# What does the sand do in a pot-in-pot refrigerator?

The Wikipedia article Pot-in-pot refrigerator says that sand in the outer pot somehow affects the cooling efficiency of refrigerator. I don't understand what the sand does in this situation.

For example, compare filling the outer pot with water vs. filling the outer pot with both water and sand. Would these have some significant difference in cooling efficiency, meaning how fast heat is transferred between the inner pot, the middle layer and the outside air?

• The sand acts as a wick for the water, distributing the cooling effect more evenly around the inner container. – Dave Tweed Apr 25 '16 at 13:14
• what do you mean with cooling efficiency? – mart Apr 25 '16 at 14:16
• @mart I mean how fast is heat transfer from inner pot to middle layer and then finally to the air. – fronthem Apr 25 '16 at 14:18
• Have used a similar arrangement to cool beer bottles and cans on the beach using a wet towel and sometimes sand (towel is nicer as it does not dirty up the container). It is a surprisingly effective solution. This is more effective than having the drink in the water itself as evaporation eats up quite much energy and can cool the liquid below the water temperature. – joojaa Apr 25 '16 at 14:35

This type of cooler works by evaporation ie the latent heat of vapourisation of the water draws energy from its immediate surroundings.

You can feel this effect for yourself if you ever get a low boiling point liquid like acetone on your skin (not recommended on bare skin but you can still feel the effect through thin gloves).

Below its boiling point water can only evaporate from it's surface (in fact this is the definition of boiling) so placing it in a porous medium, in this case an unglazed pot greatly increases the surface area available for evaporation.

Again you can see this in the fact that a glass of water will take a long time to evaporate but of you pour it out onto a flat surface it will disappear quite quickly.

Clearly this works bet if the porous medium is damp rather than completely soaked as the idea is to have a thin film of water which is constantly replenished as it evaporates.

The wet sand provides additional surface area and also is an effective way of keeping the outer pot damp but not saturated while allowing conduction of heat from the inside of the vessel to the outside surface.

Looking up some data on the Intartubezz

All units are $\frac{W}{m-K}$

Water at 20° C      0.6

Sand, dry          0.15 - 0.25
Sand, moist        0.25 - 2
Sand, saturated    2 - 4


So you can see that the combination of sand and water is better than either alone.

• It would be useful to define the quantities in your table and provide a link to the source. – Chris Mueller Apr 26 '16 at 12:14
• @ChrisMueller apologies - I didn't notice that putting text in blockquote nullified the LaTex markdown. As to links -- there are a hundred equally useful found via a simple google search – Carl Witthoft Apr 26 '16 at 13:15