# Why does the speed of light change? [duplicate]

While researching the cause of refraction, I found that refraction occurs due to the change of speed of light when it goes from one medium to another (according to Huygens principle).

But I cannot get why actually does its speed change when the medium changes?

The speed of light in a medium is given by $$\frac{1}{\sqrt{\epsilon \mu}}$$, where $$\epsilon$$ and $$\mu$$ are the relative permittivity and permeability of the medium respectively (these in turn affect how large the electric & magnetic forces in the medium are).

Since the permittivity & permeability of materials vary, it's not surprising that the speed of light varies as well.

• But, if the speed of light changes, doesn't that mean that the light is accelerating? That shouldn't happen, isn't it? Feb 1 '20 at 5:13
• @PrathamHullamballi It doesn’t violate Special Relativity, if that’s what you’re wondering. The speed of light is the constant $c$ only in vacuum. Feb 1 '20 at 5:28
• @PrathamHullamballi light is certainly capable of accelerating, e.g. when a mass deflects light, the direction vector changes, and there is acceleration. Acceleration doesn't have to involve a change in speed, and even if it does, it does not have to actually lead to a greater speed. Feb 1 '20 at 5:59
• @Allure Are you talking about a Newtonian model of light deflection? In GR, particles moving on geodesics through curved spacetime are not accelerating. Feb 1 '20 at 7:00
• I think I might speak for @Allure who just answered a question posed in term of common meaning, which is not GR. Feb 1 '20 at 9:42

As Wikipedia’s article “Refraction” explains,

Light slows as it travels through a medium other than vacuum (such as air, glass or water). This is not because of scattering or absorption. Rather it is because, as an electromagnetic oscillation, light itself causes other electrically charged particles such as electrons, to oscillate. The oscillating electrons emit their own electromagnetic waves which interact with the original light. The resulting “combined” wave has wave packets that pass an observer at a slower rate. The light has effectively been slowed down. When light returns to a vacuum and there are no electrons nearby, this slowing effect ends and its speed returns to $$c$$.

• "The oscillating electrons emit their own electromagnetic waves which interact with the original light." can you please tell me if this states that there is photon photon interaction, though that is not possible (only higher order)? Feb 1 '20 at 4:52
• @ÁrpádSzendrei No, this is unfortunate wording. The article is using “interact” to mean “superpose”. Feb 1 '20 at 4:59
• Indirect photon-photon interactions via virtual electron-positron pairs become significant only at very high energy. Feb 1 '20 at 5:01
• @G.Smith “Electromagnetic waves which interacts with the original light, resulting in a combined wave which has wave packets that pass an observer at a slower rate”?? Doesn’t this leave a whole lot of explaining? How do the two interact? What is your wave? What are your wave packets and how are they slowed down? How about proven science involving the constant speed of light and refraction? As photons travel through the medium they deviate back-and-forth between the atoms do to refraction. The back-and-forth trajectory is longer than a straight trajectory and this longer path takes more time. Feb 1 '20 at 6:32
• @BillAlsept : Your line "As photons travel through the medium they deviate back-and-forth between the atoms [due] to refraction." describes either (1) a diffusion process, which we could detect by photons taking four times as long to pass through twice as much material, which is not observed, or (2) an unprecedented coincidence of scattering only in the very finely tuned directions necessary to get exactly the right amount of "zig-zagging" which we would show is impossible by passing light through an amorphous solid, like glass, which is not observed. Feb 1 '20 at 16:04