Skip to main content
Science

The Photon: Light's Strangest Particle — and Why It's Powering a Quantum Revolution

What is a photon, really? A clear guide to the quantum particle of light — its bizarre properties, whether you can 'split' one, and the photonic tech reshaping computing and security in 2026.

7 min read
Share:
A laser beam from an observatory telescope pointing up into the star-filled Milky Way
Image: ESO/Y. Beletsky (CC BY 4.0)

You are bathed in them right now. Every beam of sunlight, every glance at this screen, every signal racing through a fibre-optic cable is made of the same thing: photons — the elementary particles of light. They're among the most ordinary objects in the universe and among the strangest, and in 2026 they've quietly become the workhorses of a coming technological revolution. So let's answer it plainly: what is a photon — and why is everyone from Google to national labs racing to master it?

Editor's note: This is an evidence-based explainer. The physics is well established; the "frontier" sections describe active research and are sourced at the end.

What is a photon?

A photon is the quantum of the electromagnetic field — the smallest possible "packet" of light. And not just visible light: radio waves, microwaves, X-rays and gamma rays are all electromagnetic radiation, all made of photons, differing only in energy. The photon is also the force carrier of electromagnetism itself — the particle that, in effect, transmits the electric and magnetic forces between charged particles.

Its properties are genuinely bizarre:

  • It has no mass. A photon weighs nothing — yet it carries energy and even momentum. (That momentum is real: it's why light can nudge objects, the principle behind solar sails, and it shows up in the Compton effect.)
  • It only ever travels at one speed: the speed of light, c, about 300,000 km/s in a vacuum. A photon is never at rest. It is created moving at c and is absorbed still moving at c.
  • Its energy depends on its colour. A photon's energy is set by its frequency through Planck's relation, E = hf — higher frequency (bluer, then ultraviolet, then X-ray) means more energetic photons.
  • It's a boson, which means many photons can pile into the same state — the principle that makes lasers possible.

The concept was born in stages: Max Planck introduced energy "quanta" in 1900, Einstein used the photon idea to explain the photoelectric effect in 1905 (the work that won him the Nobel Prize), and the name "photon" was coined by chemist Gilbert Lewis in 1926.

Wave and particle — at the same time

Here's where intuition breaks down. Is light a wave or a stream of particles? The maddening answer from quantum mechanics is: both, depending on how you look.

  • Shine light through two narrow slits and it produces an interference pattern — unmistakably wave behaviour, even when you send the photons through one at a time.
  • Shine light on a metal and it knocks out electrons in a way that only makes sense if light arrives in discrete packets — the photoelectric effect, pure particle behaviour.

You can measure wave-like or particle-like properties, but never both in the same experiment. This wave–particle duality isn't a failure of our instruments; it's a fundamental feature of reality at the quantum scale, and the photon is its poster child.

Can you actually "cut a photon in two"?

You may have seen dramatic headlines about splitting a photon into two — or even into an "infinite" number of new ones. It's a real effect, dressed up in sensational language. Here's the honest version.

The process is called spontaneous parametric down-conversion (SPDC). Fire a high-energy photon into a special "nonlinear" crystal, and occasionally that photon is annihilated and replaced by two new, lower-energy photons. You're not slicing a particle in half — the original is destroyed and two fresh photons are created, with the books balanced exactly: the two daughters' energies and momenta add up to the parent's (so E = hf is conserved).

Diagram of spontaneous parametric down-conversion: a high-energy pump photon enters a nonlinear crystal and becomes two lower-energy entangled photons (signal and idler), with their energies summing to the original
Spontaneous parametric down-conversion: one 'pump' photon is annihilated and reborn as two lower-energy, entangled photons — energy and momentum perfectly conserved.

The "infinite photons" flourish comes from the deeper quantum bookkeeping: in principle the process has higher-order terms that can produce multi-photon states, so the full quantum description includes contributions from ever-larger photon numbers. In practice the two-photon case dominates — and it's the one that matters, because those two photons come out entangled. Measure one, and you instantly know something about its partner, no matter how far apart they are — Einstein's "spooky action at a distance." That entanglement, not the splitting, is the real prize.

Why photons run the quantum era

Photons turn out to be near-perfect carriers of quantum information — what physicists call "flying qubits." They move at light speed, they barely interact with their surroundings (so they don't easily lose their quantum state), and they can be entangled and sent down ordinary optical fibre. If you want to transmit quantum information, photons are the obvious messenger.

That's why the photon sits at the centre of three of the most consequential technologies in development:

1. Quantum photonic chips

The exotic optical benches of the past — lasers, crystals, mirrors filling a room — are being shrunk onto chips. Modern quantum photonic chips integrate waveguides, beam splitters, single-photon sources and detectors into millimetre-scale circuits, trading fragile bulk optics for the stability and reproducibility of semiconductor manufacturing. This is the path toward scalable photonic quantum computers.

2. Single-photon sources

Many applications need light delivered exactly one photon at a time, with every photon identical ("indistinguishable"). That's surprisingly hard. In 2025, researchers reported progress integrating deterministic, indistinguishable single-photon sources directly on-chip — including sources built from individual molecules and quantum dots — a key step toward practical quantum machines.

3. Squeezed light

By cleverly manipulating quantum noise, physicists can create "squeezed" light that beats normally unavoidable measurement limits. It already sharpens gravitational-wave detectors like LIGO, and 2025 work on wafer-scale squeezed-light chips points toward mass-producing it for quantum sensing and computing.

What it's all for

Strip away the jargon and the payoff is concrete:

  • Quantum computing — photonic qubits are one of the leading routes to machines that could crack problems classical computers can't.
  • Unbreakable communication — entangled photons underpin quantum key distribution, where any attempt to eavesdrop disturbs the photons and reveals the spy. It's the foundation of a future quantum internet.
  • Ultra-precise sensing — single photons and squeezed light enable measurements of time, gravity and distance at once-impossible precision.

The particle of light, in other words, is becoming the particle of information.

The bottom line

A photon has no mass, never slows down, behaves as both wave and particle, can be "split" into entangled twins, and — when handled with enough finesse — can carry information no eavesdropper can steal. It is at once the most familiar thing in your life and one of the deepest puzzles in physics. For a century we studied the photon to understand the universe. Now we're learning to engineer it — and in doing so, we may be building the backbone of the next era of computing, communication and discovery. Light, it turns out, was never just for seeing.


Sources

Share:
Stylized direct-air-capture towers pulling CO2 molecules from a dark sky

ScienceAnalysis

The Truth About CO₂ Emissions Nobody Wants to Hear

Carbon capture is sold as a machine that undoes pollution. An evidence-based look at what it can and can't do in 2026 — the brutal scale math, the real niche, and the danger of overselling it.

Jun 18, 20269 min