Photonics

Prof. Maria Antonietta Vincenti

Università degli Studi di Brescia
 Electromagnetic Optics:
Reflection, Refraction and Scattering
 Refraction
Water Waves E&M Waves

Waves refract where the water is shallower

Refraction involves a change in the direction of wave propagation due to a change in

propagation speed. It involves the oblique incidence of waves on media boundaries,
and hence wave propagation in at least two dimensions.
 Refrac+on
Think of refraction as a pair of wheels on an axle going from a sidewalk onto grass.
The wheel in the grass moves slower, so the direction of the wheel pair changes.

Note that the wheels

move faster (longer space)
on the sidewalk, and slower
(shorter space) in the grass.
 Reflection and Refraction
𝑥 at an interface
𝑛$
𝜖! 𝜇! 𝜖" 𝜇"
𝜃! 𝐤!
TE-pol. 𝐄!
wave Incidence and
𝐇! transmission media
𝑛$
are assumed
𝐄! 𝜃! 𝐤!
TM-pol. isotropic.
wave 𝐇!
𝑧=0 𝑧
Plane of Incidence: The plane containing the incident wavevector ki and a vector that is normal
to the interface is called the plane of incidence (in the figure above, the normal to the interface
is 𝑧̂ and the x-z plane is the plane of incidence)

TE Wave: If the E-field of the wave is perpendicular to the plane of incidence then the wave is
called a TE-wave (transverse electric polarizaBon, also known as s-polarizaBon)

TM Wave: If the H-field of the wave is perpendicular to the plane of incidence (i.e., if the E-field
lies in the plane of incidence) then the wave is called a TM-wave (transverse magneBc
polarizaBon, also called p-polarizaBon)
 Reflection and Refraction
at an interface
TE Wave

𝑥
𝜖! 𝜇! 𝐤# 𝐇# 𝜖" 𝜇"
𝜃#
𝐄# 𝐤"
𝑛$ 𝐄" 𝜃"
𝜃! 𝐤!
𝐄!
𝐇"
𝐇! 𝑧=0 𝑧

ni
k i = kix xˆ + kiz zˆ = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû , k i = w µie i = (if nonmagnetic) = w
c
ni
k r = krx xˆ + krz zˆ = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû , k r = k i = w µie i = (if nonmagnetic) = w
c
nt
k t = ktx xˆ + ktz zˆ = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû , k t = w µt e t = (if nonmagnetic) = w
c
 Reflection and Refraction
at an interface
TE Wave – First Boundary Condition: continuity of 𝐸$ at 𝑧 = 0

k i = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû

E (r ) z <0 = yE
ˆ ie - jk i ×r
+ yE
ˆ re - jk r ×r
k r = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû
E (r ) z >0 = yE
ˆ te - jk t ×r
k t = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû

(1) At z = 0 the E-field parallel to the interface must be continuous across the interface for all x:
- jki sin (qi ) x - jkr sin (q r ) x - jkt sin (qt ) x
Ei e + Er e = Et e
 Reflection and Refraction
at an interface
TE Wave – First Boundary CondiDon: conBnuity of 𝐸$ at 𝑧 = 0

The only way the above boundary condition can be satisfied for all x , and at all times t, is if all
the x dependent phase factors are the same. This condition is called “phase matching
condition”, or conservation of the transverse wavevector, or conservation of the transverse
momentum.
ki sin (qi ) = kr sin (q r ) = kt sin (qt ) kix = krx = ktx

This also implies that: angle of incidence

sin (qi ) = sin (q r ) qi = q r equals the
angle of reflection
 Reflection and Refraction
at an interface
TE Wave – Snell’s Law

ni nt
ki sin (qi ) = kt sin (qt ) w sin (qi ) = w sin (qt )
c c

ni sin (qi ) = nt sin (qt ) Snell’s law

Moreover:

- jki sin (qi ) x - jkr sin (q r ) x - jkt sin (qt ) x Ei + Er = Et (1)

