Nearly all fusion designs involve fusing isotopes of hydrogen to produce helium. There are two main isotopes of hydrogen: deuterium and tritium. All hydrogen isotopes have one proton by definition. For regular hydrogen, that’s all it has. Deuterium also has a neutron and tritium has two neutrons.

The easiest, lowest temperature, and highest yielding approach to fusion combines an atom of deuterium with an atom of tritium (D-T). When they fuse, they combine to form a helium atom (2 protons + 2 neutrons), and the extra neutron carries away most of the energy. It’s this neutron that we use to heat up a steam turbine cycle to produce electricity.

Deuterium is found in all forms of water on Earth. For every million molecules of regular H2O, 150 of these molecules will be deuterium (0.015%), and can be easily extracted. However, tritium does not occur naturally in any significant quantity – and so you have to make (or “breed”) it. If the “spare neutron” emitted by the DT reaction is used to heat up lithium, then it will make more tritium, as well as a bit more helium (See “Where will you get Tritium”).

As long as you have a little bit of tritium to get going then the rest of the fuel will be continuously manufactured in the fusion power plant itself, also called “closing the fusion fuel cycle.” Just twenty or thirty car batteries worth of lithium would be used up per year to create enough tritium to power an entire city.

The challenge is that you need to hold the fuel at extremely high temperatures, and at the right density, for long enough for net energy gain. When the right conditions are met, you start to self-heat and can reach a point called “ignition” where the helium atoms produced in one reaction will deposit enough energy to drive the next reaction – a bit like striking a match. Nearly all fusion power plant designs work when this self-heating or “ignition” point can burn enough of the fuel to generate lots of energy – similar to when that match lights a wood fire. 

We can’t get to ignition using deuterium alone, also known as a D-D reaction, with any known kind of driver (i.e., laser, magnet, etc.) on Earth. The temperatures needed for D-D reactions are too high, approximately 500 million degrees C, and the reactions are too inefficient. So, they require more energy to create, and they release less energy in the process compared to D-T fusion.

In contrast, the D-T reaction is the most efficient fusion reaction possible. D-T fusion ignition occurs at much lower temperatures than D-D designs, a “mere” 150 million degrees C, and is more efficient, releasing about 3.5x as much heat energy. So, the fusion conditions require less energy to create, and the resulting fusion reactions create more heat energy output—it’s a win/win if you’re trying to create electricity.

There are proposals for other fuels that utilize alternate light elements that don’t create neutrons as part of the process, such as deuterium and helium-3 (D-³He) and proton-boron (p-¹¹B). These fuels would mostly produce charged particles, which could enable direct electricity conversion.

But these options need much hotter, better-confined plasmas, and suffer significant x-ray losses that cool the system and reduce confinement. Additionally, they require ultra-pure fuel, which takes more energy to produce, thus reducing the net energy from the entire system. In the case of D-³He, there’s practically none on Earth. The closest source of naturally produced ³He is the moon.

While a neutron-free reaction does simplify the plant design and operation, in practice they’re not truly neutron-free: D-³He makes neutrons via unavoidable D-D reactions, and p-¹¹B will do so if there’s even trace deuterium. Because most of the output is charged particles and x-rays that stop within microns of the interior wall, they create intense surface heating and mechanical stress. So, despite creating fewer neutrons, these systems must overcome extreme first-wall heat-flux and materials challenges before they can rival the D-T pathway demonstrated today.

D-T fusion designs create many neutrons, as explained above. This is good in the sense that they can then breed their own tritium for fuel. But these neutrons are damaging to anything they hit, and so precautions have to be taken to shield the outside environment from them. Luckily, concrete does the job very nicely and inexpensively. Curious about how we’ll protect the fusion chamber from the effects of neutrons, read “What about the first-wall problem?”