So I was on a plane and tiny crystals of ice formed on the window glass. This window is formed by a few screens of glass (maybe or maybe not) with vacuum inside. I think the frost was due to some water vapor in the outer vacuum chamber, since at times also some condensed vapor appeared.

So there is vapor with the cold glass wall: the vapor condenses; when the glass becomes too cold the vapor freezes; then at a point sun shines on the glass and some condensation begins to appear (I'd say through sublimation of the frost).

Now the question is explaining the following: this newly obtained vapor condenses again on the glass wall (which still has some frozen crystals), but it condenses only at a certain distance from the frozen crystals. The pattern of the condensation encircles the ice crystals, which are surrounded by a region with no condensed vapor.

I have two hypotheses/ideas:

a) the region surrounding the crystal has no vapor because there the probability of being caught by the ice is higher than that of being caught by the glass: the ice is still very cold (maybe it can be lower than zero since it is sublimating and not melting) while the glass is now above zero (the vapor condenses on it in liquid droplets);

b) the glass surrounding the ice crystals is hotter because it receives heat from the crystals, which receive heat from the sunrays..

There could be other possibilities like:

vapor released from the crystal has high kinetic energy and therefore has a mean free path (like 1 mm of half mm) before being caught by the cold glass wall;

or the glass nearby the ice is colder and for some reason this impedes condensation (maybe if it is below 0°C ? Bah this does not sound correct..)

The point a) is not very consistend with the fact that the ice is sublimating and with the fact that there is zero condensation in the near surrounding of the ice; the point b) could work if the sublimation has a low rate so that the ice loses its heat mainly by contact with the glass..

So does anyone has any other idea of why condensation does not happen nearby the ice?

  • 1
    $\begingroup$ I've seen that too, but never thought of! two more explanations: c) the ice condensed on surface causes a local static electric charge, that repel new ice from forming; d) The air in the gap moves around bringing water vapour, the existing crystals are obstacles and around them the air flows less (like "mountains protecting from the wind") thus no new vapour is brought in their vicinity $\endgroup$ – patta Apr 11 at 11:35
  • $\begingroup$ Again, to have condensation on a surface, you need "nucleation points", as water condenses preferably on other water or at scratches/dust. Doesn't like to condense on clean glass. Then your a) works well, as the existing crystals attract water and inhibit formation of crystal on the glass nearby (energetically less advantageous) or your cycle sublimation/condensation remove nearby small crystals $\endgroup$ – patta Apr 11 at 11:37
  • 2
    $\begingroup$ Related question by OP $\endgroup$ – Aaron Stevens Apr 11 at 11:41
  • 1
    $\begingroup$ @patta maybe depends on the plane. For instance the second picture of the answer in the post you linked seem to have a vacuum chamber in the outer layers. Yeah about pressure I am not sure it would crush the glass: for instance a glass bell can be used to produce vacuum $\endgroup$ – AoZora Apr 11 at 13:58
  • 2
    $\begingroup$ There is never a vacuum or any other fixed pressure between the glass panes of a plane. The large variation in air pressure during a flight would put too much stress on the glass panes. There is always a tiny hole that connects the air gap between the panes with the inside of the plane, such that the pressure difference between outside and inside is handled fully by the outside pane. But I don't think that is relevant for the original question - although ice crystals tend to form first around the tiny hole. $\endgroup$ – fishinear Apr 11 at 15:26

Your Answer

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

Browse other questions tagged or ask your own question.