I made a mistake. I put my ice cream in the refrigerator instead of the freezer and left it there for a night. I realized my folly the next day.

I put it in the freezer. When I took it out, I noticed it had a bunch of ice crystals inside it and on its surface. This intrigued me (not that it is a new discovery of mine or anything). I was left with the question: Are these ice crystals the product of added water, or is it just the phase change of water already present in the ice cream.

Okay, so I have to hypotheses:

Hypothesis 1

When the ice cream was defrosted, it became more liquid and as such, the bonds of the chemicals were weakened. This allowed the water that was mixed with the milk to break loose a little. When the ice cream was put back inside the freezer, it cooled down at too quick of a pace, making it so that some of the water didn't have time to fall back in line before it underwent a phase change.

Hypothesis 2

While the ice cream was hotter and more liquid, its permittivity increased. Atmospheric water, of which there is probably much of inside a refrigerator, entered the semi-melted ice cream and over-saturated it with water. When it was put back into the freezer, this over-saturation led to the crystallization of excess water.

So, which one of these hypotheses are most likely, if not surely, correct or most correct? Or is none of them correct? Or is the phenomenon caused by a combination of the two hypotheses?


1 Answer 1


Industrial ice cream makers make it frozen much faster than household freezers. They freeze faster and ice cream is disturbed to take out heat and to keep the emulsion well dispersed. As result, frozen water based emulsion droplets are very small.

When melted and standing, emulsion slowly separates either and oily phases.

She put in household freezer, it is getting frozen very slowly. That helps further emulsion separation and ice have enough time to form big crystals.

Note that it happens for frozen vegetables as well, if melted and refrozen again. Without any added water.

  • $\begingroup$ Okay, so the first hypothesis was the most correct? Thanks for the answer btw! $\endgroup$
    – A. Kvåle
    Oct 23, 2020 at 16:46
  • 1
    $\begingroup$ Not really,there is no too quick freezing. Too slow freezing together with prior emulsion separation. $\endgroup$
    – Poutnik
    Oct 23, 2020 at 17:53
  • $\begingroup$ Yeah that's true, forgot that part of the hypothesis. $\endgroup$
    – A. Kvåle
    Oct 23, 2020 at 21:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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