Ei e + Er e = Et e
 Reflection and Refraction
at an interface
TE Wave – Second Boundary Condition: continuity of 𝐻% at 𝑧 = 0

k i = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû

H (r ) z <0 (
ˆ
= ki ´ y
ˆ e ) hi
Ei - jk i ×r E
(
+ kˆr ´ yˆ r e - jk r ×r ) hi
k r = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû
H (r ) z >0 ( E
) ht
= kˆt ´ yˆ t e - jk t ×r k t = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû
(2) At z = 0 the H-field component parallel to the interface must be continuous for all x
Ei - jki sin (qi ) x Er - jkr sin (q r ) x Et - jkt sin (qt ) x
- xˆ cos (qi ) e + xˆ cos (q r ) e = - xˆ cos (qt ) e
hi hi ht
é Ei Er ù E
cos (qi ) ê - ú = cos (qt ) t (2)
ë hi hi û ht
 Reflection and Transmission
Coefficients
TE Wave

éE E ù E
Ei + Er = Et cos (qi ) ê i - r ú = cos (qt ) t
ë hi hi û ht
By solving the equation on the right using the equation on the left we get:
Et ht cos (qi ) Er ht cos (qi ) - hi cos (qt )
tTE = =2 rTE = =
Ei hi cos (qt ) + ht cos (qi ) Ei ht cos (qi ) + hi cos (qt )
The two equations above can be re-written when considering µ= µ0=1 as:

Et ni cos (qi ) Er ni cos (qi ) - nt cos (qt )

tTE = =2 rTE = =
Ei nt cos (qt ) + ni cos (qi ) Ei ni cos (qi ) + nt cos (qt )
 Reflection and Refraction
at an interface
TM Wave

𝑥
𝜖! 𝜇! 𝐤# 𝐄# 𝜖" 𝜇"
𝜃#
𝐇# 𝐤"
𝐄"
𝑛$ 𝜃"
𝐄! 𝜃! 𝐤! 𝐇"

𝐇!
𝑧=0 𝑧

The definitions of the wavevectors are identical for both polarizations:

ni
k i = kix xˆ + kiz zˆ = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû , k i = w µie i = (if nonmagnetic) = w
c
ni
k r = krx xˆ + krz zˆ = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû , k r = k i = w µie i = (if nonmagnetic) = w
c
n
k t = ktx xˆ + ktz zˆ = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû , k t = w µt e t = (if nonmagnetic) = w t
c
 Reflection and Refraction
at an interface
TM Wave – First Boundary Condition: continuity of 𝐻$ at 𝑧 = 0

Ei Er k i = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû

H (r ) z <0 = yˆ e - jk i ×r - yˆ e- jk r ×r
hi hi
k r = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû
Et
H (r ) z >0 = yˆ e- jk t ×r k t = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû
ht
(1) At z = 0 the H-field parallel to the interface must be continuous across the interface for all x:
Ei - jki sin (qi ) x Er - jkr sin (q r ) x Et - jkt sin (qt ) x
e - e = e
hi hi ht
 Reflection and Refraction
at an interface
TM Wave – First Boundary CondiDon

𝜖! 𝜇! 𝜖" 𝜇"

The only way the above boundary condition can be satisfied for all x is if all the x dependent
phase factors are the same. This condition is called “phase matching”.

ki sin (qi ) = kr sin (q r ) = kt sin (qt ) kix = krx = ktx

This also implies that:

angle of incidence
sin (qi ) = sin (q r ) qi = q r equals the
angle of reflection
 Reflection and Refraction
at an interface
TM Wave – Snell’s Law

ni nt
ki sin (qi ) = kt sin (qt ) w sin (qi ) = w sin (qt )
c c

ni sin (qi ) = nt sin (qt ) Snell’s law

Moreover:

Ei - jki sin (qi ) x Er - jkr sin (q r ) x Et - jkt sin (qt ) x hi

e - e = e Ei - Er = Et (1)
hi hi ht ht
 Reflection and Refraction
at an interface
TM Wave – Second Boundary Condition

k i = ki éësin (qi ) xˆ + cos (qi ) zˆ ùû

E (r ) z <0 ( ) ( )
= - kˆi ´ yˆ Ei e - jk i ×r + kˆr ´ yˆ Er e - jk r ×r
k r = kr éësin (q r ) xˆ - cos (q r ) zˆ ùû
E ( r ) z >0 = - ( kˆt ´ yˆ ) Et e - jk ×r t

k t = kt éësin (qt ) xˆ + cos (qt ) zˆ ùû

(2) At z = 0 the E-field component parallel to the interface must be continuous for all x
- jki sin (qi ) x
= - xˆ cos (qt ) Et e t ( t )
- jkr sin (q r ) x - jk sin q x
- xˆ cos (qi ) Ei e +- xˆ cos (q r ) Er e

cos (qi )( Ei + Er ) = cos (qt ) Et (2)

