There is a longstanding joke that nuclear fusion is always 30 years away, but recent advances mean that the technology could finally be within our grasp, reports Jasmin Fox-Skelly
Almost all the elements in the universe were formed in the heart of stars. Our own Sun turns 600m t of hydrogen into helium every single second, releasing a huge amount of energy in the process. In fact, stars are perfect places for this nuclear fusion reaction to occur. In the Sun, temperatures of 15m°C, along with the crushing force of gravity, cause atoms to collide together at such high speeds that the natural repulsion that exists between the positive charges of the atoms’ nuclei is overcome, and the atoms fuse together, forming a different, heavier element.
For almost 100 years, scientists have dreamed of recreating this reaction on Earth, spurred on by a powerful vision of limitless clean energy – fusion has no carbon footprint, and unlike nuclear fission, there is no risk of meltdown. In terms of sheer scale, the potential energy released through nuclear fusion surpasses anything on Earth. Fusing atoms together releases nearly four million times more energy than a chemical reaction such as the burning of coal, oil or gas, and four times more than nuclear fission – when the nuclei of heavy atoms such as uranium or plutonium decay.
‘If you think about the amount of energy the Sun kicks out every moment, it’s astronomical, yet it has been burning for 4.5bn years,’ says Mark Henderson, a senior engineer and physicist working on the International Thermonuclear Experimental Reactor (ITER) in Southern France.
‘At the same time, if you look at the reserves of deuterium – a hydrogen isotope – in the oceans there’s enough energy to last humanity 150m years at our present rate of consumption. If we could tap into this energy source it would be basically an endless supply of energy, and it almost seems stupid for us not to try to exploit that.’
If you look at the reserves of deuterium in the oceans there’s enough energy to last humanity 150m years at our present rate of consumption. If we could tap into this energy source it would be basically an endless supply of energy, and it almost seems stupid for us not to try to exploit that.
Mark Henderson senior engineer, the International Thermonuclear Experimental Reactor, France
However, until now, generating this form of nuclear power has proven extremely difficult, requiring heating subatomic particles to hundreds of millions of degrees – far too hot for any material container to withstand.
Scientists first managed to fuse hydrogen atoms together in a lab in the early 1930s, and by the mid-1950s, ‘fusion machines’ were operating in the Soviet Union, UK, US, France, Germany and Japan. As far back as 1955, the physicist Homi J. Bhabha claimed we would have fusion power within two decades. However, the challenge then – and now – was to safely create a sustained fusion reaction that generates more power than is needed to keep the fusion reaction going.
To do so presents significant engineering challenges. There are two main approaches: inertial confinement and magnetic confinement. Inertial confinement fusion (ICF) involves heating a pellet containing a mixture of the hydrogen isotopes deuterium and tritium, using high-energy laser beams. This causes the heated outer layer to explode outward, and the inner pellet to implode and compress, fusing the hydrogen atoms together.
Magnetic confinement, on the other hand, works by heating hydrogen atoms to temperatures between 15-150m°C, as hot, or in some cases hotter than the Sun. At these temperatures, atoms form a soup-like, fourth state of matter called plasma. The plasma is compressed and held in place by magnets, preventing it from contacting any part of the chamber and causing the hydrogen isotopes to fuse, producing helium and neutrons that shoot outwards at high speeds.
The engineering challenges in building such a machine are immense.
‘We are heating something to temperatures ten times hotter than the sun, but within just a few meters, we have superconducting magnets, which must be kept at 4°K, just above absolute zero,’ says Henderson. ‘The thermal gradient is enormous, and then you have to absorb this enormous amount of power. All of this in a fairly compact volume really challenges just about every single technology we have.’
The most commonly used reactor, a doughnut-shaped magnetic confinement device called a tokamak, was first used in 1968 in the Soviet Union. Since then, it has become the dominant concept in fusion research.
So far, however, no one has managed to achieve plasma energy breakeven (Q=1) – the point at which plasma releases as much energy as is required to heat it. The current record is held by the Joint European Torus (JET) at Culham in Oxfordshire, which can generate 16MW of fusion power for 24MW of power input (a Q ratio of 0.67).
