Polarisation of EM waves
Linear and circular polarisation
• We have seen in an EM wave the electric field (E)
is the most important and this defines the
polarisation of the wave
• E and B fields are both travelling waves, each
oscillating in perpendicular planes to each other
and in phase with each other
E
B
S
An observer looking along Poynting’s Vector (S)
would see E and B varying in time
Time
The E and B fields are both maximal then both zero
simultaneously.
• In general E can be at any position perpendicular
to the direction of travel the wave, then B is
perpendicular to both.
• If the EM wave travels though a medium that only
allows E to oscillate in one direction, that is a
polarising material (e.g. Polaroid). In (and after)
the material the E of all of the light is aligned in
one direction
This is known as linearly polarised light
Recall that a vector can be resolved into perpendicular
components in any basis. Hence the E field in the wave
can be thought of as the sum of two different E fields that
are perpendicular
E
E1 E2
This means the EM wave can also be considered as
being the sum of two in-phase waves with different
polarisations
Optically active materials
In some special materials the speed of light is
different along different directions of travel in the crystal
lattice. This means the two perpendicular components of
E travel at different speeds, so will become out of phase.
NOTE: In Polaroid one component is completely
absorbed, so produces only linearly polarised light.
Consider what happens if we make a thin slab of such
an optically active material exactly thick enough that one
component of the light wave lags the other by ¼
wavelengths when the light exits from the slab.
Such a slab is a ¼ wave plate
After passing though a ¼ wave plate: when one E
component is maximal the other perpendicular to it
zero and vice versa
Circularly Polarised light
Time
E1
E
2
As time goes on for a wave passing after passing
though a ¼ wave plate the two components of the E field
will be oscillating as above. The resulting wave made
by combining the two vectors is rotating in a circle.
This is known as circularly polarised light
Depending on the relative phase of the two components
the E vector can rotate in either direction known as left-
or right-circularly polarised light.
It is known by experiment (R.A. Beth, 1935) that
circularly polarised light carries angular momentum
(as well as energy and linear momentum p = E/c)
If the two amplitude of the two waves are unequal the
light is elliptically polarised.
Polarisation of photons
At the quantum level light is made of photons, photons
have an intrinsic spin of 1, which means an angular
momentum of .
Angular momentum is a vector quantity.
In a photon the spin can point either along the
direction of travel or against it. No other choice is
allowed as the photon is massless.
In a beam of linearly polarised light the photons are
an equal mixture of left- and right-polarised. There is
no net angular momentum carried by this type of
polarised light.
In circularly polarised light the angular momentum of
each photon points in one direction only.
In optics the name left-circularly polarised light is
given to light with an apparent anti-clockwise sweep of
the E vector when looking towards the source, the
photon spins point in the direction of motion.
In right-circularly polarised light the E vector
apparently rotates clockwise looking towards the
source, and the photon spins point in the opposite to
the direction of motion.
Notes: In particle physics the naming convention is
reversed! A left-handed particle has its spin in the
opposite direction to the direction of motion.
Radio engineers also look away from the source when
defining clockwise and anti-clockwise, so reverse the
definitions!
[See HECHT section 8.1.5]
The importance of this is that one can make beams of
photons with all of the spins pointing in one direction.
A Y3 Physics Lab experiment (Optical pumping) uses
this technique.
Polarisation by reflection
When light is reflected from a dielectric surface we
want to look at the polarisation of the reflected and
transmitted waves.
See Hecht section 4.3.2
Assume B is parallel to the plane of incidence.
Take light reflecting from glass. n=1 in the air, take ni
as the general case and nt in the glass.
Assume linearly polarised light with:
B fields all point into the page, parallel to the
surface.
E fields are perpendicular to the B field and the
direction of the wave.
k is the direction of travel of the waves (also
direction of Poynting vector)
At the surface the electric field is continuous (from
[4.1]), all B fields are parallel to the surface:
Bi + Br = Bt [6.14]
The E field is also continuous so looking parallel to the
surface:
− Ei cos θ i + Er cos θ r = − Et cos θt [6.15]
Remember that E=B(c/n) from Maxwell’s equations
(assuming the relative permeability of both materials are
~ 1) from [6.14]:
ni Ei + nr Er = nt Et [6.16]
Also using θi = θr with [6.15] and [6.16] we get:
E n cosθ i − ni cosθ t
R= r = t [6.17]
Ei nt cosθ i + ni cosθ t
This is the amplitude reflection coefficient for the
case that E is in the incident plane.
Note: the amplitude of the reflected wave will be zero
when nt cos θi = ni cos θt
n sin θ i
Using Snell’s law : t = , we can obtain the
ni sin θ t
condition for no reflection just in terms of the angles:
sin θ i cos θ i = cos θ t sin θ t [6.18]
This has the solution : θ i + θ t = π / 2
This is the Brewster condition
Note the reflection
vanishes when its
Poynting vector would be
parallel to the electric
field of the transmitted
wave.
Brewster condition
When a wave is reflected the reflected wave is caused
by re-radiation from the wave in the other material.
When the E field in the other material is parallel to the
direction the reflected wave would propagate in the
amplitude goes to zero ( E ⋅ S = 0 ).
Using Snell’s law again we can show the Brewster
Angle, the angle for which there is no reflection for light
polarised perpendicular to the plane of the surface:
n
tan θ B = t
ni
Remember energy transmitted is (1-R2) = 1 at the
Brewster angle, so the transmission efficiency is 100%.
Plot of R against θ for perpendicularly
polarised light:
1 1
0.5
r( θ ) 0
0.5
1 1
0 θ π
2
n − ni
At normal incidence θi = 0, R = t , at glancing
nt + ni
incidence θi = π/2, R = −1
Case of E transverse to plane of reflection
In the other orientation, with the E vector parallel to
the surface of the material, called the transverse
polarisation, we find another amplitude coefficient. See
for example Hecht section 4.3.2:
n cosθ i − nt cosθ t
R⊥ = i
ni cosθ i + nt cosθ t
NOTE: ni and nt have been reversed compared to the
other polarisation [6.17].
This reflected amplitude coefficient is always
negative, so R⊥ > 0 . So waves reflected at the
Brewster angle are 100% transverse polarised.
In general light reflected from a dielectric surface (e.g.
glass or water) is partially polarised.