Physical Optics

18 March 2021

Dr. Tsung Sheng Kao

Office: TKP building R416B

Department of Photonics & Institute of Electro-Optical Engineering
National Chiao Tung University
 Schedule
Week Date Content
1 02/25 Course Schedule & Introduction of Photonics
Comprehensive quiz; Development of photonics; Introduction of waves
2 03/04 Maxwell’s Equations & Electromagnetic Waves
Maxwell’s equations; Electromagnetic waves; Electromagnetic optics
3 03/11 Electromagnetic Optics
Electromagnetic optics; course review
4 03/18 Geometric Optics
Geometric optics; Matrix optics
5 03/25 1st exam (mainly on Electromagnetic optics)
6 04/01 Gaussian Optics
Fundamentals; ABCD law; Different orders of Gaussian beams
7 04/08 Reflection & Refraction
Thin film optics; Total internal reflection & Brewster angle; Polarization, birefringence and
nonlinear optics
8 04/15 Optical Coherence and Interference
Coherence; Interference
9 04/22 Resonators
Parallel and spherical mirrors; Gaussian waves in resonators, 2D/3D resonators
10 04/29 2nd exam (mainly on Geometric optics, Gaussian optics, Light on interfaces)
 Electromagnetic wave
• A single-frequency electromagnetic wave
exhibits a sinusoidal variation of electric
and magnetic fields in space.
• Electromagnetic waves transport energy 𝜆
through empty space, stored in the
propagating electric and magnetic fields.
𝐄

Magnetic field variation is

perpendicular to electric field. Electric field variation 𝐄

Magnetic field variation 𝐇 𝜕2𝐄

2
𝛁 × 𝛁 × 𝐄 = 𝛁 𝛁 ∙ 𝐄 − 𝛻 𝐄 = −𝜇0 𝜀0 2
𝜕t
1 𝜕2𝑢
Wave equations 2
𝛻 𝑢− 2 2 =0
In free space 𝐇 𝑐 𝜕t
𝑐0
𝑐= 2
1 𝜕2𝐄 2
1 𝜕2𝐇
𝑛 𝛻 𝐄− 2 2 =0 𝛻 𝐇− 2 2 = 0
𝑐 𝜕t 𝑐 𝜕t
Electric field, 𝐄(𝒓, 𝑡) Magnetic field, 𝐇(𝒓, 𝑡)
𝛁∙𝐄=0 Gauss’s Law 𝛁∙𝐇=0 Gauss’s Law for magnetism
𝜕𝐇 𝜕𝐄
𝛁 × 𝐄 = −𝜇0 Faraday’s Law 𝛁 × 𝐇 = 𝜀0 Ampere’s circuital Law (Maxwell)
𝜕t 𝜕t
 EM waves on dielectric media
• A single-frequency electromagnetic wave
exhibits a sinusoidal variation of electric
and magnetic fields in space.
• Electromagnetic waves transport energy 𝜆
through empty space, stored in the
propagating electric and magnetic fields.
𝐄

Magnetic field variation is

perpendicular to electric field. Electric field variation 𝐄

Magnetic field variation 𝐇

The interactions between light (electromagnetic
waves at optical frequencies) and media.
@ 𝑡′ 𝐇
Dielectric
𝐄 Electric field→Displacement of bound charges→Polarization 𝐏 𝒓, 𝑡
media 𝐏
@𝑡 ∞
𝐏 𝒓, 𝑡 = 𝜀0 න 𝝌 𝒓 − 𝒓′ , 𝑡 − 𝑡 ′ 𝐄 𝒓′ , 𝑡 𝑑𝑡 ′ 𝑑3 𝒓′
−∞

Electric susceptibility (impulse response): 𝝌

𝐄 𝒓′ , 𝑡 𝝌 𝐏 𝒓, 𝑡
 Dielectric media

Electric field→Displacement of bound charges→Polarization 𝐏 𝒓, 𝑡

Dielectric ∞
𝐄 𝐏 𝒓, 𝑡 = 𝜀0 න 𝝌 𝒓 − 𝒓′ , 𝑡 − 𝑡 ′ 𝐄 𝒓′ , 𝑡 𝑑𝑡 ′ 𝑑3 𝒓′
media 𝐏 −∞

Electric susceptibility (impulse response): 𝝌

𝐄 𝒓′ , 𝑡 𝝌 𝐏 𝒓, 𝑡

𝐏 𝒓, 𝑡 = 𝜀0 𝜒𝐄 𝒓, 𝑡
𝐃 = 𝜀0 𝐄 + 𝐏 = 𝜀0 1 + 𝜒 𝐄 = 𝜀0 𝜀𝑟 𝐄 = 𝜀𝐄
𝜒: scalar parameter

linear (𝐏 ∝ 𝐄)
non-dispersive (independent with 𝜔)
homogeneous (independent with 𝒓; 𝐏 and 𝐄 are not spatially related)
isotropic (𝐏 ∥ 𝐄, i.e. 𝑃𝑥 ∝ 𝐸𝑥 , 𝑃𝑦 ∝ 𝐸𝑦 , 𝑃𝑧 ∝ 𝐸𝑧 )
 Dielectric media

