In the era of global climate change, and concerns about humanity's long-term reliance on fossil fuels, many think the solution lies in alternative sources of energy, including nuclear power. All our nuclear power plants are based on fission: splitting heavy atoms into lighter components in a controlled fashion. Though fission is safe when all goes well, the fuel is radioactive, waste disposal can be problematic, and as the Fukushima disaster showed there is a high cost to accidents.
Nuclear fusion is in principle cleaner and comes from a cheaper, more abundant fuel source: an isotope of hydrogen called deuterium can be extracted from water and only helium is produced as waste. From The Matrix to SimCity 2000 to political dreamers, fusion has often been seen as an inevitability for society. However, despite decades of work nuclear fusion remains a dream. As the joke goes fusion is the power of the future – and always will be.
That's not because creating a sustained fusion reaction – in which more energy is produced than is required to start and maintain the process – is impossible or even terribly hard to achieve (at least by high-energy particle physics standards). The most infamous example of a fusion reaction is the hydrogen bomb, which sacrificed control and safety for the sake of violence and death. Fusion reactors obviously need to have more stringent requirements.
To see what's needed to create a sustained reaction, let's look to the best-known fusion reactors of all: stars. In the core of a star like the Sun, strong gravitational pressure forces together hydrogen plasma – an equal mixture of protons and electrons. Extreme conditions of 15 million-degree-temperatures and high pressures mean that protons have enough energy to overcome their mutual repulsion for each other, allowing the attractive forces to kick in. When protons fuse together they are converted into neutrons and release a lot of energy.
Bluntly stated, we can't recreate those conditions, even if we wanted to. Stars have a sufficiently large mass to contain the hydrogen plasma by the force of gravity alone, but we don't have that option, so physicists have to confine plasma using electromagnetism instead. Researchers can also start with deuterium or tritium plasma instead of hydrogen to lower the energy required to start fusion. (Tritium is a hydrogen isotope consisting of a proton and two neutrons; unlike normal hydrogen and deuterium, it's unstable and therefore harder to keep around.) However, the temperature and pressure still needs to be high, so it requires a larger energy input than fusion liberates, which defeats the purpose.
Part of the difficulty lies in the nature of plasma. If you put a normal neutral gas, such as oxygen, in a container you can increase both pressure and temperature by compressing it. Plasma, on the other hand, consists of charged particles at sufficiently high temperatures to melt the container walls. Also, without maintaining conditions carefully, the electrons tend to reunite with protons, creating a neutral hydrogen gas that's useless for fusion; it's imperative that the container trapping the plasma contain no gas, for similar reasons.
Some hope lies in using elements other than hydrogen, as these contain more than one proton per nucleus. That increases the electric repulsion and in some cases can make the energy barrier to fusion even higher. While some fusion reactions involving helium, lithium, or boron are areas of active research, a major problem is that these materials are far rarer on Earth than hydrogen.
All is not lost, however. Researchers are pursuing several possible solutions to the fusion problem, mostly based on clever methods for confining or compressing deuterium. The oldest of these is magnetic confinement, in which strong magnetic fields act as the “walls” of a container.