Stern-Gerlach: The Experiment That Forced Spin Into Being

1922: a beam that should have smeared out

In Frankfurt in 1922, Otto Stern and Walther Gerlach sent a beam of neutral silver atoms, evaporated from an oven and collimated into a thin stream, through a strong, deliberately non-uniform magnetic field, then let them land on a glass detector plate. Silver's outer electron sits in a 5s orbital with zero orbital angular momentum, so any magnetic moment the atom carries comes entirely from that electron's intrinsic angular momentum — though the very concept of "spin" did not exist yet in 1922. Classically, if an atom carries a magnetic moment pointing in some direction, the force it feels in a field gradient depends continuously on how that moment happens to be oriented, so a beam of atoms with randomly oriented moments should simply smear into a continuous band on the screen, from maximum deflection one way to maximum deflection the other.

What Stern and Gerlach actually saw

Instead, the beam split cleanly into two discrete spots, with essentially nothing landing in between — direct evidence that the magnetic moment's component along the field axis takes only two possible values, not a continuum. This is space quantization: the atom's angular momentum, when measured along any chosen direction, is restricted to a discrete set of outcomes. At the time the result was interpreted as confirming the Bohr-Sommerfeld model's prediction of quantized orbital orientations; only after Samuel Goudsmit and George Uhlenbeck introduced electron spin in 1925 did physicists realize the doublet actually came from spin, not orbital motion — a famous case of a landmark experiment whose full explanation arrived after the experiment itself.

F_z = μ_z · (∂B_z / ∂z)      (force along the field-gradient axis)

μ_z = ± μ_B                   (only two allowed values, spin-1/2 valence electron)
μ_B = the Bohr magneton

The modern picture: measurement of a two-state system

In the modern picture, silver's magnetic moment comes from the outer 5s electron's spin, and quantum mechanically that spin's component along any chosen axis (call it z, set by the direction of the magnetic field) can only ever be measured as +ħ/2 or −ħ/2 — never anything in between. Before entering the apparatus, the atom's spin is in general a superposition of "up" and "down" along that axis; the inhomogeneous field correlates the spin degree of freedom with the atom's trajectory, deflecting up-spin atoms one way and down-spin atoms the other, and hitting the screen acts as a measurement that sorts every atom into exactly one of the two beams. It has become the standard textbook example for introducing quantum measurement and two-level systems.

Sequential Stern-Gerlach: incompatible observables

The classic teaching extension sends the atoms through a second magnet oriented along a different axis. Take a beam already filtered to "spin up along z" and pass it through a magnet oriented along x instead: it splits again, 50/50, as though the earlier z-measurement never happened. Filter that x-up beam again through a z-oriented magnet, and it splits 50/50 into up and down along z once more — the intermediate x-measurement has completely erased the earlier z-information. This clean demonstration of incompatible, non-commuting observables, without any of the wave-diffraction complications of a double-slit experiment, is now a standard first introduction to spin-1/2 qubits in every quantum mechanics and quantum computing course.

Why it mattered beyond spin

Stern was awarded the 1943 Nobel Prize in Physics largely for the Stern-Gerlach result and his broader development of the molecular-beam method, which became a foundational technique of atomic and molecular physics — later feeding directly into the development of the maser, the laser, and atomic clocks. Today the experiment's real legacy is pedagogical as much as historical: it is the cleanest physical system for introducing the formalism of quantum measurement, superposition and two-level systems that underlies the entire theory of qubits.

Frequently asked questions

Try it live

Everything above runs in your browser — open Stern-Gerlach Experiment and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.