28 May 2021

Making ice in a red-hot crucible

This week I finished a final goodbye read of the collected tales of Edgar Allan Poe, and encountered this interesting passage:
Place a platina crucible over a spirit lamp, and keep it a red heat; pour in some sulphuric acid, which, though the most volatile of bodies at a common temperature, will be found to become completely fixed in a hot crucible, and not a drop evaporates- being surrounded by an atmosphere of its own, it does not, in fact, touch the sides. A few drops of water are now introduced, when the acid, immediately coming in contact with the heated sides of the crucible, flies off in sulphurous acid vapor, and so rapid is its progress, that the caloric of the water passes off with it, which falls a lump of ice to the bottom; by taking advantage of the moment before it is allowed to remelt, it may be turned out a lump of ice from a red-hot vessel.
I was going to ask readers here about the thermodynamics, but a quick search of the internet today led me to a page at Physics Stack Exchange that discussed this very same passage.  The site provices additional citations of similar experiments in the nineteenth century:
M. Boutigny, by means of sulphurous acid, first froze water in a red-hot crucible; and Mr. Faraday subsequently froze mercury, by means of solid carbonic acid...
A reader there provided some clarification re the reagents -
What they were calling "sulphurous acid" back then is not what we would call an acid today. It was anhydrous sulphur dioxide which has a boiling point of −10∘C.

When liquid sulphur dioxide was poured into the red-hot vessel, due to the Leidenfrost effect, it would form itself up into globules and float on a layer of its own vapour. In this state the temperature of the globules would be just below that of its boiling of −10∘C as it evaporates away at a now greatly reduced rate. Pouring in a small amount of water, which freezes at 0∘C, while the sulphur dioxide is in this state results in it freezing within a few seconds. Once all the sulphur dioxide has evaporated off, the ice will quickly melt again before being brought up to just below its boiling point of 100∘C as it assumes its spheroidal form due to the Leidenfrost effect. If one is quick, before all the liquid sulphur dioxide has disappeared one can throw out a small lump of ice from a red-hot crucible!
Sounds like a demonstration that would make an alchemist proud.


  1. So we know that to cool something, you can take a cool substance and bring it in contact with the hot object. This seems to be the case here. But how to cool something using evaporation, expansion, dissolution or by radiating excess heat. Now, those are some of the entropy defying questions, that boggle the mind. How Does Evaporation Cause Cooling? - The search engine non-answer: Because this process requires heat energy.

    Well, to me the idea of a water molecule up and leaving, and taking all the energy with it always has been deeply unsettling. Yes, as a gas that bugger can wiggle and rotate and bounce around much more freely and energetically, but that is still just the effect not the cause. We can assume that it was statistics that pushed it out but how did the energy transfer take place? Being on the surface has a lot to do with, there are fewer attractive neighbors. But then a confluence of unfortunate vibes, lets the last attractive forces snap like a rubber band, and sends the bugger hurtling through the air. Similarly when pulling two magnets apart we feel an inevitable jerk/snap when the bond breaks. That jerk is EM-attraction being transformed into kinetic energy. The question then becomes how do EM-fields interact with each other: Opposite fields attract and cancel each other out when overlapping. When bonds break fields around the molecule reform. Is the field magnetic or electric? In short water is diamagnetic, but it is (due to it's asymetric bend) partially electrically charged. That attraction, breaking which results in the release of energy, can be demonstrated when the elctric field of a statically charged glass rod is held close to a tiny water stream, attracting and bending it.

    Could you fix the link?

    1. Fixed. Thank you for the heads-up.

      Re your first sentence, my understanding is that in the case described, you are freezing the water by bringing it into contact with a COLDER substance (the vaporized sulphur dioxide), not a hot object. I don't believe the water ever contacts the hot crucible.

      Your other queries are thought-provoking.

  2. Looking at the evaporation comment first:

    “So we know that to cool something, you can take a cool substance and bring it in contact with the hot object.”
    That’s not correct. What’s happening here is that the water being dropped in is contacting the SO2 at –10 °C — it’s the cold SO2 that does the cooling, not the hot crucible.

    “how to cool something using evaporation, expansion, dissolution or by radiating excess heat. Now, those are some of the entropy defying questions”

    That’s actually exactly how entropy works. Let’s try a thought experiment: Imagine you have a beaker full of water molecules (let’s imagine 10, though the real number would be astronomically larger than that), and each molecule has, ON AVERAGE, 8 quanta (chunks) of heat energy (yes, the real number would again be vastly more than that). The temperature of the liquid is directly related to the average number of quanta of the water molecules. We have 10 molecules with a total of 80 quanta so let’s call the present temperature 8 degrees TYWKI (8 quanta per molecule). However, although the average number of quanta per molecule is 8, they are continually exchanging and shuffling their quanta so that a few might have as many as 20 quanta while others might have only 2 or 3.

    Now let’s imagine that they need 10 quanta to break free of the liquid. Those molecules that accrued at least 10 quanta, and are on the surface, now have enough quanta to break free and fly away as a vapour molecule. Let’s imagine that happens five times. 5 molecules have escaped with 10 quanta, meaning there are only 30 quanta left in the beaker. The temperature of the beaker has now decreased to 30 quanta over 5 molecules, or only 6 degrees TYWKI — the temperature has decreased. (The boiling point is reached when the average number of quanta per molecule reaches 10)

    This is exactly why evaporation cools a liquid and it’s all based on increasing entropy. Firstly, entropy means that the quanta are not evenly distributed since they are free to be exchanged around the beaker and some molecules will end up with more than others (if only for a short time). Secondly, those molecules with enough energy to fly away will do so. Again, this is entropy — the molecules fly away because THEY CAN, and, unless you keep the lid on, they’re not coming back.

  3. Chemical bonding comments next:

    “When bonds break, fields around the molecule reform. Is the field magnetic or electric?”

    The short answer is that it’s electrogmagnetic, though that, perhaps isn’t very helpful. The bonds in molecules are primarily the result of electrostatic attraction between electrons on the outside of atoms and the nuclei at their centres. There *is* a magnetic component as well but it makes a much smaller contribution. As a simplification, you can break chemical bonds with an electrical field (e.g. lightning) but not with a magnetic field (which is one of the reasons MRI scanners are safe).


Related Posts Plugin for WordPress, Blogger...