The development of fusion, since the dawn of the nuclear age, has been the antithesis of fission. When the two processes were first studied it was fusion that looked most likely to succeed. It seemed simpler than fission, requiring materials that are abundant and offering the possibility of a self-sustaining power source, providing an inexhaustible supply of energy. The concept of electricity ‘too cheap to meter’ was one scientists used in the 1950s to describe the potential of fusion, but the phrase now haunts the fission industry as critics use it to describe the apparent failings of the nuclear industry to live up to the promises of it pioneers.

Whilst progress in fission technology progressed rapidly, its military potential ensuring it prospered in a Cold War climate, magnetic confinement fusion had few military or strategic implications and did not receive the same level of attention. Fission technology was guarded jealously, but fusion research collaboration crossed the iron curtain. In 1956 Nikita Khrushchev, then leader of the Soviet Union, visited the UK with Igor Kurchatov, the father of the Russian fusion programme. They visited the UK research site at Harwell and made a number of announcements about Soviet fusion research. The British head of research at the time, John Cockroft, was so impressed by the Soviet work and their willingness to share it, he persuaded the British government to declassify UK fusion research. Throughout the Cold War, with the arms race flat out, fusion research remained a model of East West cooperation.

But since the fall of the Berlin Wall, fusion’s diplomatic role has lost a lot of its usefulness and, with the ultimate goal of electricity generation still seemingly as distant as ever, the advocates of fusion research are having to develop new arguments to justify the programme.

The problems over the future of the fusion programme have come to a head in the last few months over the funding of the International Thermonuclear Experimental Reactor (ITER) project. Initiated by Reagan and Gorbachev, ITER is the next stage in the development of a fusion reactor based on the tokamak design originally developed by the Soviet Union. There are four parties to the ITER project, the United States, the European Union, Russia and Japan. In July 1998, following a six year design phase, the US Congress decided that continuing support for the project was not a sensible use of resources and the US effectively withdrew its portion of the financing of the next stage, to build a reactor, and instead advocated the pursuit of “low-cost approaches” to magnetic confinement fusion and to establish “a new international arrangement encompassing this and other fusion science areas”. As a result, the initial design, which would have cost $5.5 billion, is no longer feasible and a new design, costing in the region of $2-3 billion is now being developed. Despite economic difficulties Japan remains committed to the project, as does the European Union, which has agreed to extend its support for another three years. Russia cannot provide much in the way of financial support, but its expertise remains vital.

The United States has agreed to continue funding of the design phase until the end of the 1999 financial year, to achieve “an orderly closeout of our design activities”, and the other parties have agreed a three year extension to the design phase to address the question of ITER’s location as well as the development of the new design. The original aim of the ITER project was to build a machine which would achieve the goal of producing more energy than it consumed and a design criterion was to produce 1.5 GW of power; it would have been the one step between where we are now and a prototype commercial reactor.

According to Chris Carpenter, public affairs manager for the UKAEA fusion programme, the physics has largely been solved, the challenges are engineering ones.

“Clearly the costs were ambitious,” he said. “But people at ITER designed a machine to achieve its goal. Now that there is less money and political support the parameters are having to be tightened.”

In basic terms ITER is a larger version of the Joint European Torus (JET), currently the world’s largest fusion experiment, based at Culham. It has a major radius of 3 metres and in experiments in 1997 with deuterium, tritium plasma mix produced a Q ratio (power out / power in) of 0.65, the best yet achieved. JET is one of a series of designs based on a ‘doughnut’ shaped plasma [See box p30]. Smaller machines with the same proportions include the Compact Assembly (COMPASS) experiment at Culham and the ASDEX Upgrade device in Germany. The original ITER design had a major radius of 8.1 metres and extrapolation of experimental results from the existing experiments suggests it would have easily achieved a Q ratio greater than 10 and would theoretically have been able to reach ignition, a state where the alpha particles created in the fusion reaction produce enough energy to maintain the plasma at a temperature high enough for the reaction to continue.

With construction costs now having to be reduced by nearly 50% these ambitious criteria will have to be scaled down. However according to Robert Aymar, director of the ITER project, the overall objectives of demonstrating the scientific and technical feasibility of fusion should still be possible and the objective of making ITER the one step between JET and a prototype commercial reactor is still possible. The revised objectives for plasma performance are:

• Extended burn in inductively driven plasmas with the ratio of fusion power to auxiliary power Q>10 for a range of scenarios.

• Demonstrating steady-state operation using non-inductive current drive with Q>5.

• Not preclude the possibility of controlled ignition.

The engineering objectives remain similar to those of the original ITER design: to demonstrate the availability and integration of essential fusion technologies; to test components; and to test tritium breeding module concepts.

