Hot ice cream sometimes freezes faster than cold ice cream

In 1961 a Tanzanian teenager called Erasto Mpemba noticed that when he put hot ice cream in the freezer it sometimes froze faster than when he put cold ice cream in the freezer.

Kudos to him for noticing this — no ice cream that comes into my house lasts long enough to be subjected to scientific experimentation.

Mpemba mentioned the effect to his teachers, but they didn't take him seriously. In 1966 he went to secondary school and had a chance meeting with a physics professor called Denis Osborne to whom he mentioned what he'd noticed. The physics professor did take him seriously and they authored a paper together.

It was thereafter called the Mpemba Effect, but the rest of the physics world was a bit dubious. Something that is hot has faster moving molecules than something that is cold. That's what heat is. Something that is hot therefore needs to do more slowing down than something that is cold, so it should take longer to freeze. This was the understanding at the time.

Nevertheless, scientists have since confirmed the Mpemba Effect is real.

The thing is, this movement of the molecules that creates heat is an average. Not all the molecules will be moving at the same speed — some will move faster than the average and some will move slower. Merely taking an average is insufficient in this case; the distribution of the various speeds within that average also needs to be considered. The outliers are significant. Statisticians have a term for this called Kurtosis.

Coming up with a precise theoretical description has been harder, but it's now thought it has something to do with the way the statistical outliers affect hydrogen bonding.

Either that or it's merely a ruse that allows scientists to order loads of tubs of Ben & Jerry's at the expense of the institutions that employ them.