I had a question regarding anomalous expansion of water. So if we have water and we start cooling it up to 4 degrees, water contracts, but after that in the range of 0-4 degree water starts expanding due to hydrogen bonding. The problem was not in this statement but what the book said after it:

This is the primary reason as to why when during winters, when the temperature drops, not all water in the water body freezes, only the surface layer freezes, because as water is being cooled, after 4 degrees, any ice that is formed comes to the top due to it being less dense and ice being a bad conductor of heat insulates the water below it from the outside region hence preventing it from freezing altogether.

But if that is the case, shouldn't we observe a similar thing whenever we freeze water in our refrigerators , we should not have had ice cubes at all. since a similar thing happens here, and the ice tray also being a good insulator, insulating the bottom most water from freezing?

  • $\begingroup$ How do you compare the heat capacity of the water in a ice cube tray to a lake. As well as the T gradient in nature and that existing on a freezer? $\endgroup$
    – Alchimista
    Oct 4, 2021 at 9:48
  • $\begingroup$ Many mountain lakes do freeze completely. $\endgroup$
    – Jon Custer
    Oct 4, 2021 at 13:20

1 Answer 1


The volume of water in ice trays is very small. Hence it has no problem freezing completely as the icy layer doesn't get thick enough to prevent heat loss to the freezer. In a lake or sea, the thickness of the icy layer is a lot more and it provides a lot of thermal resistance. So the lakes do not freeze completely.


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.