I use an LPG (liquefied petroleum gas) gas stove at home for cooking purposes. I notice that, in the flame, the bottom part is almost always blue but the top has a yellowish-orange color. Why is that?

Edit - There was a suggestion from the community asking me whether an answer regarding the color of a candle flame. I read that and I feel it doesn't quite answer my question because the answer is based on the fact that candle flame consists of glowing soot, whereas here in the case of LPG, there isn't any unburnt carbon (soot) I think because it's a gaseous hydrocarbon fuel (propane and butane) unlike wax in a candle.

  • 9
    $\begingroup$ Does this answer your question? Why is the bottom part of a candle flame blue? $\endgroup$ Jul 25 at 4:37
  • 2
    $\begingroup$ The question Possible colors of fire? is related but does not address the question of why it's the top part that is yellow. $\endgroup$ Jul 25 at 5:13
  • $\begingroup$ @Vincent Thacker That question is about a candle flame, which is basically burning wax. In a candle flame, there is soot which glows (according to the answer in that question). Here it is LPG, which does not have soot (unburned carbon) I beleive. $\endgroup$ Jul 25 at 7:46
  • $\begingroup$ Your stove probably requires maintenance and may not be safe, see here: lpgjets.com/blogs/news/…. $\endgroup$ Jul 25 at 13:57
  • 1
    $\begingroup$ @PhysicsWizardUd yes there is soot, otherwise the flame would not be yellow-orange. Put a spoon or something in it and see how black it gets. $\endgroup$
    – OrangeDog
    Jul 25 at 15:54

The blue colour of a propane gas flame is caused by various discrete electronic transitions, mainly CH and $C_2$. If the combustion is incomplete it allows the formation of very fine soot particles that become hot enough to glow with a pseudo-blackbody spectrum, with significant emission in the yellow/red part of the spectrum. The soot is initially formed closer to the flame but convects upwards.

A propane stove producing a yellow flame should not be used, since it will be producing carbon monoxide too.

Here is a spectrum of a clean-burning butane flame (taken from the wikipedia page on "flame"; I couldn't find one for propane), showing the discrete nature of the blue optical emission.

Blue butane flame Credit: Deglr6328. Licensed under the Creative Commons Attribution-Share-Alike 3.0 Unported license.


You are obviously not old enough to have been taught about the Chemistry of the Bunsen burner flame.

Right at the centre of the flame there is a zone where LPG and air are being heated by the surrounding blue part of the flame but there is no chemical reaction in this region and so no light is emitted from this part of the flame as the temperature is about $\sim 500^\circ \rm C$ - the flame is colourless in this region.

Outside the central region of the flame is the main combustion volume where the gas reaction is very efficient and lots of heat is produced which results in a high temperature (up to $\sim 2000^\circ \rm C$) and the emission of blue light. In part the blue light is due to the for­ma­tion of rad­i­cals which are in an excited state and high­ly un­sta­ble. In moving to the ground state these radicals release energy in the form of blue light.

Some gas mol­e­cules do not burn com­plete­ly, form­ing small par­ti­cles of soot. These par­ti­cles rise high­er and are ox­i­dized by oxy­gen, producing a temperature of around $1000 ^\circ \rm C$ with the emission of "yel­low" (blackbody) radiation. This is the outer part of the flame. The unburnt carbon is the reason why cooking utensils sometimes go black on the outside.


For most flames, the color is simply a measure of the temperature of the glowing gas. The spectrum of the flame is fairly close to a black body, with a bluer peak in the spectrum indicating a higher temperature. So what you are seeing is that the flame is hotter lower down. This can be seen in gas jet flames or in match flames.

Lower down, at the site of the gas jets, the fuel is more concentrated, and the higher density of burning gas generates more heat. The extreme heat causes the gas particles in the flame to rise, and as they rise, the fuel spreads out and burns off, while the whole mixture cools. This leaves the top of the flame cooler than the bottom, and so when the bottom of a flame looks blue, its upper reaches are typically more orange.

  • 4
    $\begingroup$ The flame temperature is not high enough that you would get blackbody radiation in the blue part of the spectrum. The gas flame spectrum does not look like the Planck function. $\endgroup$
    – ProfRob
    Jul 25 at 8:11

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