Quantum Optics

Handwritten Lecture Notes

C. Bulutay

26 May 2022

Credits: The only originality of these lecture notes is my handwriting! Especially, the first
one-third closely follows Gerry & Knight IQO.
Here is a semi-complete list of references (in order of contribution):

• C. G. Gerry and P. L. Knight, Introductory Quantum Optics (Cambridge, 2005)

• M. Dignam, Lectures Notes (based on Gerry & Knight)
• Wikipedia, especially for the illustrations, as noted in the script.
• M. O. Scully and M. S. Zubairy, Quantum Optics (Cambridge, 1997)
• I. Deutsch, Lectures Notes and Lecture Videos.
• M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information (Cam-
bridge, 2000)
• J. Preskill, Lecture Notes on Quantum Computation
• B. Schumacher and M. Westmoreland, Quantum Processes Systems, and Information
(Cambridge, 2010)
• Individual papers as mentioned within the script
Computational Project Grader: Ekrem Taha Güldeste

Congratulations and many thanks to Class of Spring 2022, Bilkent University

Contents
Part-I: Quantized Electromagnetic Field 4
Formal Procedure for Quantizing a Classical Field 5
Quantization of a Single Mode EM Field 7
Quantization of EM Field: Generalization to Multimode Fields 19
The Quantum Phase 22
Coherent States 28
Pure versus Mixed States and Density Operator 38
P-Representation 46
Wigner Function 51
Q Representation 53
Squeezed States 56
Nonclassical States of Light 66
Optical Coherence Measures 69
Beam Splitters and Hong-Ou-Mandel Effect 84
Part-II: With Matter 95
Atom-Field Interactions 96
The Rabi Model 102
Ensemble of Two-Level Atoms: Density Matrix Formalism 107
Photon Echo 114
Interaction of an Atom with a Quantized EM Field 116
Jaynes-Cummings Model 124
Dressed States 133
Superradiance and Dicke Model 138
Three-Level Quantum Optics 145
Quantum Beats 146
Coherent Population Trapping 147
Electromagnetically Induced Transparency 149

2
Adiabatic Elimination in a Lambda System 153
Part-III: With Environment 156
Heisenberg-Langevin Equations 157
Quantum Cavity Optomechanics 167
Composite Systems, Direct Products and Sums 180
Reduced Probability Densities and Schmidt Decomposition 192
Kraus Representation 196
Lindblad Master Equation 200
Part-IV: Course Assessments 212
Eight Computational Projects 213
Midterm Exam with Solutions 222
Final Exam with Solutions 229

3
Part-I: Quantized Electromagnetic Field

4
 Quantum Optics - Lecture Notes

Formal Procedure for Quantizing a Classical Field

5
6
Quantization of a Single-Mode EM Field

7
8
9
10
11
12
13
14
Quantum Fluctuations of a Single-mode Field

15
16
Quadrature Operators (Single-Mode Field)

17
18
Generalization to Multimode Fields

19
20
21
Quantum Phase Operator

22
23
24
25
26
27
Coherent States

28
29
30
31
32
33
34
Phase space vs Complex Plane

More on Coherent States

35
36
37
Pure/Mixed States, Density Operators

38
39
40
Mixed States

41
Properties of Density Operators (for closed/isolated systems)

42
43
44
How to prepare a mixed state?

45
Representations of Quantum States

P-representation

46
47
48
Normal Ordering Operator

Optical Equivalence Theorem

49
50
Wigner Function

51
Properties of the Wigner Fn.

52
Q-representation

53
54
55
Squeezed States

56
57
58
59
60
61
62
63
64
65
Nonclassical States of Light

66
67
68
Optical Coherence Measures

69
70
71
72
73
74
75
76
Hanbury-Brown Twiss Experiment: Intensity Correlations

77
78
79
80
81
82
83
84
85
86
87
88
89
Hong-Ou-Mandel (HOM) Effect

90
91
92
93
94
Part-II: With Matter

95
Interaction of EM Field with Matter

96
97
98
99
Rotating Wave Approximation (RWA)

100
101
The Rabi Model

102
103
104
105
106
107
108
109
110
111
112
113
114
115
Interaction of an "Atom" with a Quantized EM Field