Electric field→Displacement of bound charges→Polarization 𝐏 𝒓, 𝑡

Dielectric ∞
𝐄 𝐏 𝒓, 𝑡 = 𝜀0 න 𝝌 𝒓 − 𝒓′ , 𝑡 − 𝑡 ′ 𝐄 𝒓′ , 𝑡 𝑑𝑡 ′ 𝑑3 𝒓′
media 𝐏 −∞

Electric susceptibility (impulse response): 𝝌

𝐄 𝒓′ , 𝑡 𝝌 𝐏 𝒓, 𝑡

𝒑 = 𝛼𝐄𝑙𝑜𝑐 𝛼: atomic polarizability

𝐏 = 𝑁 𝒑 = 𝑁𝛼𝐄𝑙𝑜𝑐 = 𝜀0 𝜒𝐄 𝐄: applied electric field

𝑁𝛼 𝑁𝛼
𝜀𝑟 = 1 + 𝜒 = 1 + → 𝑛 ≅ 𝜀𝑟 = 1+𝜒= 1+ 𝑛: refractive index
𝜀0 𝜀0

If the gas or low-atomic-density materials (𝑁𝛼 Τ𝜀0 ≪ 1) are considered (low interactions).
 Clausius-Mossotti relation
𝐄
𝐏
𝐄𝑙𝑜𝑐 = 𝐄 +
3𝜀0

𝐏 𝐏 𝜀0 𝜒𝐄
𝐏 = 𝑁𝛼𝐄𝑙𝑜𝑐 = 𝑁𝛼 𝐄 + = 𝜀0 𝜒𝐄 → 𝑁𝛼 𝐄 + = 𝜀0 𝜒𝐄
3𝜀0 3𝜀0
Surface charges
𝑁𝛼 𝑁𝛼
1 + 2( ) 𝜀𝑟 − 1 𝑁𝛼
𝜀0 3𝜀0
∴𝜒= → 𝜀𝑟 = 1 + 𝜒 = → =( )
https://en.wikipedia.org/wiki/Dipole 𝑁𝛼 𝑁𝛼 𝜀𝑟 + 2 3𝜀0
1−( ) 1−( )
3𝜀0 3𝜀0

Clausius-Mossotti relation

𝑛2 − 1 𝑁𝛼
= ( ) Lorentz–Lorenz equation
𝑛2 + 2 3𝜀0
 Maxwell’s equations in dielectric media
In dielectric materials (no free charges)
Electric field 𝐄 𝒓, 𝑡 → 𝐃(𝒓, 𝑡): electric displacement field (electric flux density) 𝐃 = 𝜀0 𝐄 + 𝐏
𝛁∙𝐃=0 Gauss’s Law 𝐃 = 𝜀0 𝐄 + 𝐏 = 𝜀0 𝐄 + 𝜀0 𝜒𝐄 = 𝜀0 1 + 𝜒 𝐄 = 𝜀0 𝜀𝑟 𝐄 → 𝜀𝑟 = 1 + 𝜒

𝜕𝐁
𝛁×𝐄=− Faraday’s Law
𝜕t
Magnetic field 𝐇 𝒓, 𝑡 → 𝐁(𝒓, 𝑡): magnetic induction field (magnetic flux density) 𝐁 = 𝜇0 𝐇 + 𝐌
𝛁∙𝐁=0 Gauss’s Law for magnetism 𝐌 : magnetization density
𝜕𝐃
𝛁×𝐇= +𝐉 Ampere’s circuital Law (Maxwell)
𝜕t
In homogeneous materials (assume that 𝐌 = 0)

𝜕𝐇 𝜕𝐄
𝛁 × 𝐄 = −𝜇0 𝛁 × 𝐇 = 𝜀0 𝜀𝑟
𝜕t 𝜕t
𝛁∙𝐃=0 𝛁∙𝐇=0
2 2 2
1 𝜕 𝐄 1 𝜕 𝐏 𝜀 𝜕 𝐄
wave equation: 𝛻2𝐄 − 2 2 = 2 2 → 𝛻2𝐄 − 2 2 = 0
𝑐 𝜕t 𝑐 𝜕t 𝑐 𝜕t
 Maxwell’s equations in dielectric media
In dielectric materials (no free charges)
Electric field 𝐄 𝒓, 𝑡 → 𝐃(𝒓, 𝑡): electric displacement field (electric flux density) 𝐃 = 𝜀0 𝐄 + 𝐏
𝛁∙𝐃=0 Gauss’s Law 𝐃 = 𝜀0 𝐄 + 𝐏 = 𝜀0 𝐄 + 𝜀0 𝜒𝐄 = 𝜀0 1 + 𝜒 𝐄 = 𝜀0 𝜀𝑟 𝐄 → 𝜀𝑟 = 1 + 𝜒

𝜕𝐁
𝛁×𝐄=− Faraday’s Law
𝜕t
Magnetic field 𝐇 𝒓, 𝑡 → 𝐁(𝒓, 𝑡): magnetic induction field (magnetic flux density) 𝐁 = 𝜇0 𝐇 + 𝐌
𝛁∙𝐁=0 Gauss’s Law for magnetism 𝐌 : magnetization density
𝜕𝐃
𝛁×𝐇= Ampere’s circuital Law (Maxwell)
𝜕t
In inhomogeneous materials (𝜒 = 𝜒(𝒓), 𝜀 = 𝜀(𝒓), 𝑛 = 𝑛(𝒓))

𝜕𝐇 𝜕𝐄 2
𝜕 𝐃
𝛁 × 𝐄 = −𝜇0 𝛁 × 𝐇 = 𝜀0 𝜀𝑟 → 𝛁 × 𝛁 × 𝐄 = 𝛁 𝛁 ∙ 𝐄 − 𝛻 2 𝐄 = −𝜇0 2
𝜕t 𝜕t 𝜕t
𝛁∙𝐃=0 𝛁∙𝐇=0 𝛁 ∙ 𝐃 = 𝛁 ∙ 𝜀𝐄 = 𝜀𝛁 ∙ 𝐄 + 𝐄 ∙ 𝛁𝜀

