Examining water’s unusual physical properties and why they matter in refrigeration.

We are surrounded by a chemical so commonplace that we regularly make the mistake of presuming that its behavior is normal and that other chemicals behave in the same way. This can sometimes lead to embarrassing mistakes and even catastrophic failures, so it is good to step back and take stock of the weirdness. The chemical is, of course, water.

Some of the weirdness makes water useful in refrigeration and air conditioning, but in other ways causes problems.

The first odd property of water is density. As water changes from liquid to solid its density suddenly reduces by about 10%. This is due to the crystalline nature of the solid, where the water molecules are less closely packed than in the liquid and it explains why ice floats on liquid water. It’s a good job that it does because otherwise ponds, lakes and rivers would freeze from the bottom up and life, as we know it, wouldn’t work.

I remember when I was small being fascinated by the milk bottles on our doorstep in winter. When the milk froze it extruded a column of what looked like ice cream about 3 in. (76 mm) high out the top of the bottle. The column still had the foil top on it if we got to it in the morning before the birds did.

What’s even more strange is that the density of water prepares for the step change at freezing—as the temperature approaches 32°F (0°C) from above the density reaches a maximum value, at about 39°F (4°C) and then starts dropping before the liquid solidifies. This provides relatively stable temperatures, above freezing, in depths of water and means that rivers continue to flow even when the air temperature falls far below freezing.

It often comes as a surprise to people that other liquids don’t behave this way when they solidify. We are often asked if pipes will burst when carbon dioxide drops to its triple point of –69.5°F (–56.4°C). The answer is “no” because the density of solid carbon dioxide is the same as the density of liquid at that temperature. When liquid carbon dioxide at saturation reaches the triple point some of it turns solid but the rest of it becomes gas. The “solid” is actually fluffy snow that gives the appearance of rapid expansion. Water doesn’t do this, at least not under any normal circumstances.

Another weirdness is in the latent heat of water. This is the amount of heat per pound of water that needs to be added to turn it from liquid to gas. The latent heat of water is huge. This is because the energy required to break the bonds holding water molecules together in the liquid form is very high.

This is also key to life; it is the reason why sweating is such an effective method of temperature control. The only liquid I can think of with a latent heat close to that of water is ammonia, but even it is only 61% the latent heat of water. These are both polar molecules, which is why the liquid bonds are so strong.

In contrast the latent heat of R-134a at atmospheric pressure is less than 10% of water. Water’s specific heat capacity is also very high—again one of the few liquids that compares is ammonia, which is actually about 5% higher than water. Also again, R-134a’s specific heat is only 30% of water’s.

The third property in this trio of weirdness is what happens to the freezing point of water when other chemicals are added. Lots of things when added to water cause the freezing point to drop, for example salt, sugar, alcohol and glycol. The solid that forms on freezing is pure water, but it freezes below the usual 32°F (0°C) freezing point.

As this pure water solid is removed from the solution the remaining additive concentration gets stronger and so the freezing point drops even further. This is the key to many of the mysteries of good and bad refrigeration. I can’t think of anything that raises the freezing point when added to water—can you?