainty; it then takes a
Gaussian wavefunction. Of which its Fourier transform is a Gaussian function
itself. The figure 4.3 is generated by a code we have made showing the time
evolution of the wavefunction ψ(x, t), given an initial width for the momentum
wavefunction Gaussian : Observe how the wavefunction - indicating our certainty
Figure 4.1: Simulated time evolution of a Gaussian wavefunction for the free par-
ticle, observe the dispersion of the wave as time progresses
of the particle’s position- gets wider and wider as time progresses. The equation
(4.17) indicates that the position wavefunction is composed of infinite number of
momentum wavefunctions. The movement of the quantum particle is therefore
expressed in terms of a wavepacket resulting from infinite number of waves inter-
fering; we have run a computer code representing the wavepacket as seen in figure
4.3 :
4.4 Probability flux and density

Since |ψ(x, t)|2 gives the probability of finding the particle at any point in space,
at a given time. It defines a positive real-valued function of space. It is known as
the pprobability density function:
ρ(x, t) = ψ ∗ (x, t)ψ(x, t) (4.18)

4.4. PROBABILITY FLUX AND DENSITY 29
Figure 4.2: The wavepacket of the free quantum particle ψ(x, t)
We also define the probability current density

∂ψ ∗ (x, t)

~ ∗ ∂ψ(x, t)
j(x, t) = − ψ(x, t) + ψ (x, t) (4.19)
2mi ∂x ∂x
∂j(x, t)
let’s now look at :
∂x
∂ψ ∗ (x, t)

∂j(x, t) d ~ ∗ ∂ψ(x, t)
= − ψ(x, t) + ψ (x, t)
∂x dx 2mi ∂x ∂x
2 ∗ ∗
~ ∂ ψ (x, t) ∂ψ (x, t) ∂ψ(x, t)
= − 2
ψ(x, t) − +
2mi ∂x ∂x ∂x
∂ 2 ψ(x, t) ∂ψ ∗ (x, t) ∂ψ(x, t)

∗
ψ (x, t) +
∂x2 ∂x ∂x
2 ∗
∂ 2 ψ(x, t)

~ ∂ ψ (x, t) ∗
= − ψ(x, t) + ψ (x, t) (4.20)
2mi ∂x2 ∂x2
We can also calculate ρ̇(x, t):
∂
ρ̇(x, t) = (ψ ∗ (x, t)ψ(x, t))
∂t
∂ψ ∗ (x, t) ∂ψ(x, t)
= ψ(x, t) + ψ ∗ (x, t) (4.21)
∂t ∂t
2
∂ψ(x, t) −~ ∂ ψ(x, t)
But from Schrödinger’s equation, we have = 2mi . Hence we
∂t ∂x2
incur:
2 ∗
∂ 2 ψ(x, t)

~ ∂ ψ (x, t) ∗
ρ̇(x, t) = ψ(x, t) − ψ (x, t) (4.22)
2mi ∂x2 ∂x2
∂j(x, t)
⇔ + ρ̇(x, t) = 0 (4.23)
∂x
This is the continuity equation in quantum mechanics , implying probability is
conserved .
4.5 The Born conditions

Max Born’s best known contribution to quantum mechanics was his proposal that
the wave function, or rather its square modulus, should be interpreted as the
probability density for finding the system in a given state at a given time. However,
he also proposed four conditions on the wave function which are used in finding
many solutions of the Schrödinger equation. As always, it’s useful to take another
look at the Schrödinger equation (in one dimension (4.14)) so we can see how
Born’s conditions fit in.
Born’s conditions to be imposed on the wave function ψ(x, t) are:
1. he wave function must be single valued. This means that for any given values
ofx and ψ(x, t) must have a unique value. This is a way of guaranteeing that
there is only a single value for the probability of the system being in a given
state.
2. The wave function must be square-integrable. In other words, the integral

of |ψ|2 over all space must be finite. This is another way of saying that it
must be possible to use |ψ|2 as a probability density, since any probability
density must integrate over all space to give a value of 1, which is clearly not
possible if the integral of |ψ|2 is infinite. One consequence of this proposal
is that ψ must tend to 0 for infinite distances.
3. The wave function must be continuous everywhere. That is, there are no
sudden jumps in the probability density when moving through space. If a
function has a discontinuity such as a sharp step upwards or downwards,
this can be seen as a limiting case of a very rapid change in the function.
Such a rapid change would mean that the derivative of the function was very
large (either a very large positive or negative number). In the limit of a
step function, this would imply an infinite derivative. Since the momentum
4.6. PROBLEMS 31
of the system is found using the momentum operator, which is a first order
derivative, this would imply an infinite momentum, which is not possible in
a physically realistic system. Such an infinite derivative would also violate
condition 4.
4. All first-order derivatives of the wave function must be continuous. Follow-

ing the same reasoning as in condition 3, a discontinuous first derivative
would imply an infinite second derivative, and since the energy of the system
is found using the second derivative, a discontinuous first derivative would
imply an infinite energy, which again is not physically realistic.
4.6 Problems
1. Calculate hpi and hEi for the free particle
2. We define the ’group velocity’ vg for a wavepacket as:
dω
vg =
dk
Show that this equals the ’classical velocity’ for the particle.
3. We define the ‘phase velocity’ vp as:

ω
vp =
k
Show it equals half of the classical velocity.
4. If the momentum wavefunction ϕ̃(p) → δ(p − p0 ). Write down ψ(x, t) explic-

itly. What do you observe ?
5. Show that in Heisenberg picture, we arrive to the same stationary states

Et
solution e−i ~
6. Show that the free particles could have both k > 0 and k < 0, i.e e−ikx is
also a solution to the free particle. What does that mean physically ?
7. Write Schrödinger’s equation for the free particle in 3 dimensions.

Chapter 5
The Uncertainty Principle
5.1 The uncertainty principle for the free parti-

cle
In the previous lecture, we have derived the expression for the wavefunction of the
free particle, for a Gaussian momentum wavefunction with width σ it is given by:
σ2 2
+∞
e− ~2 (p−p0 ) eipx/~
Z 2
p t
− 2m~
ψ(x, t) = dp e 2 √ (5.1)
−∞ ( 2π~
4σ 2 )1/2 h
This integral is a Gaussian itself. Nevertheless, the width of the position Gaussian
is inversely proportional to the momentum one, in particular :
σ(x)σ(p) ∼ ~ (5.2)
This is a form of the celebrated uncertainty relation for position and momentum,
which is directly derived from solving the free particle problem. The physical
meaning for this relation is that in order to have a well-defined position for the
wave-packet ( sharp width), we need to superimpose wide momentum wavefunc-
tions ; and vice versa ( figure 5.1). Implying the impossibility for having an
absolute accurate measurement for either momentum and position regardless of
the ’apparatus’ used. Moreover, the increase in the accuracy in one implies the
decrease in the other quantity.
1
This principle is what -alone- explains the stability of atoms, counter to what
the classical theory of electrodynamics predicts 2 . Because electrons are bound
within the atomic radius ∼ 1 ÅThey have to possess a velocity uncertainty of
1
The uncertainty principle is a fundamental property in nature unrelated to our methods of
measurement
2
In classical electrodynamics, an accelerating charge emits electromagnetic radiation
33
34 CHAPTER 5. THE UNCERTAINTY PRINCIPLE
Figure 5.1: The position wavefunction is made from interference of infinite number
of momentum wavefunctions
Figure 5.2: A real picture of Hydrogen atom; taken by special techniques. Indicat-
ing the probability of electron’s position around the nucleus, in complete agreement
with quantum theory. Reference:Stodolna, A. S., et al. Phys. Rev. lett. 110.21
(2013)
about ∼ 3 × 103 m/s or few eV c2 of energy. For the electron to fall into the
nucleus, i.e. being bound by the nuclear radius ∼ 10−5 Å; it requires an enormous
energy ∼ 22 MeV or more. Thus, electrons are forbidden from falling into the nu-
cleus. Figure 5.2 shows how the uncertainty principle, and schrödinger’s equation
agrees completely with nature, the picture illustrates a real hydrogen atom’s elec-
tron wavefunction indicating the probability of finding the electron in the vicinity
of the atom.
5.2. GENERAL UNCERTAINTY PRINCIPLE 35
5.2 General uncertainty principle

We can show - mathematically- a more general form of the uncertainty principle
for position and momentum starting from the axioms of quantum mechanics; in
particular the canonical commutation relation. We start by defining the variance
of an operator V ar(A) 3 , as :
V ar(A) = |(A − hAi)2 | = σ(A)2 (5.3)
In the position representation this equals to dxψ ∗ σ(A)2 ψ. We define the Schwartz
R
inequality :
|A|2 |B|2 ≥ |AB|2 (5.4)
For A = σ(X) and B = σ(P ); we have :
V ar(X) V ar(P ) = |hX P i|2 (5.5)
But the product XP can be expressed as :
1 1
XP = [X, P ] + (X P + P X) (5.6)
2 2
Proof :
Start by Writing :
1 1
XP = XP − P X + XP
2 2
Moreover :
1 1 1 1
XP = P X + XP − P X + XP
2 2 2 2
Gathering terms, we obtain :
1 1
XP = [X, P ] + (XP + P X)
2 2
Substituting this result in (5.5); and knowing [X, P ] = i~ we obtain:
1 1
V ar(X)b V ar(P ) ≥ | i~|2 + |(hP Xi) + hXP i|2 (5.7)
2 4
Since both hP Xi and hXP i are equal, and the expression 41 |(hP Xi) + hXP i|2 is
non-negative . We may write:
3
We shal drop the ’hat’ whenever it is understood we are talking about an operator
~2
V ar(X) V ar(P ) ≥ (5.8)
4
Taking the square root :
~
σ(X)σ(P ) ≥ (5.9)
2
In fact, this result can be generalised for any two non-compatible observables A
and B:
1
σ(A)σ(B) ≥ |h[A, B]i| (5.10)
2
This is the general form of uncertainty principle.
5.3 Uncertainty in time and energy