𝑛2 𝒓 𝜕 2 𝐄
2
1
→𝛻 𝐄− 2 + 𝛁 𝛁𝜀 ∙ 𝐄 = 0
𝑛2 𝒓 𝜕 2 𝐄 𝑐 𝜕t 2 𝜀
wave equation: 𝛻 𝐄 −
2
=0
𝑐 2 𝜕t 2 0
if 𝜀 = 𝜀(𝒓) variates at a low rate
 Dielectric media
In homogeneous media
𝛁∙𝐃=0 𝛁∙𝐇=0
2
𝜀 𝜕2𝐄
→𝛻 𝐄− 2 2 =0
𝜕𝐇 𝜕𝐄 𝑐 𝜕t
𝛁 × 𝐄 = −𝜇0 𝛁 × 𝐇 = 𝜀0 𝜀𝑟
𝜕t 𝜕t

In inhomogeneous media

𝜒 = 𝜒(𝒓) 𝜀 = 𝜀(𝒓) 𝑛 = 𝑛(𝒓)

2
𝜕2𝐃
𝛁 × 𝛁 × 𝐄 = 𝛁 𝛁 ∙ 𝐄 − 𝛻 𝐄 = −𝜇0 2
𝜕t

𝛁 ∙ 𝐃 = 𝛁 ∙ 𝜀𝐄 = 𝜀𝛁 ∙ 𝐄 + 𝐄 ∙ 𝛁𝜀

𝑛2 𝒓 𝜕 2 𝐄 1 1
2
𝛻 𝐄− 2 + 𝛁 𝛁𝜀 ∙ 𝐄 = 0 𝜀 = 𝜀(𝒓) variates at a low rate → 𝛁 𝛁𝜀 ∙ 𝐄 = 0
𝑐 𝜕t 2 𝜀 𝜀

2
𝑛2 𝒓 𝜕 2 𝐄
→𝛻 𝐄− 2 =0
𝑐 𝜕t 2
 At different interfaces

Between two lossless linear media

D1t ε1
E1t = E2t → = D1n = D2n → ε1 E1n = ε2 E2n
D2t ε2

B1t μ1
H1t = H2t → = B1n = B2n → μ1 H1n = μ2 H2n
B2t μ2
 Anisotropic media

𝑃𝑖 = 𝜀0 ෍ 𝜒𝑖𝑗 𝐸𝑗 → 𝐷𝑖 = ෍ 𝜀𝑖𝑗 𝐸𝑗
𝑗 𝑗
Birefringence
• Light propagation in an anisotropic medium
• Crystals are generally optically anisotropic, except the cubic crystals
• The dielectric constant 𝜀𝑟 depends on electric polarizations → the displacement of electrons
• The refractive index depends on the propagation direction and the state of polarization
• Any unpolarized light ray entering such a crystal breaks into two different rays with different
polarizations and phase velocities. (calcite crystal)
• Optically anisotropic crystals are called birefringent. → doubly refract an incident light
• A special direction in a birefringent crystal: the optic axis.
• According to the principal refractive indices,

NaCl crystal calcite crystal

 Dispersive media
1
Potential: 𝑈 𝒓 = 𝑈 𝑟0 + 𝛻 2 𝑈(𝑟 − 𝑟0 )2 + ⋯
2

𝐅𝑇 𝒓 = −𝛁𝑈 𝒓 → 𝐅𝑇 𝑟0 = 0
𝜆
𝑘 = 𝛻 2𝑈
restoring force 𝐄
𝐅𝑟 𝒓 = −𝑘𝐗 Dipole oscillations

Magnetic field variation is

perpendicular to electric field. Electric field variation 𝐄

Magnetic field variation 𝐇

The interactions between light (electromagnetic
Dielectric waves at optical frequencies) and media.
media 𝐇

𝑑2 𝐗(t) 𝑑2 𝐗 t 𝑑𝐗 t 2
𝑒𝐄 t 𝑘
𝐅 = 𝑒𝐄 t − 𝑘𝐗 t − 𝛾𝑚𝑣 = 𝑚𝒂 = 𝑚 → + 𝛾 + 𝜔 0 𝐗 t = 𝜔0 =
𝑑t 2 𝑑t 2 𝑑t 𝑚 𝑚
𝑑2 𝐏 𝑑𝐏 𝑁𝑒 2 𝐄 t
𝐏 = 𝑄𝐗 t = −𝑁𝑒𝐗 t → +𝛾 2
+ 𝜔0 𝐏 = Lorentz model
𝑑t 2 𝑑t 𝑚
∞ F.T.
′ ′ ′ ′ 3 ′
𝐏 𝒓, 𝑡 = 𝜀0 න 𝝌 𝒓 − 𝒓 , 𝑡 − 𝑡 𝐄 𝒓 , 𝑡 𝑑𝑡 𝑑 𝒓 → 𝐏(𝜔) = 𝜒 𝜔 𝐄(𝜔)
−∞
 Nonlinear optical media

restoring force 𝐅𝑟 = −𝑘𝐗 + 𝛼𝐗 2 + 𝛽𝐗 3 + ⋯ = −𝑘(𝐗 + 𝜀2 𝐗 2 + 𝜀3 𝐗 3 + ⋯ )

∴ 𝐏 = 𝜀0 [𝜒 1
𝐄+𝜒 2
𝐄2 + 𝜒 3
𝐄 3 + ⋯ ] nonlinear, homogeneous, isotropic, nondispersive materials

1 𝜕 2 𝐄 𝜕 2 𝐏 𝑁𝐿
wave equation: 𝛻 𝐄 − 2
= ∴ 𝐏 𝑁𝐿 = 𝜀0 [𝜒 1
𝐄+𝜒 2
𝐄2 + 𝜒 3
𝐄3 + ⋯ ]
𝑐 2 𝜕t 2 𝜕t 2
 Monochromatic waves
Both in linear and nonlinear materials, the functions of position and time are not coupled.

