Causing a fusion frisson: the ITER site in the south of France
Causing a fusion frisson: the ITER site in the south of France
Credit: ITER

Harnessing nuclear fusion as a means of power generation has for decades been the Holy Grail for atomic scientists across the world, but there are several projects underway that could deliver results much sooner than anticipated, writes Paul Breeze.

Harnessing nuclear fusion has been a dream of technologists almost from the moment that nuclear processes taking place within the sun were recognised early in the 20th century.

The first, unsuccessful attempts at fusion took place in the 1930s and the quest was taken up again in the late 1940s. Since then there have been a string of successful and increasingly large fusion reactors built. Today there are 20 in operation around the world.

Since 1958, co-operation has been an important feature of fusion research and has centred on reactors that utilise magnetic confinement to contain the superheated plasma at the heart of the fusion process.

The latest, largest and most expensive of these is the International Thermonuclear Experimental Reactor (ITER – Latin for the way), which is under construction in the south of France. ITER is expected to be the first such reactor to be capable of delivering more energy from a fusion reaction than is used to generate the reaction in the first place, a key requirement if fusion is ever to serve as a viable power source. If it keeps to schedule it will reach full-scale operation by 2030 or before.

Recently, however, an alternative approach to fusion has started to make headlines. This is based on a completely different concept called inertial confinement and if it can be perfected, it promises a demonstration during the 2020s and commercial fusion plants by 2030, sooner than the magnetic confinement path can deliver.

Rather than being international, the most advanced inertial confinement development is being carried out in the US, where it has emerged, almost unannounced, from the defence establishment. Like the magnetic confinement approach, inertial confinement has yet to produce more energy from fusion than is used to achieve a fusion reaction. However, the US programme is confident that it will achieve this milestone in the near future.

The fusion problem

Fusion is attractive because it promises almost limitless energy from a simple process that is largely free of atmospheric emissions or toxic by-products. The principle reactions that take place within the sun involve hydrogen atoms fusing to produce heavier atoms.

The mass of the resulting heavier atoms is not the exact sum of the two initial atoms, some mass has been lost and great amounts of energy gained. This is what Einstein’s formula E = mc² describes: the tiny bit of lost mass (m), multiplied by the square of the speed of light (c²), results in a very large figure (E), which is the amount of energy created by a fusion reaction.

There are two important fusion reactions in the sun and the stars. The first involves fusion of two hydrogen atoms to generate a deuterium or heavy-hydrogen atom. In the second, deuterium and hydrogen atoms fuse to create a helium atom.

However it is a third reaction between deuterium and the even heavier hydrogen isotope tritium that interests fusion scientists because it proceeds more easily than the other two and under relatively more benign conditions. A fusion reaction between these two hydrogen isotopes produces one helium atom and one neutron and it is the latter that carries most of the energy released during the fusion process. That energy must then be captured and used to generate electricity.

The potential is massive. The energy from one tonne of deuterium is equivalent to 3 x 1010 tonnes of coal. Unfortunately the prize is not easily won. The reaction will only take place in a plasma at massively high temperatures and in the case of inertial confinement, under conditions of enormous pressure. Reaching the conditions necessary for fusion to take place – and then controlling and maintaining them – have been the primary challenge of fusion research.

Magnetic confinement

When matter is heated to temperatures that approach anywhere near those of the sun, the material becomes a plasma in which the individual atoms disintegrate into a sea of atomic nuclei and electrons that are bound by electromagnetic interactions.

It was recognised early on in fusion research that such a material state could not be contained using conventional materials and the idea of magnetic containment was born. This proved much more difficult to realise that had been expected and it was Russian scientists that eventually solved the problem with a toroidal magnetic confinement which they called a tokamak. Although exploration of other approaches continued, this become the de facto design for a fusion reactor.

The two largest fusion reactors in operation today are the Joint European Torus (JET) at Culham in the UK and the Tokamak Fusion Test Reactor (TFTR) at Princeton in the US. Both started experimenting with deuterium-tritium (DT) fuel during the 1990s, and in 1997 JET set the current record for the largest amount of power generated by a fusion reactor – 16 MW.

Some 633 of these massive stainless steel forgings will be necessary for the construction of the ITER vacuum vessel sectors<br>Credit: ITER
Some 633 of these massive stainless steel forgings will be necessary for the construction of the ITER vacuum vessel sectors
Credit: ITER

The reactor consumed more than 16 MW to achieve this record although the ratio of power in to power out (the gain of the reactor), at 0.7, was close to the break-even target. However, JET could only sustain a plasma burst for 5 seconds before its ancillary services started overheating.