In modern physics, we were introduced to the quantum harmonic oscillator.
Although we shall revisit it soon, but we may recall an import result. Which is
the zero-point energy:
1
E0 = ~ω (5.11)
2
Although the harmonic oscillator is in its ground state, yet it has an amount of
energy inversely proportional to its period of oscillation. If the period is dependent
of measurement - like any quantum mechanical observable- to assign the period
with accuracy, we need to measure time with precision. This implies a large ω
of the oscillator that should be used for measurement since T ∝ ω1 . That implies
adding a lot of energy to the system - even if it was in the ground state- by the
above equation (5.11). In fact this translates to :
~
σ(E)σ(t) ≥ (5.12)
2
This principle holds for any quantum system, not just the harmonic oscillator.
However, it is easier to see in this case.
The time-energy uncertainty principle has rather deep implications, in particular
for the law of energy conservation. One can ’trade’ energy with time, by adding
energy -from nothing- to the system provided it is for short period of time. This is
an important phenomena observed a lot in particle physics. For example the weak
interaction which allow beta radiation ( neutrons decaying into protons or vice
versa). A particle called W- boson , having about 80 times the mass of proton is
created from nothing, but for a very short time ∼ 10−18 sec. Just enough to allow
the weak interaction to occur.
5.4. EHRENFEST THEOREM 37
Figure 5.3: A diagram showing the weak interaction mediated by the W boson
which is created from nothing by the uncertainty principle
5.4 Ehrenfest theorem

As we have seen, classical mechanics is obtained from averaging quantum me-
chanical results. For example the conservation of energy is an average result from
the uncertainty principle discussed above. Therefore, we need to prove that from
quantum mechanics we can arrive to classical equations of motion. This is Ehren-
fest theorem.
The classical observable ω is the expected-value of the quantum mechanical oper-
ator Ω̂:
ω = hΩ̂i = hψ|Ω̂|ψi (5.13)
Let’s take the time derivative of the expected-value d/dt(hΩ̂i):
d
hψ|Ω̂|ψi
dt
by the product rule:
∂hψ| ∂ Ω̂ ∂|ψi
Ω̂|ψi + hψ| |ψi + hψ|Ω̂
∂t ∂t ∂t
1 ∂|ψi
Now by substituting in Schrëdinger’s equation i~
Ĥ|ψi = ∂t
, and its complex
conjugate. We obtain :
d 1 ∂ Ω̂ 1
hψ|Ω̂|ψi = − hψ|Ĥ Ω̂|ψi + hψ| |ψi + hψ|Ω̂Ĥ|ψi
dt i~ ∂t i~
This is equivalent to :
d 1 ∂ Ω̂
hΩ̂i = h[Ω̂, Ĥ]i + h i (5.14)
dt i~ ∂t
This is the exact mathematical formula for Ehrenfest theorem. Now, we turn to
applying this theorem to the operators p and X, in order to recover the classical
equations of motion:
˙ = 1 h[p, V (x)]i
hpi
i~
p2
since p is time-independent, and the Hamiltonian takes the form : H = 2m +V.
Working in the position representation, the above commutator is written explicitly
as : Z Z
∗ ∂ ∂
− dxψ (V ψ) + dxψ ∗ V (ψ)
∂x ∂x
Applying the product rule to the first expression :
Z Z Z
∗ ∂ ∗ ∂ ∂
− dxψ (V ) ψ − dxψ V (ψ) + dxψ ∗ V (ψ)
∂x ∂x ∂x
Thus we get:
˙ = −h ∂V i
hpi (5.15)
∂x
Which is Newton’s second law, or one of Hamilton’s equations ( recall that F =
− dV
dx
)
∂V
h[p, H]i = −h i (5.16)
∂x
Let’s now apply Ehrenfest theorem. to the position operator X:
2
˙ = 1 h[X, p ]i
hXi
i~ 2m
since X is also time-independent, and the potential is only a function of position.
We need to use the result from lecture (3), equation (22)
df (B)
[A, f (B)] = [A, B]
dB
with A = X, B = p and f (p) = p2 ; we obtain:
˙ = 1 d
hXi h[X, p] (p2 )i
2im~ dp
⇔
1
= (i~)(2hpi)
2im~
Finally :
˙ = 1 hpi
hXi (5.17)
m
The definition for the classical velocity, or the second Hamilton’s equation:
1
h[X, H]i = hpi (5.18)
m
5.5. PROBLEMS 39
5.5 Problems
1. We define the first ionisation energy , as the energy needed to free an
electron from the vicinity of its atom. For Carbon 14, it is found; by detailed
quantum mechanical calculations and experimental verification that the first
ionisation energy is Emin = 11.3 eV.
(a) Estimate using the uncertainty principle Emin , if you know the electron
in the C(14) atom is confined to a box of x = 0.182 nm. and E = p2 /2m
(b) Provide an explanation, from the uncertainty principle for not observing
’protons’ being confined to atoms; use the last calculations on C(14) as
a guide.
2. What is the origin of electrons in beta decay ?

Beta (β − ) decay is a nuclear decay process that turn a neutron into a pro-
ton. This happens when the number of neutrons are imbalanced, hence the
nucleus decays by emitting an electron and another particle called the anti-
neutrino in order to be stable.
In the early days of nuclear physics, and because of the beta decay, it was
argued that the nucleus is composed of electrons and protons. Thus electrons
emitted in the beta decay are bound in the nucleus. This is known as the
proton-electron hypothesis.
Provide an argument against this hypothesis from the uncertainty principle.
Take the C(14) decay into N(14) as an example. The nuclear radius for
C(14) is 5.8 × 10−15 m and the energy of the emitted electron is at maximum
Emax = 0.016 MeV.
3. Einstein’s Recoiling Slit Experiment.

Einstein was known for his opposition against the quantum theory, although
he admitted its validity. One of the arguments Einstein has made against
the quantum theory - in particular the uncertainty principle- is a ’modified’
version of the double-slit experiment . This full thought experiment is con-
structed as the figure shows. However, we shall take a simpler one. The
experiment goes as follows :

Imagine having a narrow slit, with small width d, a quantum particle passes
though this slit . This let us detect its position by d, as a result, this causes
a momentum uncertainty of ∼ 2d ~
. However, Einstein suggested that we can
measure the particle’s momentum from the recoil of the screen of which the
particle collides with after passing through the slit. If the screen is free to
move in the x direction, by the laws of momentum conservation, the recoil
of the screen is related to the incident particle momentum. In fact, the ex-
tended version of this thought experiment is done experimentally cf.(Nature
Photonics 9, 120–125 (2015)).
Of course, Einstein has missed something in his thought experiment. Niels
Böhr had provided an answer to Einstein’s experiment defending the uncer-
tainty principle. Can you guess what was that argument ?
4. Prove Ehrenfest theorem in Heisenberg picture.
(pictures are from Hyperphysics website, http://hyperphysics.phy-astr.gsu.

edu/hbase/hframe.html)
Chapter 6
Basic Scattering Theory
6.1 Remarks on the quantum scattering theory

Scattering problem in quantum theory is a vast topic, it is the second problem in
quantum mechanics along with bounded states one. There are two main cases for
a an interacting quantum particle; the first is when it’s bounded to a potential.
The particle takes a discrete energy spectrum . The second case, is when the
particle is scattered of a potential, and takes a ccontinuous spectrum of
energy.
Hence, we inscribe the incident wavefunction o ψ(x, t) by a free particle wave-packet
( as plane wave). The scatterer, is a stationary potential, that elastically scatter off
the incident wave-packet ( in our simplified problem), i.e. the energy is conserved.
Then the scattered wave shall be a spherical wavefunction out ψ(~r, t), we care about
Figure 6.1: Sketch of the scattering problem in quantum mechanics. Notice that
we use the wave nature of the quantum particles
the probability for a scattering with a certain solid angle Ω. This is calculated from
41
42 CHAPTER 6. BASIC SCATTERING THEORY
the differential cross-section see figure 6.1:

dσ Nφ,θ
≡ o (6.1)
dΩ j
Where Nφ,θ is the probability of detecting the scattered particle by a detector
placed at angles φ and θ . And o j is the incident probability flux current. i.e:
o
∂ o ψ(x, t)∗

o ~ o ∗ ∂ ψ(x, t) o
j= ψ(x, t) − ψ(x, t) (6.2)
2mi ∂x ∂x
However, we shall restrict ourselves to one dimensional scattering problems, where
the differential cross section calculation is not the major problem. Rather we are
interested in the probability of transmission and reflection of the incident flux.
6.2 Scattering/Reaction channels

In general, for two particles a and b interacting in a scattering problem. We have
many possible reaction channels; listed as follows:
Reaction channel Name

a + b −→ a + b Elastic scattering
a + b −→ a + b∗ Inelastic scattering
a + b −→ c + d Rearrangement collision
a + b −→ c Absorption
a −→ b + c Decay
Table 6.1: Classification of the reaction channels in a tow-particle scattering prob-

lem
6.3 Scattering by potentials in 1-D

A moving wave-packet with an energy E approaching some potential variations
illustrated in figure . Will have the general solution for their time-independent
Schrödinger equation (TISE):
ψL (x) = Ae+ikx + Be−ikx (6.3a)

ψR (x) = Ce+ikx + De−ikx (6.3b)
With:
2mE
k2 =
~2
6.4. THE OPTICAL THEOREM IN 1-D 43
The previous set of equations can be put in a vector form as follows:

out o B S11 S12 A
Ψ=S· Ψ ⇔ = . (6.4)
C S21 S22 D
This equation resembles the scattering amplitude and S is known as the S-matrix,
and it plays an important rôle in scattering problems. As the S-matrix elements
characterises the full properties of the scattering process.
Recall the current density in (6.2) is conserved, i.e.:
jR = jL (6.5)
That implies:
|A|2 − |B|2 = |C|2 − |D|2 (6.6)
Since:
~k
|A|2 − |B|2

jL = (6.7a)
m
~k
|C|2 − |D|2

jR = (6.7b)
m
Since we have in our problem the incident flux is from the left only ⇒ D = 0. We
have thereby:
2 2
C 2
B
TL = = |S21 | RL = = |S11 |2
A A
Or similarly for flux approaching from right A = 0 :

2 2
B 2
C
TR = = |S12 | RR = = |S11 |2
B D
With T and R are the transmission and reflection probabilities respectively.

