Unit IV- QUANTUM COMPUTING

Stern-Gerlach Experiment:

This experiment demonstrates the spin states possessed by electrons in an atom.

Principle:

An electron revolving around a nucleus results in orbital magnetic moment (ml) and the spin
of an electron results in spin magnetic moment (ms). An atom with an unpaired electron, behaves like
a bar magnet. When a bar magnet is placed in an inhomogeneous magnet field, it experiences a
resultant force. The magnitude and direction of the resultant force depend on the orientation of the axis
of the bar magnet relative to the magnetic field. This force causes deflection of the bar magnet.

A silver atom is a monovalent element and it has an unpaired electron in 5s orbital. This
unpaired electron results in spin magnetic moment of the atom. Therefore, a silver atom behaves like
a bar magnet. The spin magnetic moment of an electron can have only two possible orientations in an
external magnetic field – one with spin up and the other one with spin down, with respect to the
magnetic field. When the silver atoms are passed through an inhomogeneous magnetic field, if the
atoms with spin up electrons are deflected upwards, the atoms with spin down electrons are deflected
downwards by the inhomogeneous magnetic field.

Construction:

In Stern-Gerlach experiment, silver atoms are produced by heating silver in a small electric
oven. An inhomogeneous magnetic field is produced by having a knife edged and a cylindrically
grooved magnetic poles as shown in the figure. The intensity of magnetic field is greatest at the knife
edge and decreases towards the cylindrical pole. The silver atoms are passed through a series of slits
to make a thin beam of atoms. Then, they are passed through the inhomogeneous magnetic field. Trace
of the silver atoms is recorded on a photographic film. The entire set up is enclosed in a vacuum
chamber to avoid collision of silver atoms with air molecules.
 S

Magnetic field lines ⃗ =0

𝐵 ⃗ >0
𝐵

N Trace of silver atoms on the photographic film

Inhomogeneous magnetic field

Result and Discussion:

In the absence of magnetic field, the trace of the silver atoms on the photographic film is a
horizontal straight line. In the presence of the inhomogeneous magnetic field the silver atom beam
splits into two components and the trace is as shown in the figure.

The angular momentum J of silver atoms is entirely due to spin of its valence electron. The
spin angular momentum is given by

ℎ
𝑆 = √𝑠(𝑠 + 1)
2𝜋

Where, s = ½.

The number of orientations of the angular momentum with respect to an external magnetic field
is given by (2s+1) = [2(1/2) +1] = 2. Therefore, the spin magnetic moment can have only two
orientations with respect to the magnetic field.

𝑑𝐵
Suppose, the magnetic field is inhomogeneous along Y-direction. The field gradient is . Let the
𝑑𝑦

atomic magnet with magnetic moment M, pole strength p and length l inclined at an angle θ with the
field direction is placed in such a field as shown in the figure. If the field strength at one pole is B then
𝑑𝐵
the field strength at the other pole is 𝐵 + 𝑑𝑦 𝑙𝑐𝑜𝑠𝜃. Force on one pole of the atomic magnet is pB and
𝑑𝐵 𝑑𝐵
on the other pole is 𝑝 (𝐵 + 𝑑𝑦 𝑙𝑐𝑜𝑠𝜃). The extra force 𝑝 (𝑑𝑦 𝑙𝑐𝑜𝑠𝜃) on one pole displaces the atom

as a whole. The force Fy is given by

𝑑𝐵 𝑑𝐵
𝐹𝑦 = 𝑝. 𝑙 𝑐𝑜𝑠𝜃 ( ) = 𝑀 𝑐𝑜𝑠𝜃 ( ).
𝑑𝑦 𝑑𝑦

Where, pl = M is the magnetic moment.

Due to this force, the atomic magnet will be displaced from its straight path in the field
direction. The splitting of silver atom beam into two components in an inhomogeneous magnetic field
verifies the existence of some silver atoms with spin up electrons and some silver atoms with spin
down electrons.

Polarization of Light:

Light wave: Light wave is a transverse electromagnetic wave. A light wave consists of electric and
magnetic fields vibrating perpendicular to each other and to the direction of propagation. It is the
electric vector which is more effective in a light wave. Therefore, the plane in which the electric vector
oscillates is important.

