How does light slow down?

light shining through water.
Ever wondered how light slows down when passing through a material? (Image credit: Diane Keough via Getty Images)

The phenomenon of light slowing down as it passes through a material like glass or air is one of the most fascinating areas of physics, involving a complex interaction between light and materials. There are three ways to look at the same situation, and each employs a different understanding of physics. 

All of these explanations have strengths and weaknesses, but all of them are powerful tools for understanding this fascinating interaction.

 Related: How does astronomy use the electromagnetic spectrum? 

View No. 1: It's all waves 

The first perspective comes from James Clerk Maxwell, the 19th-century Scottish physicist and all-around genius who discovered a unified theory of electricity and magnetism, and also found that light is made of waves of electricity and magnetism.

 When these waves encounter a material like glass or water, they see a whole bunch of charged particles. The molecules in the material are made of atoms, which have protons and electrons — both charged particles. And charged particles respond to electromagnetic waves passing by them by wiggling along with them.

But moving charged particles also create electromagnetic waves of their own. The result is a giant mess, with the original electromagnetic waves interfering with all the waves generated by all the charged particles in the material (and there are a lot). Thankfully, most of those waves, except the waves traveling in the original direction of the light, cancel each other out. But because the waves generated by the particles are a little delayed, the entire ensemble moves more slowly.

The end result: The light moves more slowly.

View No. 2: It's all particles 

Light is made of tiny particles known as photons. (Image credit: DrPixel via Getty Images)

Maxwell's theory is a classical picture of radiation. Nowadays, we have a much more sophisticated view based on quantum mechanics, where light is made of countless tiny particles known as photons. Photons can act individually, but when enough of them get together, they display all of the same properties as electromagnetic waves.

A fully quantum treatment of photons interacting with materials can get pretty nasty, but thankfully, we have an approach developed by the famed physicist Richard Feynman to guide us through it. We can imagine all the photons of the incoming light slamming into the material. Once inside, they begin interacting with all the charged particles. 

Those charged particles can absorb those photons and emit their own, because that's what charged particles do. But these photons are a little different. In physics, they're known as virtual photons. 

Remember that photons can do two things: They can roam freely through the universe, existing as independent entities (this is light), and they do the legwork of mediating the electromagnetic force (like the force holding a magnet to a fridge). But these photons don't roam freely; they have a job to do. So we call them "virtual" photons — they exist only in our math to help us account for the electromagnetic force.

So all of these charged particles start emitting copious amounts of virtual particles, and once again, there's a giant, confusing mess. Feynman came to the rescue. He developed a technique of averaging out all of the possible paths that those photons can take. That averaging process eliminated all the wayward photons, leaving behind only the ones traveling in the original direction of the light. But all of those interactions come at a cost: It takes time for an electron to absorb and reemit a photon, and those delays add up.

The end result: The light moves more slowly.

View No. 3: It's all polaritons 

So far, we've focused on the properties of light, viewing it through a particle-based lens and a wave-based lens. But the material is more than a simple collection of charged particles that just do whatever they are electromagnetically ordered to do.

Materials can be interesting, too. Specifically, all materials can support vibrations — little ones, big ones, ones that last a long time, ones that fade away quickly. All material is constantly in motion, and that motion affects how that material interacts with everything else. To help physicists grapple with the complexities of all the kinds of vibrations that are constantly racing through materials, they proposed an entity known as a phonon.

A phonon is another kind of fake particle, but like virtual photons, it's very useful. It allows physicists to use the language of quantum mechanics to describe the vibrations in a material. This new language comes in handy when light, which is made of photons, enters that material.

When photons and phonons get together, they create something new: a polariton. In this view, when light enters a material, it disappears. And so do the phonons in the material itself. Instead, they get replaced by polaritons. These polaritons share a lot of properties with their parents, but they have one crucial property: They travel more slowly than the speed of light.

That speed depends on the properties of the material (the phonons). In this view, it's not light that's passing through a material, with the material responding to it, but a new object, a polariton, passing through. This view is especially useful, because in many situations, it's very easy to discard all the cumbersome math of conflicting waves or bouncing photons and just deal with a straightforward, simple entity that already encodes all the information you need.

Light goes in, a polariton travels through and light goes out.

The end result: The light moves more slowly.

Learn more by listening to the "Ask A Spaceman" podcast, available on iTunes and Ask your own question on Twitter using #AskASpaceman or by following Paul @PaulMattSutter and

Join our Space Forums to keep talking space on the latest missions, night sky and more! And if you have a news tip, correction or comment, let us know at:

Paul Sutter Contributor

Paul M. Sutter is an astrophysicist at SUNY Stony Brook and the Flatiron Institute in New York City. Paul received his PhD in Physics from the University of Illinois at Urbana-Champaign in 2011, and spent three years at the Paris Institute of Astrophysics, followed by a research fellowship in Trieste, Italy, His research focuses on many diverse topics, from the emptiest regions of the universe to the earliest moments of the Big Bang to the hunt for the first stars. As an "Agent to the Stars," Paul has passionately engaged the public in science outreach for several years. He is the host of the popular "Ask a Spaceman!" podcast, author of "Your Place in the Universe" and "How to Die in Space" and he frequently appears on TV — including on The Weather Channel, for which he serves as Official Space Specialist.

