Fusion power is what makes the stars shine. Recreating the process on earth in a controlled and sustained way would bring clean, safe and abundant energy. The technological challenges are formidable – such as how to handle a plasma at extremely high temperatures – but new materials and improved computational capacity is beginning to tip the scales.

The nuclear power employed today is based on fission – the splitting of nuclei into smaller parts. When a heavy atom such as uranium is bombarded with neutrons it will split into lighter elements, release more free neutrons and large amounts of energy. The free neutrons split other nuclei, causing a self-sustaining chain reaction that has to be continously moderated and controlled in a power plant. But there are other ways to harness energy from nuclear reactions.

The vision of free energy

Fusion reactors harness the power generation taking place within the stars.

Energy can also be released when lighter nuclei  fuse into heavier elements. This process, called fusion, is what powers stars. Harnessing fusion power has been a holy grail of energy research for decades. The reasons are that fusion, unlike fission, does not generate long-lived radioactive waste, no greenhouse gases are emitted, and the amount of available fuel is basically unlimited. Hydrogen isotopes fusing into helium under immense pressure is the typical setup. There is also only a limited amount of fuel in the reactor at a time, and any disturbance would cause the process to fizzle out instead of running amok.

Plasma: fusion power’s hot potato

The major difficulty is the extreme temperatures needed for fusion to occur. In order to make the reaction happen, a plasma has to be created and maintained – a high-temperature gas where ions and electrons have separated from each other. Because of this, plenty of energy has to be put in before ignition even happens.

Once the reaction starts, the plasma has to be kept contained in order for it to continue, but no materials are able to withstand temperatures exceeding one hundred million degrees. The option that has been most reviewed is to contain the plasma in a strong toroidal magnetic field, in a so called tokamak (top image). The released energy is absorbed by a mantle. There is an alternative method called inertial confinement, that would use hundreds of lasers for compression. This method, too, would require plenty of energy and precision.

Size matters

A larger reactor would be more stable and able to reach higher temperature, but would also be a much more complex and expensive construction. Because of this, no reactors large enough to ignite and produce energy continously have yet been built. A number of research facilities have been built around the world, though – more than 200 tokamaks. The Jet project in the UK was the most important for a long time.

The German fusion reactor Wendelstein 7-X has recently verified that the twisted magnetic fields are working as intended. Photo: Max-Planck-Institut, Tino Schulz [CC BY-SA 3.0]

In September 2016, the MIT research tokamak Alcator C-Mod set a new record for plasma pressure. This was on the last day of its operations, but other facilities are taking over where Alcator left off. A smaller facility in Germany recently recorded a successful achievement. The reactor, Wendelstein 7-X, is not a tokamak but a stellarator, meaning that the toroidal confinement is twisted like a moebious strip.

A tokamak operates in pulses, but a stellarator would be able to operate continously. The magnetic field in such a confinement would be more stable and provide a more efficient process, but the complex geometry makes the construction a more challenging feat. In an important first step, the magnetic fields have been shown to behave as intended.

Iter – an international megaproject

In the south of France, a facility called Iter is under construction. The project is a collaboration between seven member entities — the European Union, India, Japan, China, Russia, South Korea, and the United States. It involves thousands of scientists, and the first plasma is supposed to be achieved in 2020. The project has dragged on, but the plan is for Iter to generate 500 MW and deliver five to ten times the energy input. In 2035, construction of a full-scale prototype, Demo, will commence.

Iter will use deuterium from sea water as fuel, together with tritium which will be produced in the reactor. An alternative would be to use helium-3 instead of tritium; this would cause less radioactivity and would make it possible to simplify the reactor design. Helium-3 is extremely rare on Earth, but abundant on the Moon. It has been suggested that it might be feasible to bring raw material back from there . (Read more in the article ”Asteroid mining closer to reality”.)

We have mentioned Iter before (in the article ”Fusion harnesses the sun’s power”), but even though that facility remains the focus of attention for fusion, the conditions are evolving.

Computational power is opening doors

Modelling the behaviour of the plasma is in many ways a more complex undertaking than the actual fusion reaction. Higher-performing supercomputers is bringing commercial fusion power closer to reality.

Being able to control the plasma is key. To do that, we need to model its behaviour, which is very challenging computationally since all the charged particles interact with each other all the time. Now, supercomputers are able to model even complex designs such as the stellarator with reasonable precision and in reasonable timeframes. The advances in computing power is making fusion power much more attainable, and improved materials are helping as well. Both China and Korea have mentioned fusion as a priority in energy research. Private companies are also emerging on the scene, developing alternative fusion power concepts that may be useful for smaller reactors: Tokamak Energy in the UK, Helion Energy in the USA and Canadian companies Tri Alpha Energy and General Fusion, to name a few.

We know that fusion works, and the advantages are obvious. What is not yet known is whether the technology can be efficient enough to be industrialized. While the costly facilities and decades of research and development may seem discouraging, the bets have to be compared to the potential win: a clean and carbon neutral energy source, available everywhere for millions of years to come.

The article was published in March 2017.