21
Part One · Quantum mechanics

The crisis in classical physics

By classical physics, every atom in your body should have collapsed about a hundred-billionth of a second after it formed. In April 1900 Lord Kelvin told the Royal Institution that the theory of heat and light was beautiful and nearly complete, apart from two small “clouds” on the horizon. Within five years those clouds had become relativity and quantum mechanics.

Part Zero gave you the classical toolkit. This chapter gathers, in one place, the five places where it fails, and fails badly. Each failure is a clue, and each clue will be solved in the chapters that follow.

The dashboard

The plate shows four classical predictions, in grey, against what experiments actually found, in amber. Look at each in turn: , , and . Or .

In every panel the classical curve is not slightly off. It has the wrong shape.

Failure one: the glowing oven

Equipartition gives every mode of light in a hot box the same average energy kBTk_BT, and there are ever more modes at short wavelengths. The predicted brightness rises without limit towards the ultraviolet, the Rayleigh–Jeans law. Real ovens and stars glow with a peak and then fade. Chapter 22 shows how Max Planck fixed it in 1900 by letting energy come only in lumps hfhf. Established

Failure two: light that knocks out electrons

Shine light on a clean metal and electrons fly out. A wave carries energy in proportion to its brightness, so brighter light should throw electrons out faster. It does not. Brighter light throws out more electrons, at the same speeds; only changing the colour changes their speed, and below a certain frequency nothing comes out at all, however bright the light. Chapter 23 takes this apart. Established

Failure three: barcodes of light

Heat a gas or pass a spark through it and it glows, but not in a smooth rainbow. It emits a few sharp colours, a different barcode for each element. Hydrogen’s visible lines sit at 656, 486, 434 and 410 nm. A classical electron orbiting a nucleus could radiate at any frequency, smeared continuously as it spirals. Nothing in classical physics picks out particular colours. Chapter 24 decodes hydrogen’s barcode. Established

Failure four: atoms should not last

Chapter 16 showed that an orbiting electron radiates and spirals inward in about 1.6×10−111.6\times10^{-11} seconds. There is a deeper problem too. Try to build the size of an atom from the constants classical physics offers, the electron’s charge and mass and the speed of light. The only length they make is the classical electron radius, e2/4πε0mec2≈2.8×10−15e^2/4\pi\varepsilon_0 m_e c^2 \approx 2.8\times10^{-15} m, nearly twenty thousand times smaller than a real atom.

Add Planck’s constant and the right size appears at once: a0=4πε0ℏ2/mee2≈0.53×10−10a_0 = 4\pi\varepsilon_0\hbar^2/m_e e^2 \approx 0.53\times10^{-10} m. Atoms are as big as they are because of hh. The ratio of the two lengths is exactly 1/α2≈18,7791/\alpha^2 \approx 18{,}779, where α≈1/137\alpha \approx 1/137 is the fine-structure constant you will meet again.

Failure five: cold solids

Equipartition predicts that every solid needs the same heat per mole to warm by one degree, 3R≈24.93R \approx 24.9 J/(mol K): the Dulong–Petit law. Copper and lead obey it at room temperature. Diamond needs only about 6.1. And as any solid is cooled towards absolute zero, its heat capacity drops towards zero too. Classical physics has no way for a vibrating atom to refuse heat. Einstein’s quantized oscillators of Chapter 19 do. Established

What classical physics got right

It would be wrong to throw the old physics away. Conservation of energy and momentum survive untouched. Maxwell’s equations survive. Hamilton’s equations survive as the shape of quantum dynamics. What fails is the assumption that energy can be divided as finely as you like, and that a particle has a definite path. Quantum mechanics replaces those two assumptions and keeps nearly everything else.

· · ·

Three questions to carry with you.

What did classical physics predict for a hot oven’s ultraviolet glow?tap
An unlimited amount of energy: the Rayleigh–Jeans law diverges at short wavelengths.
In the photoelectric effect, what changes the electrons’ speed?tap
The frequency (colour) of the light, not its brightness.
Why can’t classical constants give the size of an atom?tap
They only make the classical electron radius, about 20,000 times too small; Planck’s constant is needed to get the Bohr radius.

And one to work out. Using 3R with R=8.314R = 8.314 J/(mol K), what heat capacity does Dulong–Petit predict per mole?

C =J/(mol K)

Sources: Lord Kelvin, Phil. Mag. 2, 1 (1901). A. Pais, Inward Bound. H. Kragh, Quantum Generations. NIST data for diamond (C_p ≈ 6.1 J/mol K at 298 K). Checked by code: a₀/r_e = 1/α² = 18,779; 3R = 24.94 J/(mol K).

Next · Chapter 22
Blackbody radiation and the quantum

The universe is the most perfect blackbody ever measured.

Turn the page