Experimental fusion machines have produced short bursts of fusion power of more than ten megawatts. A new international machine under construction, called ITER, will be capable of producing 500 megawatts of fusion power after it starts operation around 2025. Although ITER will be on the scale needed for a power station, there will still be technological issues to address to produce steady, reliable electricity, so prototype power plants will then be needed to fully demonstrate economically viable electricity production. These are expected to be built in 20-30 years, depending on funding and technical progress, and would be followed by commercial reactors supplying widespread electricity in the second half of this century. However, more investment and technology innovations could accelerate this timescale.
Making fusion a commercially viable form of electricity production is extremely technically challenging. Heating a gas of fuels to 150M degrees C is just the start; choosing optimum structural materials, developing quick turn-around robotic maintenance systems and manufacturing components that will work efficiently in high vacuum, high magnetic fields and under high heat loading – are all required for economically viable fusion power plants. There are active research programmes into all these areas of work – here at UKAEA and elsewhere in Europe and the world – but questions still remain to be answered before we can build the first true fusion power stations.
Fusion provides one of the few options for supplying large amounts of continuous power to the grid in the latter half of the 21st century and it is essential that we develop it, along with other sources, particularly renewables. The public demand for carbon-free energy sources is clear and we must do our best to establish ways of electricity production that are consistent with those demands.
There is no doubt that climate change is happening – and the world urgently needs to develop and adopt carbon-free energy sources as rapidly as possible. Fusion will not solve this on its own – but the hope is that it will be a major contributor (with advanced fission and renewables) to a net zero carbon energy future in the latter half of this century.
The two fuels commercial reactors, known as ‘tokamaks’, will use are isotopes of hydrogen called deuterium and tritium.
Deuterium is abundant in seawater, accounting for 1 in every 6700 atoms. There are a variety of chemical separation methods that enable deuterium to be extracted from water.
Tritium is produced from lithium. Future fusion power plants will be designed to manufacture tritium from lithium and feed the fuel into the reactor. Lithium is a light alkali metal found in several different minerals in the earth’s crust and can also be extracted from brines and clays. Natural deposits are particularly found in South America and these deposits, as well as lithium in sea water, are expected to provide enough fuel for fusion power plants for tens of thousands of years at least.
As a future electricity source, fusion should be available across the globe. Fusion fuels are abundant in water – a resource almost all countries possess. And, as fusion power plants require so little fuel (a few hundred kg of deuterium and tritium per year), the economics of fusion power generation will not be driven by cost of availability of fuels.
A large fusion power station generating 1,500 megawatts of electricity would use approximately 600 grammes of tritium and 400 grammes of deuterium each day. This compares with a daily consumption of around 10,000 tonnes of fuel in a coal power station of a similar size.
The fusion energy obtained from each kg of fuel is very high (several million times higher than from fossil fuels), so the fuel costs are a very small part of the expected cost of electricity. Using present cost estimates, the fuel costs will contribute much less than 1% to the cost of electricity.
One litre of ordinary water contains enough deuterium to provide the energy content (when fused with tritium) of more than 500 litres of petrol.
Yes. We concentrate on fusion of deuterium and tritium for energy production. This has been tested in tokamaks like JET and is the easiest way we know of getting a net energy gain. There are other possible fusion reactions (e.g. Deuterium-Deuterium; Deuterium-Helium-3; proton-proton as in the Sun) that produce energy. However, they are more difficult to achieve and their usefulness in a terrestrial power source remains to be demonstrated.
In a future power plant, it is envisaged that the fusion energy is released in the form of fast-moving neutrons, which will be slowed down in special blankets within the vessel walls. The resultant heat would generate electricity in just the same way as existing power stations, where the heat is used to raise steam, driving turbines to produce electricity. The possibility of using the energy more directly has been considered but does not seem practical at this time.
It is an intrinsic property of the fusion process that it is inherently safe with low environmental impact. There is only a small amount of fuel in the plasma at any time, and overfuelling or overheating the plasma will lead to it being extinguished almost instantly. Extensive studies over the last two decades (for example the European Fusion Power Plant Conceptual Study) have shown that no plant failure or accident could result in the need to evacuate public from outside the site, and that the radioactive waste products from fusion power will not be as long-lived as those from nuclear fission.
Helium will be the main waste product expended by tokamak reactors.
Because of the large amount of energy produced per unit mass of fuel, the production of helium in tokamaks is rather low; ten million times less than the CO2 production of an equivalent fossil fuel power plant. If the whole world’s energy requirements were met by fusion, the helium production would still be small compared to the present helium production of around 30,000 tonnes per year.
The helium produced in future fusion power stations would be used within the plant and not typically released to the atmosphere.
As you might expect in a magnetic confinement system, the largest cost item is anticipated to be the superconducting magnets. The next largest cost is likely to be the buildings and land needed to house the plant. These two items together are estimated to make up more than half of the cost of a fusion power plant. Innovations in magnet technology and mass production techniques could reduce the cost of superconducting magnets over time.
A complex magnetic field structure of typically a few Tesla is used to confine and hold the hot plasma of fuels away from the vessel’s material surfaces. This allows the plasma to be heated to the very high temperatures required for fusion.
The magnetic field provides insulation some 40 times better than loft insulation and is up to ten times thicker. With such good insulation, high power heating (in the megawatt range) leads to very high plasma temperatures, above 100 million degrees Celsius.
There are three main heating methods for achieving these conditions:
- Inducing a large electrical current in the plasma (known as ‘resistive heating’);
- Injecting accelerated beams of neutral particles (neutral beam injection);
- Launching high power radiofrequency waves and microwaves into the plasma.
The effect of the magnetic fields is to confine the charged plasma particles by applying a force that opposes the motion across the field. If the magnetic field lines were to connect the ends of a linear device, particles would be able to escape rapidly to the ends. In a toroidal device, the particles primarily spiral along the field lines, travelling around the machine typically a million times before escaping, meaning fusion of nuclei happens much more readily.
Most current experimental fusion devices do not use superconducting magnets since the required experimental plasmas, lasting less than one minute, can be achieved with cheaper and less complex conventional copper coils. Next-step machines such as ITER and power plant designs, which require much longer pulses, will utilise the latest superconducting materials to make the most efficient magnetic coils possible.
Existing fusion experiments need more power to heat and confine the plasma than they produce in fusion power. In a future power plant, which would be larger and more powerful, the fusion power would be around 20 to 30 times higher than power needed to heat and confine the plasma.
Can the economic cost of fusion's environmental impact be estimated, including the costs of constructing and dismantling the power plant?
Yes. Estimates have been made using the method developed by the EU’s ExternE project and more recent work such as EUROfusion’s socio-economic studies. These consider the total environmental impact of power production, from the original extraction of materials, through to the operation and subsequent recycling/dismantling of the facility. This is done by associating a cost to everything from CO2 emissions to accidents at work. The conclusions in published work have been very favourable, with fusion estimated to be considerably less harmful than conventional oil, coal, and gas.