Topic 6 of 6
Polarisation and filter axes
A polarising filter selects a transverse oscillation direction. To calculate what passes through, use the angle between the current incident polarisation and that filter's transmission axis.
Why the transverse direction matters
A transverse wave can oscillate in different directions perpendicular to propagation. Polarisation selects a direction within that transverse plane. For plane-polarised light, the electric field oscillates along one fixed transverse direction.
In a longitudinal wave, particle oscillation is along propagation. There is no equivalent choice of transverse vibration direction for a polarising filter to select. Polarisation is therefore a phenomenon associated with transverse waves.
Project the electric-field amplitude
Let the incident plane-polarised electric-field amplitude be E0. An ideal filter transmits its component along the filter axis. If θ is the angle between the incident polarisation and that axis:
For angles from 0° to 90°, the amplitude ratio is simply cos θ. The emerging polarisation lies along the filter axis. The angle is measured between two directions in the transverse plane, not between the ray and the filter surface.
At the same frequency and in the same medium, intensity is proportional to electric-field amplitude squared. Squaring the amplitude ratio gives Malus' law:
I0 is the intensity incident on this filter, not necessarily the original source intensity if earlier filters are present.
Worked single ideal filter
Amplitude and intensity have different ratios
Use plane-polarised input intensity I0 = 80 W/m2. At θ = 30°:
I = 80 cos2 30° = 80 × 3/4
= 60 W/m2
Transmit the electric-field component along the filter axis
This is a head-on transverse view: the page is perpendicular to propagation. The incident electric-field direction is vertical. The brown dashed line is the filter's transmission axis, 30° from that direction.
From a plane-polarised input of 80 W/m2, I = 80 cos230° = 60 W/m2. The emerging polarisation lies along the filter axis. The 30° angle is between field direction and axis, not between a ray and a surface.
| Angle / ° | Amplitude ratio | Transmitted intensity / (W/m2) |
|---|---|---|
| 30 | 0.866 | 60 |
| 60 | 0.500 | 20 |
| 90 | 0 | 0 |
At 60°, amplitude halves but intensity becomes one quarter, 20 W/m2. At 90°, both vanish in the ideal model. Without a supplied E0, these data determine amplitude ratios, not an absolute electric-field amplitude.
Apply the law again after each filter
Start with the same plane-polarised 80 W/m2 input. The first filter axis is 45° to the original polarisation; the second is 90° to the original. After the first filter, the light is polarised at 45°, so the second filter's local angle is 45°.
Use the incoming polarisation at each filter
Read these successive head-on views from top to bottom. The named filter angles are measured from the original vertical direction. In each filter view, a thin grey dashed line marks its incoming polarisation; the brown line is its transmission axis. Green amplitude arrows share one scale.
The intensity sequence is 80 → 40 → 20 W/m2. A single ideal filter at 90° to the original input would give zero. With the intermediate filter present, the second filter receives a different polarisation direction and must be evaluated from that new incident state.
I2 = 40 cos2 45° = 20 W/m2
E1/E0 = 1/√2
E2/E0 = (1/√2)2 = 1/2
The final amplitude is half the original, so final intensity is one quarter of 80 W/m2. A single filter at 90° to the original input would transmit zero. The intermediate filter changes the polarisation reaching the final filter; the complete two-filter output still contains less intensity than the original input.
Optional check Plane-polarised light of intensity 80 W/m^2 passes through ideal filter axes at 45 degrees and then 90 degrees to its original polarisation. What intensity emerges from the second filter?
Distinguish an unpolarised input
Unpolarised light has no single fixed transverse electric-field direction. An ideal first polariser transmits half its time-averaged incident intensity and produces a plane-polarised output. Later filters use Malus' law on that output.
An already plane-polarised input is not automatically halved by every filter: its transmitted fraction depends on the angle. Real absorption and incomplete extinction are additional effects beyond the ideal filter calculations here.