 Reflec3on and Transmission
Coefficients
TM Wave

hi cos (qi )( Ei + Er ) = cos (qt ) Et

Ei - Er = Et
ht
By solving the equation on the right using the equation on the left we get:
Et ht cos (qi ) Er ht cos (qt ) - hi cos (qi )
tTM = =2 rTM = =
Ei ht cos (qt ) + hi cos (qi ) Ei ht cos (qt ) + hi cos (qi )
The two equations above can be re-written when considering µ= µ0=1, as:

Et ni cos (qi ) Er ni cos (qt ) - nt cos (qi )

tTM = =2 rTM = =
Ei ni cos (qt ) + nt cos (qi ) Ei ni cos (qt ) + nt cos (qi )
 Reflection and Transmission
Coefficients
For non-magnetic materials
𝝁 = 𝝁𝟎

ni cos (qi ) - nt cos (qt ) ni cos (qt ) - nt cos (qi )

rTE = rTM =
ni cos (qi ) + nt cos (qt ) ni cos (qt ) + nt cos (qi )

ni cos (qi ) ni cos (qi )

tTE = 2 tTM = 2
nt cos (qt ) + ni cos (qi ) ni cos (qt ) + nt cos (qi )

TE Wave TM Wave
 Reflectance and Transmittance

𝜖! , 𝜇! 𝜖" , 𝜇" Consider a finite beam of uniform

!
𝑠𝜃
𝑐𝑜
intensity, equal to the intensity of the
𝐴

"! "
input plane wave, 𝐼! = 𝑅𝑒 𝑆! =
#$!

𝐴 𝑐𝑜
Here we assume the input medium be
lossless.

𝑠𝜃"
𝜃# 𝐴 𝜃"
𝜃! Reflectance or Reflectivity
𝑅 = 𝑃% /𝑃!

Transmittance or transmittivity
𝑇 = 𝑃& /𝑃!
𝐴
𝑐𝑜
𝑠𝜃
!

Power of the input beam Power of the reflected beam Power of the transmitted beam
𝐸! # 𝐸% # 𝐸& #
𝑃! = 𝐼! 𝐴! = 𝐴 𝑐𝑜𝑠𝜃! 𝑃% = 𝐼% 𝐴% = 𝐴 𝑐𝑜𝑠𝜃! 𝑃& = 𝐼& 𝐴& = 𝐴 𝑐𝑜𝑠𝜃&
2𝜂! 2𝜂! 2𝜂&

' 𝜂! 𝑐𝑜𝑠𝜃"
𝐸# ' '
𝑅= '
= 𝑟 𝑇 =1−𝑅 = 𝑡
𝐸! 𝜂" 𝑐𝑜𝑠𝜃!
 Snell’s Law
ni sin (qi ) = nt sin (qt )

If ni < nt then θt < θi and the transmitted wave bends towards the normal

If ni > nt then θt > θi and the transmitted wave bends away the normal
 Total Internal Reflection
& Critical Angle
ni > nt

If qi is increased, then qt
will eventually become
90°. The value of qi for
which qt is 90° is called
the criBcal angle qc

æp ö
ni sin (qc ) = nt sin ç ÷
è2ø

æ nt ö
qc = a sin ç ÷
n
è iø

If qi is increased beyond qc the wave is not transmitted but is completely (100%) reflected at the
interface back into the medium of incidence.
This phenomenon is called TOTAL INTERNAL REFLECTION, and it happens for both TE and TM
waves
 Total Internal Reflection
Phase Matching
ni > nt and θi > θc
The phase matching condition gives:
kix = krx = ktx

ktx = kix = ki sin (qi )

w2
ktx2 = ki2 sin 2 (qi ) = ni2 sin 2 (qi )
c2
w2 w2
The dispersion relation in the medium “t” is: kt2 = 2
nt
2
ktx2 + ktz2 = nt2
c c2

w2 w2 é 2 2 2 w
ktz2 = n 2
- ktx2 = 2 nt - ni sin (qi ) ù ktz = nt2 - ni2 sin 2 (qi ) This value is NEGATIVE
c 2 t
c ë û c when qI > qc