However, that may be set to change, with a host of government funded and private enterprises seeking to achieve the elusive Q=1. The biggest and most ambitious of these is the International Thermonuclear Experimental Reactor (ITER), a $25bn experiment in Saint-Paul-lez-Durance, France. ITER is a joint project of the EU, China, India, Japan, South Korea, Russia and the US.
The most common fusion reactor is a doughnut-shaped magnetic confinement device called a tokamak, first used in 1968 in the Soviet Union.
So far, no one has managed to achieve plasma energy breakeven (Q=1) – the point at which plasma releases as much energy as is required to heat it. The current record is held by the Joint European Torus at Culham, which can generate 16MW of fusion power for 24MW of power input (a Q ratio of 0.67).
A tape containing yttrium-barium-copper oxide (YBCO) forms a cable which can produce a magnetic field four times as strong as that employed in any existing fusion experiment, enabling a >10-fold increase in the power produced by a tokamak of a given size.
The ITER tokamak will have ten times the volume of the largest machine operating today, with a giant magnet capable of lifting an aircraft right out of the ocean at its centre. For 50MW of injected heating power, it will produce 500MW of fusion power, enough, if converted to electricity, to power a city the size of Liverpool, giving it a Q ratio of 10. The pilot reactor will not capture the power it produces as electricity but will prepare the way for the machine that can.
So far, the project has been hindered by delays and ballooning costs, but ITER finally reached a long-sought milestone in July 2020, when researchers began the ‘assembly phase’ of its pilot plant, joining the various components provided by the partner countries. After being switched on in 2025, there will be two decades of experiments, with ITER’s commercial fusion power plant not expected until 2054.
‘We know we are running behind schedule because of the complexity of building the machine, and also because of Covid,’ says Henderson. ‘Right now, out of the nine sectors which go into making the tokamak we have one already in place, with two more in the process of being manufactured. These nine pieces of the donut will be welded together and completed in 2023, before we add in the heating and diagnostic systems in 2024 in time for our first attempt at plasma in 2025.’
The sheer scale of ITER inevitably requires high costs and long timelines, meaning it could not feasibly put fusion power on the grid until sometime after 2060, quite possibly too late to substantially mitigate global warming.
‘To be honest in my opinion fusion is the long-term answer, but it is not the short-term answer,’ says Henderson. ‘It is going to take a good number of years before commercial reactors exist around the world, and we really need to get off our addiction to fossil fuels now.’
Image: Monty Rakusen/getty
In the UK, the government has invested £220m into the Spherical Tokamak for Energy Production (STEP) project, which aims to begin plant construction around 2030, with the plant operating as soon as 2040. The plant is pitched as an important part of efforts to hit the UK’s target of net zero emissions by 2050.
Although STEP’s power output goal is more modest than ITER – a net gain of 100MW – unlike ITER, it will be able to provide actual energy to the national grid.
Also unlike ITER, the STEP reactor features a more compact spherical tokamak design, which looks more like a cored apple than a donut. While the fundamentals of the two designs have all the same features – vacuum chambers, magnetic fields, a hole in the centre – everything is pushed closer to centre in the STEP machine, allowing for smaller magnets, potentially saving millions of pounds. A more compact design also offers quicker build times.
However, there are still some deep scientific and engineering challenges that remain before STEP can put power into the grid.
‘The step from current experiments to a fusion power plant is a big one both scientifically and technologically,’ says Nick Walkden, head of the executive office at the United Kingdom Atomic Energy Authority, which runs the STEP programme.
‘The science and engineering challenges needed to make fusion a reality need deeply innovative and fundamental solutions. For example, how do you keep a burning fuel going at 150m°C for hours or days, but avoid damaging structural materials of the machine that need to be cooled? Or how do you repair and replace machine components without any human intervention?’
To help with the first problem, researchers are developing a ‘Super-X divertor’, a kind of exhaust system that diverts excess heat and byproducts from the plasma away from the surrounding walls of the reactor, stopping it from melting. The divertor should reduce the heat to manageable levels, such as that found in a car engine.
