How to: Instantly turn water into snow

Chalk this up under "Blogs You Ought to be Following". The Tumblr Fuck Yeah Fluid Dynamics is a great place to find succinct, clear explanations of the forces that make things flow. In particular, they're fantastic at posting explanations behind things you see in YouTube videos, both viral and obscure.

The video above — in which a nice Siberian guy tosses boiling water off his balcony and creates a cloud of snow — has been making the rounds recently. Here's how Fuck Yeah Fluid Dynamics explains it:

Several effects are going on here. The first thing to understand is how heat is transferred between objects or fluids of differing temperatures. The rate at which heat is transferred depends on the temperature difference between the air and the water; the larger that temperature difference is the faster heat is transferred. However, as that temperature difference decreases, so does the rate of heat transfer. So even though hot water will initially lose heat very quickly to its surroundings, water that is initially cold will still reach equilibrium with the cold air faster. Therefore, all things being equal, hot water does not freeze faster than cold water, as one might suspect from the video.

The key to the hot water’s fast-freeze here is not just the large temperature difference, though. It’s the fact that the water is being tossed ...

Read the rest


  1. I wanna see someone compare the crystals formed from this, compared to natural snowflakes.

    My only party trick involving freezing things is putting filtered water bottles in a snowbank overnight. Can do it in a freezer too, but snow is always more fun.. they seem to have a roughly 1 in 5 chance of not developing crystals, so when you pick it up and shake it a little bit the whole bottle will quickly turn into ice. Not solid ice though, it’s a pleasant slush.

  2. surely there is also some dispersive advantage to starting out with hot water (unrelated to eventual phase change), lowered surface tension, some portion already in vapor phase, larger micro-phlogistonal N-ray heat transfer Fermi surface interface (“God made the bulk, but the devil created the surface”), surely?  (“that’s already discussed in the article you @#$! and stop calling me Shirley”)

    1. You did read that paragraph in the link you provided, eh?  From what I’ve parsed, it hasn’t been shown.

      Me, I want to ask René Descartes just how those particles were bending.

      Edited to add: If you watch the video, you’ll notice that most of the water actually hit the ground. What water did freeze, actually produced ice fog, rather than snow.

      If you want expertise on turning water to some form of snow, visit your local ski slope or talk to some pseudomonas syringae.
      Bonus: ski resorts usually use water pulled from a local river or lake – nearly freezing cold – to make the snow.

    2. As a friend once said “argue in the extreme”

      When placed into a -200°C environment, water at .00001°C which is cold, but not freezing, will freeze faster than water at 99.99999°C which is hot.  Therefore hot water will not freeze before cold water.

      Remember to disprove a theory one only needs one instance where the theory can be disproved.

      1. The way I understand it, in practical situations, is that if you put cold water in a freezer, you will get a nearly equal amount of ice in a given amount of time.  But if you put the same amount of boiling (or near-boiling) water in a freezer, you will get a smaller amount of ice in a shorter amount of time.

        I haven’t done any experiments or anything.  But it makes sense to the analytical me, which has observed that dishes rinsed in very hot water air-dry faster.  The droplets evaporate more quickly exposed to room-temperature air.  Ergo, very hot water placed in a freezer will evaporate quickly, leaving the rest of the liquid free to freeze faster.

        I’m not arguing the science that says that cold water will reach higher temperatures more quickly than hot water will, just that the evidence pointing at hot water exhibiting a solid state more quickly still makes sense.

        1. Interesting point, thanks. I should have added a 1L size for each and stated freeze means all 1L = 0°C. (all 1L remaining)

          I do agree hot water looses its heat quicker, but (in my example) the hot water has to loose 99.99999°C and the cold only .00001°C. 

          The link provided above looks to support the argument with a specific set of parameters. Theories don’t work that way. Thanks for your comment. 

          1.  Strawman. We’re talking about room temperature water vs boiling water, the parameters of the experiment are very specific.

        2.  I wonder if 1L of boiling water is more pure H20 than 1L of regular tap water, thus leading to a more efficient freezing process?

  3. Back in the Nineties, I was staying in Downtown Minneapolis with a minus Forty degree night outside. Our hotel was across the street from a local TV station. My room was directly across from the rooftop set where the weatherman was demonstrating how cold it was by pulling a pitcher of boiling water from a microwave and tossing the contents into the air to make a mini-blizzard of instant snow. It was beautiful and awesome (and damned cold, too.)

  4. I can confirm this works in air temps as “warm” as -28F (-33 C), I’ve done it before multiple times.

  5. In Spain, I tried the reverse – throwing a bucket of ice in the air on a really hot day. Results were very different.

  6. The explanation completely neglects evaporative cooling, which is significant.  All else being equal, water at 100C is evaporating much faster than water at 20C, which means it’s cooling much faster, faster still than a simple calculation based solely on temperature differential and conductive cooling would indicate.  

      1. Yeah, as if insulting me is going to make me change my mind that gratuitous vulgarity isn’t humorous.  And you haven’t seen the movie, have you?  

Comments are closed.