Unpolarised light: A light wave which has its electric vector oscillating in all possible directions
perpendicular to the direction of propagation of light is called an unpolarised light.

Polarised light: A light wave which has its electric vector oscillating geometrically restricted is called
an unpolarised light.

Linearly polarised light: A light wave whose electric vector is restricted to a plane perpendicular to
the direction of propagation of light is called a linearly polarised light or plane polarised light.

Circularly polarised light: A light wave whose tip of electric vector traces a circle perpendicular to
the direction of propagation is called a circularly polarised light.
Elliptically polarised light: A light wave whose tip of electric vector traces an ellipse perpendicular
to the direction of propagation is called an elliptically polarised light.

A plane polarised light can be converted into either a circularly or an elliptically polarised light by
suitable methods.

Production of plane polarised light: Plane polarised light can be produced by,

(i) Reflection
(ii) Refraction
(iii) Scattering
(iv) Selective absorption
(v) Double refraction

Polariser: A polariser is an optical device that transforms unpolarised light into polarised light.

Analyser: An analyser is an optical device which is used to identify the plane of vibration of a polarised
light.

Half wave plate: A half wave plate is a thin birefringent crystal of right thickness that introduces a
𝜆
path difference of between two orthogonal electric vibrations one making 0° and the other one
2

making 90° with the optic axis of the half wave plate.

Quarter wave plate: A quarter wave plate is a thin birefringent crystal of right thickness that
𝜆
introduces a path difference of between two orthogonal electric vibrations one making 0° and the
4

other one making 90° with its optic axis.

Production of plane polarised light:

Direction of
Unpolarised light Polarised light propagation

Polariser

Fig.1 Production of plane polarised light.

An unpolarised light which has the electric field vector oscillating in all possible directions is passed
through a polariser. The polariser allows only that part of light whose electric vector vibration is
restricted to only one plane and blocks all other vibrations. A linearly polarised light is obtained on the
other side of the polariser as shown in Fig.1.

Detection of plane polarised light:

Clockwise rotation
of analyser

Detector
Plane polarised
light
Analyser

Fig. 2 Detection of plane polarised light

An analyser is introduced in the path of light as shown in Fig.2. The analyser is rotated through
one full circle either clockwise or anticlockwise. If the intensity of transmitted light is zero twice, then
the light is plane polarised.

Production of circularly polarised light:

θ = 45°
θ

Plane polarised Circularly

light polarised light
Quarter
wave plate

Fig.3 Production of circularly polarised light.

A beam of plane polarised light is incident on a quarter wave plate such that the electric vector makes
an angle θ = 45° with the optic axis of the quarter wave plate as shown in Fig.3. The resultant light is
a circularly polarised light.

Detection of circularly polarised light:

Clockwise rotation
of analyser

Detector
Circularly
polarised light
Analyser

QWP

Fig. 4 Detection of circularly polarised light

An analyser is introduced in the path of light as shown in Fig.4. The quarter wave plate (QWP)
converts light into plane polarised light if it is circularly polarised. The analyser is rotated through one
full circle either clockwise or anticlockwise. If the intensity of transmitted light is zero twice, then the
light is circularly polarised.

Production of elliptically polarised light:

θ θ ≠ 45°

Plane polarised Elliptically

light polarised light
Quarter
wave plate

Fig.5 Production of circularly polarised light.

 A beam of plane polarised light is incident on a quarter wave plate such that the electric vector
makes an angle θ ≠ 45° with the optic axis of the quarter wave plate as shown in Fig.5. The resultant
light is an elliptically polarised light.

Detection of elliptically polarised light:

Clockwise rotation
of analyser

Ellipticallyly Detector
polarised light
Analyser

QWP

Fig. 6 Detection of elliptically polarised light

A quarter wave plate and an analyser are introduced in the path of light as shown in Fig.6. The
quarter wave plate (QWP) converts light into a plane polarised light if it is elliptically polarised. The
analyser is rotated through one full circle either clockwise or anticlockwise. If the intensity of
transmitted light is zero twice, then the light is elliptically polarised.

