I am a glassblower, and I would like to trap smoke in a glass bubble, my hope is to create an infinitely lingering smoke inside.

Is this possible?


Smoke is made up of particles, so eventually they will settle. When they settle, they will probably stick to the surface of the glass and become impossible to stir up into smoke again.

You might be able to do this with a vapor. Because vapors are in the gas phase, they will not settle unless chilled. Unfortunately, a vapor will dissipate over time, resulting in a consistent color inside the bubble and not the "smokey" appearance that I assume you want.

There might be a way to use two gasses that won't mix, but even if that worked they would likely cause health problems if the bubble cracked open.

  • $\begingroup$ What if one uses a vapor, and cools to condensate, then reheats the glass bubble? $\endgroup$ – Cristi Stoica Sep 24 '13 at 21:21
  • $\begingroup$ I hadn't thought of that. It would require almost constant heating and cooling of the bubble, but if Isak is okay with that, I guess you would only need a suitable vapor. $\endgroup$ – Wutaz Sep 24 '13 at 21:24
  • $\begingroup$ I've seen a similar principle (condensing in one location, vaporizing in another) used in continuous cloud chambers. In principle it might work for this effect, but the engineering of it will be difficult, and thermodynamics requires that you power the system. $\endgroup$ – user10851 Sep 25 '13 at 4:32
  • $\begingroup$ Thanks for the feedback, the constant heating and cooling of the bubble may work, however it depends on the temperatures required to create the effect, as glass is very temperature sensitive and constant heating and cooling could cause it to crack. I do like the idea of a continuos cloud chamber, I imagine the heating and cooling could take place outside of the bubble and the smoke-like vapor could constantly be fed into it whilst it collected $\endgroup$ – Isak Lystad Sep 25 '13 at 5:44
  • $\begingroup$ Condensation and continued the process outside of the glass bubble. $\endgroup$ – Isak Lystad Sep 25 '13 at 5:45

Any mixture of gases (or smoke particles suspended in air) becomes a uniform mix in minutes. So whatever sample you trap in a glass bubble will be featureless in probably less than a minute. It will be indistinguishable from a uniform stain of the glass itself

As Wutaz writes, smoke particles will likely start sticking to the surface. At best that will result in the appearance of frosted glass, chances are the particles sticking to each other will look like deposited soot.

Iodine vapor has a strong color.
(Iodine compounds used as antiseptic are yellowish, but the vapor of pure Iodine is Magenta in color.)

But again, the mix of Iodine vapor and air will become a uniform mix in minutes, if not faster. Also, you can't contain the toxic Iodine vapor in the furnace. It's clearly too dangerous to try something like that in a shop for handcrafted glasswork.


Is it necessary for the smoke to keep moving?

If not, you can fill the glass bubble with a clear, viscous fluid like corn syrup or glycerin, and then with a straw, blow some food coloring into the middle of the fluid.

Then seal the glass bubble. Heat it slightly so the food coloring diffuses slightly into the shape of smoke. Then let it cool and put in a place where it won't get disturbed. It should last awhile, depending on how viscous the fluid is.


I wish to attack this question from a more fundamental point of view.


Your bubble will quickly attain thermal equilibrium. To obtain structures such as smoke in thermal equilibrium, you need intermolecular forces. However, a characteristic property of gases is that thermal motion dominates intermolecular forces – and thus destroys the structures very quickly. Therefore, you cannot have long-lasting smoke-like structures in gases in your bubble – unless you ensure that it is not in thermal equilibrium, e.g., by permanently heating it from one side.

Long Version

In most respects, your glass bubble is an isolated system. As such it quickly reaches thermodynamic equilibrium¹, which in turn balances between:

  • Mixing different species of molecules. This is simply because there are much more mixed states than ordered states. (This is the maximisation of disorder with which entropy is often explained in popular science.)

  • Ordering forces such as

    • Forces that are specific to certain species of molecules (either acting only between the same species of molecules or only between different species of molecules). For example oil forms droplets in water due to this.
    • Forces between the inside of the bubble’s wall and some species of molecules. You can see this with some substances that tend to coat the wall.
    • External force fields like gravity. For example, this causes two liquids of different density to form layers. This includes electromagnetic fields or acceleration forces (if you centrifuge your glass bubble), which are probably not relevant for you.

If you increase the temperature of the system, mixing will become more dominant (this is due to the thermal energy overcoming the ordering forces). In particular, in a gas, thermal mixing dominates intermolecular forces. This is what distinguishes a gas from a liquid: If intermolecular forces dominated, condensation would happen and we would have a liquid (or solid). Thus the only way to obtain an ordered state in thermal equilibrium in a gas is by external force fields. (I cannot find any sources on this, but IIRC, gravity is not strong enough to stratify any pair of gases at room temperature on the scales we are talking about.)

Now, anything resembling smoke is obviously not mixed and predominantly a gas. It is also not structured by an external force field (which would cause layers or similar). Therefore it cannot be the thermal equilibrium. From another point of view, all processes causing smoke are clearly far from thermodynamic equilibrium. For example, the process of burning a candle is based on restoring a chemical equilibrium (which supplies the energy) and comes with considerable inhomogeneities in temperature.

What remains is that you disturb the equilibrium somehow. You can do this by:

  • Cycling between cooling and heating. For example, you could have a liquid (or solid) in your bubble that evaporates when heated, which would cause smoke to form.

  • Heating a part of the bubble and cooling another part. For example, you could heat the bottom of your bubble and have the rest cool down by the surrounding room. This could cause some liquid in a puddle at the bottom to evaporate, form smoke-like structures, condense on the bubble’s wall into droplets which return to the puddle by gravity. However, unless you carefully tune this, you will likely only see a foggy bubble. Think of supplying just the right amount of heat to a closed pot of water.

  • Shaking the bubble. However, this can only cause layered structures to homogenise and does not form structures.

¹ with the exception of some chemical (and atomic and nuclear) bindings which need much higher temperatures to be broken. However, the temperatures required for breaking those bindings will also melt your glass (otherwise they would been broken in production), so we can disregard them. Also, these are usually strongly exothermic one-way processes.

  • $\begingroup$ I checked for a few key words and they're in this, so good sign there! Reading bits it's clear you're at the point where you're getting the concepts and trying to explain everything with formal terms - check out that comment about nuclear strong force at the bottom! I think "online" the pedants who go "well actually..." encourage a lawyer-style of writing. Try and make sure it's something you can turn on and off as it makes things very long.... 2nd to lastly - it can often be worth it (time wise) to find someone else who drew a picture or wrote about it and link that. $\endgroup$ – Alec Teal Nov 18 '18 at 9:13
  • 1
    $\begingroup$ Real lastly, remember your audience, you have pedants, people who know this already and are at least the level of "done one of the many big-ass bookend-worthy-if-hardback almost-A4-sized heavy 1st year physics textbooks" we all have on our shelves, or will never open one in their lives probably. A little inaccuracy saves lots of explanation (cf: lies to children) $\endgroup$ – Alec Teal Nov 18 '18 at 9:15

How about aerogel? It has been nicknamed "frozen smoke" and has been made at home.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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