How can I make the flame in a wicked lamp incandescent when using azeotropic ethanol (95% ethanol with 5% water) as the primary fuel in a 'hurricane lamp'? (Ethanol burns with an almost invisible flame. It is not suitable as a lamp-fuel for light.) Would incandescence be improved by dissolving a little table salt in the azeotropic ethanol? If yes, and if sodium ions in the flame are emitting radiation in the yellow part of the visible spectrum, what happens to the chlorine ions? Would that be an unsuitable solution from a chlorine toxicity viewpoint? Or would incandescence be improved by dissolving a little tallow in the azeoptrope? The problem is to be solved in mid-nineteenth-century setting in which ethanol is freely available and kerosene, paraffin and turpentine are not available.

  • 3
    $\begingroup$ I added a little sodium chloride to my homemade alcohol lamp and it looked like this: chemistry.stackexchange.com/a/164168/79678. The chlorine goes off in the exhaust or reacts with flame species. Interesting challenge. $\endgroup$
    – Ed V
    Commented Aug 31, 2023 at 23:08
  • 1
    $\begingroup$ In case of the ultimate goal is the light itself, there are camping lamps based on LPG flame and thorium dioxide(?) fabric, that makes the shining. It could be used for ethanol too. $\endgroup$
    – Poutnik
    Commented Sep 1, 2023 at 6:30
  • $\begingroup$ Thanks, Poutnik, but the setting is nineteenth century. $\endgroup$
    – JohnH
    Commented Sep 1, 2023 at 7:36
  • $\begingroup$ How about naphtha? The recipe of Greek fire is long lost, but naphtha has been suggested as a plausible ingredient. Pine oil as well. $\endgroup$
    – Ed V
    Commented Sep 1, 2023 at 12:12
  • $\begingroup$ @Ed V, Thanks for pine oil idea. But I wonder whether it would be soluble in ethanol/water base-fuel? (Ethanol is feely available in the historical setting. Alcohols are the driver, except for the almost invisible flame if used as fuel without additives.) $\endgroup$
    – JohnH
    Commented Sep 1, 2023 at 12:24

1 Answer 1


Yes, a very little $\ce{Na+}$ goes a long way to make a bright yellow flame. Little would dissolve in EtOH with only 5% water, but that little should be more than sufficient.

Though the chlorine likely would react to a very small extent to form halogenated hydrocarbons, pre-1900, houses were quite drafty, and I don't believe much would accumulate, unless the lamps were used in closed-end tunnels or mines. Another issue with burning EtOH is that incomplete combustion produces acetaldehyde. Hot brass fixtures could act catalytically, creating larger amounts.

Consider "wood alcohol" (methanol) as a fuel, too. Due to tax laws, it is usually cheaper than ethanol, and was made by destructive distillation of wood waste. (Originally, I had misread the question and assumed that would be the fuel of choice.)

Also consider the limelight, which approached carbon-arc lamp brightness and intensity. "During the American Civil War in July and August 1863 calcium lights were used during the siege of Fort Wagner." Perhaps EtOH + forced air would be sufficient to get that lime glowing, though not as intensely as in oxy-hydrogen flame.

  • $\begingroup$ If the alcohol is 140-proof liquor (70%alcohol by volume) containing some fusel oil, would that also burn in a Hurricane lamp? If yes, would that (with some NaCl) produce a brighter flame because of the presence of the higher alcohol? Any comment on my tallow idea? Thanks in advance. $\endgroup$
    – JohnH
    Commented Sep 1, 2023 at 6:18
  • $\begingroup$ Tallow might work -- you could easily test it at home. BTW, the wealthy, in the 19th century, used whale oil, or fish oil ("hooligan oil" among the Northwest Native American (First Nations, if you prefer). en.wikipedia.org/wiki/Whale_oil#Applications and fws.gov/story/salvation-fish See also "conspicuous consumption". $\endgroup$ Commented Sep 1, 2023 at 18:08
  • $\begingroup$ Thanks, DrMoishe. That adds to my body-of-knowledge. $\endgroup$
    – JohnH
    Commented Sep 2, 2023 at 19:21

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.