Chapter 5 Electromagnetic Optics: 5.1 Electromagnetic Theory of Light
Chapter 5 Electromagnetic Optics: 5.1 Electromagnetic Theory of Light
Lin
In free space
∂E
∇ × H = ε0 (5.1-1)
∂t
∂H
∇ ×E = −µ 0 (5.1-2)
∂t
∇ ⋅E = 0 (5.1-3)
∇ ⋅H = 0 (5.1-4)
1 Farad
ε 0 : Electric permittivity = × 10 −9 ( ) in MKS units
36π Meter
Henry
µ 0 : Magnetic permeability = 4π × 10 −7 ( ) in MKS units
Meter
1
EE 485, Winter 2004, Lih Y. Lin
∂D
∇ ×H = (5.1-7)
∂t
∂B
∇ ×E = (5.1-8)
∂t
∇ ⋅D = 0 (5.1-9)
∇ ⋅B = 0 (5.1-10)
D = ε 0E + P (5.1-11)
B = µ 0H + µ 0 M (5.1-12)
P : Polarization density (macroscopic sum of the electric dipole
moments that the electric field induces)
M : Magnetization density (macroscopic sum of the magnetic dipole
moments that the magnetic field induces)
In a non-magnetic medium (assumption of this course), M = 0
→ B = µ 0H (5.1-13)
Boundary conditions
Assume no free charges or surface currents.
In a homogeneous medium, E , H , D , B are continuous.
At the boundary between two dielectric media, the tangential components of E and
H, and the normal components of D and B are continuous.
2
EE 485, Winter 2004, Lih Y. Lin
Anisotropic media
P and E are not necessarily parallel.
If the medium is linear, non-dispersive, and homogeneous,
Pi = ∑ ε 0 χ ijE j i, j = 1, 2, 3 denote the x, y, and z components
j
Di = ∑ ε ijE j
j
Dispersive media
t
P (t ) = ε 0 ∫ x(t − t ' )E (t ' )dt ' (5.2-17)
−∞
3
EE 485, Winter 2004, Lih Y. Lin
Nonlinear media
If the medium is homogeneous, isotropic, and nondispersive, P = Ψ (E ) for all
position and time, Ψ is a nonlinear function.
Wave equations:
2 1 ∂ 2E ∂ 2 Ψ (E )
∇ E − 2 2 = µ0 (5.2-20)
c0 ∂t ∂t 2
→ Principle of superposition no longer applicable
Maxwell’s equations
∇ × H = jωD (5.3-2)
∇ × E = − jωB (5.3-3)
∇⋅D = 0 (5.3-4)
∇⋅B = 0 (5.3-5)
D = ε0E + P (5.3-6)
B = µ0H (5.3-7)
4
EE 485, Winter 2004, Lih Y. Lin
∇ 2U + k 2U = 0 (5.3-15)
k = nk 0
Inhomogeneous media
Eqs. (5.3-11) ~ (5.3-14) remain applicable, ε → ε(r ) .
For locally homogeneous, ε(r ) varies slowly with respect to wavelength
ε(r )
→ k = n(r )k0 , n(r ) =
ε0
Dispersive media
P = ε 0 χ ( ν )E (5.3-16)
D = ε(ν)E (5.3-18)
ε(ν) = ε 0 [1 + χ(ν)] (5.3-19)
→ The only difference between non-dispersive medium and dispersive medium is
that ε and χ are frequency-dependent. The Helmholtz equation applicable with
ε( ν )
k = n (ν ) k 0 , n (ν ) = (5.3-20)
ε0
E0 1 µ0 η
= ≡ η = 0 : Impedance of the medium (5.4-5,6)
H 0 n ε0 n
µ0
η0 = = 120π = 377Ω : Impedance of free space (5.4-7)
ε0
5
EE 485, Winter 2004, Lih Y. Lin
2
E
Magnitude of Poynting vector S = 0 = I : Intensity (5.4-8)
2η
B. Dispersion
χ = χ(ν), n = n(ν), c = c(ν) = c0 / n(ν)
6
EE 485, Winter 2004, Lih Y. Lin
Measures of dispersion
Examples:
1) For glass optical components used with white light,
n −1
V number ≡ D
nF − nC
nF, nD, nC: Refractive indices at blue (486.1 nm), yellow (589.2 nm),
and red (656.3 nm)
dn
2)
dλ λ = λ
0
dθ d dθ d dn
Prism: =
dλ dn dλ
7
EE 485, Winter 2004, Lih Y. Lin
8
EE 485, Winter 2004, Lih Y. Lin
ν
ν = ν 0 → χ' (ν 0 ) = 0, - χ" (ν) = 0 χ 0
∆ν
Near resonance:
ν0 / 2
χ( ν ) = χ 0 (5.5-18)
(ν 0 − ν) + j∆ν / 2
ν ∆ν 1
χ" (ν) = −χ 0 0 (5.5-19)
4 (ν 0 − ν )2 + (∆ν / 2 )2
ν − ν0
χ' ( ν ) = 2 χ" (ν) (5.5-20)
∆ν
∆ν : FWHM of χ" (ν)
Linear-system description
Suppose A (0, t ) = A(0, f ) exp( j 2πft )
A( z , f ) = A(0, f )H ( f ) (5.6-2)
H ( f ) = exp − α f +ν z − j (β f +ν − βν )z
1
(5.6-3)
2 0 0
0
9
EE 485, Winter 2004, Lih Y. Lin
∞
A(0, f ) = ∫ A (0, t ) exp(− j 2πft )dt
−∞
A( z , f ) = A(0, f )H ( f )
Inverse Fourier transform
∞
A ( z, t ) = ∫ A( z, f ) exp( j 2πft )df
−∞
Or by convolution:
∞
A ( z , t ) = ∫ A (0, t ' )h(t − t ' )dt ' (5.6-5)
−∞
∞
h(t ) = ∫ H ( f ) exp( j 2πft )df
−∞
10
EE 485, Winter 2004, Lih Y. Lin
τ d = z / υ g , υ g : Group velocity
1 1 dβ dβ
= = (5.6-8)
υ g 2π dν dω
1 d 2β d 2β d 1
: Dispersion coeff.
Dν = = 2π 2 = (5.6-9)
2π dν 2 dω dν υ g
Dispersion coefficient
υ g = υ g (ν ) → τ d = τ d ( ν )
dτ d d z
δτ = δν = δν = Dν zδν
dν dν υ g
Normal dispersion: Dν > 0 . Anomalous dispersion: Dν < 0 .
If the pulse has a spectral width σ ν (Hz), the spread of the temporal width
σ τ = Dν σ ν z (5.6-10)
Dν : second/m·Hz → Measure of the pulse time broadening per unit spectral width
per unit distance
Determine the shape of the transmitted pulse:
∞
A ( z , t ) = ∫ A (0, t ' )h(t − t ' )dt '
−∞
1 (t − τ d ) 2
h(t ) = H 0 exp jπ (5.6-11)
j Dν z Dν z
3 2
λ d n (λ ) sec
Dν = 0
( ) (5.6-20)
c dλ2 λ=λ m ⋅ Hz
2
0 0
In terms of wavelength:
Dλ dλ = Dν dν
λ 0 d 2 n (λ ) sec
Dλ = − 2
( ) (5.6-21)
c0 dλ λ=λ m ⋅ nm 0
11
EE 485, Winter 2004, Lih Y. Lin
12