Q&A: Nuclear fusion - Hype or hope for a cooler planet?

1. What is nuclear fusion?

2. How does nuclear fusion differ from nuclear fission/conventional nuclear power?

3. What are the advantages of nuclear fusion?

4. What are the downsides of nuclear fusion?

5. When can we expect to see the first commercially viable fusion reactor?

6. What are the technical hurdles to a functioning fusion reactor?

7. Are fusion reactors and renewables compatible?

8. Are there any other hurdles?

9. What is the EU's role in nuclear fusion?

10. What is Germany's approach to fusion energy?

11. Why does this text reflect the "scientific consensus"?

Nuclear fusion is the process through which light atomic nuclei are fused together to produce heavier nuclei, generating energy in the process. In principle, this process is what happens in the Sun, which also produces energy, in the form of light and heat, through the fusion of atomic nuclei. To replicate this process here on Earth, the fusion of two different varieties of hydrogen atoms (deuterium and tritium) is considered to be the most promising. These light nuclei lose some mass during fusion, which is released as energy.

Nuclear fusion produces an extremely hot mixture of particles, called plasma. The aim is to use this heat to turn water into steam, which can then power conventional steam turbines and generators to produce electricity and/or heat. To set the process of nuclear fusion in motion, temperatures of around 150 million degrees Celsius and very high pressure are needed to overcome atomic repulsion, which keeps atoms apart under normal conditions.

There are various types of nuclear fusion. While so-called magnetic fusion uses powerful magnets to create the conditions necessary for fusion and hold the resulting plasma in place, inertial or laser fusion uses an array of laser beams for the same purpose. All approaches are at an early stage of research and cannot yet be used on an industrial or commercial scale.

The central challenge in nuclear fusion is generating more energy than is needed to trigger it. This has not yet been achieved. In December 2022, a team at the Lawrence Livermore National Lab (LLNL) in the USA reported a positive energy balance (in the narrow sense) for the first time, and a repetition of the experiment produced even better results in 2023. Laser fusion was pursued here, where the fusion reaction is triggered by laser-generated pressure waves in a very small fusion chamber. Temperatures of more than 120 million degrees Celsius are generated in this so-called pellet, a tiny container filled with the fuel, usually a mixture of deuterium and tritium. However, numerous media reports and public debates have ignored the fact that the energy used to generate the laser pulses must also be taken into account in the overall energy balance. If this is considered, far more energy had to be used for this fusion reaction than was generated (see section 6 for the technical hurdles for a fusion reactor).

Nuclear fusion itself is a carbon-neutral source of energy (even if greenhouse gases are currently emitted during the construction of the reactors and the generation of the energy used to power the reaction). In addition, nuclear fusion is a very efficient energy source from a purely material point of view. According to the Max Planck Institute for Plasma Physics, one gram of fusion fuel could theoretically provide as much energy as eleven tonnes of coal.

Compared to wind and solar power plants, fusion reactors would have the advantage of being able to generate a great deal of energy regardless of the weather (this is referred to as secured or guaranteed power, more on this in section 7). They also require much less space. Although there are no exact figures for fusion reactors as they do not yet exist, nuclear power plants, for example, require on average only around 0.4 to 4 percent of the area of solar or wind power plants with the same output, depending on the source (Smil 2015, Lovering et al. 2022).

Two types of fusion technology are currently in development (see also section 1). One is plasma-based fusion reactors, where the kinetic energy of hydrogen nuclei in a very hot plasma is used to initiate the fusion reaction. Two designs exist here, known as tokamaks and stellarators, and there is currently intense debate among researchers as to which variant should be pursued further. Both governments and private companies are currently developing both types of fusion reactors. The forecast date of both groups for a first operational reactor differs significantly.

The ITER facility in France, which is currently being built by Europe, China, India, Japan, South Korea, Russia and the US, in 2024 postponed the target date for energy producing fusion reactions – the goal of the project - to 2039. The first demonstration reactor, which could be connected to the European power grid for testing purposes, is scheduled for operational trials around 20 years later, by in the mid-2050s, according to the European research plan. A fusion reactor that is part of an energy supplier's power plant fleet is expected much later.

But fusion start-ups argue that the ITER timetable is not a good indicator for the technology’s progress. They say the political nature of the project, the complicated coordination between many states, the integration of components delivered from across the globe, bureaucracy, as well as its massive size, make ITER inherently slow. Private companies are not held back by these constraints, allowing them to move much faster.