‘This is an innovative system to handle the excess heat from a fusion device – a show stopping problem for fusion power plants if it cannot be solved,’ says Walkden. ‘The very first experiments on this new system are being conducted this year, which will help to inform the design of STEP and other power plant prototypes.’
The International Thermonuclear Experimental Reactor (ITER), a $25bn experiment in Saint-Paul-lez-Durance, France, will have ten times the volume of the largest machine operating today. For 50MW of injected heating power, it will produce 500MW of fusion power.
Meanwhile, researchers at Massachusetts Institute of Technology (MIT) in Boston, US, have teamed up with spinoff company Commonwealth Fusion Systems to develop their own reactor called SPARC, which they say will be capable of producing electricity for the grid by 2030. SPARC is planned to be the first ever device to achieve a ‘burning plasma’ – a self-sustaining fusion reaction without the need for any further input of energy.
Central to their effort is the teams’ work developing the world’s most powerful superconducting electromagnet, made from a newly available material — a steel tape coated with a compound called yttrium-barium-copper oxide (YBCO). The tape is packaged together to form a superconducting cable called VIPER (Vacuum Pressure Impregnated, Insulated, Partially transposed, Extruded, and Roll-formed), which can conduct electricity with no resistance, and does not degrade under extreme mechanical, electrical, and thermal conditions. Cryogenic coolants, such as helium, can also easily flow through the cable to remove heat and keep it cold.
The biggest advantage of the cable is that it can be fashioned into twists and turns, making it an ideal building material for winding into coils capable of generating and containing magnetic fields of enormous strength, such as those required to make fusion devices. The cable is capable of producing a magnetic field four times as strong as that employed in any existing fusion experiment, enabling a more than tenfold increase in the power produced by a tokamak of a given size. This means that SPARC could deliver a lot of fusion energy for its size, producing a power output of 100MW – enough to power a small city. Although that is about a fifth of that of ITER, SPARC is only about 1/65 of its volume.
Its compact nature has other benefits too; components that would otherwise be so large that they would have to be manufactured on-site could instead be factory-built and trucked in, cooling systems could be scaled back, and the total cost and time for design and construction could be drastically reduced.
The other good news is that once the magnets are finished, the next step of designing the SPARC reactor is based on a relatively straightforward adaptation of existing tokamak technology, refined over the years by MIT.
The researchers recently released a series of papers showing that their fusion design should work.1 Using cutting-edge simulations on powerful supercomputers, the analysis shows that SPARC should be able to achieve a Q factor of at least two and could achieve a ratio of ten or more.
Building is expected to commence in June 2021, with operation expected by the mid-2020s. If the pilot plant is successful, a commercial power plant about twice SPARC’s diameter could produce 200MW of electricity, comparable with most modern commercial electric power plants.
The MIT team says that such a plant could exist within 15 years, but some believe that is optimistic.
‘One of the problems with a much more compact system is that although that machine will be very useful for demonstrating fusion, it is not going to be suitable for a long-term steady state operation,’ says Henderson. ‘Every high energy neutron produced by fusion takes roughly between 0.5-1m of steel and water to slow it down. That means that any size machine has to have this, what we call a blanket, to protect your coils and vacuum vessel. That has to be in there if it’s going to be a long-term steady state device.’
Nevertheless, Henderson, along with other scientists believe that it’s vital that teams pursue different paths to fusion, as designing only one type of machine has risks, and we don’t yet know the best option.
‘Whilst the race is on, the most important thing is that someone finishes the race, so that today’s generation will see the first fusion power on the grid, and tomorrow’s generation will see major deployment of fusion power,’ says Walkden.
Making fusion work will require a huge financial investment, and perhaps a leap of faith, but the rewards make it worth it.
‘I always think of John F Kennedy back in 1962, when he said my goal is to put a man on the moon and return him safely by the end of the decade,’ says Henderson. ‘All it requires is one country to say, look we’re going to do this. Fusion could last us from 75 years to 15bn years, but we need to invest money, time and expertise so we are ready.’