116
117
118
119
120
121
122
123
Jaynes-Cummings Model

124
125
126
127
128
129
Quantum Rabi Model

130
Cummings Collapse

131
132
Dressed States

133
134
135
136
137
Superradiance [Main Ref: Gross & Haroche Phys. Rep. '82]

Dicke Model

138
139
140
141
142
143
144
Three-Level Quantum Optics

145
146
147
148
149
150
151
152
Adiabatic Elimination in a Lambda System

153
154
155
Part-III: With Environment

156
Heisenberg-Langevin Equations

157
158
159
160
161
162
163
164
165
166
Quantum Cavity Optomechanics

167
168
169
Rotating Frame Transformation

170
171
172
173
174
175
 Mechanical Resonators

176
 Cooperativity in Optomechanics

Some Order-of-Magnitude on OM & LIGO

Cavity Optomechanics

177
LIGO

178
Order of Magnitude Parameters & Ratios

179
Mathematical Framework for Open Systems
Composite Systems:

Tensor product of Hilbert Spaces

180
181
Higher tensor product spaces:

Dual Space

Inner Product

182
Product Operators

183
184
185
Direct Sum

186
Matrix Rep. of Direct Sum & Direct Product

187
188
189
190
191
Marginal (Reduced) Probability Densities

192
Schmidt Decomposition

193
194
195
Kraus Representation

Open System

196
197
198
199
Lindblad Master Equation

200
Lindbladian

201
202
203
204
205
206
207
208
209
210
211
Part-IV: Course Assessments

212
 Project #0: Coupled ODE Solver in Python
PHYS 442/612 Quantum Optics

Due: Tuesday, 8 February 2022 [Till 23:59] version: 1 February 2022

Motivation: The aim of this first project is to warm you up to Python based on a scientific
computation relevant to quantum optics, so-called optical Bloch equations. This will also form
the core of some follow-up projects later on.
Mission:

• Make use of the very important packages: SciPy, NumPy, Matplotlib.

• Write your own Python code to solve the following set of coupled ODEs

d R1 (t)
R1 (t) = − + ∆ cos(Ωt)R2 (t),
dt T2
d R2 (t)
R2 (t) = − − ∆ cos(Ωt)R1 (t) − R3 (t),
dt T2
[C. Bulutay, Bilkent University]

d R3 (t) + 1
R3 (t) = − + R2 (t),
dt T1

• You can make use of built-in ODE/ODEINT python libraries.

• Your code should generate 3 different figures according to the 3 different parameter sets below.
Each figure should contain R1 (t), R2 (t), R3 (t) discriminated by different colors/linestyles.

• For all parameter sets use t ∈ [0, 30] and the initial values, R1 (0) = 0, R2 (0) = 0, R3 (0) = −1.

• Parameter Sets: [T1 , T2 , ∆, Ω] (All inputs should be hard-coded inside)

i) [50, 5, 0, 0], ii) [50, 5, 0.25, 0], iii) [50, 5, 0.25, 2].

• Grading:

– If the Python code does not work, you get 20% at most.

– If the code generates all the requested outputs, you get 75%. The remaining 25% will
be granted to the quality of the plots and coding style (inc. comments within the code).

Rules of Conduct:

• Use the following header in your code to be submitted.

• Name the file as yourname-surname-project-x.py [e.g., ceyhun-bulutay-project-0.py]

• Upload your working code (single file) to the corresponding Moodle assignment (Email
submissions not considered).

• When executed, the code should perform all the tasks, and generate the requested plots
successively without any external input.

• You can use anything already available in the Internet. Beyond that, this is a solitary exercise,
no code sharing or help from anybody.

213
 Project #1: Photonic Distributions Using QuTiP
PHYS 442 Quantum Optics

Due: Friday, 25 February 2022 [Till 23:59] version: 17 February 2022

Motivation: The aim of this project is to familiarize you with the QuTiP based on various
quantum distributions. More projects to follow will be using it.
Mission:
• Install and learn how to use the QuTiP Library from its User’s Guide and Tutorials.

