Can we chemically characterize a flame ?

Recently, while I was lighting a candle, I came across this question; can we find out the chemical composition of a particular flame? If yes then how? In other words, can we know which elements are present in flame, and not just the salts we test using the flame tests?

  • $\begingroup$ Well, I guess you'd like description of combustion reactions. It can be done but mechanisms are very complex and of course dependant on specific case. $\endgroup$
    – Mithoron
    Mar 7, 2017 at 20:08
  • $\begingroup$ I'm voting to re-open this question since later questions have been closed as duplicates with this question listed as "the original". chemistry.stackexchange.com/questions/169691/… $\endgroup$
    – Curt F.
    Dec 3, 2022 at 20:30

2 Answers 2


If you are using a gas stove with a methane ($\ce{CH4}$) flame, an interesting things occurs if you heat $\ce{CuO}$ in the gas part of the flame. I once created cupric oxide from heating a penny to glowing red and then shaking in air. Then, I inserted the dark colored penny into the flame zone, and it appeared to be cleaned!

Apparently, the hot cupric oxide is reduced by $\ce{CH4}$ to $\ce{Cu2O + Cu}$. As such, my take on a possible reaction liberating elemental copper, $\ce{Cu2O}$ and carbon:

$$\ce{CH4 + 3 CuO -> 2 H2O + Cu2O + C + Cu}$$

Now, the flame test for Carbon is red-orange and for Copper green/blue, so a mixture of colors could theoretically be observed in this experiment as the methane flame itself participates in element production.

Normally, however, this does not occur.


You can find certain elements in the flames if you are lucky enough and they do not overlap too much. Like in classic flame tests by having the light emitted from the flame pass through a prism or optical grading you can compare the emission spectrum to all known elements. But this won't be a full characterisation probably. Sodium can be seen quite easily with its yellow flame. When I had to distinguish between Strontium, Calcium and Lithium I looked other emission lines like the green ones. So with some tricks you might be able to determine two or three elements at once but I doubt you'll be able to tell whether it's sodium bromide, chloride or sulfate in the end.

  • $\begingroup$ but that is for the salts not what are components present in the flame $\endgroup$ Mar 6, 2017 at 8:40
  • 1
    $\begingroup$ Then you would be close to something like mass spectrometry where you ionize your compounds and analyze the charged fragments that break off your molecules. $\endgroup$ Mar 6, 2017 at 9:07

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