In other words, the z-component of the wavevector has become w

completely imaginary, therefore we can write the E-field ktz =-j ni2 sin 2 (qi ) - nt2 = - jktz''
(assuming TE polarizaDon) as: c

ˆ t e - jktx x e - ktz z
''
E(r ) z >0 = yE EVANESCENT FIELD in the z-direction
 Total Internal Reflection
Electric Field Profile
ni > nt and θi > θc

the E-field in the medium “t”(assuming TE polarization)is:

ˆ t e - jktx x e - ktz z = yˆ Et e - jj e - jktx x e - ktz z

'' ''
E(r ) z >0 = yE

The wave is propagating along

E(r ) z >0 = yˆ Et e - ktz'' z
cos (wt - ktx x - j ) the interface (in the x-direction) but decaying
(without spatial oscillations) in the z-direction

If θi > θc the wave is completely reflected back into the medium of incidence and we have:
w
ktz = - j ni2 sin 2 (qi ) - nt2 = - jktz''
c
Er kiz - ktz kiz + jktz'' ij
The reflection coefficient for the E-field (assuming TE wave) is: G = = = = e
Ei kiz + ktz kiz - jktz''
G = 1, j Goos-Hanschen phase-shift
 Brewster’s Angle

Today’s Culture Moment

Sir David Brewster

• Scottish scientist 1781 –1868

• Studied at University of Edinburgh at age 12
• Independently discovered Fresnel lens
• Editor of Edinburgh Encyclopedia and contributor to
Encyclopedia Britannica (7th and 8th editions)
• Inventor of the Kaleidoscope
• Nominated (1849) to the National Institute of France.

Kaleidoscope
 Brewster’s Angle

Question: Can one ever get the reflection coefficient to go to zero (very desirable to get rid of
unwanted reflections in optics)?

Er ni cos (qi ) - nt cos (qt ) H r ni cos (qt ) - nt cos (qi )

rTE = = rTM = =
Ei ni cos (qi ) + nt cos (qt ) H i ni cos (qt ) + nt cos (qi )

For a TE wave reflecBon is zero if : ni cos (qi ) = nt cos (qt ) ki cos (qi ) = kt cos (qt )

If: ni ¹ nt then kiz ¹ ktz since kix = ktx for the phase-matching condition.

Therefore REFLECTION IS NEVER ZERO FOR TE WAVES

 Brewster’s Angle

Question: Can one ever get the reflection coefficient to go to zero (very desirable to get rid of
unwanted reflections in optics)?

Er ni cos (qi ) - nt cos (qt ) H r ni cos (qt ) - nt cos (qi )

rTE = = rTM = =
Ei ni cos (qi ) + nt cos (qt ) H i ni cos (qt ) + nt cos (qi )

For a TM wave reflection is zero if : ni cos (qt ) = nt cos (qi )

For Snell’s law: ni sin (qi ) = nt sin (qt )

Using both equations, solving for the terms in qi and then adding each corresponding term one
can obtain: æ nt ö
q B = a tan ç ÷
n
è iø
 Prac3cal Example:
The Lens Flare!
If qi=0 reflectance and transmittance are:

2 2
2 æ ni - nt ö nt 2 æ ni ö nt
R= r =ç ÷ T = t = 4ç ÷
n +
è i tø n ni +
è t i ø ni
n n

For a camera lens reflection/transmission occurs at the interface between glass and air.
If ni = 1 and nt = 1.5:

R = 4% and T = 96%

Reflectance is “only” 4% but has

big implication for photography
 Practical Example:
Fresnel Equations in action!

Windows look like mirrors at night (when you’re in a brightly lit room).

One-way mirrors (used by police to interrogate bad guys) are just

partial reflectors (actually, with a very thin aluminum coating).

Smooth surfaces can produce pretty good mirror-like reflections, even

though they are not made of metal.
 Prac3cal Example:
Fresnel Equa3ons in ac3on!

Optical fibers only work because of total internal reflection.