Classical Computing:

The principles of classical physics and mathematical reasoning form the foundation of classical
computing.
Traditional computer software is designed for serial computation. It indicates that one task must be
finished before starting another.
Parallel computing is possible on a computer with multiple processors. It means it can work on multiple
tasks at the same time, and the results should be integrated.
Moore’s Law:
Moore’s law is not a natural law. It is an observation did by Gordon G. Moore.
It states that the number of transistors on a computer chip and, thus its power doubles every 2
years.
Computer hardware is getting smaller, cheaper and faster. This has been possible because of the
miniaturization in ICs (integrated circuits). It means the spacing between the transistors and other
components keeps on decreasing.
Is Moore’s law still holding?
Although Moore’s law trend continued for more than 50 years, the transistor counts doubled after 3
years, not 24 months.

Need for quantum computing:

There are two problems associated with miniaturization.
1) When the separation between the components is of the order of atomic dimensions, the uncertainty
principle and other quantum mechanical laws become prominent. In other words, the computation
based on classical physics will no longer be reliable.
2) The heat produced by one component will affect the performance of another component. The present
day computers generate a lot of heat.

Quantum computing:
A quantum computer is a machine that performs calculations based on the laws of quantum
mechanics, which is the behaviour of particles at the sub-atomic level.
The fundamental concept of classical computation and information is a “bit”. The data is stored in a
bit. The bit can take the value 0 or 1.
The corresponding parameter in a quantum computer is a quantum bit or qubit.

Single particle quantum interference

A laser that emits one photon at a time is used in the experimental setup (Fig. a). This can be
accomplished by employing a set of attenuators, which can filter out laser light and ensure that only
one photon emerges at a time. This photon is then split by a beam splitter. The beam splitter reflects
half of the light that strikes it and allows the other half to pass through. Photon detectors A and B
detect the photon with equal probabilities. Hence it can be concluded that during any one run the
photon has traveled one of the paths since it cannot be split into two. However, this assumption is not
true.
 Detector A

V
Single Photon
H Detector B
Beam splitter

Fig. (a)
Now experimental set up is changed as shown in Fig. (b). A single photon may travel
horizontally, gets deflected by a mirror, and reaches the detector. Another possibility is that it passes
vertically, gets deflected by a mirror, and reaches the detector. Hence if the photon really takes a single
path through the apparatus, both detectors would detect it with equal probabilities. However, this does
not happen. The photon always strikes detector A and never detector B. If we change the path length
by introducing a glass plate in one of the paths (say vertical one), the photon is detected by detector B
and never by detector A. It means that the photon was in a superposition state, and it travelled through
both paths simultaneously. At the second beam splitter, both the components interfered, constructively
or destructively, and get detected by one of the detectors.

Detector
Mirror B
Beam
splitter

One photon at a time

Mirror
Beam splitter

Fig. (b)
Difference between classical computing and quantum computing:

Classical Computing Quantum Computing

Conventional computing is based on the Quantum computing is based on the

classical phenomenon of electrical circuits phenomenon of Quantum Mechanics, such
being in a single state at a given time, as superposition and entanglement, the
either on or off. phenomenon where it is possible to be in
more than one state at a time.

Information storage and manipulation are Information storage and manipulation are
based on “bit”, which is based on voltage based on Quantum Bit or “qubit”, which is
or charge; low is 0 and high is 1. based on the spin of an electron or
polarization of a single photon.

The circuit behaviour is governed by The circuit behaviour is governed by

classical physics. quantum mechanics.

Conventional computing uses binary codes Quantum computing uses Qubits i.e. |0>,
i.e. bits 0 or 1 to represent information. |1> and the superposition state of both |0>
and |1> to represent information.

Transistors are the basic building blocks of Superconducting Quantum Interference

conventional computers. Devices (SQUID) or Quantum Transistors
are the basic building blocks of quantum
computers.

In conventional computers, data In quantum computers, data processing is

processing is done in the Central done in a Quantum Processing Unit or
Processing Unit or CPU, which consists of QPU, which consists of several
an Arithmetic and Logic Unit (ALU), interconnected qubits.
processor registers and a control unit.
Comparison of classical and quantum information

➢ Information, either classical or quantum, is physical.

