QUANTUM CIRCUITS PHYS C-0254

Lecture 1

Lecturer: G. S. Paraoanu
Department of Applied Physics, School of Science,
Aalto University, P.O. Box 15100, FI-00076 AALTO, Finland

1
I. INTRODUCTION

In this lecture we will review key concepts from quantum mechanics, electromagnetism,
and solid-state physics which will be essential for this course.

II. MOTIVATION

Moore’s Law – Doubling of the number of transistors per chip every 18-24 months. But
will it hold and for how long?

FIG. 1. Moore’s Law

We are already reaching some hard physical limits:

– dimension of gate ≈ 5 nm presently! Very near the molecular scale and approaching
atomic scale, where quantum effects (tunneling) will become important. Note that the
lattice constant of Si (diamond cubic crystal structure) is 0.54 nm.

2
 – Density: with 5 nm technology the densities are ≃ 108 /mm2 .

– power density, i.e., how much heat they generate per unit time and per area. Presently
approaching 6 W/mm2 = 600 W/cm2 . Compare with a light bulb ≃ 0.01 W/mm2 = 1
W/cm2 and with our Sun = 60 W/mm2 .

This is why research into alternative forms of information processing such as quantum
computing is highly relevant.

III. CLASSICAL WAVE PHYSICS

Recall that plane waves have the form

ψ(x) = eikx , (1)

where x denotes the position, and k the wavenumber. The wavelength λ relates to the
wavenumber via λ = 2π/k. Furthermore, recognizing the dimensionless quantity ϕ(x) = kx
as the phase, Eq. (1) is more compactly written as ψ(x) = eiϕ(x) . In Fig. 5 we show how
the phase ϕ(x) advances with x.

FIG. 2. Simple representation of a classical wave that puts in evidence the complex phase ϕ(x).

A very useful representation for waves is terms of the Fourier transforms:

(k,ω)-space
Let us first consider time-independent problems: we would usually write

ˆ ∞ ˆ ∞
−ikx 1
ψ[k] = dxe ψ(x) ⇐⇒ ψ(x) = dkeikx ψ[k]. (2)
−∞ 2π −∞

3
 In general, for both time- and space- dependent problems we have to extend this by the
use of a double Fourier transform:

ˆ ∞ ˆ ∞ ˆ ∞ ˆ ∞
−ikx+iωt 1
ψ[k, ω] = dx dte ψ(x, t) ⇐⇒ ψ(x, t) = dk dωeikx−iωt ψ[k, ω] .
−∞ −∞ (2π)2 −∞ −∞
(3)
Some properties:

Real coordinates (x, t) Fourier coordinates (k, ω)

ψ(x, t) → ψ(x − x0 , t) ψ[k, ω] → e−ikx0 ψ[k, ω]

ψ(x, t) → eik0 x ψ(x, t) ψ[k, ω] → ψ[k − k0 , ω]

ψ(x, t) → ψ(x, t − t0 ) ψ[k, ω] → eiωt0 ψ[k, ω]

ψ(x, t) → e−iω0 t ψ(x, t) ψ[k, ω] → ψ[k, ω − ω0 ]

Quiz: what happens with the derivative and indefinite integrals of ψ(x, t)?? - to what
functions they correspond in the (k, ω) space?

IV. QUANTUM MECHANICS

Recall the time-dependent Schrödinger equation:

d
iℏ ψ(x, t) = Hψ(x, t) , (4)
dt
where ψ(x, t) is the wavefunction. Also, the probability density is |ψ(x, t)|2 and the proba-
bility that a particle is between points a and b is

ˆ b
|ψ(x, t)|2 dx . (5)
a

• Typically H = kinetic energy + potential energy, i.e.,

p2
H= +V , (6)
2m
where p is the momentum operator, which in the position representaion reads
d
p = −iℏ . (7)
dx

4
 Importantly, p relates to the position operator x via the canonical commutation relation:

[x, p] = iℏ. (8)

• One of the most useful notations is the Dirac bra-ket; it is more abstract and more
general. Here are some exampples of how it works.