2
𝜀 𝜕2𝐄 2
1 𝜕 2 𝐄 𝜕 2 𝐏 𝑁𝐿
𝛻 𝐄− 2 2 =0 𝛻 𝐄− 2 2 =
𝑐 𝜕𝑡 𝑐 𝜕t 𝜕t 2

1 2 𝜀 𝜕2𝑇
Let the 𝐄 = 𝑅 𝑟 𝑇(𝑡) → 𝛻 𝑅− 2 =0
𝑅 𝑐 𝑇 𝜕𝑡 2

𝜀 𝜕2𝑇 𝜕2𝑇 𝑘2𝑐2 𝑑2 𝑦

→− 2 2
2
=𝑘 → 2+ 𝑇 = 2 + 𝜔2 𝑇 = 0 𝑇 ∝ 𝑒 ±𝑖𝜔𝑡
𝑐 𝑇 𝜕𝑡 𝜕𝑡 𝜀 𝑑𝑡
General solution can be obtained from the superposition
of functions at different frequencies.
𝐄 𝒓, 𝑡 = 𝑅𝑒[𝐄 𝒓 𝑒𝑥𝑝(𝑖𝜔𝑡)]
for all the physical parameters, 𝑷, 𝑫, 𝑩, …
𝐇 𝒓, 𝑡 = 𝑅𝑒[𝐇 𝒓 𝑒𝑥𝑝(𝑖𝜔𝑡)]
𝜕
Let = 𝑖𝜔 → 𝛁 × 𝐇 = 𝑖𝜔𝐃 𝛁 × 𝐄 = 𝑖𝜔𝐁 𝐃 = 𝜀0 𝐄 + 𝐏
𝜕𝑡

𝛁∙𝐁= 0 𝛁∙𝐃=0 𝐁 = 𝜇0 𝐇
1
𝚽 = 𝑅𝑒 𝐄𝑒 𝑖𝜔𝑡 × 𝑅𝑒 𝐇𝑒 𝑖𝜔𝑡 = [𝐄 × 𝐇 ∗ + 𝐄 ∗ × 𝐇 + 𝐄 × 𝐇𝑒 𝑖2𝜔𝑡 + 𝐄 ∗ × 𝐇 ∗ 𝑒 −𝑖2𝜔𝑡 ]
4
1 1 1
𝐼 = 𝚽 = 𝐄 × 𝐇 ∗ + 𝐄 ∗ × 𝐇 = 𝐒 + 𝐒 ∗ = 𝑅𝑒 𝐒 → 𝑺 = 𝐄 × 𝐇 ∗
4 4 2
 Monochromatic waves
In linear, nondispersive, homogeneous, isotropic media

𝐃 = 𝜀0 𝐄 + 𝐏 = 𝜀0 𝜀𝑟 𝐄 𝐁 = 𝜇0 𝐇

𝐄 𝒓, 𝑡 and 𝐇 𝒓, 𝑡 are fulfilled with Helmholtz eq. 𝛻2 + 𝑘2 𝑈 𝑘 = 𝜔 𝜀0 𝜇0 𝜀 = 𝑛𝑘0

In linear, dispersive, homogeneous, isotropic media

∞ ∞ ∞

𝐏 𝑡 = 𝜀0 න 𝜒 𝑡 − 𝑡 ′ 𝐄 𝑡 ′ 𝑑𝑡 ′ 𝐏 𝑡 = 𝜀0 න 𝐏(𝜔)𝑒 𝑖𝜔𝑡 𝑑𝜔 𝐄 𝑡 = 𝜀0 න 𝐄(𝜔)𝑒 𝑖𝜔𝑡 𝑑𝜔

−∞ −∞ −∞

𝐏(𝜔) = 𝜀0 𝜒 𝜔 𝐄(𝜔) → 𝜒 𝜔 = න 𝜒(𝑡)𝑒 −𝑖𝜔𝑡 𝑑𝑡

−∞

𝐃 𝜔 = 𝜀0 𝜀𝑟 𝜔 𝐄 𝜔 𝜀𝑟 𝜔 = 1 + 𝜒 𝜔 → 𝑘 𝜔 = 𝜔 𝜀0 𝜇0 𝜀𝑟 𝜔 = 𝑛 𝜔 𝑘0
 Reflection and refraction at boundaries
Phase of wave is continuous at all times across boundaries
𝑘𝑖 = (sin 𝜃𝑖 𝑥,
ො 0, − cos 𝜃𝑖 𝑛)𝑘
ො 0 𝑛𝑖
𝑘𝑖 , 𝑘𝑡 , 𝑘𝑟 all lie in the same plane (plane of incidence)
𝑘𝑟 = (sin 𝜃𝑟 𝑥,
ො 0, cos 𝜃𝑟 𝑛)𝑘
ො 0 𝑛𝑟

𝑘𝑡 = (sin 𝜃𝑡 𝑥,
ො 0, − cos 𝜃𝑡 𝑛)𝑘
ො 0 𝑛𝑡
S-polarized wave (TE)