Note that:
TL(R) + RL(R) = 1 (6.8)
6.4 The optical theorem in 1-D

Consider a free particle (V = 0), we can write the its S-matrix as:

0 1
S= , (6.9)
1 0
44 CHAPTER 6. BASIC SCATTERING THEORY
taking the canonical form. When (V 6= 0), S should be written according to the
continuity conditions. Let :
ψ(x)L = eikx + re−ikx ψ(x)R = teikx (6.10)
Satisfying the conditions ( V changes at x = 0):
dψ(x)L dψ(x)R
ψ(x = 0)L = ψ(x = 0)R |x=0 = |x=0 (6.11)
dx dx
We obtain the expression for the S-matrix:

2ir 1 + 2it
S= , (6.12)
1 + 2it 2ir∗ 1−2it
1+2it
∗
Or:
|r|2 + |t|2 = =(t) (6.13)
Which is the optical theorem in one dimension.
Chapter 7
Stationary States
7.1 Particle in a box

We discuss here an application to the mathematical axioms of quantum mechanics
studied earlier to a simple, yet important problem. The quantum particle in an
infinite potential well, or a particle in a box see figure 7.1. This is an idealisation
Figure 7.1: A diagram illustrating the particle in a box problem
for a large enough potential compared to the particle’s energy. This problem is
very important example to study discrete spectrum .
We start by a particle trapped in a potential well, of width a. The Schrödinger’s
equation for this particle is written as - in position representation-:
∂ ~2 ∂ 2
i~ ψ(x, t) = − ψ(x, t) + V (x)ψ(x, t) (7.1)
∂t 2m ∂x2
With : (
0, for 0 < x < a
V (x) = (7.2)
∞, otherwise
45
46 CHAPTER 7. STATIONARY STATES
Since the above equation is clearly separable, similar to the free particle. It resem-
bles a stationary state. We then write the eigenvalue problem:
~2 ∂ 2
u(x; E) + V (x)u(x; E) = Eu(x; E) (7.3)
2m ∂x2
The energy eigenfunctions are then equal to ψ(x, t; E) = u(x; E) e−i~ωt
7.1.1 Solution to Schrödinger’s equation

We now attempt to solve (7.3) subject to the boundary conditions:
u(x = 0) = u(x = a) = 0 (7.4)
Because there is a null probability of the particle being outside the box. And the
second condition on the wavefunction :
du(x) du(x)
|x=0 = |x=a (7.5)
dx dx
This condition comes from naturally from the analysis of the problem. The first
derivative of the wavefunction is proportional to the momentum of the particle,
we expect the particle will ’bump’ with both walls in the same manner. Although,
this violates the Born conditions discussed in lecture 5, but keep in mind that
the infinite well is an unphysical example!
Now, we rewrite Schrödinger’s equation as:
d2 u(x)
+ k 2 u(x) = 0 (7.6)
dx2
√
with k = 2mE~; this differential equation is solved by the substitution u(x) =
eRx Resulting:
u(x) = Aeikx + Be−ikx (7.7)
Where A and B are constants, we use the identity :
We observe that, using the boundary conditions we obtain :
ik Aeika − Be−ika = ik(A − B)

⇒ u(x) = C sin(kx) (7.8)

u(a) = C sin(ka) = 0
nπ
⇒k = (7.9)
a
7.1. PARTICLE IN A BOX 47
Figure 7.2: First,second, and third lowest-energy eigenfunctions (red) and associ-
ated probability densities (blue) for the infinite square well potential
√
But since k = 2mE~, we conclude that energy takes discrete values :
~2 n2 π 2
En =
2ma2
=E0 n2 (7.10)
~ π2 2
With E0 = 2ma 2 , the ground energy. And n here denotes the quantum number
for the excited states of the particle in the box. Now, we mayqwrite the energy-
eigenfunctions, after calculating the normalisation factor C = a2 eiϕ 1
r
2 nπx i(ωt+φ)
ψn (x, t) = sin( ) e (7.11)
a a
The total wavefunction is written as, by the superposition principle :

r ∞
2 X nπx i(ωt+φ)
ψ(x, t) = sin( ) e (7.12)
a n=1 a
Which is a Fourier series ( we could have obtained this solution immediately by

Fourier analysis ). The following figure demonstrates the eigenfunctions for various
excitation states, and the probability density function ρ:
1
we can always add an arbitrary phase factor eiϕ
7.1.2 Momentum eigenfucntions

We now turn into calculation the Fourier transform of u(x) in order to compute
the momentum eigenfucntions φ(p, t):
Z a
1 i
v(p) = √ e ~ px u(x)
dx |{z} (7.13)
2π~ 0 |{z}
hp|xi hx|ui
substituting with u(x) we have:

Z a
1 i
v(p) = √ dx e ~ px sin(n~px) (7.14)
πa~ 0
2
Evaluation of this integral gives:
i
!
1 − (−1)n e− ~ pa
r
aπ
v(p) = n 2 (7.15)
~ n2 π 2 − a2 ~p2
Plotting the momentum probability density function - for various excitations - :

This calculation ends the basic analysis for a particle in a box.
Figure 7.3: Momentum probability density function for the ground state and three
more excitation states.
2
The following identity was used :
ax
ax
sin bx dx = a2e+b2 (a sin bx + b cos bx)
R
e
7.2. SIMPLE HARMONIC OSCILLATOR 49
7.2 Simple Harmonic Oscillator

7.2.1 Quantization of the SHO Hamiltonian
From lecture (1) we have the classical Hamiltonian for the simple harmonic oscil-
lator (SHO) :
1 2 1
H(p, x) = p + mω 2 x2 (7.16)
2m 2
Using the postulates of quantum mechanics discussed before, we obtain- upon
quantization - the Hamiltonian operator
1 2 1
Ĥ = P̂ + mω 2 X̂ 2 (7.17)
2m 2
The Hilbert space of which X̂ and P̂ act on is
L2 (−∞, +∞; dx)
We now introduce the dimensionless Hamiltonian

1 1 mω 2
Ĥ 0 = P̂ 2 + X̂ (7.18)
2m~ω 2 ~
This operator can be factorised and written in terms of ’creation’ and ’inhalation
’operators; a† and a respectively
1
Ĥ 0 = a† a + I (7.19)
2
with:
r r
mω 1
a= X̂ + i P̂ (7.20a)
2~ 2mω~
r r
† mω 1
a = X̂ − i P̂ (7.20b)
2~ 2mω~
These operators, along with Ĥ 0 , satisfy a well-known commutation relations. 3
[a, a† ] = I (7.21a)
[a, H 0 ] = a (7.21b)
[a† , H 0 ] = −a† (7.21c)
3
The operators a, a† and H 0 along with the commutator operation [·, ·] satisfy the su(1, 1)
algebra.
We also define the number operator N ≡ a† a that acts on the eigenstates |ni
resulting an eigenvalue of n :
N |ni = n|ni
as a result we may conclude that
a|0i = 0 (7.22)
acting on the ’ground state’by the inhalation operator, kills it. Moreover
√
a|ni = n|n − 1i (7.23)
p
a† |ni = (n + 1)|n + 1i (7.24)
Hence, The Hamiltonian acting on these states will result (the energy eigen-
value)
1
Ĥ|ni = ~ω(n + )|ni (7.25)
2
Implying that the ’number states’ are the excitation states for the quantum har-
monic oscillator. The creation and inhalation operators excite or deceit it, and it
has a discrete energy spectrum of :

1
En = ~ω n + (7.26)
2
Even in the ground state, the quantum harmonic oscillator has a non-vanishing
energy. This is a direct result for the uncertainty principle in time and energy.
Figure 7.4: Energy-levels and wavefunctions of the quantum harmonic oscillator

7.2. SIMPLE HARMONIC OSCILLATOR 51
7.2.2 The eigenfunctions

Since we introduced the eigenstates for the Hamiltonian ( or the number operator
equivalently). |ni. We can use the ladder operator method to solve Scrödinger’s
equation in the position representation and find ψn (x) = hx|ni.
we start from (7.22):
aψ0 (x) = 0 (7.27)
resulting the differential equation,

~ d
x+ ψ0 (x) = 0 (7.28)
mω dx
having the solution:
mω 2
ψ0 (x) = Ae− 2~ x (7.29)
We can find the normalisation factor by:
Z +∞
mω 2 1
e− 2~ x dx = . (7.30)
−∞ |A|2
Which is a typical Gaussian, hence A is found to be,
mω 1/4
A= (7.31)
π~
Now, in order to find the nth wavefunction ψn (x), we first need to prove the
following identity
(a† )n
|ni = √ |0i (7.32)
n!
Proof:
a† (a† )2 (a† )n
√ | n − 1i = p | n − 2i = · · · = √ |0i.
n n(n − 1) n!
Thereby,
(a† )n
ψn (x) = √ ψ0 x (7.33)
n!
Writing the expression explicitly , we obtain :
1
ψn (x) ≡ hx | ni = √ π −1/4 exp(−x2 /2)Hn (x) (7.34)
n
2 n!
Where Hn (x) is the nth Hermit polynomial, that having the generating formula
(Rodrigues’s formula)
n
n x2 d 2
Hn (x) = (−1) e n
e−x (7.35)
dx
They are one of the classical orthogonal polynomials.
Note that this result can be obtained by solving immediately the Schrödinger’s
equation (using series solution, or Sturm-Liouville theorem ).
7.2.3 Coherent states

Coherent states are a very important topic in quantum mechanics. Coherent states
are quantum states that display an oscillatory behaviour similar to the one dis-
played in the simple harmonic oscillator. Formally, a coherent state is a state that
has a minimum uncertainty, and takes the form ( where a here is the inhalation
operator) :
a|αi = α|αi (7.36)
The ground state |0i is a coherent state. Since it has the minimum ucertainty :
~
h∆Xi0 h∆P i0 = (7.37)
2
This is not however the case for the stationary states |ni, as we can show that:
~
h∆Xin h∆P in = (2n + 1) (7.38)
2
Hence the states αi, the coherent states of the Harmonic oscillator are not station-
ary states. Their time evolution is important in the classical limit, as it leads to
Ernfest theorem, and one can obtain from them the Classical equations of motion
for the Harmonic oscillator. Hence the classical harmonic oscillator is the limit
for the coherent states |αi, not the states |ni. However the detailed mathematical
argument is beyond the scope of this course, the interested reader might want to
refer to any Textbook in the references for details.
7.3 Problems
7.3.1 Particle in a box
1. Using the uncertainty principle for position and momentum, estimate the
ground state energy for an infinite well of width a, compare the obtained
result with the one found in the lecture.
2. Compute hEi, hpi and hxi for the particle in a box.
3. Show thatqthe normalisation factor for the particle in a box wavefunction is

given by a2
4. Show that the eigenfunctions un (x) are orthogonal .
5. What is the ground state, first excited and second excited states energies for
an electron trapped in an infinite well of width 1 Å.
7.3. PROBLEMS 53
7.3.2 Simple Harmonic Oscillator

1. Express the position and momentum operators X and P in terms of a and
a†
2. Verify the commutation relations of the triple H 0 , a and a† stated in the

lecture notes, recall the canonical commutation relation [X, P ] = i~I.
3. Find hXi , hP i , hX 2 i and hP 2 i .
4. A quantum harmonic oscillator, in the 2nd excited state, having an energy

of 2.45eV , find its angular frequency , and period.
5. Find the eigenfunction ψ1 (x), and show that it is orthogonal to ψ0 (x) seen
in the lecture.
6. From problem (3), verify the uncertainty relation for position and momentum
.
√
7. Verify that a† |ni = n + 1|n + 1i
8. Write the operators a and a† as matrices acting on the vector states :

     
1 0 0
0 1 0
|0i = 0 |1i = 0 |2i = 1 . . .
     