• Write your own Python code to compute the following plots using the QuTiP built-in tools
where available:

1. For the thermal photon ensemble generate the photon Fock distribution 1×2 subplot
for n̄ = 0.1, 2 (as in Fig. 2.3 in Gerry & Knight, Introductory Quantum Optics pg. 28)
[C. Bulutay, Bilkent University]

2. For the coherent state generate the photon Fock distribution 1×2 subplot for n̄ = 2, 10
(as in Fig. 3.1 in Gerry-Knight pg. 47)

3. For the coherent state numerically generate the phase distribution 1×2 subplot for
n̄ = 2, 10 (as in Fig. 3.2 in Gerry-Knight pg. 48)

4. For the coherent state n̄ = 10, and the number state n = 3 generate the Q-distribution
1×2 subplot (as in Fig. 3.9 in Gerry-Knight pg. 68)

5. For the coherent state n̄ = 10, and the number state n = 3 generate the Wigner-
distribution 1×2 subplot (as in Fig. 3.10 in Gerry-Knight pg. 69)

• A single code should generate these five plots successively with appropriate labelings.

• There can be unnoticed errata in the plots of Gerry & Knight.

• The Grading and Rules of Conduct are the same as Project-0.

214
 Project #3: Nonclassical Photonic States
PHYS 442 Quantum Optics

Due: Tuesday, 8 March 2022 [Till 23:59] version: 28 February 2022

Motivation: The aim of this project is to explore nonclassical photonic states again with the help
of QuTiP library.
Mission:
• Write your own Python code to compute the following plots using the QuTiP built-in tools
where available:

1. For the squeezed vacuum state (r = 0.5) generate the photon Fock distribution (similar
to Fig. 7.8 in Gerry & Knight, Introductory Quantum Optics pg. 162)

2. Generate the photon distributions for the coherent state (|α|2 = 50), and squeezed state
[C. Bulutay, Bilkent University]

((|α|2 = 50, r = 0.5) for two different θ values (as in Fig. 7.9 in Gerry-Knight pg. 163);
use distinct linestyles/colors

3. Generate the Q-distribution 1×2 subplot for squeezed vacuum state (r = 0.5) and its
displaced form (you decide on the parameter) (as in Fig. 7.11 in Gerry-Knight pg. 166)

4. Generate the Wigner distribution 1×2 subplot for (a) equal statistical mixture of two
coherent states, (b) and even Cat state (i.e., even superposition of two coherent states)
√
both for (α = 5) (as in Fig. 7.15 in Gerry-Knight pg. 178)

5. Generate the density matrix histograms (using QuTiP Hinton) for thermal distribution
n̄ = 2, coherent state n̄ = 2, and Fock (number) states n = 2 as a 1×3 subplot. Truncate
the Fock space to a size of 10.

6. For the Fock (number) states n, generate the second-order photon coherence g (2) (τ = 0)
as a function of n, where n ∈ [1, 24]. On this plot show for any coherent state its g (2) (0)
value as a (horizontal) asymptote. Set the Fock space size to 25.

• A single code should generate these six plots successively with appropriate labelings.

• There can be unnoticed errata in the plots of Gerry & Knight.

• The Grading and Rules of Conduct are the same as Project-0.

215
 Project #3: Semiclassical Hahn Echo
PHYS 442 Quantum Optics

Due: Sunday, 3 April 2022 [Till 23:59] version: 22 March 2022

Motivation: You will study the Hahn echo for an ensemble of dephasing spins. Basic aim is to
refocus the inhomogeneously precessing (detuned) spins using a simple Hahn echo sheme. Essen-
tially it will be based on your Project-1 code (QuTiP library is not particularly necessary). This
was the warm-up of Mustafa Kahraman’s Senior Project on Dynamical Decoupling.
Mission:
• Develop a Python code on semiclassical optical Bloch equations to combat inhomogeneous
broadening due to environmental noise.

• The general setting: Start N = 500 spin vectors at t = 0 all pointing along the -y direction
[C. Bulutay, Bilkent University]

⃗ i (t) = −ŷ, i ∈ [1, N ]), and evolve each one based on its detuning ∆i till t = tR . At this
(R
instant rotate them around x-axis by π radians. Then, further evolve the spins from t = tR
to 3 T2 again under its detuning precession. For each item below, on 2 × 1 subplots show the
time evolution of the mean spin’s x and y components, i.e., R̄x (t), R̄y (t) for t ∈ [0, 3T2 ]; both
on the same plot distinguished by line style/colors. For all parts use (in some normalized
units): T2 = 60 (more precisely, T2 → T2′ ), T1 = 2 T2 , ν =0.
1. (No-rotation) Top subplot, use detuning ∆i = 1, (free-induction decay); Bottom subplot
use {∆i } as uniformly distributed between [1, 1.5] (inhomogeneously broadened FID).