ෝ
𝒏 𝒌 ∙ 𝒓: phase continuous at the interface
2𝜋 2𝜋
𝒌𝑟 𝑘𝑖 sin 𝜃𝑖 = 𝑘𝑟 sin 𝜃𝑟 = 𝑘𝑡 sin 𝜃𝑡 , 𝑘𝑖 = 𝑘𝑟 = 𝑛1 ,𝑘 = 𝑛2
𝜆0 𝑡 𝜆0
𝑬𝑟
𝒌𝑖 𝑬𝑖 ∴ 𝜃𝑖 = 𝜃𝑟 & 𝑛𝑖 sin 𝜃𝑖 = 𝑛𝑡 sin 𝜃𝑡
𝜽𝑖 𝜽𝑟 𝑯𝑟
𝑛1 , 𝜀1 , 𝜇0 𝑯𝑖
plane of incidence
𝑛2 , 𝜀2 , 𝜇0
1 𝑛𝑖
ෝ𝐸𝑖 𝑒 𝑖(𝜔𝑡−𝒌𝑖 ∙𝒓) 𝐁𝑖 =
𝐄𝑖 = 𝒚 𝐤 𝑖 × 𝐄𝑖 = (− cos 𝜃𝑖 ෝ ෝ )𝐸𝑖 𝑒 𝑖(𝜔𝑡−𝒌𝑖 ∙𝒓)
𝒙 − sin 𝜃𝑖 𝒏
𝜔 𝑐
𝜽𝑡
1 𝑛𝑟
𝑬𝑡 ෝ𝐸𝑟 𝑒 𝑖(𝜔𝑡−𝒌𝑟 ∙𝒓) 𝐁𝑟 =
𝐄𝑟 = 𝒚 𝐤 𝑟 × 𝐄𝑟 = (cos 𝜃𝑟 ෝ ෝ )𝐸𝑟 𝑒 𝑖(𝜔𝑡−𝒌𝑟 ∙𝒓)
𝒙 − sin 𝜃𝑟 𝒏
𝜔 𝑐
𝑯𝑡 1 𝑛𝑡
ෝ𝐸𝑡 𝑒 𝑖(𝜔𝑡−𝒌𝑡 ∙𝒓) 𝐁𝑡 =
𝐄𝑡 = 𝒚 𝐤 𝑡 × 𝐄𝑡 = (− cos 𝜃𝑡 ෝ ෝ )𝐸𝑡 𝑒 𝑖(𝜔𝑡−𝒌𝑡 ∙𝒓)
𝒙 − sin 𝜃𝑡 𝒏
𝒌𝑡 𝜔 𝑐
 Fresnel’s equations
To obtain the amplitudes of the reflected and transmitted waves …
For 𝐸⊥ For 𝐸∥

→ For 𝜃𝑖 =0, →

The amplitudes and phases of the reflected and transmitted waves can be determined
from the reflection and transmission coefficients. For example, 𝑟⊥ = 𝑟⊥ exp(−𝑗𝜙⊥ )
𝜃𝑝 : Brewster’s angle

→ linearly polarized light

n1=1.44, n2=1
 Internal & external reflection

Phase changes at 𝜃𝑖 > 𝜃𝑐

n1=1.44, n2=1

𝑛1 < 𝑛2

mirror

mirror
 Snell’s law

transmitted wave

reflected wave

A light wave traveling in a medium with a greater

refractive index (n1>n2) suffers reflection and
refraction at the boundary.
 Dielectric constants
𝑑2 𝐗(t) 𝑑2 𝐗 t 𝑑𝐗 t 2
𝑒𝐄 t 𝑘
𝐅 = 𝑒𝐄 t − 𝑘𝐗 t − 𝛾𝑚𝑣 = 𝑚𝒂 = 𝑚 → + 𝛾 + 𝜔 0 𝐗 t = 𝜔0 =
𝑑t 2 𝑑t 2 𝑑t 𝑚 𝑚

𝑑2 𝐏 𝑑𝐏 𝑁𝑒 2 𝐄 t
𝐏 = −𝑁𝑒𝐗 t → 2 +𝛾 2
+ 𝜔0 𝐏 = Lorentz model
𝑑t 𝑑t 𝑚

If 𝐄 𝑡 = 𝑅𝑒 𝐄𝑒𝑥𝑝 𝑖𝜔𝑡 → 𝐏 𝑡 = 𝑅𝑒 𝐏𝑒𝑥𝑝 𝑖𝜔𝑡

𝑑2 𝐏 𝑑𝐏 2
𝑁𝑒 2 𝐄 t 2 2
𝑁𝑒 2 𝐄
∴ 2 +𝛾 + 𝜔0 𝐏 = → −𝜔 𝐏 + 𝑖𝛾𝜔 𝐏 + 𝜔0 𝐏 =
𝑑t 𝑑t 𝑚 𝑚

𝑁𝑒 2 𝐄 𝑁𝑒 2 𝜒0 𝜔02
𝐏= = 𝜀0 𝜒𝐄 → 𝜒 𝜔 = =
𝑚(𝜔02 − 𝜔 2 + 𝑖𝛾𝜔 ) 𝑚𝜀0 (𝜔02 − 𝜔 2 + 𝑖𝛾𝜔) 𝜔02 − 𝜔 2 + 𝑖𝛾𝜔
𝜒0 𝑣02
𝜒 𝑣 = 2
(𝑣0 − 𝑣 2 + 𝑖𝛾𝑣)/2𝜋

𝜒0 𝑣02 (𝑣02 − 𝑣 2 ) −𝜒0 𝑣02 (𝛾𝑣/2𝜋)

