# Is the temperature of light affected by color filters?

This video demonstrates how varying color filters alter the energy of photoelectrons emitted from a light source: https://www.youtube.com/watch?v=kcSYV8bJox8

But I am confused by what these different colors having different "energies" means. Blue light is said to have more energy than red light. Does this mean that the blue light is hotter than the red light? If so, where does the heat go for lower energy colors - is it absorbed by the filter?

It seems to me this probably isn't true, and just filtering light wouldn't change it's temperature. But, if the colors are all the same temperature, then what does them having different "energy levels" mean? Is this a very specific definition relative to the emission of photoelectrons from a metal?

Just to clarify, I know that a light source will change colors as it gets hotter, I am not referring to a light source whose temperature is directly being changed like a fire, I am referring color filters like in the video.

Temperature, colour and energy all mean different things.

A single 'photon' has an associated energy, but it's misleading to think of this as a colour. Three things that are (for most intents and purposes) equivalent, are energy $$E$$, wavelength $$\lambda$$ and frequency $$f$$, which are related by

$$E = hf = hc/\lambda$$

where $$c$$ is the speed of light and $$h$$ is Planck's constant. Note that as the wavelength gets smaller, $$E$$ gets bigger.

Humans have three kinds of cone (colour sensing) cells - red, green and blue, but these do not "only" see red green and blue respectively - they are excited to varying degrees by the other wavelengths as well.

To a human, a combination of 500nm and 660nm light appears yellow, but this is physically a different situation to e.g. a yellow sodium lamp, which emits primarily ≈580nm photons.

Light from something like the sun or an incandescent lamp is even more complicated - these are near-perfect blackbodies, meaning that they emit at all wavelengths to some extent: This excites all three of our cones equally, which our brain processes as "white". The crucial point is that there is no such thing as a "white" photon, but there is such thing as a "blue" photon.

The changing colour of a blackbody as it gets hotter is due to the movement of the "peak" wavelength, i.e. the wavelength at which most light is emitted, which for the Sun is about 500nm. If the sun were to get hotter, the peak would shift to the left, exciting more blue cones and fewer red cones. Likewise, if it got colder, the peak would shift to the right (lower energy), making the Sun look redder.

The Bottom Line All wavelengths of photons carry energy, "white" light is a mixture of many wavelengths, and filtering this mixture to only get e.g. red light will make the filter heat up / reflect the energy elsewhere.

Light does not have a temperature but carries energy at different frequencies. The energy from the light causes matter to heat up when they interact. Energy that is filtered will cause the filter to heat up.

• Thank you. But, does the light from different colored filters heat matter up - say a thermometer - at different temperatures? Blue light is said to be more energetic than red light, will the blue light heat up a thermometer more than red light? If the two colors of light have the same exact heat content, then what does one having a higher energy mean exactly? Commented Oct 20, 2021 at 3:30
• @Tristan Higher frequency = shorter wavelength ->higher energy. So blue light photons have more energy than red light photons. Think about the photoelectric effect, where electrons are ejected when the light exceeds a certain frequency—regardless of intensity and duration. Commented Oct 20, 2021 at 4:08
• @Tristan Higher energy PER PHOTON, but you can also change the number of photons. Arsenic is more toxic than water but more people still die by drowning. Commented Oct 20, 2021 at 11:24
• @Tristan Just in case you came across the term, there's something called "colour temperature", and for many (artificial) light sources it is routinely reported as one of its parameters/features, in Kelvin. Its a way to describe a colour (via blackbody radiation etc.). Confusing perhaps. But a very different thing from temperature in the usual "kinetic energy of molecules" sense. Commented Oct 20, 2021 at 14:43
• Really nice answer. Commented Oct 21, 2021 at 19:03

But I am confused by what these different colors having different "energies" means. Blue light is said to have more energy than red light.

The statement "blue light has more energy than red light" is an imprecise statement. A more precise statement is "blue light has more energy per photon than red light." Saying that blue light has more energy than red light is a bit like saying that gold weighs more than water.

If so, where does the heat go for lower energy colors - is it absorbed by the filter?

When a filter absorbs light, it absorbs its energy. So both a filter that absorbs blue light, and filter that absorbs red light, absorbs energy and decreases the energy of the light.

Blue photons (= photons of ~420 nm wavelength) have higher energy than red ones (= photons of ~600 nm wavelength). That means that each individual absorbed blue photon will indeed heat the surface more, because it gives it more energy. But two red photons would heat the surface more than a single blue photon. Photons themselves don't "have" heat, heat is a collective consequence of random motion of many atoms/molecules/...

Photoelectric effect in the video is a bit different phenomenon from heating the material because photons need sufficient energy to eject electrons from the material. Red light might not eject any electrons no matter how many photons hit the material, while blue would even if you have few photons (and therefore lower heating).

Temperature of light, eg as on "3000K" light bulbs, is something different still. Blackbody gives out EM radiation. Spectrum of that radiation depends on the temperature - say something at 3000K has peak of radiation at approximately 1 micrometer wavelength (near-IR) while our sun at ~6000K has peak at green light. And that temperature designation of the light bulb simply means "this bulb gives out light that has approximately similar spectrum to the body at this temperature" (skipping details about color quality).

You can start with 3000K bulbs, then add filter to get 6000K light. First light is red-ish, second one is blue-ish. But even though after filter your photons flying out are more blue and therefore have more energy on average, this filter absorbed (or reflected; in video filter absorbed light) quite a lot of energy - many red photons had to get blocked. So, average photon has higher energy, but total energy of all photons in the light is lower. Shining unfiltered red-ish light on thermometer would heat it more than shining filtered blue-ish light on it.

This losing energy is true even if you have laser light and convert it to have half the wavelength (eg 1064 nm to 532 nm) - in that case all input photons have low energy and all output photons have high energy, but you lost many (at least half) photons in the process, so the total energy (that gets converted to heat by absorber) is the same or usually lower (with some energy lost during conversion and resulting in heating of the converter). In this case of up-conversion (photons flying out having higher energy) you can also get photoelectric effect with the converted light while original one doesn't kick single electron out of the material. This is unlike with filters that merely block some wavelengths - there, photoelectric effect of filtered light is at most the same as in original light.

Like @PabloH said, one of the big sources of confusion in your question is bringing the temperature to the discussion.

There are two very conflicting definitions: one being warm/cool color dichotomy tied to human perception only and association of reds and yellows with sun and therefore warmth and blue tint with shadows and cold. Color temperature goes the opposite way and is tied to black body radiation spectrum (as already explained in other answers).

Color theory is a very, very complex topic and has more to do with psychophysiology than physics, admittedly.