Nuclear Fusion

The Sun converts roughly 600 million tonnes of hydrogen into helium every second, releasing 4 x 10^26 watts. Replicating this on Earth, at a controllable scale, has been the ambition of fusion research since the 1950s. As of 2025, no fusion device has produced net electricity.

Nuclear fusion combines two light nuclei to form a heavier one, releasing energy because the product sits higher on the binding energy curve (more tightly bound) than the reactants. The most accessible reaction, deuterium + tritium, yields 17.6 MeV per event, roughly 4x more energy per nucleon than fission. But fusion requires temperatures above 100 million °C to overcome electrostatic repulsion between positively charged nuclei. Maintaining a plasma at those temperatures long enough to produce net energy (the Lawson criterion) remains the core engineering challenge.

Deuterium is abundant in seawater. Tritium must be bred from lithium in the reactor. Fusion produces no long-lived radioactive waste and no chain reaction risk (any disruption stops the reaction immediately).

Worked Example

The NIF laser facility in 2022 delivered 2.05 MJ to a fuel pellet and produced 3.15 MJ of fusion energy, a gain of 1.54x.

• Check net energy. Fusion output exceeded laser input: 3.15 / 2.05 = 1.54.

But the lasers consumed about 300 MJ of electricity. What is the wall-plug gain?

3.15 / 300 = 0.0105, or about 1%. The physics milestone (ignition) is real; the engineering gap to net electricity is roughly 100x. Commercial fusion would need a gain factor of at least 10-30 at the plasma level to be viable after all system losses.

Fusion represents the theoretical ideal of abundant, clean energy, but it does not yet exist as a power source. Policy must plan with proven technologies while supporting research into unproven ones.

---