Heat Engines and Cyclic Processes

Picture a gas trapped in a piston, run through a cycle on a PV diagram: it expands, maybe cools, compresses back, and returns to its starting point. Because it returns to the same state, its internal energy doesn't change over one full lap. But the path it takes matters: on a PV plot, a clockwise loop encloses an area, and that area is the net work the gas does on its surroundings in one cycle. Heat goes in at the hot parts of the cycle, heat comes out at the cold parts, and the difference between those two is exactly the work you get out.

That outgoing heat is not a flaw you can engineer away. The second law of thermodynamics says a gas can't just absorb heat and convert all of it to work in a repeating cycle, because that would leave the gas permanently more ordered with no offsetting disorder dumped elsewhere. So every real cycle has to exhaust some heat to a cold reservoir, which is why efficiency e = 1 - Qc/Qh is always less than 1, even for an engine with zero friction. Friction just makes things worse; removing it only gets you to the theoretical ceiling, not past it.

Run the exact same cycle backward, counterclockwise on the PV plot, and the engine becomes a refrigerator: now you put work in to drag heat from a cold space to a hot one, since heat won't flow that direction on its own. This backward cycle is the bridge from engines to refrigerators and heat pumps, same loop, same areas, just the arrows reversed.