     
.. .. ..
. . .
Chapter 8
Angular Momentum
8.1 The Classical angular momentum

Recall that the angular momentum for a system is given by :
X
~ =
L ~ri ∧ p~i (8.1)
i
With ∧ being the cross ( wedge) product between the position ~ri and linear mo-
mentum p~i of the ith degree of freedom in the system. For a single particle in 3 D
we give a precise definition for the angular momentum :

x̂ ŷ ẑ
~ = x y z

L (8.2)
px py p z
Figure 8.1: Illustration for the angular momentum of a classical rotating particle
55
56 CHAPTER 8. ANGULAR MOMENTUM
8.2 Quantisation of the angular momentum

One can apply the canonical quantisation for the angular momentum observable
and turn in into a vector operator, since it is defined -classically- by the coordinates
and linear momenta :
[xi , pj ] = i~δji [xi , xj ] =0 [pi , pj ] = 0 (8.3)
Hence the angular momentum operators are ( dropping the hat) :
Lx = −i~ (y∂z − z∂y )

Ly = −i~ (z∂x − x∂z ) (8.4)
Lz = −i~ (x∂y − y∂x )
Or we may write them in the spherical coordinates (r, ϕ, θ) :
Lx = i~ (sin ϕ∂θ + cot θ cos ϕ∂ϕ )

Ly = i~ (− cos ϕ∂θ + cot θ sin ϕ∂ϕ ) (8.5)
Lz = −i~∂ϕ
We also define the operator :
L2 = L2x + L2y + L2z (8.6)
It is a formidable, yet straightforward task to prove the following commutation

relations:
[Lx , Ly] = i~Lz [Ly , Lz ] = i~Lx [Lz , Lx ] = i~Ly (8.7a)

[L2 , Li ] = 0 For all i = x, y, z (8.7b)
We can also define the ( rising and lowering) operators:
L± = Lx ± iLy (8.8)
Along with Lz = L3 they satisfy a well-known commutation relations, known as

the su(2) algebra :
[L+ , L− ] = 2~L3 (8.9a)

[L± , L3 ] = ∓~L± (8.9b)
The rising and lowering operators are expressed in the coordinate representa-
tion as :
L± = ±e±iϕ (∂θ ± i cot θ∂ϕ ) (8.10)
8.3. THE SPHERICAL HARMONICS 57
8.3 The spherical harmonics

As we maintain in the coordinate representation, we wonder about the eigen-
function for the operator L3 and their properties. The operator L3 has a special
importance over the other two angular momentum operators, as the latter ones
compose the ladder( rising and lowering) operators. We start by assuming such
wavefunction :
L3 Y`m (θ, ϕ) = mY`m (θ, ϕ). (8.11)
With m and ` are eigenvalues, and m takes an integer values between ` and −`,
this shall be made clear in the next lecture. However, at the meantime, we just
accept these as given facts.
Therefore, we conclude that :
Y`m (θ, ϕ) = eimϕ y`m (8.12)
Moreover, the fact thatL± y`±` = 0 gives us the differential equation:
(∂θ − ` cot θ) y`±` = 0 (8.13)
Whose complete solution gives us the explicit expression of the functions Ym` , which
are known as the Spherical Harmonics
s
2` + 1 (` + m)! imϕ `
Y`m = (−1)m e Pm (θ) (8.14)
4π (` − m)!
With Pm` (θ) is the associated Legendre Polynomial.

The spherical harmonics form a complete orthonormal basis for the Hilbert space
:
H = L2 (S 2 , dΩ)
With S 2 being the unit sphere, and dΩ = dφ sin θdθ, the solid angle element of the
unit sphere.
8.4 Properties of the spherical harmonics

• Eigenvalue for L2 :
L2 Y`m (θ, ϕ) = `(` + 1)Y`m (θ, ϕ). (8.15)
• Orthonormality :
Z
Y`m
1
1
(θ, ϕ)Y`∗m
2
2
(θ, ϕ)dΩ = δm1 ,m2 δ`1 ,`2 (8.16)
angles
• Since the spherical harmonics form an orthonormal basis, the product of two
of them is again expressed in terms of spherical harmonics. Take the product
Y`m
1
1
(θ, ϕ) · Y`m2
2
(θ, ϕ), we can directly conclude that the resultant product is
a multiple of the spherical harmonics having M = m1 + m2 since the term
containing m is only an exponential, moreover, L taking the range |`1 − `2 | ≤
L ≤ ||`1 + `2 |. The general rule for multiplication is given by Wigner 3j-
symbols ( or Clebsh-Gordon coefficients) C(`1 , m1 ; `2 , m2 ; L, M )that shall be
studied later in the addition of angular momenta.
r
X (2`1 + 1)(2`2 + 1)(2L + 1)
Y`m 1
(θ, ϕ) · Y`m 2
(θ, ϕ) =
1 2
M,L
4π

`1 `2 L ∗M `1 `2 L
× YL (θ, ϕ)
m1 m2 M 0 0 0
(8.17)
Where the Racah symbol is expressed in terms of 3j symbol

`1 `2 L 1
= (−1)`1 −`2 −M √ C(`1 , m1 ; `2 , m2 ; L, −M ) (8.18)
m1 m2 M 2L + 1
These relations will prove useful as we discuss the addition of angular mo-
menta.
• The Herglotz generating function

If the quantum mechanical convention is adopted for the Y`m , then,
∞ X
` r
va·r
X 4π r` v ` λm
e = Y m. (8.19)
2` + 1 (` + m)!(` − m)! `
p
`=0 m=−`
with
λ 1
a = ẑ − (x̂ + iŷ) + (x̂ − iŷ) (8.20)
2 2λ
λ here is a real parameter
More properties are found in the textbooks. We list here some of the spherical
harmonics and their graphical representation:
r
1 1
Y00 (θ, ϕ) = (8.21)
2 π
8.5. ANGULAR MOMENTUM EIGENSTATES 59
r r
1 3 1 3 (x − iy)
Y1−1 (θ, ϕ) = · e−iϕ · sin θ = · (8.22)
2 2π 2 2π r
r r
1 3 1 3 z
Y10 (θ, ϕ) = · cos θ = · (8.23)
2 π 2 π r
r r
1 3 1 3 (x + iy)
Y11 (θ, ϕ) = − · eiϕ · sin θ = − · (8.24)
2 2π 2 2π r
Figure 8.2: Graphical representation for some of the spherical harmonics, the
colour coding represents the probability density calculated via |Ym` (θ, ϕ)|2
8.5 Angular momentum eigenstates

We now use more abstract method to analyse the angular momentum spectrum,
by introducing the eigenstates for L2 and L3
|β, mi (8.25)
such that :
L2 |β, mi = β|β, mi (8.26a)

L3 |β, mi = m~|β, mi. (8.26b)
Figure 8.3: The z-component of the angular momentum is quantised
Now we look at the effect of the operators L± on the eigenstates:

L3 L± |β, mi = L± L3 |β, mi + [L3 , L± ]|β, mi
= L± (L3 ± I)|β, mi
= L± (m~ ± 1)β, mi
⇒ L± |β, mi = |β, m ± 1i (8.27)
Therefore, the operators L± acting on the eigenstates rise / lower the state, just like
the creation and annihilation operators seen in the quantum harmonic oscillator.
In fact, the operator L+ rotates the angular momentum towards the zaxis , whilst
L− rotates it away from the z axis towards the −z axis.
8.6 The spectrum of angular momentum observ-

able
Now consider the following expected values :
hL23 i = ~2 m2 (8.28a)
hL2 i = hL21 i + hL22 i + hL23 i
β = a2 + b2 + ~2 m2 (8.28b)
For some numbers a and b. In order to find the explicit relation between β and
m, we ought to investigate the spectrum of the angular momentum further.
We know, that for some value mmax and mmin :
L+ |β, mmax i = 0 (8.29a)
L− |β, mmin i = 0 (8.29b)
8.6. THE SPECTRUM OF ANGULAR MOMENTUM OBSERVABLE 61
since the angular momentum will be totally alight with the wither z or z− after
successive application of L+ or L− . If we let `~ be the total angular momentum
eigenvalue, then obviously mmax = ` and mmin = −`.
Now we analyse (5b) further :
hβ, mmax |L†+ L+ |β, mmax i = 0

hβ, mmax |(L1 − iL2 )(L1 + iL2 )|β, mmax i = 0
hβ, mmax |L21 + L22 + i[L1 , L2 ]|β, mmax i = 0
hβ, mmax |L2 − L23 − L3 |β, mmax i = 0
β − ~2 mmax − ~mmax = 0
⇒ β = ~2 `(` + 1) (8.30)
Hence, we may denote the eigenstates in terms of ` instead of β, which is more

physically relevant :
|β, mi ←→ |`, mi
Such that:
L2 |`, mi = ~2 `(` + 1) (8.31)
Hence the magnitude of the angular momentum observable :
p
hLi = ~ `(` + 1) (8.32)
We may now write a full description for the angular momentum spectrum:
Figure 8.4: A vector model of the orbital quantum number
1. The orbital angular momentum eigenvalue is `, it refers to the maximum

positive or negative value the orbital angular momentum can take.
2. The z-component of the orbital angular momentum is m , sometimes , when

other angular momenta are included we refer to it by m` takes the integer
values between +` and −`.
p
3. The length of the angular momentum is ~ `(` + 1).
4. We call ` the orbital / azimuthal quantum number and m` the magnetic

quantum number.
5. There are other types of angular momenta, that shall be explored later, same
analysis will be applied to them.
8.7 Problems
1. Verify that the operator L2 commutes with all the angular momentum op-
erator components Lz , Lx andLy .
2. Verify the commutator algebra relations for L± and Lz using the commuta-
tor relations between angular momentum operator components commutation
relations.
3. Show that Y10 and Y11 are orthogonal.
4. Write the Hamiltonian operator for a system of two particles with reduced
mass µ orbiting each other, in the position representation.
5. Calculate directly the product Y11 · Y10 .
6. Write all the eigenstates for the orbital quantum number ` = 3
7. Draw the z components of the angular momentum states above
8. What is the minimal length of the momentum vector observable that carries
the quantum number m = −4~ ?
9. What is ` for the free particle ?
10. Show that we can write hϕ, θ||`, mi as Pm (ϕ)F` (θ) .