Many lasers use Brewster’s angle components to avoid

reflective losses
 Scattering
Scattering can be broadly defined as the
redirection of radiation out of the original direction of propagation
usually due to interactions with molecules and particles

Reflection, refraction, diffraction etc. are actually all just forms of scattering. The
superposition of incident and scattered waves is what is actually observed.
 When is scattering important?

Scattering is negligible whenever gains in intensity due to scattering are small

compared to:

• Losses due to extinction

• Gains due to thermal

emission

• When considering direct

radiation from a point source,
such as the sun

In the UV, visible and near-IR bands, scaUering is the dominant source of radiaVon
along any line of sight, other than looking directly at the sun.
 Types of Scattering
Elastic scattering – the wavelength (frequency) of the scattered light is the same as the
incident light (Rayleigh and Mie scattering)

Inelastic scattering – the emitted radiation has a wavelength different from that of the
incident radiation (Raman scattering, fluorescence)

Quasi-elastic scattering – the wavelength (frequency) of the scattered light shifts (e.g.,
in moving matter due to Doppler effects)
 Types of Scattering

Single scattering: photons scattered only once prevail in optically thin media, since photons have
a high probability of exiting the medium (e.g., a thin cloud) before being scattered again . Also
favored in strongly absorbing media

Multiple scattering: prevails in optically thick, strongly scattering and non-absorbing media.
Photons may be scattered hundreds of times before emerging
 Practical example:
Variation in sky brightness

NOTE: The horizon sky is

usually brighter than the
zenith sky.
This is a result of single
scattering (zenith) vs.
multiple scattering
(horizon)
 Parameters governing scattering

(1) The wavelength (λ) of the incident radiaVon

(2) The size of the scaUering parVcle

(3) The parVcle opVcal properVes relaVve to the surrounding medium

Ei ( r ) Es ( r )
 Scattering from spherical particles
Ei ( r )

Different scattering conditions can be identified depending on their geometrical size in relation
with the incident wavelength. Let’s define the adimensional parameter:

2p a
x= = ka
l

If ka << 1 the scattering is called Rayleigh Scattering

If ka ~ 1 the scattering is called Mie Scattering
If ka >> 1 the scattering is called Geometrical Scattering

When ka << 1, the particle sees a uniform E-field that is slowly oscillating in time
 Light scattering regimes

NOTE: This plot considers only single scattering by spheres. Multiple scattering and
scattering by non-spherical objects can get really complex!
 Rayleigh Scattering

Ei ( r ) z

One way to understand scattering is as follows:

i) The incident E-field induces a time-varying dipole moment in the sphere
ii) The time-varying dipole radiates like a Hertzian dipole and this is the scattered radiation
Let’s now suppose the z-directed E-field phasor for the incident plane wave at the locaBon of
the parBcle is:
E(r = 0) = zE
ˆ i
The z-directed dipole moment p induced in a sphere in the presence of E-field E is:

3æ e1 - e 0 ö
p = 4pe 0 a ç ÷ Ei
è e1 + 2e 0 ø
 Rayleigh Scattering

Ei ( r ) z

The electric field radiated by the oscillating dipole in the far field (r>>a) is:

1 𝑘'
𝐄𝐬𝐜 (𝐫) = 𝒓$ ×𝐩 ×$𝒓𝑒 !+#
4𝜋𝜖* 𝑟

If we now replace the expression of the dipole moment we get:

𝑘 ' 𝑎, 𝜖- − 𝜖. !+#
𝐄𝐬𝐜 (𝐫) = 𝐸! 𝑒 𝑠𝑖𝑛𝜃 𝒆$ /
𝑟 𝜖- + 2𝜖*
 Rayleigh Scattering
Total scanered power Ps from a dielectric sphere is:
2p p 2 2
E s (r ) 4p 4 6 æ e1 - e 0 ö 2
òò
2
Ps = r sin(q )dq df = k a ç ÷ Ei
0 0
2h0 3h0 è e1 + 2e 0 ø
2
The incident power per unit area is the Poynting vector of the incident wave: Ei
2h0
The scattering cross-section σs of a scatterer is defined as the area of a plane oriented
perpendicular to the direction of incident wave that would intercept the same total incident
power as the power Ps that the scatterer radiates:

Ps σs is also the ratio of the total scattered power

ss = 2 to the power per unit area of the incident wave
Ei (r ) 2h0 at the location of the scatterer

2
8p 4 6 æ e1 - e 0 ö
For the dielectric sphere:
ss = k a ç ÷
3 e
è 1 + 2 e 0ø
 Scattering phase functions
The scattering phase function, or phase function, gives the angular distribution of light
intensity scattered by a particle at a given wavelength.
 Practical example:
Rainbows
Rainbow: for large parDcles (x = 10,000), the
forward and backward peaks in the scanering
phase funcBon become very narrow (almost
non-existent). Light paths are best predicted
using geometric opBcs and ray tracing.