This is why the timetables of private companies are much more ambitious. US company Helios wants to produce electricity as early as 2028; Commonwealth Fusion, also based in the US, aims for the world’s first grid-scale commercial reactor in the early 2030s; German start-up Proxima Fusion plans to launch a reactor capable of continuous operation in 2031, “opening the door to the commercial application of fusion energy”.

However, ever since the first fusion reactors were planned, schedules have had to be reconsidered. In 2004, for example, the ITER researchers believed that the first demonstration reactor could go online in 2033. Eight years later, they pushed this back to the early 2040s, and in 2018, the reactor was finally announced for the 2050s. A study reviewing the progress of fusion research stated that:

"It is therefore important to state that roadmaps from enterprises, both public and private, are not considered to be credible sources for timescales of fusion. Many developers have published pathways to commercialisation that are a moonshot approach, in other words all milestones are achieved with no showstoppers, delays, and assume that there is adequate funding for all the activities at each step. […] That is not to say that that the timescales are un-achievable, but these are often optimistic in order to attract investment and are lacking in accompanying data that back up predictions.” (Griffiths et al. 2022)

Opening schedules weren’t the only things to be underestimated in the past, but also cost forecasts. In the case of the international ITER project in Cadarache in the south of France, the estimates were around EUR 5.9 billion (excluding the contributions in kind that the project partners in the participating countries are providing on their own responsibility) in 2008. Two years later, the EU Commission already assumed that it alone would have to contribute 7.3 billion euros (as the European Union bears 45.5 per cent of the total costs, this corresponds to a tripling). In 2016, ITER increased cost estimates to 22 billion euros. In May 2022, the European Parliament in a resolution expressed its concern that there was a risk of further "significant cost increases and/or further delays in the implementation of the ITER project". When project management presented a new, postponed schedule in July 2024, there was also talk of further cost increases of five billion euros.

The second type of fusion that is distinct from plasma-based fusion is laser or inertial confinement fusion. The successes of the Lawrence Livermore National Lab in the USA (see section 1) has given this approach strong tailwind. The race for the best, cheapest, most efficient, and most rapidly available concept is thus wide open. A number of start-ups with strong financial backing are aiming for rapid commercial realisation of laser fusion technology. However, estimates of future costs and timetables remain unreliable.

If the first commercially viable fusion power plants are connected to the grid from the 2040s in an optimistic scenario, or from the 2060s in a conservative one, the electricity grid in the major countries is likely to be dominated by a mixture of renewable energies and nuclear power plants. This is because the EU as well as China and the USA have announced their intention to be climate-neutral between 2045 and 2060. Several studies dealing with the future role of fusion power plants assume this scenario (Hamacher et al. 2013; Nicholas et al. 2021). Several other studies show that the respective demand for energy projected in different countries could theoretically also be covered by renewable energy alone (Breyer et al. 2022; Zappa 2019). In any case, there is a consensus among researchers that renewables will cover a very large proportion of global energy demand in a few decades. Fusion proponents argue that rising electricity demand, for example for data centres running AI, will require additional sources to complement renewables.

The question is what role fusion power plants can and should play in a future electricity system. Like gas, coal or nuclear power plants, fusion power plants are what is referred to as "base-load capable" in traditional energy systems. This means that they can meet a certain amount of energy demand at all times, regardless of weather conditions. However, the higher the proportion of renewable energy in an energy system, the less important (or even obstructive) "base load power plants" with continuous production become, as flexibility and rapid controllability become more important. The key question might be whether a power plant can step in rapidly when solar and wind are not supplying enough energy.

In principle, fusion power plants are able to adjust their output (Ward/Kemp 2015). However, ramping up and down worsens the economic efficiency of the plants. According to various studies, in the energy supply systems of a decarbonised world, fusion power plants will compete with two other technologies to serve as a supplement to the cheaper but fluctuating renewables: fission nuclear power plants on the one hand, and gas-fired power plants with CCS capture facilities on the other. However, as fusion reactors will produce comparatively expensive energy even at full load (see section 4), they may need to be subsidised or otherwise (politically) supported in order to survive on the market.