1. We obtain the wavefuntion in the position representation ψ(x, t) = ⟨x|ψ(t)⟩.

´∞ ´∞
2. Getting the normalization: −∞ dx|ψ(x, t)|2 = −∞ dx⟨ψ(t)|x⟩⟨x|ψ(t)⟩ = ⟨ψ(t)|ψ(t)⟩ =
´∞
1, where we have used the completeness relation −∞ |x⟩⟨x| = I (I) is the iden-
tity).

3. time-dependent Schrödinger equation: iℏ dtd |ψ(t)⟩ = H|ψ(t)⟩.

Note that it is possible to have a time-dependent Hamiltonian H(t). However,
if H is time-independent, we can solve the Schrödinger equation by the method
of separation of variables. For example, if we start in a state |ψ(0)⟩ which is
an eigenstate of the energy, that is H|ψ(0)⟩ = E|ψ(0)⟩, then the solution of the
E
time-dependent Schrödinger equation is |ψ(t)⟩ = e−i ℏ t |ψ(0)⟩.

A. Free particle

For a free particle by definition V (x) = 0.

Solution:
ℏ d 2 2
The time-dependent Schrödinger equation is − 2m dx2
ψ(x) = Eψ(x), and can be solved
p
by ψ(x) = const × e±ikx , with k = 2mE/ℏ2 .
Then, using the method above, we can write the solution of the time-dependent Schrödinger
equation iℏ dtd |ψ(t)⟩ = H|ψ(t)⟩ as ψ(x, t) = const × e±i(kx−ωt) , where ω = E/ℏ.
In condensed matter (solid-state) physics is often convenient to work with periodic bound-
ary conditions: this gives us a handy way to control the normalization factor. We imagine a
box of length L centered at zero and impose the condition that the wavefunction is the same
at −L/2 and L/2: ψ(−L/2, t) = ψ(L/2, t). This implies e±−ikL/2 = e±ikL/2 or e±ikL = 1 or

2πn
k= . (9)
L

5
Here n are integers - they can be positive or negative. The relation E = ℏk 2 /2m is an
specific instance of dispersion relation, where the dependence on k is quadratic, reflecting
the non-relativistic character of the Schrödinger equation.

B. Infinite square well

Consider an infinite square well potential with the boundaries


0, x ∈ [−L/2, L/2]

V (x) = (10)
∞, x ∈ (−∞, −L/2) ∪ (L/2, +∞) .


We seek the eigenenergies and eigenstates.

Solution:
ψ(x) = A sin(kx + kL/2).
The boundary conditions are

x = −L/2 → ψ(−L/2) = 0

(11)
x = L/2 → ψ(L/2) = A sin(kL) = 0 ,


where the second condition implies that k is quantized

nπ
k= . (12)
L

Note the difference in the factor of 2 with respect to the previous case: also this time
n = 1, 2, 3 · · · are positive natural numbers. Indeed k = 0 is not a solution because it
leads to a zero wavefunction (in the previous case of a free particle it leads to a constant
√
wavefunction 1/ L, which is OK), and also negative values would be redundant since they
will just change the sign of the wavefunction (and in quantum mechanics the wavefunction
is defined only up to a global phase).
With normalization we have
ˆ +∞ ˆ L/2  nπ
2 nπ 
|ψ(x)| dx = dxA2 sin2 x+ =1, (13)
−∞ −L/2 L 2
p
which implies that A = 2/L. Moreover, the eigenstates are
p  kL 
ψn (x) = 2/L sin kx + , (14)
2

6
where n = 1, 2, 3 · · · .
The eigenenergies are recovered from the time-independent Schrödinger equation

ℏ2 d2
− ψn (x) = En ψn (x) , (15)
2m dx2

where we find En = ℏ2 k 2 /2m.

FIG. 3. 1-D infinite square well potential.

Important observations:

– there exists a minimum non-zero energy E1 = ℏ2 π 2 /2mL2 , corresponding to the ground

state ψ1 . So a particle in a box always has some kinetic energy! This is very different
from classical physics.