  • Atlan0001
    Then you have the classical, SPACETIME, otherwise also known as LIGHT-TIME or simply "histories"! In closed system, the ceiling horizon fastest speed there is! In open system, the floor horizon slowest speed there is! An objective real in space and time will always be out front in space and time of the SPACETIME (LIGHT-TIME) subjective relative.

    Going away from anywhere in SPACE and TIME, the objective real, and therefore its clock, will be farther out front, to ever far farther out front, in space and time from anywhere than its SPACETIME (LIGHT-TIME) subjective relative hologram.

    Oncoming to anywhere in SPACE and TIME, the objective real, and therefore its clock, will still be farther out front, to far farther out front, in space and time, than its SPACETIME (LIGHT-TIME) subjective relative hologram gradually speeding up in SPACETIME (in LIGHT-TIME) in closing upon the objectively real, itself closing upon points of objective reality anywhere at all. The triangle of three points (including the subjectively relative hologram and its subjectively relative clock and clock time) will close at the 0-point closure of the two objective reals . . . including the two objectively real clocks and their objectively real clocked times.

    A SPACETIME (LIGHT-TIME) holographic constant of speed that slows down in objective expansions going away from object points anywhere and everywhere . . . and equally but oppositely speeds up in objective contractions oncoming to object points anywhere and everywhere. But always until the very last instant of time, and then only in oncoming to the closing of objectively real points, so slow (in any open or opening system whatsoever) as to be unable to catch up to any objective real in SPACE and TIME until the meeting point of objective reals.

    An amazing thing to be universally the fastest speed there is in a closed or closing system yet universally be the slowest speed in any open or opening system . . . to be nothing -- both slowing down going away and speeding up oncoming -- but future-past // past-future "'histories' on the clock!"
  • 24launch
    What I'm curious about then is why does light bend or refract when passing through these materials? These 3 explanations for why it slows down don't address that phenomenon. I do have QED, by Richard Feynman and while I've made several attempts to read it, just didn't get terribly far. I'm betting he addresses it.
  • Atlan0001
    24launch said:
    What I'm curious about then is why does light bend or refract when passing through these materials? These 3 explanations for why it slows down don't address that phenomenon. I do have QED, by Richard Feynman and while I've made several attempts to read it, just didn't get terribly far. I'm betting he addresses it.
    A maze, a hall of mirrors. Measure the speed of light going in to be 300,000kps. Measure it to be coming out 300,000kps. And speed through any discrete distance inside the hall of mirrors to be the same. But take the total measured time through the measured space of the maze of the hall of mirrors and the result will be a light speed slower than the constant of the speed of light.

    In baseball, a fastball thrown at 100+ miles per hour appears to the batter to rise upwards on its way to the plate. One thrown at 90 miles per hour appears to come straight in to plate, while one thrown at 80 miles per hours appears to the batter to be dropping as it is coming into the plate. The speed of light mixing with the bend of gravity giving three different views to the batter of pitch bend in travel to the plate. The pitch is known to travel faster out of the hand of the pitcher but appears to the batter to arrive faster at the plate than when leaving the hand of the pitcher. The speed of light spent the distance to the plate trying to catch up to the speed of the ball, not catching up to the ball -- and merging with it -- until the very last instant of the ball's arrival at the plate.

    We talk about the measurement of the speed of light, when the speed of light follows the position and velocity of whatever the object and/or object event, never, ever, surpassing it in space and time. Only a past "history" surpassing. A rearview mirror on an auto might have the words "the following vehicle in this mirror is closer to you than it looks in this mirror" (it, the following vehicle -- you might call it a "future" now -- is outrunning the speed of light's transmission of that vehicle's positions and velocities, speeding only its "past" history). Relative to light at the speed of light (always measuring a "past" history), the distant objective reality itself (no matter how close or far in position it is and no matter its velocity), is always a "future" now (a "future" placement), again "at any distance", in space and time.

    Relativity, the base Theory of Relativity, does not divide into subjective pasts and objective futures. Whereas hyperspace theories make space pliable (shape shifting space), Relativity tries to make space hard (hard! space) and time differentially pliable (thus dealing in time distortions). The first has space subjectively pliable and time objectively rigid. The second, Relativity, has space objectively rigid and time subjectively pliable. The first deals in many fractal-zoom universes (or "many worlds," plural). The second deals in a naked singularity of universe 'Relativity'.