Primary rainbow: single internal reflecBon

Secondary rainbow: double internal reflecBon
 Practical example:
Glory and Fogbow
Fogbow: spikes in scattering phase function present but not
sharp as for rainbows. Hence the separation of colors (due
to varying refractive index) is not as vivid as a normal
rainbow. A whitish ring centered on one’s shadow (i.e.
opposite the sun) is seen. Arises when water droplets have
a size characteristic of fog and clouds rather than rain.

Glory

Glory: opposite end of the phase function from the corona.

Seen as a ‘halo’ around one’s shadow when looking at a fog
bank with the sun at your back. Also seen from aircraft.
Glories have vivid colors if the range of drop sizes in the fog
is relatively narrow, otherwise they are whitish.
 Practical example:
Corona

Corona: for intermediate values of the size parameter (x), the forward scattering peak
is accompanied by weaker sidelobes. If you were to view the sun through a thin cloud
composed of identical spherical droplets (with x = 100 or less), you would see closely
spaced rings around the light source. The angular position of the rings depends on
wavelength, so the rings would be colored. This is a corona. Because few real clouds
have a sufficiently narrow distribution of drop sizes, coronas are usually more diffuse
and less brightly colored.
 Diffraction
DIFFRACTION
refers to what happens to a wave when it hits an obstacle.
 Huygens’ Principle

The key to understanding diffraction is a very simple observation first due to Huygens
in 1678. Say a wave arrives at an opaque screen with a little hole in it. On the other
side of the screen, the wave equation must still be satisfied with boundary conditions
given by the motion of the wave in the hole. In other words, the solution is identical to
a situation where there was a source in the hole. The wavefront diagram looks like
this:
 Huygens’ Principle

The great thing about this way of thinking about diffraction is that, since the wave
equation is linear, you can use this trick for any number of holes. You simply add the
amplitude for the waves produced from a “source” at each hole:

Calculating the amplitude by adding point sources in this way is known as Huygens’
principle. Huygen’s principle works even if the holes are very close together. In fact, it
works if they are connected, so instead of a hole, it’s a slit.
 Double hole diffraction

Using Huygens’ principle, we can easily calculate the diffraction pattern from a plane wave
passing through any number of holes. Say there are two holes separated by a distance d. The
solution will be as if there are N sources separated by a distance d.
Say one source has field E = E0ejwte-jkr. The diagram for two sources separated by a distance d is:

P At a distant point P at an angle to the sources

r1 the field is:
S1
d q r2 𝐸0 = 𝐸* 𝑒 12" 𝑒 31𝒌𝒓 1 + 𝑒 316
q
S2 '78
Where. Δ = 𝑠𝑖𝑛𝜃 + 𝜑
dsinq 9

with j the difference in phase between the two sources. This phase difference will be j = 0 if
the sources are illuminated by a plane coming perpendicular to the separation between the
sources. If the wave comes at an angle, j ¹ 0.
 Single slit diffraction
Let’s now consider a slit of width a and light of wavelength l. When the light impinges on the
slit, the pattern of the resulting wave can be calculated using the Huygens’ principle:

“All points on a wave front can be treated

as point sources of secondary waves with speed, frequency,
and phase equal to the initial wave”

So what happens when a wave hits a slit aperture? It depends on its size!

The transmined wave is concentrated in the

forward direcBon, and at near distances the
λ << a
wave fronts have the shape of the aperture.
The wave eventually spreads out.

The transmitted wave behaves similar to a

λ >> a
wave from a point source.
This effect is called diffraction.
 Single slit diffraction

Where are the minima in the diffraction pattern?

