Why is fusion not part of addressing low-carbon energy production today? Put simply, because it is very challenging. To make fusion happen, a gas of hydrogen fuel has to be heated to extreme temperatures (>100mn °C). Keeping this very hot gas (actually, a plasma) well-confined and stable enough to sustain the conditions for fusion is hard. The most advanced way of making and confining this plasma are in facilities called tokamaks – where powerful magnetic fields are used to hold the plasma in place in a ring-shaped container.

Fusion has been proven to work in large laboratory experiments, such as the Joint European Torus (JET) facility sited at UKAEA’s Culham Campus. It is now being developed on an industrial scale with the international ITER project now being built in the south of France by a consortium of 35 countries. Alongside the government-funded research programmes, there is a growing private sector fusion industry with more than 40 fusion companies, who have together attracted over $6bn in investment.

Numerous landmark results have been achieved by the fusion community in the last two years, which have increased optimism. Just last month, we announced that JET produced 69 MJ of fusion energy. While the reaction was only sustained for five seconds due to the limitation of the copper coils used in JET, the experiments matched predictions made in advance, which gives confidence for next-step machines. Around the same time, researchers in China demonstrated a 1,000 seconds long fusion pulse using super-conducting magnets, clearly showing the sustainment of a fusion reaction. Finally, the Lawrence Livermore National Laboratory demonstrated fusion ignition, where more energy from fusion was released than the laser energy absorbed to drive fusion. While all are important results, we are still some way from commercial demonstration of net electricity production.

Numerous landmark results have been achieved by the fusion community in the last two years, which have increased optimism.

Delivering a viable fusion power plant is one of the holy grails of science and engineering, but it represents an unprecedented technical challenge. Commercial fusion requires many engineering challenges to be met simultaneously: (1) the sustainment of a controlled plasma over long timescales with fusion-born helium dominating the plasma heating; (2) the controlled exhaust of heat and helium ‘ash’ from the plasma core; (3) the development of structural materials which have to sustain, for many years, high temperatures (>150mn °C) and pressures in the presence of exceptionally intense neutron fluxes, without generating unmanageable volumes of radioactive waste; (4) the ability to generate its own fuel; and (5) the requisite high availability and efficiency of the machine and its systems to produce a viable levelised cost of electricity.

Fusion is different from most other technologies in that a full test is only possible in a complete power plant, and the cost and timescale of each step means that a succession of small-increment full physical prototypes is unrealistic. This means that comprehensive in-silico design, digital prototypes, and finally models of components and systems, are needed to support qualification of the solutions.

A number of nations are prepared to take on these challenges in the quest to deliver the first fusion power plants, with many governments publishing ambitious strategies in the last two years. The most ambitious public programmes are the UK fusion strategy, which aims to build a compact prototype powerplant called STEP; the US White House decadal vision for fusion and a subsequent public-private partnership programme; and the Chinese Fusion Engineering Test Reactor – all of which aim for prototype plants commencing operations around 2040.

Despite the challenges being so complex, it is still worth pursuing multiple different pathways to reaching the goal of putting fusion on the grid. There are a growing number of fusion companies pursuing a diverse range of approaches to fusion, of which 93% are aiming to deliver a fusion prototype device during the 2030s. And many are making impressive progress. The penetration of fusion into the market is almost impossible to predict when looking beyond the 2040 timescale, but one could extrapolate from other comparable large technologies, such as early adoption of oil and gas, or early adoption of conventional nuclear fission powerplants. If it were valid to use such previous growth rates as a guide, fusion could reach 10% of global energy demand by 2100 – equivalent to supplying all of Europe – or as much as 40% if capital costs were 30% cheaper.

The past two years have been a watershed moment for fusion, which has engendered excitement in the sector. The combination of producing clear national strategies, meeting major technical landmarks and escalating investment is driving the sector at renewed pace, with a marked growth in those pursuing fusion. In the UK, the next phase of STEP will see UK Industrial Fusion Solutions seek long-term partnerships with an engineering partner and a construction partner to build a full prototype powerplant in Nottinghamshire, a major opportunity for the conventional energy industry to be at the forefront of delivering fusion too.

Fusion has many hurdles to overcome, but it offers such huge potential in providing abundant, low-carbon, baseload energy, including high grade heat, that it is surely worth pursuing the mastery of these challenges. One of the forefathers of fusion, Lev Artismovich, once said that ‘fusion will be ready when society needs it’. Today there is renewed optimism that fusion can deliver on its promise.

The views and opinions expressed in this article are strictly those of the author only and are not necessarily given or endorsed by or on behalf of the Energy Institute.

• For more detail about recent developments in the development of fusion, see ‘The nuclear fusion race – one small step nearer?'
• Further reading: ‘The new business of fusion’. A profile of the private-sector nuclear fusion industry mainly based in the US.