Chapter 9
The Hydrogen Atom
In classical mechanics, the Hamiltonian H for two bodies interacting via a - time
independent- potential is given by :
2
X (pi )2
(H = + V (r) (9.1)
i=1
2µ
Where, µ = MM+m m
is the reduced mass, and r is the radial separation between the
bodies. pi is the canonical momentum, that we may decompose into two parts :
L
pr =linear momentum pt = angular momentum (9.2)
r
Hence, we may write (9.1) as :
p2 L2
H= + + V (r) (9.3)
2µ 2µr2
since we know that the moment of inertia I = µr2 we can therefore write: recall that L =
Iω
p2 L2
H= + + V (r) (9.4)
2µ 2I
The potential for the Hydrogen atom is the Coulomb potential, given by the for-
mula:
ke2
V (r) = − (9.5)
r
Hence we write the Hamiltonian function as :
p2 L2 ke2
H(p, r) = + − (9.6)
2µ 2µr2 r
63
64 CHAPTER 9. THE HYDROGEN ATOM
9.1 Canonical quantisation of the H-atom Hamil-

tonian
We have the state |Ψi of the H-atom. We the define the Hamiltonian operator Ĥ
from quantising the Hamiltonian function in (9.6) that is :
p̂2 L̂2 ke2

Ĥ = + − (9.7)
2µ 2µr̂2 r̂
Due to the spherical symmetry, it is logical to project the state |Ψi into the
configuration space in spherical polar coordinates, hereby the the Hilbert space
is :
H : (L, dµ)
where dµ = dr d2 dφ sin φ2 dθ , the volume element in the spherical polar coordi-
nates and the wavefunction:
ψ(r, φ, θ) = hr, φ, θ|ψi
The H-atom is surely a stationary state Ψ(r, φ, θ; t) = ψ(r, φ, θ)e−iωt .Hence the
time-independent Schrödinger’s equation is written as :
~2 2 ke2
− ∇ ψ(r, φ, θ) − ψ(r, φ, θ) = Eψ(r, φ, θ) (9.8)
2µ r
It was found -mathematically- that the wavefunction can be separated into three
parts:
ψ(r, φ, θ) = R(r)P (φ)F (θ) (9.9)
Where R(r) is the radial function, and P (φ)F (θ) make up the spherical Harmonics
Y (φ, θ).
9.2 Quantum numbers

Each of the functions above is an eigenfunction for some observable about the
hydrogen atom with an associated quantum number:
R(r) −→ n = 1, 2, 3 . . . Principle quantum number

F (θ) −→ ` = 0, 1, 2 . . . , n − 1 orbital quantum number
P (φ) −→ m` = −`, −` + 1, . . . , +` megnatic quantum number
9.3. SOLUTION OF THE TISE FOR THE HYDROGEN 65
9.3 Solution of the TISE for the Hydrogen

In fact, the H-atom is the only physical problem in quantum mechanics that can be
solved exactly without using perturbation theory or other approximation methods.
The TISE (9.8) can written as three differential equations:

1 d 2 dR(r) 2µ
+ 2 Er2 + ke2 = `(` + 1)

r (9.10a)
R(r) dr dr ~

sin θ d dF (θ)
sin θ + Cr sin2 θ = −Cφ (9.10b)
F (θ) dθ dθ
1 d2 P (φ)
= Cφ (9.10c)
P (φ) dφ2
Solving the above equations to obtain the full expression for the wavefunction:
r
ψn,`,me ll (r, φ, θ) = Ar` e−r/a0 r r` Ln` ( ) · Ym` (φ, θ) (9.11)
a0
Where:
• Ar` , a normalisation constant.
• a0 , Bohr radius and it is equal to ≈ 0.53Å.
• Ln` ( ar0 ), the associated Laguerre polynomial.
• Ym` (φ, θ), the associated spherical harmonics, the eigenfunction for the oper-
ators L̂ and L̂z .
We can find the energy spectrum for the Hydrogen atom from 9.11 :
µ e2

1 13.6 eV
En = − = − (9.12)
82o ~2 n2 n2
With o the vacuum permittivity.

As expected, the spectrum is discrete, but the spacing between the energy levels
gets smaller and smaller as the principle quantum umber increases, figure 9.3
If we wish to find the ionisation energy for hydrogen atom ( i.e. the energy
required to free the electron from the atom) we let n → ∞ we obtain:
E∞ = 13.606 eV = Ry
it is equal to the Rydberg energy . For generality we can approximate the energy
spectrum for any atom having Z electrons and µ reduced mass by:
µ
E ∼ −Z 2 Ry (9.13)
me
Figure 9.1: Energy levels of the idealised H-atom and the well-known spectral
series associated with electron transitions
We ought to emphasise this is merely a hand-waving approximation ! The true

energy spectrum for atoms ( even the H-atom) is far more complicated, as we shall
see later when we study the Real Hydrogen atom.
9.4 Degeneracies in the ideal H-atom

Since (9.11) depends on three quantum numbers , ` and m` , but the energy spec-
trum only depends on n, we have degenerate states in the idealised H-atom. Where
we can have multiple wavefunctions having the same energy. For example, in the
first excited state n = 2 we have the following wavefunctions:
ψ2,0,0 ψ2,1,0 ψ2,1,−1 ψ2,1,1 ,
and so on . However, the number of electrons that can occupy each energy level is
determined by Pauli exclusion principle: stating that no two electrons in the
atom can wave an overlapping wavefunctions. In other words each wavefunction
can describe one electron only. Meaning no electrons in the atom can have all of
their quantum numbers identical.
9.5 Spin quantum number

In addition to the quantum numbers (n, `, m` ) there is a forth quantum number
of the electron, corresponding to its spin, which is an internal degree of freedom
electrons are found to possess. The spin quantum number ms can take one of two
values ± 12 of multiples of ~. So far, this new quantum number does not seem to
9.6. PROBLEMS 67
affect the energy spectrum. However, we shall see later that it player a röle in an
interaction inside the H-atom affecting the energy spectrum.
9.6 Problems
1. If you know that the ground state radial function of the H-atom is given by:
R10 = e−r/a0
(a) Normalize it .
(b) Construct ψ100 .
(c) Calculate hri , hr2 i, hxi and hx2 i, what do you observe?
2. Derive the Rydberg formula for the H-atom. Then obtain similar formulas
for D-atom and positronium (electron orbiting its anti particle) .
3. Having Pauli exclusion principle in mind, what is the maximum number of

electrons one can have in each shell ( for a given n ) ?
4. The H-atom emission spectrum is observed to be described in the picture

below
(a) If there is only one electron in the H-atom, how do you explain the
existence of multiple emission lines ?
(b) Explain the spectrum in detail, using the Rydberg formula.
(c) Spectroscopy is used to distinguish between the isotopes of elements (
H2 and D2 for example). Predict the D2 spectrum and draw it, compare
it the given H2 spectrum.
5. Compare the energy spectrum of the H-atom, with the one in a 3-D box of
dimension 1 Å.
6. Show that:
∂U 1
h i = h 2i
∂r r
If U is the Coulomb potential.
7. Is it possible for electrons to exist in the nucleus ? Justify your answer

using the calculation of probability for the electron to be detected within the
nuclear radius rnuclear ∼−15 m.
(a) Start by using the wavefunction ψ100 , and show it is valid for all values
ofr up to r = 0.
(b) Assume that the wavefunction is constant over the small volume of the
‘spherical nucleus’ then calculate the probability P ≈ Vnuclus × |ψ100 |2 .
(c) The previous step can be done in more ‘sophisticate way’ by expanding
the wavefunction around the origin using the parameter = 2rnuclear /a
(d) substitute for a0 and rnuclear to estimate the probability.
8. The associated Laguerre polynomials are given by the following ‘ generating

formula ’: p
P p d
L(x)q−p = (−1) Lq
dx
With L(x)a is the Laguerre polynomials give by:
q
d
L(x)q = e x
(e−x xq )
dx
Use these equations to generate the first two associated polynomials. Are
they orthonormal over the interval [0, +∞] and weighting function w(x) =
xp e−x ?
Chapter 10
Spin
10.1 Transformation of the wavefunction

Given a wavefunction defined at the point (in 1-D) x0 , then we define the in-
finitesimal translation of the wave function as a transformation preformed on
it, by shifting the point x0 by an infinitesimal parameter .
ψ(x0 ) −→ ψ(x0 + )
Using the Taylor series expansion of the translated wavefuntion around the point
x0 we can write :
dψ
ψ(x0 + ) = ψ(x0 ) + + O(2 ). (10.1)
dx
since is an infinitesimal parameter , the order terms of 2 are considered vanishing,
hence:
dψ
ψ(x0 + ) ∼ ψ(x0 ) + (10.2)
dx
i
It won’t affect the expansion if we multiplied and divided by ~
:

i ~ dψ
ψ(x0 + ) ∼ ψ(x0 ) + , (10.3)
~ i dx
~ d
we hereby identify i dx
as the m operator, P̂ . Thus,
i
ψ(x0 + ) ∼ ψ(x0 ) + P̂ (ψ) . (10.4)
~
This equation basically tells us that the translation is generated by the momen-
tum operator , or the momentum operator is the generator of translation.
69
70 CHAPTER 10. SPIN
Same argument can be made for 3-D, in Cartesian coordinates ~r = (x, y, x) , being
~ That is,
translated by δr.
~
ψ(~r) −→ ψ(~r + δr)
By the same argument done before, the multivariate Tylor expansion :

