48

Electric sparks tend to appear blue or purple or white in color. Why?

math_lover
  • 4,526
  • 10
  • 47
  • 92

3 Answers3

60

Air is normally a bad conductor of electricity, but with enough voltage it can be converted to plasma, which is a good conductor. In a plasma, the electrons constantly bind to and leave atoms. Each time an electron binds to an atom, it emits the energy in light. As a result, the plasma glows the color of a photon with that energy. There are a few different energy levels that get involved, so the spectrum has a few different peaks. The final color depends on the gas you use. For example, neon looks red or red-orange. Air ends up looking blue, so electricity passing through air makes it glow blue.

Reinstate Monica -- notmaynard
  • 109
DanielLC
  • 1,669
  • 11
  • 11
  • 12
    +1 good answer: but do you know which atmospheric gas is mainly responsible for the color? – Selene Routley May 27 '15 at 05:00
  • 2
    @WetSavannaAnimalakaRodVance I think you will find an answer by looking at the composition of earth's atmosphere – Xeren Narcy May 27 '15 at 05:12
  • 3
    You have described recombination radiation, but this forms a continuum. What are the particular transitions in what elements that cause the colours seen? – ProfRob May 27 '15 at 16:41
  • 1
    I know that in nitrogen discharges, the main emitter in the visible band is the radiative decay of N2(C3Pi) radiating downward to either N2(B3Pi) or N2(A3Sigma) (I forget which one). I don't know how these compare with oxygen emission however. – Godric Seer May 27 '15 at 17:40
  • @XerenNarcy True, but one also needs to know the amplitude of each ionization event thus some means of working out the relative strength of all the lines. I don't think it is as straightforward as looking at all the spectral lines and saying "it must be that one". It's the kind of thing I'd work out experimentally by taking a gawk at the light in question through a spectrometer. I'd really learn something if you could prove me wrong by telling me that it is simple (this is not altogether my field, so I don't fully know these things). – Selene Routley May 28 '15 at 01:18
  • @WetSavannaAnimalakaRodVance Yes, you're right, (not my area either). Abundance is just one factor, I assume it's the major one, though that's only an intuitive argument. – Xeren Narcy May 28 '15 at 01:28
  • 1
    @RobJeffries you're right, the light emitted by a spark is a discrete spectrum, mostly from electrons which are excited to a higher energy state (but don't leave the atom or ion) when they drop back to a lower energy one. So it's not recombination radiation... Edit: ah, like you describe in your answer :-) – craq May 28 '15 at 11:43
  • 1
    To be more clear. The ionisation energies of N and O are about 14eV. Recombination of electrons and ions produce a continuum of radiation, mostly in the ultraviolet. Airglow is due to discrete lines caused by the transitions of bound electrons in ions and atoms. – ProfRob May 29 '15 at 08:31
  • But not all sparks are blue. I've personally observed orange sparks in air by what I think is sufficient voltage and very low amperage. You can see this if you get a variable gap spark tester and hook it up to a car's ignition system with the gap set so that the spark just barely has enough power to jump the gap. – Robert S. Barnes Mar 21 '16 at 08:34
  • So, is it impossible to have an electric spark in vacuum? – Gabriel Sandoval Feb 19 '19 at 04:37
40

The deexcitation of nitrogen and oxygen, the primary components of air, are that of blue/purple.

See http://en.wikipedia.org/wiki/Ionized-air_glow for pictures of nitrogen and oxygen in gas discharge tubes.

pixels
  • 509
  • Not only does that wiki article contain the described pictures, it contains a good answer to the question as well! – Jason C May 31 '15 at 02:06
22

The answer is that electrical excitation of air molecules is able to produce lots of excited singly ionised nitrogen ions.

The electronic structure of singly ionised nitrogen has a number of allowed radiative transitions, where the outer excited valence electrons can rearrange themselves into lower energy configurations.

The most prominent turn out to be those transitions corresponding to emitted photons at 443, 445 and 463 nm, and it is these that are responsible for the blue airglow.

ProfRob
  • 130,455
  • 5
    +1 for the first answer that mentions the relevant wavelengths. – Guntram Blohm May 27 '15 at 22:14
  • Why do some air gap sparks appear orange/yellow? I've personally observed this and have asked a question about it here: http://physics.stackexchange.com/questions/244630/why-are-some-air-gap-sparks-orange/244637?noredirect=1#comment537013_244637 – Robert S. Barnes Mar 22 '16 at 13:43