Shine light on certain metals and electrons pop off the surface — the photoelectric effect, observed since the 1880s. It should be a simple energy story: brighter light carries more energy, so brighter light should knock electrons off with more force. Turn up the brightness on the wrong colour of light, though, and nothing happens at all, no matter how intense the beam gets. Switch to a dimmer light of the right colour, and electrons leave immediately, with no delay for the light to "build up" enough energy. Classical physics, where light is a continuous wave, had no explanation for either result.
Three results a wave couldn't explain
A wave model of light predicts that a dim beam should still eventually eject electrons — it would just take longer, as the wave's energy slowly accumulates at the metal's surface. Experiments found the opposite: emission is instantaneous, with no measurable lag even at very low intensity. A wave model also predicts that any colour of light, given enough brightness, should eventually knock electrons loose. Instead, each metal has a sharp threshold frequency — below it, electrons never come off, regardless of how bright the light is. And a wave model predicts that a brighter beam should eject electrons with more kinetic energy. Instead, brightness only changes how many electrons come off; their individual energy depends entirely on the light's frequency, not its intensity.
Einstein's fix: treat light as a stream of countable packets
In 1905, Einstein proposed that light itself is quantised — delivered not as a continuous wave but as discrete packets, later called photons, each carrying an energy proportional to its frequency. A single photon strikes a single electron and transfers all of its energy at once. If that energy exceeds the amount needed to free the electron from the metal — the metal's work function — the electron leaves immediately, carrying off whatever energy is left over. Below the threshold frequency, no single photon carries enough energy to free an electron, no matter how many photons arrive per second. That explains the instant emission, the sharp frequency threshold, and why more light means more ejected electrons rather than more energetic ones: more photons, each still delivering the same fixed energy per hit. Robert Millikan, initially sceptical, spent nearly a decade testing the prediction and confirmed it precisely in 1916. Einstein won the 1921 Nobel Prize in Physics specifically for this explanation — not, as often assumed, for relativity.
What we're still unsure about
The photoelectric effect settled that light behaves like discrete particles in this situation. Other experiments — most famously the double-slit — show light behaving like a continuous wave, producing interference patterns a stream of simple particles shouldn't create. Quantum mechanics reconciles both results mathematically, treating light as neither a classical wave nor a classical particle but something whose behaviour is only fully captured by probability amplitudes. That machinery predicts experimental outcomes with extraordinary precision. What it doesn't settle — and physicists still openly disagree about — is what to say light actually is between measurements, which is why multiple competing interpretations of quantum mechanics still coexist, a century later.
This sits inside Photoelectric Effect & Wave-Particle Duality, one of seven topics in Modern Physics, one of five domains in Physics, one of seventeen subjects the app can quiz you on.