Why Nucleation for Water to Freeze?

Most recent answer: 12/06/2014

Q:
Why do water molecules need an impurity to form crystals (or freeze) ? I get that it does need an impurity to start the process of forming the first crystal and that all the other crystals follow to form afterwards, but why is this impurity needed? I mean what does it do to the molecules of water? And can water molecules form without this impurity?
- Nada (age 14)
A:

Below about -40°C, it turns out water will just freeze on its own without any special nucleation site. Yet a liitle below freezing, say -5°C, it takes more or less forever without some nucleation. So to answer your great question we need an explanation that works over this whole range. An accurate explanation may involve a bit more background than you were expecting.
(For related questions see: .)

Water, like other materials, in the very long run settles into what's called the lowest free-energy phase on its own. Part of that free-energy is ordinary energy, so settling in is exactly like water settling into the bottom of a bowl after sloshing around a bit. Part of minimizing the free energy is maximizing something called the entropy, which is a measure of how many different states the parts can be in. When water evaporates, the molecules actually pick up more energy but they get so many extra states by running around in the big atmosphere that their free energy still goes down.

Above 0°C, the extra entropy that the water gets from tumbling around in a liquid beats the energy it could lose by freezing into a crystal, so it stays liquid. Below 0°C, the energy loss is more important, so the stable phase is the crystal.  So now we ask, why doesn't the crystal just go and form at say -5°C?

The reason is that the first few molecules that lower their energy by lining up into a crystal-like pattern don't lose as much energy as they would by joining a full-size crystal. The surface of that cluster is still surrounded by liquid water, not lined up. Yet by lining up the molecules lose just about as much entropy as they would lining up in a big crystal. So a molecule gains free energy by joining a very small crystal. That means that these tiny crystals aren't stable. Until a medium-sized crystal forms, the molecules have no way of starting on the steady downward path in free energy toward a big crystal. 

When the water is cold enough, the energy loss of just a very few molecules lining up is already enough to beat the entropy loss, so the crystal can just start to form on its own. At say -5°C it takes so many molecules lining up, gaining free energy, that it almost never happens on its own. That's why some sort of dust or bubble, already lining up some of the molecules, is needed to get the freezing started.

Mike W.


(published on 12/06/2014)