~ i ~
ψ(~r + δr) ∼ δr ∇(ψ) . (10.5)
~ i
We know that the ’linear momentum operator is defined by P~ ~i ∇ , where the gra-
dient operator here is in the Cartesian coordinates .
Now, we can run the same argument for multi-particle system with f number of
degrees of freedom in generalised coordinates : (q 1 , q 2 , . . . q f ), the transformation
of generalised coordinates is not restricted translations, but also rotations ( if some
of the generalised coordinates correspond to angles for example. ). But such trans-
formation is written in the form : ∂q∂ i , that is, it depends on the derivatives of the
coordinates - for a general configuration space- .
10.2 External and internal degrees of freedom

for a system
The f degrees of freedom the system has, can be linked to a set of generalised
coordinates of the configuration space. And the transformation of the wavefunc-
tion for that system’s degrees of freedom is written in terms of ( derivatives of the
generalised coordinates), or we can say that such transformations are translations
in space and time, in addition to 3-D rotations.
Examples : The These degrees of freedom are called external degrees of freedom. The observables
linear and angu- that are explicit functions of generalised coordinates are linked to these degrees of
lar momentum, freedom. In other words, the operators corresponding to those observables act -
the velocity and effectively- on the same Hilbert space.
energy...
However, in quantum mechanics, there exist another type of degrees of free-
dom, that do not ( explicitly) depend on space and time. ( or configuration
space). Rather they form a separate Hilbert space, with operators correspond to
observables that are not an explicit functions of generalised coordinates, nor time.
Although they can affect the external degrees of freedom observables, through
physical process called coupling. But this is only when we consider the product
of the two Hilbert spaces - the system as a whole-.
This is better understood via the transformations we have discussed earlier. As
a space/ configuration space transformation is preformed, the internal degrees of
10.3. MATHEMATICAL DESCRIPTION FOR INTERNAL DEGREES OF FREEDOM71
freedom of the system are not affected by themselves. This is the main difference
between internal and external degrees of freedom of a system.
In fact, there is a theorem in mathematical physics called the Coleman Mandula
theorem, that separates space and time translation symmetries and internal ones
( one cannot be obtained from the other.) unless supersymmetry is used, this is
beyond the scope of our study.
The distinction between, scape and time (explicitly) dependent observables, and
internal degrees of freedom is crucial in the understand of elementary particle
physics in one hand or quantum information in the other. There are many ex-
amples for observables originating from internal degrees of freedom of quantum
particles. . Electric charge, magnetic dipole moment ( from free electron), quan- We should em-
tum numbers like: leptonic or baryonic quantum numbers, strangeness...etc phisise that they
are degrees of
freedom and
10.3 Mathematical description for internal de- not internal
structure , as
grees of freedom ( elementary)
As we mentioned above, the internal degrees of freedom for a quantum system particles are as-
has an independent Hilbert space that we dealt with earlier ( we can always go to far-as we know
L2 for external ones). Now, we need to study how such space can be defined, or dimentionless .
constructed.
We start by defining the state ket |ψ, decomposed into the basis for the Hilbert
space {|αi i}.
XH
dim
|ψi = αi |αi i (10.6)
i=1
Such that, the dimension of the Hilbert space is the same as the number of the
internal degrees of freedom described by that space. The coefficients αi are the
probability amplitudes of detecting the system having the state αi , i.e. usually the
quantum system is in a state of superposition, until it is measured. Identically to
what is studied before.
We can picture a state space for these degrees of freedom and define transforma-
tions just like we have done in the external ones 10.3 Since the magnitude of |ψi
should be 1, i.e.
||ψi| = 1,
any transformation made upon this ket is ought to a rotation in the state space.
If that space is quantised, the rotation is only possible in integer multiples , and
via a ladder ( raising and lowering) operators similar to what we have seen in
the (orbital) angular momentum. Hence all these internal degrees of freedom are
physical realisations of the same mathematical structure discussed earlier in the
72 CHAPTER 10. SPIN
Figure 10.1: The quantum state can be abstractly represented as a vector in the
state space
angular momentum.
These facts are very powerful, because it allows us to apply the same techniques,
and thought processes to many physical systems ( with modifications that appears
naturally ). Depending on the nature of the physical problem ( the symmetry, the
number of degrees of freedom ..etc). To illustrate this power, this technique ( which
is known formally as representation theory) is used extensively in Fundamental
physics. The whole standard model of particle physics is fundamentally built
upon the same idea of representation, quantum electrodynamics and quantum
chromodynamics ..etc
10.4 Discovery of electron’s magnetic dipole mo-

ment
In 1925, the idea of ’spin’ angular momentum for electron was first proposed George
Uhlenbeck and Samuel Goudsmit to explain hyperfine splitting in atomic spectra.
However, in 1922, Otto Stern and Walther Gerlach had shown that electrons act
like tiny magnetic bars ( they have a magnetic dipole moment), and that magnetic
moment takes particular values only.
We are going to construct step-by-step the theory of quantum spin via the obser-
vations of S-G experiment, and discuss the mathematical properties of our con-
struction. Following the lead of the philosophy developed above.
S-G apparatus is basically a magnetic field that is free to be in any direction,

x, y, z. As a beam of electrons is passed through the S-G apparatus, the beam
10.4. DISCOVERY OF ELECTRON’S MAGNETIC DIPOLE MOMENT 73
74 CHAPTER 10. SPIN
splits either in the positive or negative direction ( say the B field is aligned with
the z axis ). The split is due to a change of energy of the beam, because electron’s
have magnetic dipole moment, and it takes a ± values or 1/2 values only, otherwise
... If the electron has an integer multiples of some magnetic dipole moment, not
1/2 the splitting would be at least in three ways, for ± direction and renaming of
the beam passing through unaffected. Quantitatively we write the change of the
energy is given by :
∆H = µ ~
~ · B. (10.7)
~ = Bz only the projection of the magnetic dipole moment in the z direction
Since B
counts, we hence conclude that :
1
µz = ± · const. (10.8)
2
Same effect is observed if the B field was aligned with any axis not just the z
axis. Hence the observable of the dipole magnetic moment is associated with a
vector operator~ˆµ that has an eigenvalues of 1/2 of some constant. This constant
was found experimentally, and called the Born magnetron µB for the electron,
but generally it is noted by γ the gyromagnetic ratio, being more careful, Born
magnetron and the gyromagnetic ratio are not exactly equal, due to relativistic
effects, and Thomas precession γ = gs · µB where gs is called Landé g-factor .
In fact we write :
~ˆ
S
~ˆ = −gs µB
µ (10.9)
~
ˆ the spin operator .
We call the operator~S
10.5 Electron’s spin

Although~Sˆ does not correspond to a direct observable ( nobody can detect the spin
directly separately from the magnetic dipole moment) But seems very natural to
study the spin rather than the magnetic dipole moment directly. Although there is
no classical analogy to the spin, as electrons do not spin around themselves ( they
are dimensionless points ) but spin is a form of angular momentum corresponding
to an internal degree of freedom for the electron.
So far, from S-G experiment, we can conclude that :
1
hŜz i = ± ~ (10.10)
2
10.5. ELECTRON’S SPIN 75
We hall stick with z but any component gives as similar results as the Sz . And the
~ is just to reserve the dimension of the spin, since it is an angular momentum.
We let , for convince denote :
1 1
hŜz i = ms ms = − ~, + ~ (10.11)
2 2
Note how similarities in notation with the orbital angular momentum is appear-
ing. Now, we turn into studying the spin operators more, via repeating the S-G
experiment in the following way:
Start with S-G apparatus in the z direction, take the half of the initial beam that
is aligned in the +z direction ( having ms = +1/2~) passed though another S-G
apparatus but aligned in the x direction, it shall split the beam once again. This
time in the ±x directions. One would expect that by these two apparatuses; we
were able to measure two observables associated with Ŝz and Ŝx . Nevertheless,
experiments have shown that this is not true, as taking the 1/4 of the beam that
passed via the +z first and +x second, to a third S-G apparatus aligned in the z.
We expect only one beam to pass in the +z direction, but this doe not happen.
The beam splits into two a third time ! This does not happen if we passed it (
initially) into three consecutive S-G apparatuses aligned in the z direction. Hence,
we cannot measure the spin in to direction simultaneously. This is mathematically
written as :
[Ŝz , Ŝx ] 6= 0 (10.12)
In fact the commutation relation for the spin operators take the form for the indices
i, j, k taking the values x, y, z:
[Ŝi Ŝj ] = i~kij Ŝk (10.13)
1
Which is the same as for the L̂’s that we studied before. We can therefore,
adopting the philosophy of previous lectures define the following operators:
Ŝ± = Ŝx ± iŜy (10.14)
and:
Ŝ 2 = Ŝx2 + Ŝy2 + Ŝz2 (10.15)
With the eigenstates :
|s, ms i (10.16)
But the eigenvalue s take one value only s = 12 ~ and ms as we have seen takes the
values ms = ±1
2
~:
Ŝz |s, ms i p
= ms ~|s, ms i
(10.17)
Ŝ|s, ms i = ~ s(s + 1)|s, ms i
1
The symbol kij is called the Levi-Civita symbol and it is equal to 0 if i = j and +1 for even
permutation and −1 for odd permutations of the indices
76 CHAPTER 10. SPIN
And the ladder operator 2 :

p
Ŝ± |s, ms i = ~ s(s + 1) − ms (ms ± 1)|s, ms ± 1i (10.18)
The three operators Ŝ± and Ŝz form the well-known su(2) algebra commutation
relations:
[Ŝ± , Ŝz ] = ∓~Ŝ±
(10.19)
[Ŝ+ , Ŝ− ] = 2~Ŝz
The Hilbert space for the spin is very simple, as it is 2-D only spanned by the
basis | 21 , + 12 i, | 21 , − 12 i or denoted by |χ+ i, |χ− ii, respectively.
X
|ψi = αλ |χλ i (10.20)
λ=±
representing a superposition of the spin states.
10.6 Infel-van der Warden Symbols

