The Tax on Every Transaction

The first law says energy is conserved. Every joule that goes into an engine is still somewhere afterwards. This is exactly true and it makes the ordinary phrase about using energy up sound like a mistake, which raises an obvious question: if nothing is lost, what exactly are we buying when we buy fuel?

The answer is concentration. Energy in a tank of petrol is packed into chemical bonds at enormous density. Energy in the air a hundred metres behind a moving car is the same quantity spread across an immense volume at a temperature barely above ambient. Both are energy; only one can be made to do anything. What a fuel supplies is not energy but energy in a useful arrangement, and what every conversion destroys is the arrangement.

This is the second law, and it is worth stating carefully because it is usually stated badly. It is not a mechanical rule forbidding certain motions. It is a statistical claim: there are overwhelmingly more ways for energy to be spread out than gathered, so systems left alone drift towards spread-out states for the same reason shuffled cards are not in order. Order can and does increase locally (a growing organism, a freezing pond, a refrigerator) always at the cost of a larger increase elsewhere.

Sadi Carnot worked out the practical consequence in 1824, remarkably, before the concept of energy had been properly formulated. He reasoned about what would have to be true if perpetual motion were impossible, and derived a ceiling on the efficiency of any heat engine that depends only on the two temperatures it operates between. Not on the working fluid, not on the mechanism, not on the cleverness of the designer. A limit derived two centuries ago still binds engines that had not been imagined when it was written.

Which explains something that looks like waste and is not. A modern combined-cycle gas plant reaches about sixty per cent efficiency, and the remaining forty per cent leaves as heat through a cooling tower. That is not sloppiness awaiting better engineering. Rejecting heat to a cold reservoir is how the cycle closes: a plant with nowhere to dump heat produces no work at all. Improving efficiency means raising the hot temperature, which runs into the melting point of turbine blades, not eliminating the cold side.

The most counterintuitive consequence runs the other way. A heat pump can deliver three or four units of heat for each unit of electricity it consumes, which sounds like a violation and is not. It is not converting electricity into heat; it is using electricity to move heat that already exists from outside to inside. There is thermal energy in air at any temperature above absolute zero, and pumping is far cheaper than creating. The figure quoted is a ratio of transport to work, not an efficiency of conversion, and confusing the two is why the technology sounds too good to be true.

Engineers summarise the three laws as: you cannot win, you cannot break even, and you cannot get out of the game. It is a joke, and as a summary of what constrains every energy system ever built, it is also close to complete.