Seven international partners, representing more than half the world’s population, are now building the world’s biggest fusion reactor, ITER, at Cadarache in Southern France. The machine will be finished in ten years. The immediate goal – the baseline performance – is to produce 500MW of fusion power with less than 50MW of input power, an amplification of at least ten times. This will take ITER firmly into the regime where the plasma is self-heated by the energetic helium produced in fusion reactions.
There is much more to commercial power generation, however, than a demonstration of scientific feasibility. Several components of the future fusion reactor, in particular the systems that convert neutron power to electrical power, have yet to be tested at any scale.
In a fusion reactor, the plasma must be surrounded by a blanket of lithium that absorbs energy from the neutrons and breeds tritium fuel. The blanket is not simply a block; it is a complex system to extract heat and tritium that operates at high temperatures in a harsh neutron environment. Tritium is extracted from the blanket regularly and used to fuel the plasma to keep the radioactive inventory low.
For economic viability, the blanket must operate robustly for many years. Because this component is both critical and unique it will contain much of the intellectual property associated with the commercial development of fusion.
The right material
The first step towards wall and blanket development is to develop the materials: structural materials, breeder materials and high heat flux materials. These materials must retain structural integrity in very challenging conditions.
It is also highly desirable that these materials do not become long-lived radioactive waste under neutron bombardment. Several promising candidate materials are being developed. However, they have not yet been tested under bombardment by fusion energy neutrons (approximately 14MeV).
The fusion community is developing a facility – the International Fusion Materials Irradiation Facility (IFMIF) – to test small samples of the promising materials. The test samples will be irradiated in a beam of neutrons for several years to evaluate the changes in structural properties.
Blanket and wall structures
Clearly, developing the materials is a necessary but not sufficient step; an integrated wall and blanket system is needed. Many blanket designs are being developed and, although these designs have much promise, it is not given that they will result in a commercially viable system.
Indeed, this is probably the critical path for fusion. Until blanket viability is shown, fusion power is only a possibility. Several of the blanket designs will face their first nuclear tests in the later stages of ITER operation. Definitive tests require a continuous fusion neutron flux of 1-2MW/m2 for several years.
Even the ITER tests will not deliver this flux and thus will not provide a conclusive prototype. If full power wall and blanket testing is deferred until after ITER has completed its demonstration of scientific feasibility, fusion power could be postponed for decades.
It has been suggested that wall and blanket designs can be tried on the first generation of power generating fusion reactors – often called DEMOs. In current fusion plans these are to be built 30 years from now. I find this suggestion problematic. It may be that leaving blanket testing to this stage is not possible from a licensing point of view. But, even if it is possible, it would surely slow progress and carry an unacceptable degree of technical risk.
A viable solution
It is my belief that a bold step could alleviate the risk and significantly accelerate the development of blanket and wall structures. What is needed is a compact, affordable fusion device that can deliver reactor-level neutron flux over many square metres.
The ‘spherical tokamak’ – an alternative machine design with a tighter plasma configuration – is just such a device. The UK fusion programme at Culham has pioneered spherical tokamaks and currently operates the MAST (MegaAmp Spherical Tokamak) that has achieved near-fusion plasma conditions at very modest scale.
Both Culham and Oak Ridge Laboratory in the US have developed conceptual designs of component test facilities based on spherical tokamaks. These facilities would be dedicated to testing whole components of the blanket and wall at full power for many years. They are driven systems that do not deliver power to the grid. Both UK and US designs are compact and affordable. Culham is currently proposing to upgrade MAST to demonstrate that the plasma performance of the component test facility can be achieved.
A pragmatic approach to fusion would be to build component test facilities in parallel with ITER. A vigorous programme of wall and blanket development on these test facilities coupled with ITER’s programme could pave the way for the first demonstration reactors (DEMOs) in the 2030s.
There are no plans for a component test facility in the current international programme but there should be. Until blankets have been built and operated, predictions of the timescale of fusion’s entry into the energy market are necessarily imprecise.
It is time for the international community to recognise this reality and begin development of a component test facility. Shortening the time scale to commercial fusion by even a decade has enormous consequences for a world hungry for energy.