Why One Experiment Forced Physics to Accept That Nature Is Fundamentally Digital

In 1922, Otto Stern and Walther Gerlach fired silver atoms through a magnet. They expected a smear. Nature gave them two dots.

That single result — two clean bands instead of one fuzzy streak — rewrote physics. It proved that angular momentum doesn’t just come in discrete magnitudes (as Bohr had already suggested). It comes in discrete orientations in space. You can’t point a magnet at just any angle. You can only point it up or down.

The apparatus they built was brutally simple: a hot oven, a narrow slit, a custom magnet with a sharpened pole edge, and a glass plate. No lasers. No cryogenics. No computer. Just a 1,000°C furnace and a question: do atoms have a magnetic direction?

The Experiment That Classical Physics Couldn’t Explain

The classical prediction was unambiguous. Silver atoms have one unpaired electron, which gives each atom a magnetic moment. In a classical world, that moment can point in any direction. As atoms pass through the magnet’s field, each one gets pushed by an amount proportional to its orientation. Random orientations produce a continuous smear on the detector.

Stern and Gerlach ran the experiment expecting to confirm this. Instead, they found two sharp, separate bands. Half the atoms deflected up. Half deflected down. Nothing in between. No continuum. Just two states.

The only way to explain it was to assume that angular momentum — and therefore magnetic moment — is quantized. Not just in size, but in its projection onto any axis you choose to measure. This became known as spatial quantization.

What “Spin” Actually Means (It’s Not Spinning)

The result was already strange, but it got stranger. In 1925, two years after the experiment, George Uhlenbeck and Samuel Goudsmit proposed that electrons have an intrinsic angular momentum they called spin — with quantum number s = ½.

The name is unfortunate. Electrons are not tiny spinning tops. They are point-like particles with no measurable radius. Yet they carry angular momentum as surely as a rotating planet, and they carry magnetic moment as surely as a current loop. The math works. The physical picture is simply not classical.

For silver atoms, the 46 electrons in filled inner shells pair up with opposite spins, canceling completely. The 47th electron does not. Its spin — its intrinsic angular momentum — is the only thing determining whether the atom deflects up or down in the magnet.

The magnetic moment follows the formula:

μ = g · μ_B · m_s

where μ_B is the Bohr magneton (a fundamental constant), g ≈ 2 for electrons, and m_s = ±½ are the only two allowed values. Those ±½ values are the entire explanation for the two bands.

Why Two Bands Instead of Three or Five?

The number of bands equals 2s + 1, where s is the spin quantum number. For spin-½ particles like electrons and silver atoms, 2(½) + 1 = 2. For spin-1 particles, you’d get three bands. For spin-3/2, four bands, and so on.

When you rotate the magnet 90 degrees — measuring spin along a different axis — you still get two bands. The quantization axis changes, but the spin itself doesn’t. This was deeply unsettling to physicists in 1922: it implied that the property itself is binary, regardless of how you look at it.

The 50-50 Split and What It Means for Randomness

One of the most striking features of the result is how clean the split is. Roughly equal numbers of atoms go up and down, no matter how carefully you prepare the beam. This is not an artifact of the apparatus or the temperature. It is fundamental.

Quantum mechanics interprets this as the spin state being random before measurement. The atoms are not secretly all spin-up or all spin-down; they exist in a superposition of both states until the measurement forces them to pick one. The 50-50 statistics emerge naturally from the mathematics of superposition.

This randomness is not due to ignorance. It is not that we don’t know the spin — it genuinely does not have a definite value until measured. This is what Einstein famously objected to (“God does not play dice”) and what decades of subsequent experiments have repeatedly confirmed.

The Quantum Measurement Problem, Hiding in Plain Sight

The Stern-Gerlach experiment is the clearest real-world demonstration of the quantum measurement problem. Before measurement: a beam of atoms with no definite spin orientation. After measurement: two discrete spots, each corresponding to a definite state.

What happens in between? The theory doesn’t say — it only predicts the statistics. This gap between the smooth evolution of the wave function and the abrupt “collapse” at measurement is still debated today, nearly a century later.

In modern quantum computing, this measurement process is the readout. A qubit — the quantum equivalent of a classical bit — is typically a spin-½ particle (often a trapped ion or a superconducting circuit). Reading its value means sending it through a Stern-Gerlach-like field and detecting which path it takes. The two-state system that Stern and Gerlach stumbled onto in 1922 is now the foundation of machines that might reshape computing.

Modern Applications: From NMR to Quantum Computers

Magnetic Resonance Imaging (MRI): The principle of manipulating spin states with external fields underlies all of nuclear magnetic resonance. Your doctor can see inside your body because hydrogen nuclei (protons, spin-½) align in a strong magnetic field, absorb a radio pulse, and emit a signal that varies with their chemical environment. Stern and Gerlach started this.

Atomic Clocks: The most precise timekeeping devices on Earth rely on selecting and measuring specific spin states of atoms. The definition of the second is now tied to the hyperfine transition frequency of cesium-133 — a spin property.

Quantum Computing: Google’s Sycamore processor and IBM’s quantum machines use superconducting qubits that are, at the physical level, devices that behave like spin-½ particles. State preparation and measurement in these systems are direct descendants of what Stern and Gerlach did in Frankfurt in 1922.

Spintronics: Modern hard drive read heads use spin-dependent tunneling — the phenomenon that electrons tunnel through barriers at rates that depend on their spin orientation. This technology, which stores petabytes of data worldwide, is pure Stern-Gerlach physics.

Otto Stern’s Quiet Nobel

Otto Stern received the Nobel Prize in Physics in 1943 — twenty-one years after the experiment. The award was partly for this work and partly for his measurement of the proton’s magnetic moment. He was the first to prove, experimentally, that quantum mechanics was right about spatial quantization.

The prize came late and under dark circumstances. By then, Stern had fled Nazi Germany (he was Jewish) and was working in the United States. He had left behind his laboratory, his students, and much of his career. But the experiment survived. Two dots on a glass plate, fired from a furnace in 1922, that still light the way toward understanding matter at its most fundamental level.

See It for Yourself

The Stern-Gerlach Experiment visualization on ElysiaTools lets you run the experiment interactively. Adjust the magnetic field gradient, change the oven temperature, watch how the beam velocity affects separation distance, and compare the classical prediction (continuous smear) against the quantum result (two sharp bands) side by side. The visualization uses the same physics equations Stern and Gerlach solved by hand nearly a century ago.

The apparatus has changed. The lesson hasn’t: nature is not analog. It comes in pieces, and sometimes those pieces split in two and go their separate ways.