How does a refrigerator cool food?

How does a refrigerator cool food?

Does a refrigerator manufacture cold, or does it move heat? If the kitchen becomes slightly warmer while the food becomes colder, that observation offers us a clue.

Inside the refrigerator, a fluid called refrigerant passes through an evaporator at low pressure. There it boils at a low temperature and absorbs heat from the compartment. The refrigerant carries that energy away as a gas.

Why boiling, though? A cold fluid could simply be run through the box and warmed up. It could — but notice how much less it would carry. Warming a liquid takes a modest amount of heat per degree, and you only have so many degrees to play with before the fluid is no colder than the food. Turning that same liquid into vapor takes an enormous amount, all of it absorbed at a single temperature. For R-134a, long the standard household refrigerant, boiling a kilogram absorbs about 217 kilojoules, comparable to what you would need to cool that same kilogram of liquid through more than a hundred degrees — degrees it does not have. So the phase change, not the temperature change, is what actually carries the heat out of your food. Everything else in the machine exists to make that boiling happen in the right place and then be undone somewhere else.

Which raises the real question. If boiling absorbs heat, and the refrigerant boils inside the box, what stops it from simply staying a gas forever? It must be condensed again, and condensing gives the heat back — so it has to condense somewhere you do not mind heating, namely the kitchen. But heat only flows from hot to cold. The refrigerant left the compartment at around freezing; it cannot warm a twenty-degree room. Something must make it hotter than the room first.

That is the compressor's job, and it is worth stating precisely, because "it makes the gas hot" hides the mechanism. Boiling point is not a fixed property of a substance — it depends on pressure. Squeeze the refrigerant hard and you raise the temperature at which it will condense, up above room temperature; let the pressure drop and you lower the temperature at which it will boil, down below the temperature of your milk. The compressor and the expansion valve are two ends of the same trick. A refrigerator is one substance with two boiling points, one deliberately set beneath the food and one deliberately set above the kitchen.

Why then must the outside coils feel warm? They release not only the heat removed from the food but also the energy the compressor put in. The cycle does not destroy heat; it transports it from a cooler place to a warmer one by doing work.

The refrigerant's changes of pressure and phase make heat transfer possible at useful temperatures. Leaving the refrigerator door open cannot cool a closed kitchen: the appliance returns the extracted heat plus the compressor's added energy to the same room. Follow the energy and the arithmetic is unforgiving — every joule taken from the compartment comes back out of the coils, and the electricity comes out with it, so an open fridge is a small electric heater with extra steps.

Which brings up the figure that reliably sounds like a swindle. The ratio of heat moved to work consumed is called the coefficient of performance, and for a household refrigerator it is somewhere around two or three — meaning the machine shifts two or three joules of heat for every joule of electricity it draws. Air-source heat pumps for a house do better still. Is something being created from nothing? No, and noticing why is the whole point: the electricity is not being turned into cold. It is paying for transport. Energy is conserved exactly — heat out of the coils equals heat from the food plus work supplied — and the second law is respected too, since work was done to push heat the wrong way. The ratio can exceed one because it is not an efficiency in the usual sense; it compares a quantity moved against a quantity spent, not output against input. There is still a ceiling, set by Carnot and by the temperature gap the machine works across: the smaller the gap, the more it can move per joule, which is why keeping the coils clean and the room cool matters, and why a freezer costs so much more to run than a fridge. Real appliances land well below that ceiling.

One last thing the tidy diagram conceals: which substance you loop around it has been repeatedly and consequentially wrong. Early refrigerators used ammonia, sulfur dioxide and methyl chloride, and a leak could kill a household, which is why the chlorofluorocarbons introduced in the 1930s were greeted as a triumph of safety — non-toxic, non-flammable, chemically inert. That inertness turned out to be the problem. Being unreactive, CFCs survived long enough to reach the stratosphere and destroy ozone there, which the Montreal Protocol addressed in 1987 by phasing them out. Their replacements, the HFCs such as R-134a, leave ozone alone but are potent greenhouse gases: a kilogram of R-134a released does the warming work of something over a thousand kilograms of carbon dioxide, which is why the 2016 Kigali Amendment set about phasing those down in turn. Most new domestic refrigerators have moved to isobutane, a hydrocarbon with negligible warming potential — and which is flammable, so the charge is kept small. Three rounds now of solving one hazard and inheriting the next.

• Why frost on evaporator coils reduces efficiency, and how a frost-free model gets rid of it.
• How absorption refrigerators cool with a gas flame and no compressor at all.
• Why a heat pump's advantage over a resistive heater shrinks on the coldest days you most need it.

Every term the collection defines is gathered in the glossary.