2. Now, include a rotation pulse at tR = 30, by instantly rotating spins with π radians
around the x-axis. Top subplot: use the detuning {∆i } ∈ [1, 1.5]. Bottom subplot: keep
the bottom subplot of part-1 (to compare the effect of Hahn echo).

3. Replace the bottom (no-rotation-pulse) subplot of part-2 with {∆i } ∈ [1, 5] and a rota-
tion pulse at tR = 30, to compare the amplitudes of the spin echo, in each case.

4. Rotation error tolerance: replace the bottom subplot of part-2 where at tR = 30 the
rotation angle around x-axis is uniformly distributed for the spins over [0.9π, 1.1π]; use
{∆i } ∈ [1, 1.5] for this part (and the next part).

5. Replace the bottom subplot of part-2, this time with two π/2 rotations at the instants
tR1 = 30, tR2 = 50, and compare the amplitudes and the instants of the spin echo.

• Your code should generate successively five 2 × 1 subplots according to the tasks given above.
Indicate on each plot the instant of rotation pulse(s) with a legible downward red arrow(s).

• The Grading and Rules of Conduct are the same as Project-0.

216
 Project #4: Coherent Population Trapping
PHYS 442 Quantum Optics

Due: Tuesday, 12 April 2022 [Till 23:59] version: 3 April 2022

Formulation: By modifying your coupled differential equation solver in Project #0, develop a
Python code to demonstrate the coherent population trapping for a three-level Λ atomic system;
where the states (a, b, c) are ordered as: Eb < Ec < Ea . The atomic wave function can be written
in the interaction picture as

|ψ(t)⟩ = Ca (t)|a⟩ + Cb (t)|b⟩ + Cc (t)|c⟩,

where, the coefficients Ca (t), Cb (t), Cc (t) satisfy (dot denotes a time derivative),

i
Ċa = (ΩR1 Cb + ΩR2 Cc ) ,
[C. Bulutay, Bilkent University]

2
i
Ċb = ΩR1 Ca ,
2
i
Ċc = ΩR2 Ca ,
2

here ΩR1 is the Rabi frequency for the laser coupling the states |a⟩ ↔ |b⟩, and ΩR2 is the Rabi
frequency for the laser coupling the states |a⟩ ↔ |c⟩. We further assumed that both lasers are
resonant, i.e., ω1 = ωa − ωb and ω2 = ωa − ωc .
Let the initial atomic state be a superposition of the two lower levels |b⟩ and |c⟩ as:
   
θ θ
|ψ(t = 0)⟩ = cos |b⟩ + sin e−iβ |c⟩
2 2

Parameters and Tasks:

a) ΩR1 = ΩR2 ≡ 1, θ = π/2, β = π. Compute and plot the populations |Ca (t)|2 , |Cb (t)|2 ,
|Cc (t)|2 . (Off the record, check your numerical results with the analytical ones) On the same
plot show |Ca (t)|2 + |Cb (t)|2 + |Cc (t)|2 , to verify your numerics whether it comes out to be 1.

b) Redo part (a) with the initial conditions: Ca (0) = 1, Cb (0) = 0, Cc (0) = 0.

c) Redo part (a) with β = π/2.

• Your code should generate successively three different figures according to the three different
parameter sets given above. Each curve should be discriminated by different color/linestyle
described by a legend.

• The Grading and Rules of Conduct are the same as Project-0 (additionally, see the last item).

• There is an extra up to 3-day post deadline submission option, but a with 20% deduction!

217
 Project #5: Stimulated Raman Adiabatic Passage
PHYS 442 Quantum Optics

Due: Friday, 22 April 2022 [Till 23:59] version: 12 April 2022

Mission: Study by yourself from the posted lecture notes the topic of STIRAP, then develop a
Python code that solves the relevant set of coupled differential equations which demonstrates how
and when STIRAP works as intended. Consider a three-level atomic system, labeling the states
as i = 1, 2, 3, where intially only i = 1 state is populated. The aim of STIRAP is to transfer the
initial population with aid of Stokes (S) and Pump pulses (P ), to the i = 3, (ideally) without
populating i = 2 at any time.
Parameters:

• Assume a Λ level scheme, taking (in some normalized units) ω21 = 2.0, and ω23 = 1.5, where
[C. Bulutay, Bilkent University]

1 ↔ 3 is dipole forbidden while other transitions being allowed.