The first minimum is at an angle such that the

P light from the top and the middle of the slit
destructively interfere.

q y
a
a l
L a/2
q d = sin q min =
d 2 2
d
l
sin q min =
a

The location of the m-th minimum is:

ml
sin q min,m =
Screen a
 Single slit diffrac+on
To analyze diffraction, we treat it as interference of light from many sources. In other words, we
can model the slit as N point sources with spacing of d/N and then we can let N go to infinity.

The phase difference b between the first and last source is given by b/2p = dsinq/l » dq⁄l.

q
If we consider the phasor for each wavelet we can draw:
d
Fa is the phase shift
Aslit for the single source
b » Nfa P

Asource L>>d
(rays are parallel)
DestrucBve interference occurs when the polygon closes, i.e. b=2p

d sin q l
b 2p = 1 = q»
l d
fa
 Single slit diffraction
Now if we let N à¥, the polygon becomes the arc of a circle, whose radius is
determined by the relation between angle and arc length:

Aslit
= R sin ( b 2 ) Aslit
2 b 2 Asource
R
sin ( b 2 ) 2p
if R = Asource b then Aslit = Asource b»
l
dq
b 2

2
2 é sin ( b 2 ) ù
I P = A = I0 ê ú
ë b 2 û
 N-slit interference
To obtain the expression of the intensity for the radia;on from N slits we can combine

2
é sin ( Nf 2 ) ù
Mul;-slit Interference: I N = I1 ê ú
ë sin ( )û
f 2
and
2
é sin ( b 2 ) ù
Single-slit Diffrac;on: I1 = I 0 ê ú
ë b 2 û
to obtain
2 2
é sin ( Nf 2 ) ù é sin ( b 2 ) ù
Total Interference PaCern: I = I 0 ê ú ê ú
ë sin ( f 2 ) û ë b 2 û

Where:
φ/2π = δa/λ = (a sinθ)/λ ≈ aθ/λ
β/2π = δd/λ = (d sinθ)/λ ≈ d θ/λ
φ= phase between adjacent slits
β = phase across one slit
 Diffraction Gratings
Examples around us.
CD – grooves spaced by wavelength of visible light.

The color of some butterfly wings and reflection from other animals.
They are not pigmented! The colors come from interference of the reflected light from the
pattern of scales on the wings or shells – a grating with spacing of order the wavelength of
visible light!
 Diffraction Gratings

eye

Light graBng
source

Figure shows examples of atomic

spectra for the Sun, Hydrogen,
Helium, Mercury and Uranium.
 Diffraction Gratings
2
é sin ( Nf 2 ) ù
I N = I1 ê ú
ë sin ( f 2 ) û
The slit spacing determines the location of the peaks (NOTE: d is the inverse of the slit
spacing, i.e., d = periodicity!)

The angular dispersing power q ( l ) of the grating can be then inferred from the
grating formula.

𝑑𝑠𝑖𝑛𝜃: = 𝑚𝜆, 𝑚 = 0, ±1, ±2, ±3, …

Note that the positions of the principal interference maxima are the same for any
number of slits!

The number of slits/beam size determines the width of the peaks (narrower peaks
easier to resolve).
 Diffraction Gratings
We can also derive a generic equation for the diffraction grating that also takes into account
incident angles different from normal incidence.
The generalized grating formula then becomes:
𝑑 𝑠𝑖𝑛𝜃!; + 𝑠𝑖𝑛𝜃: = 𝑚𝜆, 𝑚 = 0, ±1, ±2, ±3, …
2nd order
1st order
𝜃'
0th order
𝜃-
𝜃*
𝜃!;
𝜆

𝑑
 Diffraction Gratings

Diffraction gratings rely on N-slit interference. They consist of a large number of evenly
spaced parallel slits.

How effective are diffraction gratings at resolving closely-spaced ‘spectral lines’? Are
these two lines distinguishable using a particular grating?

Rayleigh’s criterion
The minimum wavelength separation a grating can resolve occurs
when the λ2 peak coincides with the first zero of the λ1 peak

So, the Rayleigh criterion can be written as:

l
D ( sin q )min =
Nd
ml
The location of the peak is sin q =
d
d l D (l ) 1
Therefore D ( l ) = D ( sin q )min = = 58
min m mN l mN