∴ 𝜒′ 𝑣 = 2 𝜒′′ 𝑣 = 2
(𝑣0 − 𝑣 2 )2 +(𝛾𝑣/2𝜋)2 (𝑣0 − 𝑣 2 )2 +(𝛾𝑣/2𝜋)2
 Dielectric constants
𝜒0 𝑣02 𝜒0 𝑣02 (𝑣02 − 𝑣 2 )
𝜒 𝑣 = 2 𝜒′ 𝑣 = 2 (real)
(𝑣0 − 𝑣 2 + 𝑖𝛾𝑣)/2𝜋 (𝑣0 − 𝑣 2 )2 +(𝛾𝑣/2𝜋)2
−𝜒0 𝑣02 (𝛾𝑣/2𝜋)
𝜒′′ 𝑣 = 2 (imaginary)
(𝑣0 − 𝑣 2 )2 +(𝛾𝑣/2𝜋)2

If 𝑣 ≪ 𝑣0 → 𝜒 ′ ~𝜒0 , 𝜒 ′′ ~0, the material exists at low frequency

If 𝑣 ≫ 𝑣0 → 𝜒 ′ = 𝜒′′~0, the material exists at high frequency

If 𝑣 ≈ 𝑣0 → 𝑣02 − 𝑣 2 = (𝑣0 − 𝑣)(𝑣0 + 𝑣) ≈ 2𝑣0 (𝑣0 − 𝑣)

𝜒0 𝑣0 /2 𝜒0 𝑣0 𝑣0 − 𝑣 𝜒0 𝑣0 𝑣0 − 𝑣 /2
→𝜒 𝑣 ≈ 𝜒′ 𝑣 = =
𝑣0 − 𝑣 + 𝑖𝛾/4𝜋 𝑣0 − 𝑣 2 + (𝛾/4𝜋)2 𝑣0 − 𝑣 2 + (∆𝑣)2
−𝜒0 𝛾𝑣0 /4𝜋 −𝜒0 (𝑣0 ∆𝑣/2)
We let ∆𝑣 = 𝑣 − 𝑣0 𝜒 ′′ 𝑣 = =
𝑣0 − 𝑣 2 + 𝛾/4𝜋 2 (𝑣0 − 𝑣)2 +(∆𝑣/2)2

If 𝜒 is small,
𝜒 𝜒 𝜒′ 𝜒′′
𝑁 𝑣 = 𝑛 𝑣 + 𝑖𝜅 𝑣 = 𝜀(𝑣) = 𝜀0 + 𝜒(𝑣) ≈ 𝜀0 1 + = 𝑛0 + = 𝑛0 + +𝑖
2𝜀0 2𝑛0 2𝑛0 2𝑛0
 Dielectric constants
𝜒 𝜒 𝜒′ 𝜒′′
𝑁 𝑣 = 𝑛 𝑣 + 𝑖𝜅 𝑣 = 𝜀(𝑣) = 𝜀0 + 𝜒(𝑣) ≈ 𝜀0 1 + = 𝑛0 + = 𝑛0 + +𝑖
2𝜀0 2𝑛0 2𝑛0 2𝑛0

𝜒′ 𝜒′′
𝑛 𝑣 = 𝑛0 + ,𝜅 𝑣 = 𝛽: propagation constant, 𝛽 = 𝑛𝑘0
2𝑛0 2𝑛0 𝛼: absorption coefficient/extinction coefficient

2𝜋𝑁(𝑣) 𝛼 2𝜋 𝑛 𝑣 + 𝑖𝜅 𝑣 𝛼 2𝜋 𝑖𝜅 𝑣 𝛼 4𝜋𝜅 𝑣
𝑘= =𝛽+𝑖 → =𝛽+𝑖 → =𝑖 →𝛼 𝑣 =
𝜆 2 𝜆 2 𝜆 2 𝜆

2𝜋 𝜒′′ 2𝜋𝑣 −𝜒0 𝛾𝑣0 /4𝜋 −𝜒0 (𝑣0 ∆𝑣/2)

𝛼 𝑣 = =( )𝜒′′(𝑣) 𝜒 ′′ 𝑣 = =
𝜆 𝑛0 𝑛0 𝑐0 𝑣0 − 𝑣 2 + 𝛾/4𝜋 2 (𝑣0 − 𝑣)2 +(∆𝑣/2)2
 Dielectric constants
𝜒′ 𝜒0 𝑣0 𝑣0 − 𝑣 𝜒0 𝑣0 𝑣0 − 𝑣 /2
𝑛 𝑣 = 𝑛0 + 𝜒′ 𝑣 = =
2𝑛0 𝑣0 − 𝑣 2 + (𝛾/4𝜋)2 𝑣0 − 𝑣 2 + (∆𝑣)2

2
𝑣𝑖2 𝜆2
𝑛 ≈ 𝑎 + ෍ 𝜒0𝑖 2 = 𝑏 + ෍ 𝜒0𝑖 2
𝑣𝑖 − 𝑣 2 𝜆 − 𝜆2𝑖
𝑖 𝑖
 Dispersion relation
B C
Cauchy dispersion equation 𝑛=𝐴+ 2
+ 4
= 𝑛−2 (ℎ𝑣)−2 +𝑛0 + 𝑛2 (ℎ𝑣)2 +𝑛4 (ℎ𝑣)4
𝜆 𝜆

Sellmeier equation 2
𝐴1 𝜆2 𝐴2 𝜆2 𝐴3 𝜆2
𝑛 =1+ 2 + +
𝜆 − 𝜆12 𝜆2 − 𝜆22 𝜆2 − 𝜆23
 Absorption
Transparent spectral windows
• High freq. → induced absorption from the transitions between electronic energy levels
• Low freq. → phonon absorption from the lattice vibrations