• Use the same Gaussian profiles for S and P (other than a time delay, Td ) as

 
ΩS (t) = Ωmax exp −t2 /Tw2 , ΩP (t) = Ωmax exp −(t − Td )2 /Tw2 .

• Initially set pulse widths the same as the delay, i.e., Tw = Td .

• Start the simulation from sufficiently early times so that the tails of the Gaussians are not
appreaciable.

Tasks:

1. Choose suitable Ωmax , and T = Tw = Td , while preserving single-photon resonance, i.e.,

ωS = ω23 , and ωP = ω21 to yield a successful STIRAP. Ωmax ≪ ω23 , ω21 is necessary!

2. Redo the first part under single-photon detuning while still assuring two-photon resonance,
i.e., ω23 − ωS = ω21 − ωP ̸= 0.

3. Demonstrate independently (for each item a separate plot) what happens for:

• intuitive pulse ordering, i.e., Td < 0

• insufficient pulse overlap between S and P

• no two-photon resonance, i.e., ω23 − ωS ̸= ω21 − ωP

4. Redo (especially item 1) without rotating wave approximation to show that the results are
not significantly affected. For all tasks, show that total probability is conserved at all times:
|C1 (t)|2 + |C2 (t)|2 + |C3 (t)|2 = 1.

• Select suitable time axis ranges to clearly show population transfer regions.

• The Grading and Rules of Conduct are the same as Project-0 (with 20% deduction there is
an extra 3-day post deadline submission option)
218
 Project #6: Open Four-Level System
PHYS 442 Quantum Optics

Due: Friday, 13 May 2022 [Till 23:59] version: 5 May 2022

Mission: You will employ the QuTiP Library to solve an open four-level ladder system.
Details:
• Model the non-interacting system Hamiltonian as four levels equally-spaced in energy, by ω0
(Let h̄ = 1). Set the initial state of the system to its ground state.

• In the first part of the project you will pump this system (characterized with a rate, ΓP ) by
an external monochromatic radiation that is resonant with the energy spacing, ω0 . It will
drive stimulated emission and absorption among states.

• Being an open system, you need to include decoherence processes that affect all states; for
[C. Bulutay, Bilkent University]

simplicity assume the same rate ΓD .

• Assume that radiation and decoherence processes only couple the neighboring states, i ↔ i±1

• Model either of these processes with Lindblad (collapse) operators, and use the Lindblad
Master Equation solver in QuTiP.

• In your code make use of the following matrices which are the generators of SU(2) group, a
fancy name, but they are just the extension of Pauli matrices to a 4-level system:
 √     
0 3 0 0 √0 0 0 0 3/2 0 0 0
 0 0 2 √0     0 
J+ =   , J− =  3 0 0 0  , J3 =  0 1/2 0 .
0 0 0 3   0 2 √0 0   0 0 −1/2 0 
0 0 0 0 0 0 3 0 0 0 0 −3/2

• Note that, we are not interested in keeping track of what happens to the radiation that forms
the reservoir.

Tasks:

1. Set the pumping rate, ΓP sufficiently high, with a total simulation time such that it displays
the saturation of the system, i.e., all four states reaching the 1/4 population.

2. Carrying over the simulation from the saturated state, turn off the pumping and let the
system fully relax.

3. For both parts, plot the time evolution of the populations of each state. For the time axis a
logarithmic scale could be necessary. Separately, plot the real and imaginary parts of the 4×4
density matrix of the system at several representative time instants, including saturation and
full relaxation. You may or may not use a built-in density matrix plotting tool from QuTiP.

The Grading and Rules of Conduct are the same as Project-0 (with 20% deduction, there is an
extra 3-day post deadline submission option).

219
 Project #7: Mechanical squeezing in an
optomechanical system by pump modulation –
Code
PHYS 442 Quantum Optics

Due: Tuesday, 24 May 2022 [Till 23:59] version: 25 April 2022

Motivation: The aim of this project is to familiarize you with optomechanics through reproduc-
ing a recent research paper from literature. This will also enable to apply your prior knowledge on
squeezing.

