The Curve Everything Falls Towards

Plot the binding energy per nucleon against atomic mass and you get a curve that rises steeply from hydrogen, peaks around iron and nickel, and declines slowly out to uranium. Almost everything about nuclear energy is contained in that shape.

Binding energy is how tightly each particle in a nucleus is held, and it shows up as a mass deficit: a bound nucleus weighs slightly less than its separated constituents. The missing mass is the binding energy, related by the most famous equation in physics. Any rearrangement that moves nucleons towards the peak of the curve leaves them more tightly bound, and the difference is released.

Two directions follow. Uranium sits well above iron on the mass axis and below it on the binding axis, so splitting it releases energy. Hydrogen sits far below iron in mass and low on binding, so fusing it releases energy, considerably more per unit mass. Both processes fall towards iron, which is also why stars stop producing energy once their cores reach it, and why the subsequent collapse is what scatters heavier elements across the galaxy.

So why, given that fusion is the better reaction and its fuel is effectively unlimited, has it taken seventy years and counting? Because two nuclei carry positive charge and push each other apart with a force that grows as they approach, while the strong force that would bind them acts only at extremely short range. Getting past that repulsion demands enormous kinetic energy (which means temperature) plus enough density and confinement time for collisions actually to occur.

The sun manages at around fifteen million kelvin, which is modest by these standards, because it compensates with crushing density, the gravitational confinement of two times ten to the thirtieth kilograms of overlying material, and a timescale of billions of years during which an extremely slow reaction rate is perfectly adequate. A reactor has none of that. It must reach above a hundred million kelvin, hold a plasma in a magnetic field because no material vessel survives contact, and produce net power on a timescale of seconds rather than aeons.

One more thing the curve does not explain and everyone gets wrong: radioactive waste. The intuition is that the isotopes lasting longest are the most dangerous, and it is backwards. Activity and half-life trade off directly. Something decaying over tens of thousands of years is, by definition, emitting slowly right now. The intense early hazard of spent fuel comes from short-lived fission products that are largely gone within decades. The long-lived tail poses a genuine containment problem over geological time and a far smaller hazard per gram at any moment.

The same confusion runs through discussion of radiation exposure generally, which is why dose units exist. A count of decays is uninterpretable on its own: an alpha emitter is harmless in your hand and severely damaging in your lung, and a gamma source is the reverse. Dose weights raw activity by the energy actually absorbed and by the biological effect of the particle type. A figure without those weights is not a measure of anything.