𝜒 𝜔 = 𝜒 ′ 𝜔 + 𝑖𝜒 ′′ 𝜔
→ wavenumber 𝑘 = 𝜔 𝜀𝑟 𝜀0 𝜇0 = 𝑘0 1 + 𝜒 = 𝑘0 1 + 𝜒 ′ + 𝑖𝜒 ′′
𝜀(𝜔) = 𝜀 ′ (𝜔) + 𝑖𝜀 ′′ (𝜔)

𝛼 𝛽: propagation constant (phase velocity), 𝛽 = 𝑛𝑘0

𝑘 =𝛽+𝑖
2 𝛼: absorption coefficient/extinction coefficient

2
→ 𝐼 𝑧 = 𝐸0 exp(𝑖𝑘𝑧) = 𝐸0 2 exp(−𝛼𝑧)

𝑛 = 𝜀𝑟 = 1+𝜒

𝜒′ 𝑘0 𝜒′
if 𝜒 ≪ 𝑛0 and 𝜒 ≪ 1 → 𝑛 ≈ 𝑛0 +
′ ′′
→ 𝛼≈−
2𝑛0 𝑛0

low absorbing materials

 Optical properties of conductors
In metals and doped semiconductors (conductive materials, free charges & current density)
𝜕𝐃
𝛁×𝐇= + 𝐉 → 𝛁 × 𝐇 = 𝑖𝜔𝐃 + 𝐉 for monochromatic waves
𝜕𝑡
𝐃 = 𝜀0 𝜀𝑟 𝐄 𝐉 = 𝜎𝐄 𝜎: conductivity

𝛁 × 𝐇 = 𝑖𝜔𝜀0 𝜀𝑟 𝐄 + 𝜎𝐄 = 𝑖𝜔𝜀0 𝜀𝑟 + 𝜎 𝐄 = 𝑖𝜔𝜀0 𝜀𝑒 𝐄 𝜀𝑒 : effective dielectric constants

Complex refractive index, 𝑁 = 𝑛 − 𝑖𝜅, 𝑁 = 𝜀𝑒 𝑖𝜔𝜀0 𝜀𝑟 + 𝜎 = 𝑖𝜔𝜀0 𝜀𝑒

𝜎 𝜎
2 2 2 2
𝜎 𝜀𝑟 + = 𝜀𝑒 → 𝜀𝑒 = 𝜀𝑟 − 𝑖
→ 𝑁 = (𝑛 − 𝑖𝜅) = 𝑛 − 𝜅 − 2𝑖𝑛𝜅 = 𝜀𝑟 − 𝑖 𝑖𝜔𝜀0 𝜔𝜀0
𝜔𝜀0
real: 𝑛2 − 𝜅 2 = 𝜀𝑟
𝜎 𝜎
imaginary: 2𝑛𝜅 = →𝜅=
𝜔𝜀0 2𝑛𝜔𝜀0
2
𝜎
𝑛2 − = 𝜀𝑟 → (4𝜔2 𝜀0 2 )𝑛4 − (4𝜔2 𝜀0 2 𝜀𝑟 )𝑛2 − 𝜎 2 = 0
2𝑛𝜔𝜀0
𝜎 2
→ 𝑛4 − (𝜀𝑟 )𝑛2 − ( ) =0
2𝜔𝜀0
 State of polarization

𝐸0𝑥 = 𝐸0𝑦 , 𝜙 = ±𝜋

-45֯

Linearly polarized EM wave

• Polarization: behavior of the electric field vector in the EM wave → plane of oscillation
• A light beam can be linearly polarized by passing the beam through a polarizer.
• We can define the 𝑥 and 𝑦 components of an electric field.

𝐸𝑥 = 𝐸0𝑥 𝑐𝑜𝑠(𝜔𝑡 − 𝑘𝑧) 𝐸𝑦 = 𝐸0𝑦 𝑐𝑜𝑠(𝜔𝑡 − 𝑘𝑧 + 𝜙)

• A phase difference 𝜙 can arise if one of the components is delayed. → retardation

• For example, linearly polarized EM wave:

ෝ𝐸𝑥 + 𝒚
𝐄=𝒙 ෝ𝐸𝑦 = 𝒙
ෝ𝐸0𝑥 𝑐𝑜𝑠 𝜔𝑡 − 𝑘𝑧 − 𝒚
ෝ𝐸0𝑦 𝑐𝑜𝑠 𝜔𝑡 − 𝑘𝑧 → 𝐄 = 𝐄0 𝑐𝑜𝑠 𝜔𝑡 − 𝑘𝑧

𝐸0𝑥 = 𝐸0𝑦 , 𝜙 = ±𝜋 ෝ𝐸𝑥0 − 𝒚

𝐄0 = 𝒙 ෝ𝐸𝑦0
 Circularly polarized wave
𝜋
𝐸𝑥0 = 𝐸𝑦0 = 𝐴, 𝜙 =
𝜃 = 𝑘∆𝑧 2
𝐸𝑥 = 𝐴𝑐𝑜𝑠(𝜔𝑡 − 𝑘𝑧)
𝐸𝑦 = −𝐴𝑠𝑖𝑛(𝜔𝑡 − 𝑘𝑧)
Right circularly polarized EM wave 𝐸𝑥2 + 𝐸𝑦2 = 𝐴2

Elliptically polarized
Linearly polarized Circularly polarized
 Polarization states
For 2𝐸𝑥 = 𝐸𝑦

−𝜋 3 1 1 0 1 1 3 𝜋
− 𝜋 − 𝜋 − 𝜋 𝜋 𝜋 𝜋
4 2 4 4 2 4

For 𝐸𝑥 = 𝐸𝑦

−𝜋 3 1 1 0 1 1 3 𝜋
− 𝜋 − 𝜋 − 𝜋 𝜋 𝜋 𝜋
4 2 4 4 2 4
 Malus’s Law

What is the intensity of light

passing through two polarizers?
What are they ways to rotate the
polarization states?

