I was chatting with my kids about how incandescent lightbulbs work, clarifying that they don't so much "burn very slowly" (their original understanding) so much as they emit photons "just like everything does when it gets hot" (my understanding of thermal electromagnetic radiation). Somehow water came up in the ensuing discussion and it got me wondering whether "water" could ever been seen to glow?

I found a thread on some more general forums on the topic that was a bit muddled but afaict one of the more proficient answers included this explanation:

Thus, the answer to “Will X glow at heat Y?” is only dependent on the Y. The only reason X might be an issue is that X may or may not exist at the temperature in question. (It’s kind of like Cecil’s answer on melting wood - long before it melts, it stops being what we’d call wood.)

…implying that at high temperatures the water would thermolyze into what I assume would just be "glowing hydrogen" and "glowing oxygen". But a later answer in that thread cites sources to claim that:

[…] you should be able to heat water to about the same color temperature as an incandescent light bulb, 2700-3300°K, before it starts dissociating.

Is this true? Can water molecules be hot enough to glow without breaking up? And if so, do they simply glow like e.g. hot steel would — or would its molecules still not behave like a "blackbody" under those conditions?

  • 2
    $\begingroup$ Highly related: Why doesn't diamond glow when hot? $\endgroup$
    – Ruslan
    Oct 7, 2020 at 19:02
  • $\begingroup$ I assume you mean glow in the visible spectrum? All objects already "glow", giving off electromagnetic radiation that corresponds to their temperature. $\endgroup$ Oct 7, 2020 at 22:34

1 Answer 1


According to the Wikipedia article on water splitting, more than half of the water molecules are split into oxygen and hydrogen at 3000 °C, so the remaining ones will be radiating as intact molecules. I don't see any reason why water at that temperature would deviate from the blackbody spectrum, but I don't know for sure. But I know that it's close enough at room temperature to make infrared thermometers fairly accurate as measuring devices.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.