Not always we can have the luxury of selecting the Cartesian coordinates in order
to study the spin alignment, sometimes we need to work with any suitable set of
coordinates ( cylindrical, spherical . . . ) . Hence we need to link the spin operator
vector to a more mathematically-rigorous argument other than the observations
made from S-G experiment.
In fact, we may write the spin operator vector S ~ˆ in terms of a set of symbols,
Usually Infel-van known as Infel-van der Warden Symbols ( σ 1 , σ 2 , σ 3 ):
der Warden Sym-
bols used in ad- ~ˆ = ~ ~σ
S (10.21)
vanced texts dif- 2
fer from our no- Where, ~σ = (σ 1 , σ 2 , σ 3 ) is called the Pauli vector , and these symbols are known
tation by the fac- in the physics literature as Pauli spin matrices , Due to their relation - in this
tor of i, i.e. they context- with the spin operator.
usualy call iσ i as It goes without saying that the spin operators really have inherited the mathemat-
the symbol ical properties ( commutation relations) from the Infel-van der Warden Symbols,
or -as we shall call them from now on- the Pauli spin matrices. So, naturally there
exist the two symbols:
σ ± = σ 1 ± iσ 2 (10.22)
That along with σ 3 form the su(2) algebra commutation relations:
[σ 3 , σ ± ] = ±iσ ±
(10.23)
[σ + , σ − ] = iσ 3
2
These operators are a direct result from the quantisation of the spin space
10.7. MATRIX REPRESENTATION OF SPIN STATES 77
Also note that

[σ i , σ j ] = 2iεij
k σ
k
Moreover, the pauli spin matrices have an important mathematical property called
the Clifford algebra relation:
{σ i , σ j } = σ i σ j + σ j σ i = 2δ ij (10.24)
We call the operation {·, ·} the anticommutator. From this property we can
easily prove that:
(σ 1 )2 = (σ 2 )2 = (σ 3 )2 = δ ij (10.25)
So far, we only dealt with abstract, mathematical entities, and their properties
(commutation relations), In order to use them in physical calculations, we need
to find a proper representation for them in order to be realised in the physical
world. We shall find that there are many possible representations for the Pauli
spin matrices, that have a direct physical application and meaning. In fact, spin
of the electron is only a simple application of the representations for the Pauli spin
matrices, others go as deep as the electroweak interaction in the standard model
of particle physics, and supersymmetry! However, there are other representations,
that seems to be of an interest of mathematicians mainly like the quaternions (a
higher form of complex numbers that has 3 imaginary units) .
10.7 Matrix representation of spin states

In the last sections, we have introduced the spin operators and their eigenstates.
We discovered the algebra of spin operators, and derived the Pauli spin matrices.
However, we only stated their properties abstractly without defining a particular
representation for them, now we aim to realise the spin algebra in a simple repre-
sentation.
Consider the eigenkets |χ− i and |χ+ i.They form a basis for a 2-D Hilbert space of
the internal degree of freedom, we have called the ‘ spin’ . It is natural to introduce
a canonical representation for these kets as the column vectors :

1
|χ+ i ≡ (10.26)
0

0
|χ− i ≡ (10.27)
1
78 CHAPTER 10. SPIN
Now, the action of the spin operators Ŝ1 , Ŝ2 , Ŝ3 and the ladder spin operators
Ŝ+ , Ŝ− on the kets is :
~ ~ ~
Ŝ1 |χ± i = |χ∓ i Ŝ2 |χ± i = ±i |χ∓ i Ŝ3 |χ± i = ± |χ± i
2 2 2
Ŝ± |χ∓ i = ~|χ± i Ŝ± |χ± i = 0 (10.28)
We can easily express the operators above as matrices, and with the help of the
identity:
~ = ~ ~σ
S (10.29)
2
We may write the explicit expression of the Pauli matrices:

0 1
σ1 =
1 0

0 −i
σ2 =
i 0

1 0
σ3 = .
0 −1
And:

0 1
σ+ =
0 0

0 0
σ− =
1 0
We shall only deal with the index-down matrices and drop the ket notion on the
eigenstate, calling them. For convenience and consistency with quantum mechanics
textbooks.
Sometimes, the states χ+ , χ− are denoted by α and β, respectively. The spinor χ
is defined as:
χ+
χ = (10.30)
χ−
Moreover, we can define the Hermitian conjugate of the spinor:
χ † = χ∗+ χ∗−

(10.31)
That satisfies:
χ †χ = 1 (10.32)
We can calculate the expected value for an operator Ω̂ acting on the spin Hilbert
space by:
hΩ̂i = χ † Ω̂ χ (10.33)
10.8. GEOMETRIC REPRESENTATION 79
10.8 Geometric representation

There is another representation for spin, connected to the ‘ spin vector ’ pointing
in the 3-D space. In spherical polar coordinates, one may write the spin states as:
1 1
χ+ = ei(δ−ϕ/2) cos θ χ+ = ei(δ+ϕ/2) sin θ (10.34)
2 2
Where ϕ, θ are the polar and azimuthal angels, respectively and δ is an arbitrary
phase.
In order to see why this representation is correct, we start by evaluating the prob-
ability of detecting the particle spinning up, w.r.t. the z direction:
1
|χ+ |2 = cos2 θ (10.35)
2
Similarly for the down direction:
1 1
|χ− |2 = 1 − χ+ |2 = 1 − cos2 θ = sin2 θ (10.36)
2 2
It is clear now, what is the rôle of the magnitude of the spin states, in terms of θ.
However, we need to discuss further the rôle that the phase plays.
Lets start by calculating the expected values of the spin operators:
1
hŜx i = sin θ cos ϕ
2
1
hŜy i = sin θ sin ϕ (10.37)
2
1
hŜz i = cos θ
2
Now, assume we wish to rotate the coordinates by an angle γ around z. We have
the rotation matrix : iγ/2
e 0
R(γ) = (10.38)
0 e−iγ/2
Applied to the spinor χ is equal to:
i(δ− 1 (ϕ−γ))
cos 12 θ

e 2
R(γ)χ = 1 (10.39)
ei(δ+ 2 (ϕ−γ)) sin 12 θ
Notice that, in order for the spinor to return to its original state, before rota-
tion, one needs not to make a 2φ rotation. Rather a rotation by 4π. This is the
main characteristic of spinors, that makes them ‘ very’ different from vectors, and
manifesting itself in terms of the phase factor in the geometrical representation.
80 CHAPTER 10. SPIN
Figure 10.2: A spinor transformation can be thought of as a vector on a Möbius

band.
Sometimes we denote this characteristic by saying that the ‘ group’ of spin trans-
formations double covers the ‘group’ of spatial transformations.
In order to picture this in a deeper way, one can think of the spinor’s internal space
as a Möbius band - illustrated in the figure 1. A vector on the Möbius band needs
to be transported along the band twice, in order to return to its initial state.
10.9 Spin in constant magnetic field

A particle with a spin - an electron- for example is put in a constant magnetic
filed, such that the direction of the field is parallel to the z-component of the spin.
The Hamiltonian for such system is given by:
~ ·S
H = −γ B ~ (10.40)
such that γ = e/m, the ratio between the electron’s charge and its mass.And
~ ·S
B ~ = BSz .
It is clear that:
[H, Sz ] = 0 (10.41)
Implying that there exist eigenstates for H and S simultaneously. Since we already
know the eigenstates for Sz , and represented by the spinor χ . We then write :
Hχ = Eχ (10.42)
or:
−γB Sz χ = Eχ. (10.43)
Since, Sz χ = ± 21 ~χ The eigenenergies are:
E± = ∓µB B (10.44)
10.10. ELECTRON PARAMAGNETIC RESONANCE EPR 81
e~
The constant µB = 2m e
is Born magneton. It is necessary to add another constant
gs as we have seen earlier to this equation, known as the Landé g-factor, because
the electron precesses in the magnetic field. We then have:
E± = ∓gs µB B (10.45)
10.9.1 Stationary states

Since we have found the energy spectrum for an electron in magnetic field, we may
now write the time evolution of the state χ, using the equation:
χ(t) = e−iωt χ(0) (10.46)
We have then, for A and B are normalisation constants:
χ(t) = Ae−iωt χ+ + Beiωt χ− (10.47)
or: i 1 γBt
i 21 γBt 1 −i 12 γBt 0 Ae 2
χ(t) = Ae + Be = 1 (10.48)
0 1 Be−i 2 γBt
10.10 Electron paramagnetic resonance EPR

From the above analysis, we have learnt that an electron in a magnetic field could
occupy one of two energy states, depending on ms :
Ems = ms gs µB B (10.49)
A transition from one energy state to another, is obtained by absorption / emission

of photon of energy equal to ∆E = gs µB B :
hνr = gs µB B (10.50)
If an ensemble of electrons in the magnetic field is exposed to photons of frequency

ν = nur , then the electrons shall absorb them, otherwise no absorption will occur
-only elastic scattering-. This peak of absorption is known as electron param-
agnetic resonance or EP R. Moreover, νr is known as the resonance frequency.
This phenomena is very important in may areas , like measuring the value of
the g factor, and detecting free radicals in biological systems.
In order to understand the reason for detecting absorption lines rather than the
emission lines in EPR, we turn to calculating the population of electrons in the
82 CHAPTER 10. SPIN
Figure 10.3: Energy states of electron in a magnetic field B.
Figure 10.4: EPR absorption resonance for ν = 9388.3M Hz
upper energy level nupper and the lower level nlower , using Maxwell-Boltzmann
statistics, under a thermodynamic temperature T :

nupper Eupper − Elower ∆E hνr
= exp − = exp − = exp − (10.51)
nlower kT kT kT
Where k is Boltzmann constant .

We observe that at room temperature T ∼ 300K and typical microwave frequency
νr ∼ 9.7GHz the ratio is about nupper /nlower ≈ 0.998. That means the upper
population is slightly less than the lower one, implying transitions from the lower
to upper energy states is more probable than the reverse transitions.
10.11. PROBLEMS 83
10.11 Problems
~ the vector product with itself is equal o
1. Show that for Spin vector S,
~ ∧S
S ~ = i~S
~
~ ∧S
hint: use the expression for the vector product S ~ = εk Si Sj ~ek
ij
2. prove that (σ i )2 = δij
3. We denote the hermitian Pauli matrix by σi , show that it is in fact equal to

σi.
4. Show that,
σ i σ j = δ ij + iij
kσ
k
. hint: use both the Clifford algebra and commutation relations the Pauli
matrices obey.
5. Are Pauli matrices unitary ??
6. What is the action of σ ± , σ 3 of the states |χ± i ?
7. Use the definition of the spin vector in terms of the Pauli matrices to prove
that
3
hŜ 2 i = ~2
4
8. Given a vector ~a, find the dot product:
~a · ~σ
9. Prove that :
det(σ i ) = 1
10. We define the Dirac gamma matrices as :

k 0 −iσ k
γ =
iσ k 0
Show that the Clifford algebra relation also holds for the gamma matrices :
{γ i , γ j } = 2δ ij I
84 CHAPTER 10. SPIN
11. Given the following 2 × 2 matrices :

0 1
σ1 =
1 0

0 −i
σ2 =
i 0

1 0
σ3 = .
0 −1
Find their eigenvalues and eigenvectors They form the matrix representation
for the Pauli spin matrices, moreover, the results obtained will form the
explicit expression of the eigenstates |χ± i in column form.
12. Given
−1 −i
Ω̂ =
i 2
an operator acting on the spin Hilbert space:
(a) Show that it could correspond to an observable.