– energy levels are quantized – not every energy is allowed!– and form a discrete ladder.

– En ∝ 1/L2 . The larger the box, the smaller the gap between levels. Eventually, as
L → ∞ we reach the continuum again.

7
 C. The quantum harmonic oscillator

The quantum harmonic oscillator has the Hamiltonian

ℏ2 d2 1
H=− 2
+ mω 2 x2 . (16)
2m dx 2
We wish to find eigenenergies and eigenstates.

Solution:
En = (n + 1/2)ℏω, where n = 0, 1, 2 · · · .
−mω 2
p   1/4
ψn (x) = Nn e− 2ℏ x Hn mω
ℏ
x , where Nn = √1
2n n!
mω
πℏ
, and Hn is a Hermite poly-
nomial of degree n.
 
2 dn 2
Explicitly, Hn (z) = (−1)n ez dz n
e−z , so
H0 (z) = 1
H1 (z) = 2z
H2 (z) = 4z 2 − 2
H3 (z) = 8z 3 − 12z

Introduce the ladder operators a (lowering) and a† (raising)

p mω   q
• a= 2ℏ
x+ i
mω
p x= ℏ
2mω
(a + a† )

p mω   q
• a† = 2ℏ
x− i
mω
p p = i ℏmω
2
(a† − a)

(a† )n
• The number operator N = a† a, N |n⟩ = n|n⟩, where |n⟩ = √ |0⟩.
n!
Note: a|0⟩ = 0.
Therefore, in terms of the number operator, the Hamiltonian is H = ℏω(N + 1/2).

• Commutation relations:
[a, a† ] = 1
[N, a† ] = a†
[N, a] = −a

Important Observations:

• Energy levels are equally spaced by ℏω.

8
 • There exists a minimum energy of ℏω/2 which corresponds to the ground state, i.e.,
zero-point motion energy.

Example: Calculate the variance ⟨(∆x)2 ⟩ = ⟨x2 ⟩ − ⟨x⟩2 for the vacuum state |0⟩, where
∆x ≡ x − ⟨x⟩.
Solution:
q
ℏ
With x = 2mω (a + a† ), we find ⟨x⟩ = 0, and ⟨x2 ⟩ = ℏ
2mω
. Therefore, ⟨x2 ⟩ = x2ZPF , where
q
ℏ
xZPF = 2mω is the zero-point fluctuation.

D. Spin-1/2 particles

It comes from the Stern-Gerlach experiment where a beam of silver atoms running through
a non-homogeneous magnetic field is split into two beams.
ℏ
Angular momentum theory for spin-1/2: S = 2
σ σ = (σx , σy , σz )

s = 1/2:

| 1 , 1 ⟩

m = 1/2
2 2
|s, m⟩ = (17)
| 1 , − 1 ⟩ m = −1/2

2 2

where s refers to the fact that this is a system with total spin 1/2: S 2 | 21 , 21 ⟩ = ℏ2 s(s +
1)| 12 , 12 ⟩ = 43 ℏ2 | 12 , 12 ⟩ .

• In quantum information we use simplified notations:


| 12 , 12 ⟩ ≡ |0⟩
qubit states (18)
| 21 , − 21 ⟩ ≡ |1⟩

General qubit state: |ψ⟩ = cos 2θ |0⟩ + eiϕ sin 2θ |1⟩ .

• Eigenvectors-eigenvalues:

Question: What is the analogue of zero-point fluctuations for spin-1/2?

9
 0 −i 1 0
01




σx = 10 σy = i 0 σz = 0 −1
(x) (x) (y) (y)
σx |χ± ⟩ = ±|χ± ⟩ σy |χ± ⟩ = ±|χ± ⟩ σz |0⟩ = |0⟩, σz |1⟩ = −|1⟩
(x) (y)
|χ± ⟩ = √12 ±1
1 |χ± ⟩ = √12 ±i
1





|0⟩ = 10 , |1⟩ = 01

E. Many-particle quantum systems

In many-particle systems it is necessary to concatenate the Hilbert spaces of each particle.