➢ It is transmitted by physical means.
➢ It is stored in physical system.
Quantum information Classical information
It is encoded to some property of a It is encoded to some property of a
quantum system like the polarization of a physical system obeying the laws of
photon or spin of an electron. classical physics.
It is processed using quantum gates. It is processed using classical gates.
The fundamental unit of information is a The fundamental unit of information is a
qubit. bit.
It is difficult to store, transmit and It is easy to store, transmit and process.
process.
There is no way to copy unknown It is easy to make copies of classical
information. information.
In general, the measurement of It can be measured without disturbing it.
information destroys it.

Quantum Superposition

Quantum superposition is a phenomenon associated with quantum systems such as nuclei,

electrons and photons, for which wave-particle duality and other non-classical effects are observed. A
quantum system can exist in more than one state at the same time. The result of the measurement is
the observation of some definite state with a given probability.
Quantum superposition is easily demonstrated using a coin. A coin has a 50/50 probability of landing
as either heads or tails while flipped in the air.

What state is the coin in while it is in the air? Is it heads or tails?

We can say that the coin is in a superposition of both heads and tails. When it lands, it has a definite
state, either heads or tails. The word “state” means any particular way that a system can possibly be
described. For example, the coin can be either heads, or tails, or a combination of heads or tails while
flipped in the air. All of these cases are called states of the coin system. The measurement destroys the
superposition.
Qubit:

A qubit, like a bit, also makes use of two states |0⟩ and |1⟩ to hold information.

Mathematically, qubits |0> and |1> can be represented as column matrices:

1 0
|0> = [ ]; |1> = [ ]
0 1

However, unlike classical bits, a qubit, |Ψ> can also be in a superposition state of |0> and |1> states.

It can be written as |Ψ> =α |0> + β |1>

where α and β are generally complex numbers which represent the probability amplitudes of the states.

When a qubit is measured, it only results in either |0> or |1>.

Summation of probabilities

The probability of measuring the qubit in state |0> is |α|2, and the probability of measuring the qubit in
state |1> is |β|2.

Since the total probability of observing all the states of the quantum system must add up to 100%, the
amplitudes must follow this rule:

|α|2 +|β|2 =1

This is called a normalization rule.

Physical Realization of Qubits:

In a classical computer, the 0- and 1-bit mathematically represent the two allowed voltages
across a wire in a classical circuit. Semiconductor devices called transistors are used to control what
happens to these voltages.

“What is a qubit made out of?”

➢ Energy levels of an atom: Consider the electron in a hydrogen atom. It can be in its ground
state (i.e. an s orbital) or in an excited state. So we can also store a qubit of information in the
quantum state of the electron, i.e., in the superposition.
Ground state |0>
Excited state |1>
 ➢ Spin: Elementary particles like electrons and protons carry an intrinsic angular momentum
called spin. Their spins can be used as qubits with |0 >= | ↑>, |1 >= | ↓>
➢ Polarization of Photon: A linearly polarized photon can be either horizontally or vertically
polarized with respect to some direction in which the photon is moving. Quantum researchers
can create photons one at a time and encode qubits of information into their polarization.

\Bloch Sphere Representation

Bloch sphere is a physical representation of all possible qubit states. It is a sphere of unit radius
and the state of a qubit can be represented by a vector in this sphere. |0> is at the north pole, |1> is at
the south pole, as shown in Figure.
Using the spherical coordinate system, an arbitrary position of the state vector of a qubit can be written
in terms of the angles θ (elevation, the state vector makes from the z-axis) and ϕ (azimuth, the angle
of projection of the state vector in the x-y plane from the x-axis) it makes in the Bloch sphere as:

𝜃
𝜃 𝜃 cos ( )
|𝜓 > = cos ( ) |0 > + 𝑒 𝑖𝜙 sin ( ) |1 > = [ 2 ]
2 2 𝜃
𝑒 𝑖𝜙 sin ( )
2

Note:

• For φ = 0 and 𝜃 = 0, the state |𝜓 >corresponds to |0> and is along +z-axis.