Both JET and TFTR are experimental, pilot-scale fusion reactors and achieving break-even is a matter of scale. It requires a big reactor to achieve a gain of much more than one and get significant power generation. That will be the job of the next fusion reactor based on the tokamak design, ITER.

However the work at the smaller reactors is far from over. JET is being upgraded to extend its operating range to carry out more pre-ITER experiments, particularly on plasma stability.

The plasma in the tokamak flows along lines of magnetic force. The temperature at the centre of the JET plasma reached 170 x 106 ºC. Inside the hottest regions the plasma is bubbling like a boiling liquid and this creates eddies that make it both unstable and inefficient. Controlling and reducing the turbulence inside the plasma is one of the keys to an efficient fusion reactor and work at JET from 2015 to the end of the decade should help advance the understanding of plasma turbulence.

There are also design problems that have yet to be solved before ITER can start to operate, such as the material used for the lining of the reactor chamber. In JET, these are carbon tiles, but the carbon absorbs tritium so an alternative must be found. The favoured replacement is beryllium tungsten and this will be tested at JET. Further work on the operating modes for the reactor will also help ITER. In essence, JET is the model for ITER.

ITER has had a long gestation. It was conceived in 1988 under the auspices of the International Atomic Energy Authority and initially involved the EU, Japan, Russia and the US. An engineering design was completed in 2001, the Cadarache site selected in 2005 and the ITER agreement was signed in 2007 by the now seven members, following the addition of China, India and South Korea.

The project will have a plasma volume of 800m3, ten times larger than JET and it will have a thermal power output of 500 MW, which is 30 times larger than JET has achieved. It is hoped that at this size, the reactor should be able to achieve a gain factor of ten, so 50 MWth will drive an output of 500 MWth.

However, ITER is not a commercial demonstration project. It has been designed to prove that it can generate 500 MW of fusion power for 400 seconds. A commercial plant will need to be able to operate round the clock for weeks if not months on end.

If it was being designed today, then perhaps ITER would have been more ambitious, but most of the basic parameters were set during the 1990s when the costs and risks involved in trying to build a plant that would generate electricity seemed huge.

So the plant will be virtually commercial size but will not have steam or turbine generators. More significantly, it will not have a full-sized system for capturing the energy generated by the fusion reactions in the plasma. There will be test modules within the reactor shroud but the full energy capture system will have to wait for the first demonstration plant.

So ITER is another experimental reactor, but even so it is probably the most challenging project being built on the earth today, at least in the opinion of Michel Claessens, head of communication and external relations for the Office of the Director-General at ITER.

With seven members and a total of 34 countries that will jointly build the project, many of the components will not be built by one fabricator but by manufacturers in different countries. Scheduling the construction and maintaining the level of quality control for a project of this complexity will be a Herculean task. But if all this can be mastered then in theory the first experiments will take place at ITER in 2020.

Inertial confinement

While the research into fusion based on magnetic confinement edged forward there was, behind the facades of defence research establishments in the US and elsewhere a completely different approach to the fusion being pioneered. However the defence-related nature of much of this work has meant that until very recently little was known about what is called inertial confinement.

While a reactor such as ITER will contain a plasma that maintains conditions for fusion continuously within its heart, inertial confinement instead uses a series of small, discrete fusion reactions, each producing a burst of energy. This has been likened to a piston engine in which energy is generated is a series of small impulses rather than in a continuous stream.

The concept is relatively simple – if developing it into a commercial power station design is not. A small capsule containing a few hundred micrograms of a DT mixture is subjected to a massive pulse from a system of multiple lasers focused onto its surface. The laser energy striking the surface of the capsule causes the outer surface to explode in a pulse of x-rays and this creates an equal and opposite shock wave which travels into the capsule, heating and compressing the DT mixture to the point where the fusion reaction takes place at its centre.

Once the fusion reaction starts it radiates outwards through the whole capsule, travelling faster than the material itself can expand so that the whole charge of fuel is consumed and energy released. The inertia consequent on the mass of the atoms of the DT mixture prevents them from expanding as fast as the fusion front advances, hence the name inertial confinement.

It is possible to imagine this being achieved in a single shot experiment but to turn the concept into a power station, the process must be repeated endlessly. For a practical plant there would be about 15 of these fusion explosions each second. Yet this is exactly what a major programme in the US is proposing. What is more, the US government has built a plant, called the National Ignition Facility (NIF) that has the capability to prove the practicability of the process.

The new road to fusion

NIF is an expensive and ambitious project that has come about partly as a result of the Comprehensive Test Ban Treaty designed to eliminate nuclear weapons testing.

