Tokamaks remain the main area of development in magnetic fusion, with the Iter device now being built in France by an international consortium taking centre stage. Iter director general Bernard Bigot explained how the multinational project has been reorganised to meet escalating costs, slipping schedule commitments and earlier management deficiencies. In his presentation to the Fusion Power Associates Annual Meeting in Washington, USA, in December, he noted that the Iter Council would establish a new baseline for the project by its mid-2016 meeting. In the meantime, Iter construction and component manufacturing is proceeding at full speed.

Tony Taylor, vice president at General Atomics, called Iter an essential element for fusion energy development, burning plasma science and fusion technologies. He said tokamak confinement systems have the most advanced scientific base and have made extensive progress since they were first introduced in 1969. Iter will significantly advance the science and technology of fusion. And most important, the tokamak is the only magnetic fusion concept that is "ready".

Princeton Plasma Physics Laboratory (PPPL) is conducting systems studies of a Fusion Nuclear Science Facility (FNSF) intended to fill the gap between Iter and Demo, a demonstration fusion power plant. FNSF will be able to operate continuously for 1 to 15 days. To lay the groundwork for Demo it will be used for: fusion neutron exposure (fluence and dpa); materials development (structural, functional, coolants, breeders, shields, etc); operating temperature and other environmental variables; tritium breeding; tritium behaviour, control, inventories and accounting; long plasma durations at required performance; plasma-enabling technologies; demonstration of plant operations; subsystems; and availability, maintenance, inspectability and reliability advances.

In materials development and testing the goal will be to establish a database of components in the fast neutron environment and in the overall environment, before moving to larger operations in Demo and routine electricity production. Demo will reach power plant levels of neutron damage and will need a new class of materials that can survive in the fusion environment.

FNSF will be smaller than Demo to reduce costs and facilitate a break-in programme. The team plans a conventional aspect ratio of four; a conservative tokamak physics basis with extensions to higher performance; a 100% non-inductive plasma current; low-temperature superconducting coils made of advanced Nb3Sn; and helium cooling in the blanket, shield, divertor and vacuum vessel. Net electricity generation is not a target, but the machine could be used to demonstrate electricity generation, Charles Kessel from PPPL said.

It will move from initial shakedown operations, to deuterium operations, and finally to about 23 years of deuterium-tritium operations. It will be designed to bridge what Kessel called the "tremendous" gap between Iter and Demo. The PPPL team is trying to identify what FNSF must demonstrate, identify the R&D programme to prepare for FNSF operation, and establish the connection between FNSF and Demo, as well as future fusion power plants.

Is a viable tokamak power plant possible?

Other presenters were less optimistic about the potential for Iter, or the tokamak concept, to lead to practical fusion energy. Robert L. Hirsh, who directed the US fusion energy programme in the 1970s and is now a senior energy advisor at Management Information Services Inc (MISI) in Washington, DC, said it is "time to face reality" that the Iter tokamak will never be commercially viable. It can, however, provide valuable knowledge and experience.

One big issue with the economics of tokamak fusion power is the time required for superconducting toroidal field coils to warm up and cool down, Hirsh said. China’s Experimental Advanced Superconducting Tokamak (EAST) took about 18 days to cool from room temperature to 4.5K after a December 2006 quench. The Iter cool-down is estimated at roughly 30 days. A 30-day heat up/cool-down outage in a commercial power system "would have a major, negative impact on plant economics", Hirsh said.

Possible regulatory concerns

Regarding waste, runaway reactions are not possible in fusion devices. And the radioactive waste from a tokamak consists of activated metals, which would be shorter-lived and less hazardous than used fission fuel. However, Hirsh said radioactive waste handling, storage and disposal would still be a major regulatory concern.

Assuming that the blanket of an Iter-class tokamak must be replaced every three years due to radiation damage, the radioactive waste produced in a continuously operating Iter tokamak would be 675t/year. This is more than a fission reactor, which produces approximately 150t/year of spent fuel. While its radioactivity and toxicity is much greater and longer-lived than Iter fusion waste, nuclear regulators require waste from a fusion reactor to be handled in a similar manner, or at least under very stringent standards, Hirsh said.

Aside from waste, Hirsh identified three major regulatory concerns: superconducting magnet quench, plasma disruptions and tritium containment.

