I just built a DIY spectroscope using a CD and after making some tests, I noticed a really strong emission line located in the orange-yellowish zone of the fire spectrum (specifically, burning wood).

I think that the strong line could be actually sodium, but wouldn't that mean that wood contains a lot of sodium in order to see such a strong emission line?

  • 3
    $\begingroup$ There's a related question on the Physics site about orange colour in a gas flame. There's some blackbody radiation, but the consensus is that the yellow lines are sodium. It doesn't take much salt to produce those lines. $\endgroup$
    – PM 2Ring
    May 10, 2019 at 0:16
  • 7
    $\begingroup$ It's sodium. It's everywhere. This is why a lot of natural fires skew orange-yellow - it's the sodium pulling the color towards the yellow region. Even trace amounts will make that line, you don't need a ton of it. BTW, the line is a doublet, you can see it if your device has good spectral resolution. $\endgroup$ May 10, 2019 at 3:29
  • 7
    $\begingroup$ Congratulations on building your own DIY spectroscope! I propose you try to snap a photo through it and add it to your question, real data rocks! You might also see if adding a little salt or baking soda (sodium bicarbonate) to the fire enhances the yellow line. You also might consider moving your question to Chemistry, Physics, or Biology Stack Exchanges if people close your question here. If you can mention that you are truly interested in looking at spectra of planets or stars or the Moon next, that would keep your question on-topic here. $\endgroup$
    – uhoh
    May 10, 2019 at 3:48
  • 2
    $\begingroup$ FWIW, you can probably also see a violet line from potassium. Traditionally, potassium hydroxide was made from wood ash. And sodium hydroxide from seaweed ash, which as you might guess has a lot of salt. ;) $\endgroup$
    – PM 2Ring
    May 10, 2019 at 4:20
  • 1
    $\begingroup$ Throw some salt on the fire to confirm. $\endgroup$ May 12, 2019 at 14:41

1 Answer 1


Well, I've pointed my spectrometer at my fireplace, and I can confirm the presence of a relatively strong sodium line(s). It just takes trace amounts of sodium to produce this signature, it cannot be mistaken.

I also see an emission band around 765~766nm, of unknown origin, though ProfRob points out below that it's likely to be potassium.

Don't mind the dots, those are hot pixels, my spectrometer is using a 100um slit and requires long exposure times. The sensor is an inexpensive OV2640, spectral lines calibration is within <0.7nm but I'm still in the process of performing the spectral responsivity calibration. And it might still be slightly out of focus.

This instrument uses an actual transmissive grating, 500lines/mm, in the Littrow configuration and -1st order.

enter image description here

  • 3
    $\begingroup$ 766nm is probably potassium. $\endgroup$
    – ProfRob
    Dec 4, 2023 at 19:02
  • 1
    $\begingroup$ Potassium makes sense. Looks good. $\endgroup$ Dec 4, 2023 at 19:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .