Can a refrigerator be used to cool a room? Most people know that it cannot. But why not? What follows are two simple arguments for why you can't use a refrigerator to cool a room, and a longer explanation of how a refrigerator can work in its normal way of functioning.

The ultimate reason why a refrigerator can't be used to cool your kitchen is the Second Law. One short explanation is this: In all spontaneous processes, the total entropy must increase. In the normal operation of a refrigerator, the inside of the refrigerator gets cooler, meaning that the entropy of the contents of the refrigerator decreases. If the inside of the refrigerator is the system, S_sys<0. But the Second Law tells us that the total entropy, S_tot=  S_sys + S_surr, must be greater than zero, so the entropy of the surroundings must increase more than the entropy of the system decreases (S_surr > - S_sys). So the kitchen, the refrigerator's surroundings, must get warmer, even as the inside of the refrigerator gets colder. As the book points out, q=TS; this means that even if S_surr is approximately equal to -S_sys, the positive q of the kitchen will be larger than the negative q of the refrigerator because the kitchen is at a higher temperature. In other words, more heat goes into the kitchen than comes out of the refrigerator. Where does this heat come from? From the electrical energy that we get out of the plug in the wall, of course; we'll discuss this more in a bit.

Putting more heat into the room than you get out of the refrigerator is fine for normal operation of the refrigerator. But when you open the door to the refrigerator, you've removed the barrier between the system and the surroundings. It's all surroundings, or perhaps all system. And according to the Second Law, the only thing that can happen to this new refrigerator-kitchen system is for entropy to stay the same or to increase. And an increase in entropy means, in the absence of any phase changes or chemical reactions, an increase in temperature; the refrigerator-kitchen can't get cooler. The best it could do is not get any warmer (S=0), if the refrigerator were unplugged or turned off. But if the refrigerator is plugged in, the thermostat will make the refrigerator run, and this process (like any spontaneous process) will generate entropy and heat, warming up the refrigerator-kitchen system.

The source of energy to make the refrigerator-kitchen warmer is, of course, the electricity that the refrigerator runs off of. An alternative argument for why you can't cool a kitchen with a refrigerator follows from simply noting that the refrigerator is plugged in. The First Law states that energy is conserved. If we assume that our kitchen is reasonably isolated from the larger surroundings (closed doors and windows), we can see that energy entering the room must stay in the room. This might imply that the room might not get warmer, if we had some nice way of storing all that energy, but refrigerators aren't really built for storing energy--they're actually built to put out heat, as we saw in the first argument above. But even if they were designed for storing energy, the Second Law demands that no spontaneous conversion of energy for storage?say from electrical current to the kinetic energy of a flywheel, or from electrical current to the chemical potential energy of a recharging battery?can occur without an increase in entropy. This means that some of the energy will be released as heat. So if energy is coming into the room (our system), even for purposes of just storing the energy, the room will get warmer. Attempts to cool the room by opening the refrigerator door won't change that; energy is still coming into the refrigerator-kitchen system.

We now have two more-or-less independent ways of concluding that the room has to become warmer, both of which rely on the Second Law. But both explanations raise other interesting questions: How is entropy inside the closed refrigerator decreased even while the entropy of the surroundings are increased? How is the energy from the electrical plug used to generate heat so that the interior of the refrigerator can be cooled? The answer to both of these questions, is, of course, the workings of the refrigerator itself. An examination of these workings gives us more chances to examine applications of the laws of thermodynamics, and to arrive at the same conclusions by a more thorough understanding.

The key to the workings of a traditional refrigerator is the refrigerant. Chlorofluorocarbons, hydrofluorocarbons and ammonia have all been used as refrigerants; the important thing is that they have a phase change between the gaseous and liquid phases at the right temperatures and pressures. The refrigerant is cycled through a closed loop; part of the loop is essentially "inside" the refrigerator, and part is "outside" the refrigerator, in the kitchen. In looking at the workings of this loop, it is usually convenient to regard the refrigerant as the system, and to regard the kitchen or the refrigerator interior as the surroundings, depending upon which part of the loop we're interested in.

On the outside half of the loop, we want the refrigerant to transfer heat, and entropy, to the surroundings. An exothermic reaction might accomplish this, but a phase change such as condensation would be much simpler. We could get a gas to condense just by cooling it, but this would mean we'd have to rely on the kitchen temperature being lower than the boiling temperature of the refrigerant, thanks to the Second Law. As we'll see later, we want a refrigerant that has a low-pressure boiling temperature that is lower than the target temperature for our refrigerator?and if the refrigerator has a freezer compartment, this means we actually want it to be lower than the target temperature there, usually -20ûC.

Instead of requiring the kitchen to be at -30ûC (in which case we wouldn't need a refrigerator in the first place!), we condense the refrigerant by applying pressure. Boiling points decrease with increasing pressure, because the difference in entropy between a more compressed gas and a liquid is smaller than that between a less compressed gas and a liquid. Remember that G = H - TS. For a condensation, H < 0, so to get a negative free energy for condensation, we just have to make sure the -TS term isn't too positive. S < 0 for a condensation, so -TS is positive; G < 0 then requires that either T has to be low, or we need to make S a smaller negative number. Compression of the gas makes S less negative, so condensation can happen at a higher temperature. From the standpoint of the Second Law, this condensation is spontaneous because the entropy of the surroundings increases more than the entropy of the system, our refrigerant, decreases. More precisely,