Freezing of Hot and Cold Water

Every year at about this time, in some corner bar, the argument will arise as to whether hot or cold water will freeze first when outside temperatures drop out of sight. This usually leads to a five dollar bet and a trip outside with two cups of water--one hot and one cold.

Professor T. Neil Davis, who began this column in 1976, has dealt with the question here before, but it bears repeating because of the continuing interest which it generates and the fervor of the advocates on both sides of the argument.

If a person is scientifically inclined, he tends to scoff at the proposition that hot water can freeze before cold, when subjected to the same sub-zero temperature. But a person who has had actual experience in the matter will insist that it happens.

It does, but only when the conditions are right. Dr. Davis, in his own experiments, found that, under most circumstances, the cold water froze first. But in one trial, it did not. Two Styrofoam cups, with only the water surface exposed to the air (the sides being insulated by more Styrofoam) were placed in a freezer at 5°F (-16°C). One contained near-boiling water while the other held water at 66°F (19°C). The hot water froze first.

In heat-conducting metal containers with their sides exposed, the cold water always froze first. But conductivity of the container is not the only factor involved. An article in the September, 1977 Scientific American makes the point that dissolved gasses retard the cooling rate of water, and boiling removes the gasses. This means that, all else being equal, hot water cools at a faster rate than does cold. Also, hot water circulates more vigorously than cold, bringing more water into contact with exposed container walls and to the water surface.

The single most important factor when hot water freezes first, though, is evaporation. This is underscored by Davis' finding that the only time he could get hot water to freeze first was when only the water surface was exposed to the air, and the heat lost through the walls of the container was insignificant. Evaporation is a cooling process, and hot water evaporates more rapidly than cold, thus removing heat at a faster rate. In addition, evaporation removes water from the container, resulting in a smaller mass to cool.

In enclosed containers, evaporation is not a factor. Why is it then that commonly the hot water pipes freeze and burst in a house and the cold water pipes do not? One reason is that heating has removed the dissolved gasses and the cooling rate is increased, just as it is in open containers.

Dr. Davis makes the further point that relatively pure water with the gasses and impurities removed will cool to several degrees below 32°F (0°C) before it actually freezes--a phenomenon known as supercooling. When it does freeze, however, it does so very rapidly, expanding in volume and often breaking pipes. Although cold water pipes may actually freeze first, damage does not as often result because the rate of ice buildup is comparatively slow, allowing the water to flow down the center and be forced out of the ends of the freezing section of pipe.

So under most circumstances, it will be the cold water that freezes first, but don't bet the bar bill on it unless you've taken the trouble to boil it beforehand--especially if your opponent insists on using Styrofoam cups with the sides covered.