The main point of this chapter is that macroscopic phenomena such as pressure and heat transfer are caused by collisions of molecules. FIGURE 20.15a shows two frames from a simple “movie” of a collision between two gas molecules. FIGURE 20.15b is the same movie played in reverse; this is also a possible molecular collision. Nothing in either movie looks wrong, and no measurements we might make would reveal that either of these two collisions violates any laws of physics. We can’t label one of these movies as “real” and the other “impossible.” In other words, interactions at the molecular level are reversible.

FIGURE 20.15 Two movies showing that molecular collisions are reversible.

Full Alternative Text

Now contrast this with movies of any of the processes we described earlier. A movie in which a puddle of water turns into an ice cube while a nearby thermometer shows a reading of 25°C is clearly running backward. It shows an impossible process. But what has been violated in the backward movie?

Processes such as phase changes and heat transfer are molecular processes. If molecular interactions are reversible, why are macroscopic processes—which consist of molecular interactions—irreversible? If reversible collisions transfer thermal energy from hot to cold, why don’t they ever transfer thermal energy from cold to hot? This is a more exact framing of the question we need to answer.

As we begin to consider this question, you may recall that the thermal energy of a macroscopic object is the total energy of all the moving atoms and molecules that make up the object. However, knowing the thermal energy tells us nothing about how the energy is distributed among the molecules. FIGURE 20.16a shows a container of gas in which one molecule is moving while all the others are at rest. The thermal energy of the gas is simply the kinetic energy of the one moving molecule. This is one possible way that the energy could be distributed, but this is not a stable, equilibrium situation. The moving molecule will collide with other molecules, giving them kinetic energy while the original fast molecule loses energy. Eventually, we expect that every atom in the gas will be moving, something like the situation shown in FIGURE 20.16b. The thermal energy is the same in both cases; it’s just distributed differently.

FIGURE 20.16 Two different distributions of the thermal energy of a gas.

Let’s think about what happened here. Initially, the system’s thermal energy was concentrated in a single molecule. As time went on, collisions and interactions caused the system’s energy to spread out, or disperse, until it was shared among all the constituents. And this spreading will not spontaneously reverse. Ordinary molecular collisions cause a concentrated energy to spread out, but we never see a system spontaneously concentrating dispersed energy into a single constituent. The individual collisions may be reversible, but the dispersal of energy is an irreversible process.

All thermal interactions involve the spreading of energy. Suppose a hot object and a cold object are brought into thermal contact, as in FIGURE 20.17. The thermal energy is initially concentrated in the hot object; it is not shared equally. As we’ve seen, collisions at the boundary transfer energy from the more-energetic molecules on the warmer side to the less-energetic molecules on the cooler side until the thermal energy is shared equally among the constituents of the combined system. After some time, thermal equilibrium is reached, which means that no additional sharing or spreading is possible.

FIGURE 20.17 A thermal interaction causes energy to spread until it is shared equally among all the molecules.

The spreading of thermal energy happens spontaneously—it’s simply what happens when macroscopic systems interact. You can imagine the heat-transfer arrows of Figure 20.17 being reversed, making the cold side colder and the hot side hotter. But that would concentrate the thermal energy rather than disperse it. The concentration of thermal energy does not happen spontaneously, and thus heat is never transferred spontaneously from a colder object to a hotter object.