• For φ = 0 and 𝜃 = 180o the state |𝜓 > corresponds to |1> and is along -z-axis.
• When 𝜃 = 90o, |𝜓 > is in the x-y plane.
1
For φ = 90o, |𝜓 > = ( |0⟩ + 𝑖|1⟩ ), is a superposition state along +y-axis.
√2
1
For φ = -90o, |𝜓 > = ( |0⟩ − 𝑖|1⟩ ), is a superposition state along the -y-axis.
√2
1
For φ = 0o, |𝜓 > = ( |0⟩ + |1⟩ ), is a superposition state along the +x-axis.
√2
1
For φ = 180o, |𝜓 > = ( |0⟩ − |1⟩ ), is a superposition state along the -x-axis.
√2

For a classical computer, the two logical states 0 and 1 are represented by the poles of a sphere.

In contrast, the state of a qubit can be represented by any point on the sphere. Since there are infinite
points on the sphere, a qubit in principle has more capacity to store information compared to a classical
bit.

Note: Bloch sphere represents the state of only one qubit. There is no generalization of the Bloch
sphere for multiple qubits.

Difference between classical “bits” and “qubits”:

A fundamental difference between classical bits and qubits is the way they operate.

The classical bits can be deterministically set in a “0” or a “1” state.

It is read (or measured) any number of times as long as it is powered.

Reading a classical bit does not destroy its state. A bit retains its state as long as it is powered.
The qubits are probabilistic.

They are in a superposition state of |0> and |1> with different probabilities.

They possess characteristics of both states simultaneously, at all times, until measured.

The qubits lose their internal state when they are measured.

The framework of quantum mechanics for quantum computing:

Wave function in Dirac notation:

A wave function (say 𝜓) represents the physical state of a system.

According to Paul Dirac, the state of a system is described by a vector, called a state vector, in Hilbert
space ℋ. Depending on the degree of freedom (i.e. the type of state) of the system being considered,
ℋ may have infinite-dimensional.

[Hilbert space 𝓗: It is a complex vector space. It has all the properties of linear vector space like
vector addition and scalar multiplication. In addition, it satisfies inner product operation.

An inner product is a generalization of the dot product It is a method of multiplying vectors together
in a vector space, with the result being a scalar.]

If 𝜓 is a wavefunction, then in Dirac notation 𝜓 is represented as | 𝜓 >, which is called a ket vector.

Example: Suppose 𝜓 = 𝐴 𝑒 −𝑖 𝑘 𝑥

Dirac notation | 𝜓 > = 𝐴 𝑒 −𝑖 𝑘 𝑥

Note: Only the notation of 𝜓 is changed. The form of the wave function remains unchanged.

If 𝜓* is the complex conjugate of 𝜓, then 𝜓* is represented as < 𝜓|, which is called a bra vector

Hence,

< 𝜓| = 𝐴∗ 𝑒 𝑖 𝑘 𝑥

Basis: In quantum mechanics, the “basis vectors” can be thought of as a set of mutually perpendicular
vectors, one for each “dimension” of the space in which the state vector is expressed. The magnitude
of a basis vector is one. There is a one-to-one correspondence between basis vectors and dimensions
of the space.

Matrix form of a wave function:

Consider a discrete and complete basis that is made up of an infinite set of kets | 𝜙1 > , |𝜙2 >,
| 𝜙3 > ⋯ etc.

The state vector |𝜓 > can be written as a linear combination of kets | 𝜙1 > , |𝜙2 >, | 𝜙3 > ⋯ etc
as follows:

|𝜓 > = 𝑎1 | 𝜙1 > + 𝑎2 |𝜙2 > +𝑎3 | 𝜙3 > + ⋯ + 𝑎𝑛 | 𝜙𝑛 > = ∑∞

𝑛=1 𝑎𝑛 |𝜙𝑛 >

Where the coefficients a1, a2, a3…. an, represent the projection of |𝜓 > onto | 𝜙𝑛 > . 𝑎𝑛 is the
component of |𝜓 > along the vector | 𝜙𝑛 > .