  • Regulators are particularly concerned about superconducting magnet quenching, which is a low-probability event with an explosive energy release. A superconducting magnet quench on Iter, for example, could release more than 40GJ, or the equivalent of 10t of TNT. Hirsh pointed out that this is the same explosive power as a World War II era Blockbuster Bomb. The threat of an explosive superconducting magnet quench will require an Iter-class tokamak to be adequately contained. Given the size of the tokamak, a blast-proof containment structure would be extremely expensive, he said.
  • Plasma disruptions are another area of regulatory concern. Tokamaks operate within limited parameters. Outside this, sudden losses of energy confinement, known as disruptions, can occur. These disruptions cause major thermal and mechanical stresses to the structure and walls. Hirsh quoted physicist Sarah Angelini from Columbia University, who said: "In a large-scale experiment such as Iter, disruptions could cause catastrophic destruction to the vacuum vessel and plasma facing components." Regulators will focus on disruptions, identify all possible triggers and potential cascades, and require fail-safe protection, Hirsh said.
  • Tritium diffuses through solid materials, especially at high temperature. Vacuum and energy injection ports on tokamaks will allow tritium leakage into the reactor hall, as will equipment breakdowns and damage to the vacuum vessel. NRC’s tritium dose limits for radiation workers and the general public are significantly lower than the levels of radiation exposure known to cause health effects in humans. Regulators will require expensive fail-safe protections.

The public has been told that fusion power will be economic, safe and environmentally attractive, Hirsh said, and warned that this could backfire. Utilities also will be acutely aware of any NRC restrictions and concerns, Hirsh said, and their interest could quickly evaporate.

Tokamaks also face major operability questions when considered as power generators rather than research devices.
Hirsh pointed to divertor durability during commercial operations as a major issue. Recent research indicates that no solid material, including tungsten, can operate under expected Iter conditions for a reasonable period of steady-state operation. Hirsh said that a 2015 US DOE fusion workshop concluded that the knowledge base of tokamak divertor physics cannot specify a divertor solution, "in fact, we do not know that a solution exists, even in principle". Without that, Iter-class tokamaks will not operate for very long, Hirsh said.

Stellarator demo

Presenters from Europe and Japan said that they were looking at alternate concepts for Iter’s follow-up Demo. Chief among the concepts under consideration are stellarators, spherical tokamaks (a hybrid that combines features of both the tokamak and the stellarator) and inertial confinement options.

The 10 December ‘first plasma’ at Germany’s Wendelstein 7-X, the world’s largest stellarator, a week before the FPA meeting, provided a fitting backdrop to several presentations on stellarators’ potential to serve as the basis for a future Demo device. W7-X, a superconducting stellarator, is numerically optimised for transport and magnetohydrodynamic (MHD) stability. Its maximum expected heating pulse is 30 minutes, said Mike Zarnstorff from PPPL.

A number of stellarator characteristics may be needed to make fusion commercially viable, Zarnstorff said. These include: no disruptions; no current drive and low recirculating power with higher fusion gain; steady-state magnetic fields and plasma; and sustained high pressure (with beta 5% or above). However, the stellarator configuration must be optimised to achieve these characteristics and this is the focus of aggressive stellarator research programmes in the EU and Japan.

Those countries have mapped out a path to Demo that includes the stellarator option, and both have large superconducting stellarator experiments: Europe has W7-X (above) and Japan has the LHD, which has operated since 1998.

Japan’s Demo strategy is to develop the tokamak and the stellarator/heliotron in parallel. In 2027, Japan will assess progress of both and it will decide on a Demo approach and construction schedule around 2030.

Japan plans to begin deuterium experiments in LHD in February 2017, and is currently upgrading it to include neutral beam injection, electron cyclotron heating, ion cyclotron range of frequency heating and advanced diagnostics. The goals of the deuterium experiments are to maximise confinement performance, study isotope effects on plasma confinement, demonstrate confinement of high-energy ions and validate modelling for extrapolation. Additional research will cover MHD stability at high-beta and low collisionality, divertor optimisation and plasma wall interactions.

Like Japan, the EU hopes to make a decision on Demo by 2030. Over the next 14 years, the EU intends to develop the basis for a W7-X-like fusion power plant; develop and demonstrate power-production scenarios for a stellarator Demo; validate models and the design approach; develop and demonstrate a steady-state divertor; and produce 10MW of power for 30 minutes.

The US role in stellarator research is more oriented to basic research, improving the numerical modelling and validation of 3D physics understanding, improved 3D optimisation, and designs for stellarator pilot plants and scoping studies for PPPL’s Fusion Nuclear Science Facility.

PPPL in the USA is interested in research on both the spherical tokamak and the stellarator that could lead to a Demo-scale power plant, said PPPL’s Stewart Prager. The lab also is pursuing liquid metal research as a novel solution to the first wall problem. He noted a large gap in the world stellarator programme: there are only two large stellarators and only one is optimised. This leaves an opening for PPPL to take a position that includes forefront stellarator research. Stellarators, he said, are "arguably the most physics-optimised fusion concept".