The facility will provide experimental data to support this treaty which is why it has been able to attract $5 billion of US government funding. However, NIF will also have two other purposes – as a tool for fundamental scientific research and to prove the viability of power generation from fusion based on inertial containment.

The heart of NIF is its laser system. The facility has 192 lasers which are capable of delivering as much as 5 MJ of energy in 20-nanosecond pulses. So far it has operated at 1.8 MJ, equivalent to a power delivery of 500 TW. The lasers initially generate infra-red light but this is converted, first to visible light and then to ultra-violet before it strikes the target. That target is a tiny capsule called a hohlraum which is about 2 mm in diameter and contains 150 mg of the DT mixture. It is this tiny charge that is subjected to around 500 TW of power.

The importance of NIF from a power generation perspective is that the laser power is of the scale necessary to build a 1000 MW power station. It can therefore simulate at full scale the capacity for inertial containment to deliver energy for electricity generation.

NIF started operating in 2009 and has carried out a series of ignition experiments since then. Ignition, in this context, is the point at which the capsule of DT produces more energy that the laser pumps into it.

During the first experiments, the results were around 50 to 60 times short of the target required by ignition.Over the past three years it has crept closer to the target, which is now only a factor of two or three away. Once ignition is reached, the fusion reaction becomes self-sustaining because it generates the energy necessary to maintain the temperature and pressure required. So while they cannot say when ignition will be achieved, the scientists as NIF are confident that they will achieve it.

An artist's concept of a LIFE power plant with the exterior cut away to show the fusion chamber<br>Credit: Lawrence Livermore National Laboratory
An artist’s concept of a LIFE power plant with the exterior cut away to show the fusion chamber
Credit: Lawrence Livermore National Laboratory

If ignition can be demonstrated, then a power plant based on this principle is possible. Work to define what this power station will look like has already started and forms the basis of the Laser Inertial Fusion Energy (LIFE) project. The design for the LIFE plant has been developed through a collaboration between technologists, electric utilities, power plant vendors, regulators and environmental groups. Its aim is to build a demonstration power plant within ten years of ignition being achieved at NIF using components that can be fabricated today by technology companies.

NIF will continue to be the benchmark for testing LIFE concepts but the belief is that if ignition can be achieved, then a LIFE plant can be built. The demo plant would initially be designed to produce 400 MW of electrical power but with the ability to be scaled up to 1000 MW. Based on current estimates, this plant could be operating in the early part of the next decade, with commercial plants available by 2030.

If LIFE could achieve this target, then it would be a remarkable milestone. Before that, however, there are some major hurdles to cross. These include integrating all the components of a LIFE plant from the laser to the fuel delivery system to the heat extraction and power generation. Operating the cycling, piston engine type of ignition has yet to be demonstrated at power plant scale. And there is one technological hurdle that faces both LIFE and the first full-scale fusion power plant based on magnetic containment – the design of the blanket system.

The blanket system is the layer that surrounds the plasma chamber in the case of a tokamak reactor and the ignition chamber in a LIFE-style power plant. It has to serve two functions: the first is to slow down the very high energy neutrons that emerge from the fusion reaction, absorbing their energy and converting it into heat that can be used to generate electricity. The second is to manufacture tritium. Fusion reactors are expected to breed their own fuel and this will take place inside the blanket.

Precisely what the blanket will look like remains a matter for speculation but whatever form it takes, it will contain lithium because this will be the source of tritium. When a lithium atom is exposed to neutrons such as those generated by fusion of deuterium and tritium it reacts to form an atom of tritium and an atom of helium.

This tritium must then be harvested from the blanket ready to provide fuel for further fusion. Liquid lithium could itself form the coolant inside the reactor, cycling through a heat exchanger to generate steam. Alternatively some other coolant such as helium might be used and the lithium contained within a ceramic rather than in liquid form. Molten salts containing lithium might also be used.

The future

So what does the future hold for fusion? Optimistically, a fusion plant based on inertial confinement might deliver a commercial plant by 2030, although based on experience with other complex projects, the timeline is likely to be a little longer than this.

Meanwhile, ITER hopes to demonstrate commercial plant scale fusion by around that time too. If ITER progresses as expected then work on the first demonstration plant, referred to as DEMO in the fusion industry, will be well under way by then.

It is often said that a commercial fusion plant is always 30 years away. While there is clearly still a long way to go and nobody has yet demonstrated that fusion can produce electricity rather than simply consuming it, that threshold does seem palpably closer today than at any time in the past.

More Power Engineering International Issue Articles
Power Engineering International Archives
View Power Generation Articles on