Mission:
• Study the paper by C-H. Bai, et al. Phys. Rev. A 101, 053836 (2020). Mind you, as always
[C. Bulutay, Bilkent University]

there can be unnoticed errata in the text/plots of the paper.

• You need to go over the formulation of the paper and fill in the gaps (to be documented in
the Report part of your assignment) before you start the next item.

• Develop your own Python code to compute the following plots:

1. Reproduce Figs. 2, 3, and 10 of this paper.

2. As a bonus, you may try other variations in the Hamiltonian as long as you discuss why
they may occur in your Report.

• A single code should generate these plots successively with appropriate labelings.

• For this project, very long run times will be penalized. Optimize the numerics of your code
accordingly.

• Only upload your code file to the correct assignment in Moodle.

• The Code part of this final project will contribute 10% to your overall course grade.

• The Grading and Rules of Conduct are the same as Project-0 (with 20% deduction, there is
an extra 3-day post deadline submission option).

220
 Project #7: Mechanical squeezing in an
optomechanical system by pump modulation –
Report
PHYS 442 Quantum Optics

Due: Tuesday, 24 May 2022 [Till 23:59] version: 25 April 2022

This is the report part that accompanies the mechanical squeezing in an optomechanical system
project.

Report Guidelines:

• It should contain your theoretical formulation, especially to fill in the gaps in the Theory
[C. Bulutay, Bilkent University]

part of the paper.

• Give a physical discussion of the generated Figures in the Code part of the project; include
your relevant plots to support your arguments.

• Put section titles matching with the original paper.

• No need for a separate Abstract/Introduction/Conclusion sections, but cite your References.

No page limit.

• Plots should be captioned that concisely convey the motivation for each one together with
the relevant part of the parameter set.

• It must be a single PDF file, typeset in LATEX(preferred) or any other word processor.

• Only upload your Report file to the correct assignment in Moodle.

• The Report part of this final project will contribute 5% to your overall course grade.

• The Grading and Rules of Conduct are the same as Project-0 (with 20% deduction, there is
an extra 3-day post deadline submission option).

221
 Bilkent University, Department of Physics

PHYS 442: Quantum Optics

Midterm Examination

Duration: 100 minutes Date: 29 March 2022

Give full details of the derivations.

1. (35 points) In the small squeezing limit (|ξ| ≪ 1), the corresponding operator can be
approximated as
1 ∗ 2 †2

Ŝ ≃ 1̂ + ξ â − ξâ .
2
where ξ ∈ C is the squeezing parameter.
(a) Work out ⟨ξ, n|ξ, n⟩ (i.e., the norm) of the squeezed number state |ξ, n⟩ ≡ Ŝ|n⟩,
where |n⟩ is the number state, and show that the results will be valid in this approx-
imation only up to linear order in |ξ|.
For the squeezed number state |ξ, n⟩, up to linear order in |ξ| obtain:
(b) ⟨n̂⟩, ∆n
(c) ⟨Ê(t)⟩, ∆E(t), where Ê(t) is the electric field operator in the Heisenberg picture.

2. (30 points) Recall that the Q-Representation of density operator ρ̂, which is defined
via the coherent state matrix elements as
1
Q (α, α∗ ) = ⟨α |ρ̂| α⟩ .
π
Work out the Q-Representations of
(a) single-mode coherent state, |α0 ⟩,
(b) single-mode number state, |n⟩,
(c) single-mode filtered thermal state at temperature T .

3. (35 points) Consider a 50:50 beam splitter that transforms incoming mode operators
âA , âB to outgoing operators âC , âD as
    
âC 1 −1 1 âA
=√
âD 2 1 1 âB
If two indistinguishable photons are simultaneously incident to the input ports of
this beam splitter as â†A â†B |0, 0⟩ = |1, 1⟩AB , with |0, 0⟩ being the vacuum state,
(a) Obtain the corresponding output of the beam splitter expressed using |nC , nD ⟩CD
number states of the output ports.
(b) Based on this result, suggest a simple experimental set up to explore the indis-
tinguishability of two photons, and explain how it can be used for this purpose.
(c) Redo part (a), now with two photons incident from each input port, |2, 2⟩AB .

