THEORYMERIDIAN

Quantum Spin Lab · Stern-Gerlach

Why did the beam
split in two?

Frankfurt, 1922. Silver atoms fly through a magnet whose field is stronger on one side. Little magnets pointing every which way would spread into a smear. The atoms land in two bands instead.Try both: classical magnets, then silver atoms.

Ideal beam model
01 / Oven, slits, magnet, plate
Atoms in flightSilver on the plateCentre line, no deflection

Simulated from the modelEach atom is drawn at random: its speed from the oven, its magnetic moment from the model you chose. Paths in the picture are bent about a thousand times more than in reality; the plate inset has a true scale. The bench is an illustration, not a drawing of the 1922 apparatus.

Atoms landing in the middle

0%
Atoms on the plate
0
Upper band
—
Lower band
—
Typical deflection at the plate
—
What the plate shows
nothing yet

Start here: choose what flies through the magnet (sends 200 atoms)

Send some atoms and watch where they land.

01What the atoms are

In tesla per metre. About 1000 T/m was reached in 1922. A stronger gradient pushes the bands farther apart. Changing it clears the plate.

03Send atoms

Each atom leaves the oven with its own speed. Slow atoms spend longer in the magnet and are pushed farther.

One outer electron does it. In a silver atom the inner electrons cancel out. The magnetic moment is carried by the spin of one electron, and along the axis of the magnet that spin has only two possible answers. Two answers, two paths, two bands.

Real experiments

Two bands on glass,
and nothing between.

The split was seen before anyone had proposed electron spin. The explanation used today came a few years after the experiment.

Real experiment

Frankfurt, February 1922

Otto Stern and Walther Gerlach sent silver atoms from an oven near 1000 °C past a knife-edge pole 3.5 cm long and caught them on a glass plate at its end. With the field on, the deposit split into two bands a fraction of a millimetre apart, with almost no silver between them.

W. Gerlach and O. Stern, Z. Phys. 9, 349 (1922); English translation: arXiv:2301.11343.

History

The explanation came later

In 1922 the two bands were read as orbits that take only certain directions. In 1925 George Uhlenbeck and Samuel Goudsmit proposed that the electron itself carries angular momentum: spin. The ground state of silver has no orbital angular momentum, so the two bands are the two spin answers of its outer electron.

G. E. Uhlenbeck and S. Goudsmit, Naturwissenschaften 13, 953 (1925); B. Friedrich and D. Herschbach, Physics Today 56(12), 53 (2003).

Real experiment

Hydrogen, 1927

T. E. Phipps and J. B. Taylor repeated the experiment with hydrogen atoms: one electron, no orbital angular momentum in the ground state. The beam again split in two.

T. E. Phipps and J. B. Taylor, Phys. Rev. 29, 309 (1927).

The reasoning, in the open

Look under the model.

The calculation

A magnetic moment in a field that changes along z feels a force F = μz ∂B/∂z. An atom of mass m crossing a magnet of length L at speed v is pushed sideways by

z = μz (∂B/∂z) L² / (2 m v²)

at the end of the magnet, where the plate stood in 1922. Silver atoms: μz is +μB or −μB, with equal chance. Classical magnets: μz = μB cos θ with every direction equally likely, so cos θ is spread evenly from −1 to +1.

Atoms leave the oven with a spread of speeds, and the beam favours fast ones: on average m v² = 4kT. Slow atoms are deflected more, so each band has a tail on its outer side. With L = 3.5 cm, 1000 T/m and 1000 °C, an atom of average energy is displaced by 0.081 mm, which puts the bands about 0.16 mm apart. "In the middle" means within a fifth of that displacement of the centre line.

Read the numerical model · Validation record

Sources
Limits of this model

One field gradient for the whole beam. In the real magnet the gradient changed across the beam, which gave the deposit its lip shape. No slit width, no scattering, no background on the plate. The magnetic moment of the electron is taken as exactly one Bohr magneton. The classical magnets are the prediction that failed: nothing in nature behaves that way.

Keyboard, motion and privacy

Use Tab to reach each control, arrow keys for sliders, and Enter or Space for buttons. Pause freezes the illustration only. Motion starts paused when your system asks for reduced motion. Results are written as text. The page runs entirely in your browser with no accounts, trackers or external code. See the privacy notice.