Hence, |𝜓 > can be represented as a column vector (column matrix) given by

𝑎1
𝑎2
𝑎
|𝜓 > → .3
.
[𝑎 𝑛 ]

[A column matrix is a matrix having all its elements in a single column. The elements are arranged
in a vertical manner. The order of a column matrix having n elements is n x 1]

The bra vector < 𝜓| can be represented by a row vector (row matrix) :

< 𝜓| → [ 𝑎1∗ 𝑎2∗ 𝑎3∗ … … . . 𝑎𝑛∗ ]

[A row matrix is a matrix having all its elements in a single row. The elements are arranged in a
horizontal manner. The order of a row matrix having n elements is 1 x n]

Remark:

A ket |𝜓 > is normalized if < 𝜓|𝜓 > = ∑𝑛|𝑎𝑛 |2 = 1

If |𝜓 > is not normalized, we can multiply it by a constant ‘a’ so that

< 𝑎 𝜓|𝑎 𝜓 > = |a|2 < 𝜓|𝜓 > =1

Hence, the normalization constant ‘a’ = 1/ √< 𝜓|𝜓 >

Inner product:

If 𝜓 = 𝜓(𝑥) and 𝜙 = 𝜙(𝑥) are two wavefunctions, then their inner product can be defined as
 (𝜓, 𝜙) = ∫ 𝜓 ∗ (𝑥) 𝜙(𝑥)𝑑𝑥

In Dirac notation (𝜓, 𝜙) is written as < 𝜓|𝜙 >.

Since the inner product (scalar product) is a complex number in quantum mechanics,

< 𝜓|𝜙 > = < 𝜙|𝜓 >*

This property can be demonstrated as follows:

< 𝜙|𝜓 >* = ( ∫ 𝜙 ∗ (𝑥)𝜓(𝑥)𝑑𝑥 )* = ∫ 𝜓 ∗ (𝑥) 𝜙(𝑥)𝑑𝑥 = < 𝜓|𝜙 >

For any state vector |𝜓 >, < 𝜓|𝜓 > is real and positive.

If the state | 𝜓 > is normalized, < 𝜓|𝜓 > = 1

Therefore, < 𝜓|𝜓 > = 0 only if | 𝜓 > = 0 .

Matrix form of inner product:

Let |𝜓 > = 𝑎1 | 𝜙1 > + 𝑎2 |𝜙2 > +𝑎3 | 𝜙3 > + ⋯ + 𝑎𝑛 | 𝜙𝑛 >

𝑎1
𝑎2
𝑎
= .3
.
[𝑎𝑛 ]

|Φ > = 𝑏1 | 𝜙1 > + 𝑏2 |𝜙2 > +𝑏3 | 𝜙3 > + ⋯ + 𝑏𝑛 | 𝜙𝑛 >

𝑏1
𝑏2
= 𝑏.3
.
[𝑏𝑛 ]

𝑏1
𝑏2
< 𝜓|𝜙 > = [ 𝑎1∗ 𝑎2∗ 𝑎3∗ … … . . 𝑎𝑛∗ ] 𝑏.3
.
[𝑏𝑛 ]

= 𝑎1∗ 𝑏1 + 𝑎2∗ 𝑏2 + 𝑎3∗ 𝑏3 + ………+ 𝑎𝑛∗ 𝑏𝑛

Condition for orthogonality:

Two ket vectors | 𝜓 > and | 𝜙 >, are said to be orthogonal if their inner product is zero.

i.e. < 𝜓|𝜙 > = 0

Condition for orthnormality:

Two ket vectors | 𝜓 > and | 𝜙 >, are said to be orthonormal if they are orthogonal and if each of them
is normalized.

i.e. < 𝜓|𝜙 > = 0, < 𝜓|𝜓 > = 1, < 𝜙|𝜙 > = 1

Operator:

An operator 𝐴̂ is a mathematical rule that when applied to a ket vector | 𝜓 > transforms it to another
ket vector |𝜙 > of the same space and when it acts on a bra vector < 𝜓| transforms it to another bra
vector < 𝜙|.

i.e. 𝐴̂ | 𝜓 > = |𝜙 > , 𝐴̂ < 𝜓| = < 𝜙|

Linear operators can be represented as square matrices in quantum mechanics.

Unity operator ( 𝑰̂ ) : It leaves any ket vector unchanged.

i.e. 𝐼̂ | 𝜓 > = |𝜓 >

Identity matrix(I):

An identity matrix is a square matrix in which all the elements of principal diagonals are one, and all
other elements are zeros. If any matrix is multiplied by the identity matrix, the result will be given a
matrix.