222
Quantum Optics Midterm Solutions

223
224
225
226
227
228
 Bilkent University, Department of Physics

PHYS 442: Quantum Optics

Final Examination

Duration: 120 minutes Date: 26 May 2022

Lab visit attendance will contribute by 5% to your Final exam grade. Over 100+10 points.

1. (30 points) Write the Hamiltonians/Equations for the following quantum optics mod-
els adhering to the established notation
(a) Rabi, (b) Quantum Rabi, (c) Jaynes-Cummings, (d) Dicke, (e) An optomechan-
ical Hamiltonian under radiation pressure coupling, (f) Lindblad Master equation.

2. (30 points) For some optomechanical system, the quantum fluctuation parts of the
photonic (δâ), and mechanical/phononic (δ b̂) annihilation mode operators after lin-
earizing the Heisenberg-Langevin equations take the form
  κ √
δ â˙ = −i∆δâ + iG(t) δ b̂ + δ b̂† − δâ + κâin ,
2
˙ γm √
δ b̂ = −iωm δ b̂ + iG∗ (t)δâ + iG(t)δ↠− δ b̂ + γm b̂in ,
2

where ∆, ωm , G(t), κ, γm are known, and âin and b̂in are respectively, the photonic
and phononic input noise operators characterized through the correlators as

⟨âin† (t) âin (t′ )⟩ = na δ(t − t′ ) , ⟨âin (t) âin† (t′ )⟩ = (na + 1) δ(t − t′ ),
⟨b̂in† (t) b̂in (t′ )⟩ = nb δ(t − t′ ) , ⟨b̂in (t) b̂in† (t′ )⟩ = (nb + 1) δ(t − t′ ),

where na and nb are the mean number of photons and phonons, respectively. In terms
of these mode and noise√operators the corresponding√quadrature operators take the
form δ X̂a = (δâ + δ↠)/ 2, and δ Ŷa = (δâ − δ↠)/(i 2), and etc.
(a) Cast these into a matrix (coupled) differential equation in terms of quadrature
operators as
˙
R̂(t) = M(t)R̂(t) + N(t),
where   √ 
δ X̂a κ X̂ain
  √ 
 δ Ŷ   κ Ŷain 
R =  a, N = √ ,
 δ X̂b   γm X̂bin 
√
δ Ŷb γm Ŷbin
and determine the corresponding the entries of the matrix M(t).
(b) Work out the entries of the noise correlation matrix K(t, t′ ) = N(t) NT (t′ ) .

229
 3-level system energy and detunings

3. (35 points) Consider a three-level quantum system where the excited state |e⟩ is
individually dipole-coupled to two lower ground states |g1 ⟩ and |g2 ⟩ as described by
the Hamiltonian matrix
 Ω1 iωL1 t 
0 0 2 e
H = ℏ 0 ω1 − ω2 Ω22 eiωL2 t  ,
Ω∗1 −iωL1 t Ω∗2 −iωL2 t
2 e 2 e ω1

where energy of |g1 ⟩ is set to zero, and Ω1 , Ω2 are the Rabi frequencies of the
two coupling lasers with frequencies ωL1 , ωL2 . See the figure for the detunings
δ1 = ω1 − ωL1 , δ2 = ω2 − ωL2 , δ ≡ δ1 − δ2 , ∆ ≡ (δ1 + δ2 )/2, where all the involved
parameters are known.
(a) Obtain the transformed H matrix in the special rotating frame enabled by appying
the unitary transformation Û (t) = eiĤ0 t/ℏ
δ 
2 0 0
H0 = ℏ  0 ω1 − ω2 − 2δ 0 .
0 0 ω1 − ∆

(b) If we express a general state vector in the rotating frame as

 
C1 (t)
C2 (t) ,
Ce (t)

then by invoking the time-dependent Schrödinger equation, obtain an expression for

Ċe (t),
(c) For |∆| ≫ |δ| one can set Ċe (t) ∼ 0. This way obtain an effective 2×2 Hamiltonian
matrix that governs the dynamics of |g1 ⟩ and |g2 ⟩ in the rotating frame, which is
known as the adiabatic elimination.

4. (10 point - bonus) If you agree, please write this statement and undersign: “I would
like to appear on the Lecture notes Class of 2022 Group Photo.”

230
Final Exam Solutions 26 May 2022

231
232
233
234
235
236
237