The Photoelectric Effect: How Light Turned Out to Be Countable
There is a version of physics history in which quantum mechanics begins with something exotic. It does not. It begins with a metal plate, a lamp, and a stubborn experimental fact: dim blue light frees electrons while brilliant red light does not. Every sensible expectation says brighter should mean more energetic. It does not, and following that failure carefully leads directly to the photon - and to the photomultiplier tubes that make every neutrino observatory on this site possible.
The Observation That Did Not Fit
Heinrich Hertz noticed in 1887 that a spark gap discharged more readily when illuminated with ultraviolet light. Philipp Lenard studied it properly by 1902 and found something odd. Increasing the light's intensity produced more electrons, as expected, but did not increase their energy. Changing the light's colour did change their energy - and below a certain frequency, no electrons appeared at all, however intense the beam.
In the wave picture of light, energy is spread continuously across the wavefront and an electron should be able to accumulate it gradually. Bright enough light of any colour should eventually free electrons, after a delay if necessary. Neither the threshold nor the absence of a delay makes sense in that framework.
The failure is specific and unavoidable. It is not that the classical prediction was slightly off; it is qualitatively the wrong shape. That is what makes the photoelectric effect a genuinely decisive experiment rather than a matter of improved precision.
The threshold turns out to be a property of the metal. Each material has a work function - the minimum energy needed to remove an electron from its surface - and light below the corresponding frequency simply cannot supply it, no matter how much of it arrives.
Einstein's Explanation
In 1905 Einstein proposed that light energy is not spread continuously but arrives in discrete packets, each carrying energy proportional to the frequency. An electron absorbs one packet or none. If that packet carries more than the work function, the electron escapes with the surplus as kinetic energy. If it carries less, nothing happens, and adding more packets does not help because they are absorbed one at a time.
Every feature of the observation follows immediately. The threshold is where a single quantum equals the work function. Intensity controls how many quanta arrive and therefore how many electrons come out, but not their individual energy. And there is no delay, because absorption is a single event rather than a gradual accumulation.
The proposal was radical enough that even Max Planck, whose own work had introduced quantisation to describe thermal radiation, was reluctant to accept light itself as quantised. Robert Millikan set out to disprove it experimentally and spent about a decade on increasingly precise measurements, which confirmed Einstein's relation exactly and produced a good value for Planck's constant along the way.
Einstein received the 1921 Nobel Prize in Physics specifically for this work, not for relativity. The citation reflects how central the result was to establishing that energy at small scales is discrete.
Where It Shows Up in Practice
A photomultiplier tube is the photoelectric effect turned into an instrument. A photon strikes a photocathode and frees a single electron, which is then accelerated into a series of electrodes, each releasing several more, until a measurable pulse emerges. This is why detectors like Super-Kamiokande and IceCube can register individual photons of Cherenkov light.
Solar cells use a closely related process. The distinction is that the photoelectric effect proper ejects an electron from a surface into vacuum, while in a semiconductor the photon promotes an electron across the band gap without it leaving the material. The second is called the photovoltaic or internal photoelectric effect, and the quantum reasoning is the same.
Digital image sensors, night vision tubes and light meters all rest on the same principle. It is one of the few results in fundamental physics that translated into everyday technology almost immediately and has stayed there.
The threshold behaviour also matters practically. A photodetector's material determines which wavelengths it can see at all, which is why detector design starts with matching the photocathode to the light the experiment expects.
What the Effect Does and Does Not Establish
It establishes that light delivers energy in countable units and that the energy per unit depends on frequency alone. That is a statement about how energy transfers, and it is exact.
It does not establish that light is only a particle. Interference and diffraction remain equally real, and the modern picture treats light as a quantum field whose behaviour is wave-like in propagation and particle-like in absorption. Presenting the photoelectric effect as having proved light is made of tiny balls oversimplifies it.
It also does not imply that any illuminated surface produces useful power. The effect describes the mechanism of energy transfer, not its magnitude. Whether a given surface and a given light source yield anything worth harvesting is a separate engineering question with a quantitative answer.
That distinction is worth carrying into any discussion of ambient energy. A mechanism existing is not the same as a mechanism being useful at a given scale, and the neutrinovoltaic research question - whether ambient radiation and thermal fluctuations at an engineered material surface can drive a measurable current - is precisely a question about magnitude rather than about whether energy transfer is possible in principle.
Frequently asked questions
What exactly is the photoelectric effect?
The ejection of electrons from a material when light strikes it. The defining feature is that it only happens above a threshold frequency, and the electrons' energy depends on the light's frequency rather than its intensity.
Why couldn't classical physics explain it?
In the wave picture, energy arrives continuously, so bright enough light of any colour should eventually free electrons. The observed frequency threshold and the absence of any delay contradict that qualitatively, not just numerically.
What did Einstein actually propose?
That light energy arrives in discrete quanta with energy proportional to frequency, absorbed one at a time. That single assumption reproduces the threshold, the intensity behaviour and the instantaneous response.
Is this what Einstein won the Nobel Prize for?
Yes. The 1921 Prize cites his work on the photoelectric effect, not relativity, which reflects how decisive it was in establishing that energy is discrete at small scales.
Does it prove light is a particle rather than a wave?
No. It shows that light transfers energy in countable units. Interference and diffraction remain real, and the modern description is a quantum field that propagates like a wave and is absorbed like a particle.