PPPL has also recently completed a major upgrade of its National Spherical Tokamak Experiment (NSTX-U), which will help with establishing sustained, high-performance plasmas, advancing both toroidal confinement physics for Iter and beyond, and the spherical tokamak as a candidate for a next-step fusion facility.

A key question is whether deuterium-tritium (DT) stellarator operation needed before developing plans for Demo?
It is clear that DT operation would reduce risk, but add a step to the process. Participants at a March 2015 meeting in Japan debated whether validating integrated models using Iter and non-DT large experiments could reduce these risks. The general conclusion was that DT operations would be needed, but that a final decision would depend on research advances. The EU Demo plan calls for DT stellarator operation before a decision is made on the design concept.

Fusion materials

The environmental attractiveness and economic competitiveness of all fusion power will directly depend on the materials used in power plants, said Steve Zinkle from Oak Ridge National Laboratory. Environmental attractiveness will require materials that can protect the public and the environment from radioactive releases and accidents. The waste should be low-hazard and short-lived, with reduced activation materials and low-tritium sequestration materials. Economic competitiveness will depend on high-performance and long-lived materials, short repair and outage times and high-thermodynamic efficiency.

The major fusion materials development challenges include:

  • Plasma-facing components: Will tungsten work, and how long can it survive without embrittlement?
  • Tritium containment: Can materials be developed that will prevent tritium leakage and allow for on-line extraction and fuel reprocessing?
  • Non-structural materials, including plasma diagnostics (e.g. optical fibers, electrical insulators), plasma heating feedthrough insulators and next generation magnet systems and ceramic breeders. Many of the materials now used have no operating experience in a DT environment as expected in Demo. Structural systems options would work well for the first year, but new, longer-lived options are needed for commercial systems.

In structural materials, the fusion community can leverage work being done for fission power plants, and research and development NNSA is doing for inertial confinement fusion and the weapons programme. However, in other areas, there is not such a strong synergy, particularly in tritium containment and non-structural materials.



Tritium containment presents a particular problem. Strong neutron bombardment creates microscopic cavities that retain significantly higher amounts of tritium than unirradiated materials. Kessel pointed said PPPL’s Tokamak Fusion Test Reactor had a "massive" number of cavities that trapped more than 100 times as much tritium as expected. This level of tritium retention could be a public safety hazard in a machine as large as Demo, or even Iter.
Researchers also are working on low-activation steels. Kessel said it is now possible to computer design high-performance steels. Computational thermodynamics modeling has identified potential new thermomechanical treatment processes for commercial 9-12% chromium steels but some, such as hot rolling, may be hard to implement on some product forms and cannot be used in weldments.
EU researchers have embarked on a programme to design next-generation reduced-activation ferritic-martensitic steels. This includes developing a computational model for the selection of 9%Cr steels optimised for high-temperature applications. Mechanical properties and microstructural investigations are under way, with irradiation tests planned for 2016-2017.
Fusion researchers are looking at a broader range of possible materials, including advanced ceramicsand materials developed for other applications. Kessel pointed to SiC/SiC composites now being qualified for jet turbines by a joint venture of GE and Snecma. The first deployments will be the Airbus 320neo in 2016 and the Boeing 737 MAX in 2017. Developers estimate that higher temperatures and lower weight will produce fuel savings of about 15%.
Two new SiC fibre and CMC fabrication plants for the new composite will be built in the USA. Successful use of the composites in the airline industry should spur development of improved SiC fibres and lower cost composites for other applications, including possibly fusion, Kessel said. Currently, the composites cost between 100 to 1000 times more than metals for the same applications.
Kessel pronounced himself "bullish" on structural materials, noting the high performance structural materials available for nuclear environments. There is a high level of confidence in the suitability of such materials for fission neutron environments, but uncertain suitability for fusion beyond about 5MW-yr/m2.

Japan maps Demo requirements
In a report published in September, a high-level team of Japanese fusion scientists defined the goal of Demo as demonstrating fusion energy to be economically and socially competitive with other power plants. It should be aimed at steady and stable electricity generation at levels beyond several hundred megawatts, amenable to commercialisation and able to breed enough tritium for a self-sufficient fuel cycle.
Hiroshi Yamada, science advisor to Japan’s Ministry of Education, Culture, Sports, Science and Technology, led the that established the technology bases required for Demo. The team’s report, Development of Strategic Establishment of Technology Bases for a Fusion Demo Reactor in Japan, was published online on 26 September 2015 and is available in English.
Yamada said that the ministry has called for development of alternative concepts such as helical and laser systems, along with tokamaks, "in a strategically linked manner".