A study that specifically analysed the possible role of fusion power plants in a future energy system dominated by renewables came to the following conclusion:

"While there remains a clear motivation to develop fusion power plants, this motivation is likely weakened by the time they become available." (Nicholas et al. 2021)

A reactor that can supply so-called base load energy could be obsolete in the post-fossil energy future, the research team warns. It therefore urgently recommends that fusion research and the design of future reactors pay more attention to their flexible controllability. German research academies also said in late 2024 that an energy system dominated by solar and wind energy does not require baseload power stations to guarantee supply security.

Some studies also see a possible role for fusion reactors in the production of hydrogen, which is to play a key role in a climate-neutral energy system and will then be needed in very large quantities (Gi et al. 2020). Fusion proponents also argue that the technology could not only serve to generate electricity, but could also be used to supply heat to industrial processes.

The development of fusion science and technology in Europe gained momentum with the 1957 Euratom treaty, which established a European atomic energy community. Since then, Euratom has coordinated European fusion research by funding initiatives such as the Joint Undertaking Fusion for Energy and EUROfusion, which aim to accelerate the development of fusion plants. The ITER agreement, signed by Euratom along with China, India, Japan, South Korea, Russia and the US, is responsible for managing and constructing the ITER facility in France. The goal of the project is to prove that a plasma-based fusion reactor can produce ten times the thermal power injected into the plasma.

Fusion for Energy is tasked with delivering Euratom’s contributions to the ITER project. Another key initiative is the International Fusion Materials Irradiation Facility (IFMIF), a joint European-Japanese project that will be constructed in Japan and set to operate in parallel with ITER. IFMIF will test and select materials capable of withstanding the extreme conditions inside a reactor, which include very high temperatures and high-energy neutrons. EUROfusion, on the other hand, represents a pan-European consortium with 30 members across 29 countries collaborating on fusion research.

The European fusion research follows a long-term strategy set out 2018 in the European research roadmap which charts the path for achieving fusion energy connected to the grid in the second half of the century. ITER is pivotal to this roadmap, because it is meant to prove the scientific and technological feasibility of this approach.

Despite fusion research dating back to the 1950s, to date only the US and the UK have established regulatory frameworks. The EU Commission has lamented that the absence of national and international rules creates a “regulatory vacuum,” which could potentially lead to the application of very strict (and hence expensive and obstructive) fission power-plant regulations to fusion power plants, which would complicate development. This has been the case for the ITER facility, which was classified as a nuclear reactor facility and thus subjected to a fission-based regulatory framework. However, there are compelling reasons why a fission regulatory approach would not be appropriate for fusion (as pointed out in section 2).

Ursula von der Leyen, president of the European Commission, affirmed need for action. “We need to put in place a specific regulatory framework for nuclear fusion,” she said, adding “we need to send a clear political signal that this is also a secure investment for private capital.” Later, the EU adopted the Net Zero Industry Act, which included a list of net-zero technologies and also covered nuclear fusion under “other nuclear technologies.” The act sets the target of producing 40 percent of net-zero technologies deployed in Europe on the continent by 2030, and to capture 25 percent of the global market value for these technologies. It also intends to deal with the challenges in scaling up manufacturing capacities in these technologies.

The Fusion Industry Association, representing the private fusion sector, advocates for fusion to be listed as a distinct subcategory of net-zero technologies. According to the association, this distinction is crucial, because conflating fusion and fission could result in unintended complications in both regional and international markets.

Scientists have accumulated a surprising amount of specific knowledge about how effective various technologies can be. Their findings can be found, for example, in the IPCC’s Sixth Assessment Report, which runs to more than 2000 pages.

To highlight that the selection of sources is central to this project, we have deviated from our usual editorial guidelines in this Q&A, and inserted sources in academic style - to remind readers that these texts represent the scientific consensus as accurately as possible.

With this aim in mind, we ranked sources in the following order, which values relevance much more than a very recent publication date:

1)     Wherever possible, the texts rely on the IPCC, which provides highly reliable summaries and assessments of the state of research.

2)     A second-best are thorough meta-studies (studies that evaluate many other studies), as well as synthesis reports from large research consortia or organisations, where a broad circle of participants and intensive review processes are common.

3)     Only in third place, we used individual studies - limited to publications in recognised research journals that guarantee a peer review process, meaning each publication is checked by competent specialist colleagues.

This Q&A and others in this series are based on texts published by our German-language sister project Klimafakten, which were written by expert journalists, and double checked by relevant experts. Two foundations supported this editorial project, which was overseen by our colleague Toralf Staud: the Marga und Kurt Möllgaard-Stiftung and the Deutsche Bundesstiftung Umwelt.