• A linear polarizer will only allow electric field oscillations along some preferred direction,
called the transmission axis, to pass through the device.
• Dichroic crystals are good polarizers because they are optically anisotropic and attenuate
EM waves with fields that are not oscillating parallel to the transmission axis.
the intensity of a linearly polarized light passing
𝐼 𝜃 = 𝐼(0)𝑐𝑜𝑠 𝜃 2 through a polarizer to the angle between the
transmission axis and the electric field vector

Types of linear polarizers: metallic wire grid, dichroic polaroids, thin films (Brewster’s angle)
 Anisotropic media

𝑃𝑖 = 𝜀0 ෍ 𝜒𝑖𝑗 𝐸𝑗 → 𝐷𝑖 = ෍ 𝜀𝑖𝑗 𝐸𝑗
𝑗 𝑗
Birefringence
• Light propagation in an anisotropic medium
• Crystals are generally optically anisotropic, except the cubic crystals
• The dielectric constant 𝜀𝑟 depends on electric polarizations → the displacement of electrons
• The refractive index depends on the propagation direction and the state of polarization
• Any unpolarized light ray entering such a crystal breaks into two different rays with different
polarizations and phase velocities. (calcite crystal)
• Optically anisotropic crystals are called birefringent. → doubly refract an incident light
• A special direction in a birefringent crystal: the optic axis.
• According to the principal refractive indices,

𝑛𝑜 < 𝑛𝑒

𝑛𝑜 > 𝑛𝑒
NaCl crystal calcite crystal
 Uniaxial crystals
• An EM wave entering an anisotropic crystal splits into two orthogonal linearly polarized waves.
→ travel with different phase velocities; experience different refractive indices
• These two orthogonally polarized waves in uniaxial crystals are called ordinary (o) and
extraordinary (e) waves.
• o-wave: same phase velocity in all directions; the electric field is perpendicular to the phase
propagation direction.
• e-wave: phase velocity depends on its direction of propagation and polarization state; E field is
not necessarily perpendicular to the phase propagation direction.
• The two waves propagate with the same velocity only along the optical axis.
• The o-wave is always perpendicularly polarized to the optic axis and obeys the usual Snell’s law.

• Two polaroid analyzers are placed with their transmission

axes, along the long edges, at right angles to each other.
• The ordinary ray goes through the left polarizer.
• The extraordinary wave goes through the right polarizer.
• The two waves therefore have orthogonal polarizations.
 Retarding plates
• The optic axis is parallel to the plate faces.
𝑛∥ > 𝑛⊥ • A linearly polarized wave is incident at
normal incidence on a plate face. → E is
parallel to the optic axis (slow axis); E is at
right angles to the optic axis (fast axis)

2𝜋
𝜙= 𝑛 −𝑛𝑜 𝐿
𝜆 𝑒

𝐿 is the thickness of the plate.

Circular polarization from linear polarization

Consider a linearly polarized light that is incident on a quarter-wavelength plate, such

that the polarization is 45°to the slow axis. Show that the output beam is circularly
polarized.

Both the 𝑥 and 𝑧 components of the electric field emerging from the retarder plate are
propagating along the y-axis with the same cos(𝜔𝑡 − 𝑘𝑦) harmonic term after emerging
from the crystal but with a phase difference 𝜙. We can write the field components along
fast (𝑥) and slow (𝑧) axes as

𝐸𝑥 = 𝐸⊥ 𝑐𝑜𝑠(𝜔𝑡) 𝐸𝑧 = 𝐸∥ 𝑐𝑜𝑠(𝜔𝑡 − 𝜙)

When the incident polarization is at 45°, then 𝐸∥ = 𝐸⊥ = 𝐸0 . Putting 𝜙 = 𝜋/2 for a

𝜋
quarter-wave plate, using 𝑐𝑜𝑠 𝜔𝑡 − = 𝑠𝑖𝑛(𝜔𝑡), and then 𝑠𝑖𝑛2 (𝜔𝑡) + 𝑐𝑜𝑠 2 (𝜔𝑡) = 1,
2

2 2
2 2
𝐸𝑥 𝐸𝑧
𝑠𝑖𝑛 (𝜔𝑡) + 𝑐𝑜𝑠 (𝜔𝑡) = 1 → + =1
𝐸0 𝐸0

𝜆0
For ¼ waveplate, 𝑑 𝑛𝑒 − 𝑛𝑜 = (4𝑚 + 1)
4
 Quartz half-wave plate
What should be the thickness of a half-wave quartz plate for a wavelength 𝜆 =
590 nm (𝑛𝑜 = 1.5442, 𝑛𝑒 = 1.5533)?

2𝜋 1ൗ (590 × 109 𝑚)
𝜙= 𝑛𝑒 −𝑛𝑜 𝐿 = 𝜋 → 𝐿 = 2 ≈ 32.4 𝜇𝑚
𝜆 (1.5533 − 1.5442)

𝜆0
For ½ waveplate, 𝑑 𝑛𝑒 − 𝑛𝑜 = (4𝑚 + 1)
2
 Optical chirality

Thalidomide