1 0
𝐼≡ [ ]
0 1

Example: The matrix form of ket vectors |0> and |1> can be written as:

1 0
|0> = [ ] and |1> = [ ] , then
0 1

1 0 1 1𝑥1 + 0𝑥0 1
I |0> = [ ][ ]=[ ]= [ ]
0 1 0 0𝑥1 + 1𝑥0 0

1 0 0 1𝑥0 + 0𝑥1 0
I |1> = [ ] [ ]= [ ]= [ ]
0 1 1 0𝑥0 + 1𝑥1 1
Hermitian matrix:

A Hermitian matrix is a square matrix composed of complex numbers, and it is equal to its conjugate
transpose.

1 𝑖
Example: M = [ ]
−𝑖 1

1 𝑖
The conjugate transpose of the matrix is 𝑀𝐻 𝑜𝑟 𝑀† = [ ]
−𝑖 1

Here, M = 𝑀†

Hence M is a Hermitian matrix.

Unitary matrix:

Unitary Matrix is a square matrix of complex numbers. The product of the conjugate transpose of a
unitary matrix, with the unitary matrix itself, gives an identity matrix.

Example:

1 2 −2 + 𝑖
U=3 [ ]
2+𝑖 2

̅=1 [ 2
Taking conjugate 𝑈
−2 − 𝑖
]
3 2−𝑖 2

If we take the transpose of the above matrix, it is called a Hermitian matrix.

1 2 2−𝑖
𝑈† = [ ]
3 −2 − 𝑖 2

1 2 2−𝑖 1 2 −2 + 𝑖 1 0
∴ 𝑈†. 𝑈 = [ ] [ ]=[ ]=I
3 −2 − 𝑖 2 3 2+𝑖 2 0 1

1 0
Similarly, 𝑈. 𝑈 † = [ ]=I
0 1

Hence, U is a unitary matrix.

Pauli matrices:

The Pauli matrices are a set of four 2x2 complex matrices. They are used to represent spin angular
momentum. These matrices are Hermitian and Unitary.
These matrices are very powerful in quantum computing as they can be used to represent quantum
logic gates. They can set the rotational parameters for qubits. These matrices go by a variety of
notations.

0 1 0 −𝑖
𝜎1 ≡ 𝜎𝑥 ≡ 𝑋 ≡ [ ] 𝜎2 ≡ 𝜎𝑦 ≡ 𝑌 ≡ [ ]
1 0 𝑖 0

1 0
𝜎3 ≡ 𝜎𝑧 ≡ 𝑍 ≡ [ ]
0 −1

Quantum Gates:

Classical computer circuits consist of wires and logic gates. The wires carry information around
the circuit, while the logic gates manipulate information, converting it from one to another. Classical
computers manipulate bits using classical logic gates, such as OR, AND, NOT, NAND, etc.

Similarly, quantum computers manipulate qubits using quantum gates which are usually represented
as unitary matrices. A gate which acts on k qubits is represented by a 2k X 2k unitary matrix. The
number of qubits in the input and output of the gate has to be equal. The action of the quantum gate is
found by multiplying the matrix representing the gate with the vector which represents the quantum
state.

Single qubit gates:

Pauli-X gate

In classical computers, the NOT gate takes one input and reverses its value. For example, it changes
the 0 bit to a 1 bit or changes a 1 bit to a 0 bit. It is like a light switch flipping a light from ON to OFF,
or from OFF to ON.
Pauli-X gate is a quantum analogue of the classical NOT gate.
• The application of this gate rotates the qubit by 180o along the x-axis. It transforms |0⟩ to |1⟩
and vice versa.
• The matrix form of X-gate is obtained as follows
X = |0 > < 1| + |1 > < 0|
1
=[ ] [0 1] + [0] [1 0]
0 1
0 1 0 0
=[ ]+ [ ]
0 0 1 0
0 1
= [ ]
1 0
• Circuit representation
 • Dirac notation
X |0⟩ = |1⟩
X |1⟩ = |0⟩
• When the qubit is in a superposition state |𝜓⟩ = 𝛼 |0⟩ + 𝛽 |1⟩ , then
X |𝜓⟩ = 𝛼 |1⟩ + 𝛽 |0⟩
𝛼
• In matrix form |𝜓⟩ = [𝛽 ]