V , W Hilbert spaces |v⟩ ∈ V , |w⟩ ∈ W
V ⊗ W = tensor product |v⟩ ⊗ |w⟩ ∈ V ⊗ W
But how do we write the wavefunctions? Say we have two particles, is the wavefunction
|v⟩1 |w⟩2 , |w⟩1 |v⟩2 or
√1


2
α|v⟩1 |w⟩2 + β|w⟩1 |v⟩2 ? This topic is a bit more advanced and we are only interested
here in the key concepts: appreciating that the formalism for bosons is very simular to the
algebra of the harmonic oscillator that you are used to and that the Pauli exclusion principle
is a natural consequence of the fermionic algebra.
In nature there are only two types of particles:
bosons → symmetric wavefunction

fermions → anti-symmetric wavefunction

Good news, we do not necessarily need to work with cumbersome symmetrized or anti-
symmetrized wavefunctions. Instead, a compact way of writing the wavefunction is provided
by the Fock space:

|n1 , n2 , · · · ⟩
Bosons:
[ai , a†j ] = δij
[ai , aj ] = [a†i , a†j ] = 0
N = i a†i ai
P
√
a†i | · · · , ni , · · · ⟩ = ni + 1| · · · , ni+1 , · · · ⟩
√
ai | · · · , ni , · · · ⟩ = ni | · · · , ni−1 , · · · ⟩
ai | · · · , ni = 0, · · · ⟩ = 0

10
|n1 , n2 , · · · ⟩ = √ 1
(a† )n1 (a†2 )n2
n1 !n2 !··· 1
· · · |0, 0, · · · ⟩

Fermions:
ci , c†j = δij
ci , cj = c†i , c†j = 0
N = i c†i ci
P

c†i | · · · , ni , · · · ⟩ = (1 − ni )(−1)
P
nj
j<i | · · · , ni+1 , · · · ⟩
P
nj
ci | · · · , ni , · · · ⟩ = ni (−1) j<i | · · · , ni−1 , · · · ⟩
ci | · · · , ni = 0, · · · ⟩ = 0
c†i | · · · , ni = 1, · · · ⟩ = 0 −→ Pauli exclusion principle.
|n1 , n2 , · · · ⟩ = (c†1 )n1 (c†2 )n2 · · · |0, 0, · · · ⟩

V. ELEMENTS OF SOLID STATE PHYSICS

Electrons – they are fermions; i.e., Pauli exclusion principle applies.

– at T = 0, we fill all the states until we use all of the electrons.
For example, take a wire (just because we have a single k-vector · · · ).
At T ̸= 0 the distribution of electrons is described by the Fermi-Dirac distribution:

conduction valence gap

band band

No states available for electrons

FIG. 4. Energy bands of a metal and insulator.

1
fFD (E) = . (19)
exp [(E − EF )/kB T ] + 1

11
 FIG. 5.

Two limits:

• degenerate limit: fFD (E) ≈ Θ(EF − E).

• non-degenerate limit: fFD = exp [−(E − EF )/kB T ]

when E − EF ≫ kB T .

Density of states:

Consider the free electron wavefunction with periodic boundary conditions, but this time
generalized to three dimensions:

eikx ·x eiky ·y eikz ·z 1 ⃗

ψ(⃗r) = √ · √ · √ = √ eik·⃗r . (20)
L L L V
Here, V = L3 is the volume in three dimensions and E = ℏ2 k 2 /2m is the dispersion
relation, where kx = 2πnx /L, ky = 2πny /L, and kz = 2πnz /L.

12
 We want to calculate the number of states per volume within an energy interval dE. This
is known as the density of states, see Fig. 6.
 3
2π
The volume element in k-space is V3D = L
and the volume of shell between k and
k + dk is Vdk = 4πk 2 dk.

FIG. 6. k-space schematic for calculating the density of states.

