14
Part Zero · Foundations

Classical waves and optics

Hold two polarizing filters, the kind in good sunglasses, one behind the other, and turn the back one until they are crossed. No light gets through. Now slide a third filter between them, turned halfway. Light comes back. Adding a filter, something that can only remove light, lets more light through.

That small paradox is the best preparation for quantum measurement you can have, and it is pure classical wave physics. This chapter collects the wave ideas you will need: interference, diffraction, wave packets and polarization.

Two sources, one pattern

Waves add. Where two crests meet you get a bigger crest; where a crest meets a trough they cancel. The plate is a ripple tank with two sources pulsing in step, set 3.00 wavelengths apart. Calm lanes fan out where the waves always cancel.

Thomas Young showed light doing this in 1803: light through two slits makes bright and dark fringes, spaced by Δy=λL/d\Delta y = \lambda L/d. For green light, 550 nm, slits 0.25 mm apart and a screen a metre away, the fringes are 2.2 mm apart. That was the evidence that light is a wave. Established

Diffraction is a Fourier transform

Far from an opening, the pattern of light on a screen is the Fourier transform of the opening’s shape (Chapter 6). A single slit gives a broad central peak with weak side lobes. Many evenly spaced slits, a grating, give sharp bright lines. Use 2 slits.

More slits make the bright lines narrower but not further apart: their positions depend only on the spacing, dsin⁡θ=mλd\sin\theta = m\lambda. That is why a grating spreads white light into a sharp spectrum, and why spectroscopy, which reads the composition of stars, is possible.

DeeperFor NN slits of width aa and spacing dd, with β=πasin⁡θ/λ\beta = \pi a\sin\theta/\lambda and γ=πdsin⁡θ/λ\gamma = \pi d\sin\theta/\lambda, the intensity is I∝(sin⁡ββ)2(sin⁡Nγsin⁡γ)2I \propto \left(\frac{\sin\beta}{\beta}\right)^2\left(\frac{\sin N\gamma}{\sin\gamma}\right)^2. The first factor is the transform of one slit, the second of the comb of slits.

The spot that should not exist

In 1818 Augustin Fresnel entered a prize competition with his wave theory of light. Siméon Poisson, a judge, pointed out what he thought was an absurd consequence: the shadow of a round disc should have a bright spot at its very centre, because waves bending round the edge all arrive there in step. Dominique François Arago did the experiment and found the spot. Fresnel won. Established

The plate shows the shadow of a disc, with the spot at its heart.

Phase velocity and group velocity

A single pure wave has no beginning or end, so it cannot carry a message. A message rides on a wave packet, a bump made of many frequencies. The crests inside move at the phase velocity ω/k\omega/k; the bump itself moves at the group velocity dω/dkd\omega/dk. They differ whenever waves of different frequencies travel at different speeds.

On deep water, ω=gk\omega = \sqrt{gk}, so the group moves at half the phase speed: watch crests appear at the back of the packet, run forward and vanish at the front. For a free quantum particle the relation is ω∝k2\omega \propto k^2, the group moves at twice the phase speed, and the packet slowly spreads.

Polarization and the three filters

Light is a wave whose electric field wiggles across its direction of travel. A polarizer passes only the part of the field along its axis. If the light is already polarized at angle θ\theta to the axis, the passing field is reduced by cos⁡θ\cos\theta, and the intensity by cos⁡2θ\cos^2\theta. That is Malus’s law, found by Étienne-Louis Malus in 1809. Established

Now the paradox. Unpolarized light through the first filter keeps half its intensity. A second filter at 90° passes cos⁡290°=0\cos^2 90° = 0. But put a middle filter at 45 ° and the light that reaches the end is 12cos⁡2m cos⁡2(90°−m)\tfrac12\cos^2 m\,\cos^2(90°-m), which is 12.5%. The middle filter does not let light “sneak through”; it re-prepares the light in a new direction, from which the last filter takes a share.

Replace intensities by probabilities for single photons and this is exactly the behaviour of quantum spin in a Stern–Gerlach experiment, which you will meet in Part One.

What you now have

Superposition, interference, diffraction as a Fourier transform, phase and group velocity, and polarization with Malus’s law.

· · ·

Three questions to carry with you.

Why do more slits in a grating make sharper lines?tap
The positions of the bright lines depend only on the slit spacing, but each line’s width shrinks as 1/N because more waves must all agree for full brightness.
How can a third polarizer increase the light through two crossed ones?tap
It re-prepares the light at an intermediate angle, so each filter passes a fraction instead of the last one passing nothing.
Which velocity carries a signal?tap
The group velocity dω/dk, the speed of the wave packet’s envelope.

And one to work out. Polarized light meets a filter at 60° to its polarization. What fraction of the intensity passes?

fraction =

Sources: E. Hecht, Optics. T. Young, Phil. Trans. R. Soc. 94, 1 (1804). A. Fresnel, Mémoire sur la diffraction de la lumière (1819). E.-L. Malus (1809). Checked by code: three-polarizer transmission 1/8 at 45°; Young fringe spacing 2.2 mm.

Next · Chapter 15
Electromagnetism I: Maxwell's equations

Four equations explain every phone, motor and MRI scanner.

Turn the page