Is there a single solvent that can dissolve chemicals/elements that when burned, will produce the color of its respective chemical/element?

If so, what chemical/element can be dissolved in this solution and burned to produce vibrant colors that can correspond to any of the following colors:

  1. Blue
  2. Green
  3. Pink
  4. Violet

A hue formed by Blue and Green OR Pink and Violet will also work.


A single solvent? Probably not. Different elements have dramatically differing solubilities in most solvents, and so you would have to tune the solvent to the element at least to some extent.

However, for the specific task of identifying a solvent that would (a) burn and (b) dissolve chemicals suitable for giving either a blue/green color or a pink/violet color, then yes, I think it should be feasible.

Aqueous ethanol solutions at a concentration above about $20\%$ w/w are readily flammable at room temperature, and the aqueous base should provide a suitable base solvent for many salts. Vodka might be one good readily-available source of an ethanol-water solvent for this purpose.

To achieve the pink/purple color, potassium chloride should be a good option, as it's freely soluble in water and the potassium gives a purple color in the ethanolic flame.

Getting the blue color might be a bit trickier, as the most common element giving a blue flame is copper, and most copper salts aren't especially soluble in water at neutral pH, or in ethanol. What might work is to acidify the ethanol solution with a high-concentration vinegar stock (I've seen products available at up to $45\%$ acetic acid), and then add the copper salt. Copper sulfate pentahydrate is probably a good option there, as it's commercially available as, e.g., an algaecide. You'd have to experiment with the recipe to find something that burns readily with the desired color, but that doesn't smell too strongly of vinegar.

  • $\begingroup$ Did you mean to refer to potassium chlorate or will potassium chloride work for this process? $\endgroup$ – joseph levenson Aug 11 '20 at 15:21
  • $\begingroup$ @josephlevenson As best I understand it, potassium chloride should suffice. The color is coming from the potassium ion, not any combustive/oxidative properties of the counterion. $\endgroup$ – hBy2Py Aug 11 '20 at 15:28
  • $\begingroup$ Great information, Thank you! I will let you know how it turns out. If you know of any other reactions that can be used to reveal gender, i would be really interested to hear your ideas. I have a very large property where fire or fumes/smoke are no threat. your input is greatly appreciated. $\endgroup$ – joseph levenson Aug 11 '20 at 15:40
  • 2
    $\begingroup$ I would prefer lithium salts for a red flame. Potassium produces a purple color, but it is not intense. Lithium salts produce much more intense red flames. And copper chloride is more volatile than copper sulfate. So I would recommend copper chloride to get a blue flame. $\endgroup$ – Maurice Aug 11 '20 at 16:33
  • $\begingroup$ @Maurice I have experimented with the lithium and it does produce a very nice and aggressive flame. I have placed an order for the copper chloride. Should this be dissolved in the same manner as explained in the original answer? I will be doing this later in the day while the sun is still slightly shining and would require the same burning effect that the lithium gives off. $\endgroup$ – joseph levenson Aug 11 '20 at 17:06

@ Joseph. If I were you I would start in advance by producing a regular vertical flame (nearly colorless) with a methane or propane or butane burner. A big stable flame, as long as possible. And then, just when you want the rest of the world know the signal, I would spray an aqueous solution of lithium chloride or of copper chloride horizontally through the flame (at the base of the flame). And repeat the spraying operation as often as desired. Each time it will produce a puff of colored light, which is as big as the length of the flame. It does not last more than half a second. But if you are two or three people equipped with such a spray near the flame, each one may send a puff one after the other. The emitted light will nearly look like a continuous red or blue-green flame. This is not dangerous, as only aqueous solutions are used.

  • $\begingroup$ That is what I wanted to suggest. $\endgroup$ – Alchimista Aug 11 '20 at 20:07
  • $\begingroup$ @maurice Unfortunately I feel as though i am running short on time and would not be able to build a burner large enough to display for a group of people back in a field. Do you think i can put a few inches of alcohol in a wide and short steel drum and use this for the same process? Would this flame be hot enough for use with aqueous solutions? can i try to diffuse oxygen into the alcohol for a hotter burn or maybe add some kind of agitation? $\endgroup$ – joseph levenson Aug 12 '20 at 11:08
  • $\begingroup$ @ Joseph. What does mean "a group of people back in a field" ? Three, five, ten, twenty people ? Back ? Back ? In a field ? Who are these people ? Customers ? Members of the family ? Colleagues of the army ? Second, it is not necessary to build a big burner. What for ? How far is the "group of people" ? A regular burner gives a flame 10 to 20 cm high. Is it not high enough ? I must recognize that I am not a specialist of chemical shows. I have never tried to produce huge colored flames. My demonstrations are made for a group of maybe twenty students (one class) in a classroom. $\endgroup$ – Maurice Aug 12 '20 at 16:44
  • $\begingroup$ @ Joseph. You may try to use a wide steel drum as you say. But I would not play with the fire in adding oxygen, air with or without agitation. Be prepared to extinguish the fire by covering it with a blanket, in case suddenly all the alcohol gets vaporized and burns all of a sudden, producing a huge flame. $\endgroup$ – Maurice Aug 12 '20 at 16:45
  • 1
    $\begingroup$ @ Joseph. OK. So, as the birth is not for tomorrow afternoon, why don't you try both colors with whatever drum you want ? $\endgroup$ – Maurice Aug 12 '20 at 21:13

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