Vdk k2 dk
The number of states in this shell is 2 · V3D
= π2
· L3 , where the factor of 2 comes from
the electron spin.
E = ℏ2 k 2 /2m =⇒ dk = √ 1 m
2 dE
2mE/ℏ2 ℏ

Therefore, the number of states in the interval dE per unit volume is

k 2 dk 1  2m 3/2 √
N3D (E)dE = = EdE . (21)
π2 2ℏ2 ℏ2

Exercise: To check that you understood how this works, do this calculation by con-
sidering the model of a particle confined in a box rather than a free particle with periodic
boundary conditions. Note that in this case kx = πnx /L, etc. and you will have to consider
only one octant since kx , ky , kz are all positive.

VI. ELECTROMAGNETISM

Maxwell’s equations (in SI units):

13
 ⃗
⃗ = J⃗ + ∂ D
∇×H Faraday’s Law
∂t
⃗
∂B
⃗ =−
∇×E Ampère’s Law
∂t (22)
⃗ ·D
∇ ⃗ =ρ Coulomb’s Law

⃗ ·B
∇ ⃗ =0 Gauss’ Law ,

where J⃗ is the current density and ρ is the charge density.

Electromagnetic potentials:
⃗ defined as B
* magnetic vector potential A ⃗ =∇
⃗ × A.
⃗

* electric potential V .
⃗ = −∇V
Electric field in terms of potentials: E ⃗ − ∂ ⃗
A.
∂t

Constitutive relations:
⃗ = εE
D ⃗ ε = ε0 εr = electrical permittivity,
ε0 = 8.854 × 10−12 F/m = vacuum permittivity
εr = relative permittivity

c= √1 = speed of light in vacuum

ε0 µ0
⃗ = µH
B ⃗ µ = µ0 µr = magnetic permeability
µ0 = 4π × 10−7 H/m = vacuum permeability.
µr = relative permeability

ϵ′′
For the case of AC fields: ϵ → ϵ = ϵ′ − iϵ′′ & tan δ ≡ ϵ′
= loss tangent.

Other fundamental relations:

• Ohm’s law J⃗ = ⃗σ · E
⃗

• Continuity equation: − ∂ρ
∂t
⃗ · J⃗
=∇
˝ ‚
• Gauss-Ostrogradsky theorem: − ∂t
∂
V
dV · ρ⃗ = S
⃗ i.e, the rate of decrease of
d⃗s · J,
positive charge = total current flux flowing out of the closed surface.

14
 FIG. 7.

VII. THERMODYNAMICS

• First law of thermodynamics: ∆U = Q − W

This is conservation of energy, where ∆U denotes the change in internal energy, Q
is the heat supplied to the system, and W is the work done by the system onto the
environment.

• Second law of thermodynamics: δQ = T dS (for reversible processes)

This says that the total entropy of an isolated system can never decrease. δQ denotes
the amount of heat transferred and dS is the change in entropy produced by the
transferred heat.
Corollary: It is impossible to construct a cyclic engine that produces work from the
energy extracted from a single reservoir (Planck).

• Third law of thermodynamics.

The entropy approaches a constant value when T → 0

lim S = const .
T →0

Equipartition Theorem
In thermal equilibrium, energy is shared equally between the degrees of freedom ( 21 kB T
per degree of freedom).
Example:
1
Ideal gas: E = 2
+ 21 mvy2 + 12 mvz2 = average energy =
mvx2 1
k T
2 B
+ 12 kB T + 12 kB T =
p q
3
k T
2 B
=⇒ vrms = ⟨v ⟩ = 3kmB T .
2

p2
Harmonic oscillator: E = 2m
+ 21 kx2 = average energy 21 kB T + 12 kB T = kB T
2
where kB = 1.38 × 10−23 ms2 Kkg – Boltzmann’s constant.

15
 Further Reading
Any introductory textbook on quantum mechanics, solid state physics, and electromag-
netism should suffice. For example,

• The Open University: SM358 The Quantum World

Science Level 3 Books 1-3

• David J. Griffiths – Introduction to Quantum Mechanics

• Charles Kittel – Introduction to Solid State Physics

• Martin Sibley – Introduction to Electromagnetism

There is a plethora of information, lecture notes, and video lectures on the internet!

16