(b) Decompose it in terms of the Pauli spin matrices.
(c) Find its eigenvalues.
(d) Compute hΩi.
13. Since the spin operator is a vector in the 3-D space. Show that if it is rotated
by an angle γ around the z-axis, the commutation relations algebra remains
invariant, for the rotated operator .
14. Calculate the EPR frequency for an experiment with B = 100 Gauß. where
µB = 9.274 × 10−24 Joule/Tesla. and the material used is Holmium with
gs = 1.97.
Appendix A
The Virial Theorem
Given a system having α particles, with associated position ~rα and momenta p~α .
We define the virial function as:
X
ς= p~α · ~rα (A.1)
α
It would be interesting to look for the time derivative of this function:
dς X ~
= ṗα · ~rα + p~α · ~ṙα (A.2)
dt α
Since we are dealing with many-particle system. We can take the time average for
the previous expression
R τ dς
dς 0
dt
h i = R τdt (A.3)
dt 0
dt
ς(τ ) − ς(0)
=
τ
Now, if the system has a periodic motion of a period τ . The time average for the
derivative of the virial function will vanish. even if the system does not admit a
periodic motion, the virial function ought to be bounded, hence one can integrate
dς
over a sufficiently large interval such that the time average h i will approach zero.
dt
Hence, we have ( at least as an approximation): recall that F~ = ṗ~
X X
h p~α · ~ṙα i = −h ṗ~α · ~rα i (A.4)
α α
85
86 APPENDIX A. THE VIRIAL THEOREM
We can now identify the LHS being twice the kinetic energy , the RHS is the force
dotted with the position :
1 X~
hT i = − h Fα · ~rα i (A.5)
2 α
This is the Virial theorem , the expected value for the kinetic energy for a system
is equal to its virial function.
It is interesting to look at forces that arise from central potential taking the form
:
V = krn+1 (A.6)
Hence, by the virial theorem eq (A.5):
1 d
hT i = hr · (krn+1 )i (A.7)
2 dr
1
= h(n + 1)krn+1 i
2
n+1
= hV i
2
For Columb and gravitational potentials, n = −2. Therefore we have :
1
hT i = − hV i (A.8)
2
Appendix B
Summery of Hilbert spaces
B.1 Representations of Hilbert space

We list a summery for the most basic properties of Hilbert spaces, in both repre-
sentations. Discrete and continuous
The list goes as follows:
Continuous representation Discrete representation

R P
|ψi = ψ(x)|xidx |ψi = n ψn |en i
hx0 |xi = dxδ(x0 − x)

R
hem |en i = δmn
ψ ∗ (x)ψ(x)dx = 1
R
|ψn |2 = 1
P
hψ|ψi = hψ|ψi = n
hφ|ψi = hφ|xihx|ψidx = φ∗ (x)ψ(x)dx φ∗n ψn

R R P P
hφ|ψi = n hφ|en ihen |ψi = n
Note that we sometimes denote the basis |ei i by |ii. Here we consider the
vector |ψi to be normalised.
B.2 Operator properties

We start by defining the normalised ket :
X
|ψi = ψi |ii
i
With the property :

|ψj |2 ≡ P (j)
87
88 APPENDIX B. SUMMERY OF HILBERT SPACES
Hence the square modulus of the component is equal to what-to-be considered as

probability. Note that any of the following argument will also apply to continuous
basis representation according to the list above. We can write an operator Ω̂ as :
X
Ω̂ = ωi |ωi ihωi |
i
This is knows as the spectral decomposition , where ωi and |ωi i are the eigen-
values and eigenbasis respectively. The expected-value for Ω̂ is written as:
hΩ̂i = hψ|Ω̂|ψi
Proof:
Expanding the above expression :
! ! !
X X X
ψk∗ hωk | ωj |ωj ihωj | ψi |ωi i =
k j i
XX XX
⇔ ψk∗ ωj hωk |hωj i ψi hωj |hωi i =
k j
| {z } j i
| {z }
=δkj =δij
X X
⇔ ωj |ψj |2 = ωj P (j)
j
|{z} j
=ψj∗ ψj
Similarly, we can define the standard deviation :

2
σ 2 (Ω) = hΩ̂2 i − hΩ̂i
And the expected value for a function of Ω̂:

X
hf (Ω̂)i = f (ωj )P (j)
j
That implies that the eigenvalue of the function of the operator is the function of
the eigenvalue itself.
We now attempt to find the matrix element for an operator Ω̂ not expressed in
the eigenbasis. Starting by :
|φi = Ω̂|ψi
expanding this expression:
X X
φk |ki = Ω̂ ψi |ii
k i
B.2. OPERATOR PROPERTIES 89
Taking the jth component of the LHS, by projecting on the basis |ji:
X
φj = ψi hĵ|Ω|ii
i
Moreover, the ket |φi is written as:

X XX
φj = ψi hĵ|Ω|ii
j j i
We identify the matrix elements of the operator by

XX
Ωji = hĵ|Ω|ii
j i
We can simply make the transition to the eigenbasis :
Ωij = hωi |Ω̂|ωj i
But Ω̂|ωj i = ωj |ωj i , then we have , by linearity:
Ωij = ωj hωi |ωj i =
⇔ ωj δij
Hence the operator is diagonalised by the eigenbasis.
90 APPENDIX B. SUMMERY OF HILBERT SPACES
Bibliography
[1] James Binney and David Skinner. The physics of quantum mechanics. Oxford
University Press, 2013.
[2] Paul Adrien Maurice Dirac. The principles of quantum mechanics. Num-
ber 27. Oxford university press, 1981.
[3] Alan Robert Edmonds. Angular momentum in quantum mechanics. Princeton

University Press, 1996.
[4] Stephen Gasiorowicz. Quantum physics. John Wiley & Sons, 2007.
[5] Kurt Gottfried and Tung-Mow Yan. Quantum mechanics: fundamentals.

Springer Science & Business Media, 2013.
[6] Sadri Hassani. Mathematical physics: a modern introduction to its founda-

tions. Springer Science & Business Media, 2013.
[7] Lev Davidovich Landau, Evgenii Mikhailovich Lifshitz, JB Sykes, John Stew-
art Bell, and ME Rose. Quantum mechanics, non-relativistic theory. Physics
Today, 11:56, 1958.
[8] John Von Neumann. Mathematical foundations of quantum mechanics. Num-

ber 2. Princeton university press, 1955.
[9] Jun John Sakurai and Jim Napolitano. Modern quantum mechanics. Addison-
Wesley, 2011.
[10] A. S. Stodolna, A. Rouzée, F. Lépine, S. Cohen, F. Robicheaux, A. Gijs-

bertsen, J. H. Jungmann, C. Bordas, and M. J. J. Vrakking. Hydrogen atoms
under magnification: Direct observation of the nodal structure of stark states.
Phys. Rev. Lett., 110:213001, May 2013.
[11] Cohen Claude Tannoudji, Diu Bernard, and Laloë Franck. Mécanique quan-
tique. tome i. 1973.
91
92 BIBLIOGRAPHY
[12] Angus Ellis Taylor and David C Lay. Introduction to functional analysis,
volume 2. Wiley New York, 1958.
[13] Nouredine Zettili. Quantum mechanics: concepts and applications. John

Wiley & Sons, 2009.
View publication stats

Lecture Notes in Quantum Mechanics: June 2017

Uploaded by

Copyright:

Available Formats

Lecture Notes in Quantum Mechanics: June 2017

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture Notes in Quantum Mechanics: June 2017

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Lecture Notes in Quantum Mechanics

Book · June 2017

Geometric flows and implications to quantum gravity View project

The user has requested enhancement of the downloaded file.

Department of Physics and Astronomy

1 Review of Classical Mechanics 1

2.14 Commuting operators . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3 Postulates of Quantum Mechanics 19

5 The Uncertainty Principle 33

6 Basic Scattering Theory 41

7.2.2 The eigenfunctions . . . . . . . . . . . . . . . . . . . . . . . 51

9 The Hydrogen Atom 63

A The Virial Theorem 85

B Summery of Hilbert spaces 87

Review of Classical Mechanics

1.1 The Hamiltonian

3. A dipole in a magnetic field B has the following Hamiltonian:

5. Lastly, a very important Hamiltonian, is the Hamiltonian for a simple har-

1.2 Hamilton’s equations

1.2.1 Poisson brackets

This operation is called Poisson brackets, it shall prove importance in quanti-

It is interesting to Poisson bracket the Hamiltonian with components of ~q and p~,

1.3 Phase space

1.3.1 Example: phase-space for SHO

5. Discuss conserved quantities, and Neother’s theorem in light of Hamiltonian

2.1 Abstract vector spaces

α|ψi + β|φi, (2.1)

We can represent the components of the above vector in what-so-called a spike

2.2 Functions as vectors

Figure 2.2: Spike digram of a vector having infinite components

classical graph of a real-valued function f (x): in this way, a function is just a

Figure 2.3: Classical graph of a function f (x)

2.3 Dual spaces and inner product

2.3.1 The Bra-Ket notation

2.3.2 Normed spaces

2.4 Hilbert space

2.4.1 Basis of a Hilbert space

The coefficients cj = hej |ψi is sometimes called Fourier coefficients. We can

The basis satisfy4 :

2.5 The space L2

2.6 Outer product

2.7 Linear operators

2.7.1 The algebra of operators

(αÂ + β B̂)|ψi = α(Â|ψi) + β(B̂|ψi) (2.17)

Â|ψi = λ|ψi (2.20)

3. Self-adjointness ( Hermitian operators)

• If an inverse for the operator exists then:

An operator is called self-adjoint , if it is equal to its hermitian conjugate:

2.8 Spectral theorem

Â|uj i = λj |uj i (2.22)

Then the operator Â is decomposed as follows:

As for uncountably-infinite dimensional Hilbert spaces , which we refer to by:

2.9 Projection operators

2.9.2 Projection operators

2.10 Unitary operators

The basis vectors are:

2.11.2 The differential operator

f (x) = Ae+λx + Be−λx (2.35)