Anna-Marie Blundell, of John Paul College, asks :-

A magazine reports that hot water freezes quicker than cold water. I tried this in our freezer but there was no difference. Is it true?

Colin Hooker, a thermal physicist who is interested in weird physical problems, responded.

In some circumstances the statement is true. This has been known for centuries but systematic study of this behaviour seems to have begun after an observant third form student in Tanzania, Erasto Mpemba, discovered this himself while making ice cream in 1963 and later asked about it during a visit to his school by a physicist. The physicist carried out experiments confirming this behaviour and both he and Erasto Mpemba published separate accounts in a British physics journal in 1969.

Now the results of several sets of experiments have been published and in some of these (depending on the type of containers used, their surroundings and the sets of initial temperatures used) the hot water froze earlier than the cold water. In other cases it didn't. The main reason for hot water freezing first is that hot water evaporates rapidly absorbing the energy required to vaporise water from the water itself, cooling it. The water becomes denser on cooling and sinks, setting up a persistent circulating "convection" current which stirs the water helping it cool throughout.

In cold water this current is not so strong and, as water is a bad conductor of heat, the average temperature of the water in the container does not drop as fast.

Two other factors, which are not so important, contribute. Hot water lost by evaporation leaves less water to be cooled. Also water can be cooled below its freezing point of 0C before ice will start to form. Water that was initially hot "super cools" by less than the water that started out cold (for reasons not completely understood), allowing ice to form earlier on the initially hot water.