Then the action of X gate is

𝛼 𝛽
X [𝛽]= [ ]
𝛼

Pauli -Y gate

• The application of this gate rotates the qubit by 180o along the y-axis.
• It transforms |0⟩ to 𝑖|1⟩ and |1⟩ to −𝑖 |0⟩.
• Matrix for of Y-gate
0 −𝑖
Y= [ ]
𝑖 0
• Circuit representation

• Dirac notation
Y |0⟩ = 𝑖 |1⟩
Y |1⟩ = −𝑖 |0⟩

Pauli - Z gate

• The application of this gate rotates the qubit by 180o along the z-axis.
• It leaves |0⟩ unchanged and flips the sign of |1⟩ to - |1⟩.
• Matrix for of Z-gate
1 0
Z =[ ]
0 −1
• Circuit representation
 • Dirac notation
Z |0⟩ = |0⟩
Z |1⟩ = −|1⟩

Hadamard gate

• It is one of the most important gates for quantum computing. If the qubit starts in a definite |0⟩
or |1⟩ state, the Hadamard gate puts each into a superposition of |0⟩ and |1⟩ states.
1 1 1
• Matrix representation H ≡ [ ]
√2 1 −1
• Circuit representation

• Dirac notation

1
H |0⟩ = ( |0⟩ + |1⟩ )
√2

1
H |1⟩ = ( |0⟩ − |1⟩ )
√2

Multiple qubits gates:

CNOT gate:

• It is a Controlled NOT (CNOT) gate.

• It acts on two qubits.
• It performs the NOT operation on the second qubit only when the first qubit is |1> otherwise
leaves it unchanged.
• Matrix representation

• Dirac notation
CNOT |00⟩ = |00⟩ ; CNOT |01⟩ = |01⟩
 CNOT |10⟩ = |11⟩ ; CNOT |11⟩ = |10⟩

Quantum circuits:

A quantum circuit is required to carry out computations on a quantum computer. It consists of a series
of operations referred to as quantum gates. These quantum gates, which are assigned to certain qubits,
change the quantum states of some of the qubits, causing those qubits to perform the calculations
required to solve a problem.

(1)
|0>

|0> X

CNOT gate leaves it unchanged as the first qubit is |0>.

In the second step, X-gate flips the second qubit to |1>.

Hence the output is |01>.

(2)

The states change from the start to the end after every gate:

1
The Hadamard gate changes |0> to ( |0⟩ + |1⟩ ).
√2

Therefore,

1 1
|00>→ (|0>+|1>) |0> → (|00>+|10>).
√2 √2

In the last step, the first qubit is the control qubit.

Case 1: |00>
The control qubit is |0>, so the target qubit remains unchanged.

Case 2: |10>

The control qubit is |1>, so the target qubit flips from |0> to |1>.

1
Hence the output is (|00>+|11>).
√2

Accounting for the extra-ordinary capability of quantum computing:

The main advantage that a quantum computer has over a classical computer is parallelism. A
quantum computer can perform operations on all of the states simultaneously because qubits can be in
a superposition of states.

Let us consider two systems.

System 1: With 2 bits

This can represent 4 different values.
Possible states are [00, 01, 10, 11]
Particular state-value ∈ {0,1,2,3}, one of the 4 possible values

System 2: With 2 qubits

This can represent infinite different values (vector space) formed from 4 different basis state,
00 ≡|00⟩, 01 ≡ |01⟩, 10 ≡ |10⟩, 11 ≡|11⟩
Possible states: Infinite
particular state-value: α |00⟩ + β |01⟩ + γ |10⟩ + δ |11⟩
such that |α|²+|β|² +|γ|² +|δ|² = 1

Hence, an n-bit classical system can be in one of the 2n possible states at a time, and all it needs is the
value of these n-bits to be fully recognized.
An n-qubit system can be in a superposition of all of those states 2n states at a given time and needs the
value of coefficients of all of the 2n-1 states (considering that summation is 1) basis to be fully
recognized.
Consequently, a quantum computer can compute with 2n values in a single step. This enormous
parallelism is one reason why quantum computers are so powerful