“From the preliminary work done to date I believe that the team will be able to approach the revised design targets, with a device that will be about two thirds the linear size of the reference design and could produce about half the output power,” said Aymar.

Tight aspect tokamaks

The history of science and engineering is littered with blind alleys, discoveries made by accident, a piece of lateral thinking usurping years’ of painstaking research and brilliant ideas being shelved only for a competitor to take advantage. Who would bet against the holy grail of fusion energy ultimately being achieved not primarily through the construction of bigger, more powerful and expensive machines, but by someone making a few calculations on the back of an envelope and producing a new design? It may be romantic, but it could already have happened.

The conventional tokamaks of COMPASS, JET and ITER have an aspect ratio of between 2.5 and 3.5. This is the ratio between the major and minor radii of a plasma. In a tokamak the plasma is a sphere, circulating a central core. The major radius is measured between the central core and the plasma edge, the minor radius measures the size of the plasma ring [See diagram on p30]. Theoretical physics suggested that a tighter aspect ratio, making the plasma more spherical in shape, would allow a greater plasma pressure for a given magnetic field. Alan Sykes, a physicist at Culham, did some calculations and concluded that an aspect ratio of 1.3 would mean that the magnetic field needed to hold the plasma in place could be an order of magnitude less than on a conventional tokamak of the same scale. The UKAEA decided to build an experiment, the Small Tight Aspect Ratio Tokamak (START).

START was built out of spare parts which engineers at Culham took from other projects within the laboratory.

“It is really just an aluminium can with a rod down the middle,” said Sykes.

START first began operation in 1991 and the results have created considerable excitement. START has achieved world record results for plasma pressure for a given toroidal field. Known as beta, START produced a value of about 40%, the previous record, held by the Doublet IIID tokamak at General Atomics in San Diego, was 12%.

Unsurprisingly there is now a major research effort into tight aspect tokamaks taking place with the construction of the Mega Amp Spherical Tokamak (MAST) at Culham and the National Spherical Torus Experiment (NSTX) at Princeton in the United States. Both tokamaks are currently being commissioned with MAST expected to be running experiments by the end of 1999. START was built on a shoestring, MAST is a piece of state of the art engineering and offers an increase in plasma volume of an order of magnitude, with a specially constructed stainless steel vessel, a higher standard of vacuum conditions and sophisticated power supplies and diagnostics.

The US provided a 40 kV neutral beam injector to heat the plasma within START. For the MAST tokamak two 70kV power sources are needed.

One of the most significant advantages of spherical tokamaks is that the current needed to create the toroidal field is low enough for conventional materials such as copper to be used in the central column, as opposed to ITER which will use superconducting niobium tin. START required a current of 0.5 MA, pulses of which were created using capacitor banks, while MAST will require 2 MA and will divert power from the national grid. The fact that superconducting materials are not needed to create the forces necessary offers many simplifications in engineering design. Both MAST and NSTX should provide a good indication of whether spherical plasmas offer a significant breakthrough or whether there are major problems which are as yet unseen.

The development of spherical tokamaks is part of the concept improvement section of the fusion programme at Culham, which runs in parallel to the work on conventional tokamaks, in particular the ITER project, and long term technology work on materials such as superconductors and tritium production processes.

Despite the success of START, Sykes fully supports the ITER project.

“Spherical tokamaks still need a lot of experimental work,” he said. “We need to learn how plasmas behave in a device of a power plant scale, even if when it comes to commercial electricity production a spherical design offers reduced costs.”

Maintaining support

Even before the US pulled out of the ITER project, support for fusion research had been falling for many years. Compared with 1980, in real terms US federal funding in 1997 was only 25% of the peak value. After forty years and $13 billion, governments and the public are becoming impatient and want some return on the investment. With fusion power still decades away, it is a possibility that if public and political support for the programme continues to wane, the ultimate goal may never be reached.

G Kulcinski and J Santarius of the Department of Engineering Physics at the University of Wisconsin, presented a paper, Reducing the Barriers to Fusion Electric Power, to the 1997 conference Pathways to Fusion Power, in which they identified six reasons for the falling off of support. Firstly there is a perception in the US that there is no desperate need for a new energy source. The movement in recent years towards small, highly efficient gas turbines is likely to have a major impact on other forms of generation such as coal, fission and renewables and will dominate the energy market for the next twenty years.

Secondly the large, expensive prototype devices necessary for the current fusion programme appear increasingly out of step with developments in other fields of scientific research. The third problem, linked to the second, is that fusion science has failed to develop an economically viable power plant concept.

“We have reluctantly come to the conclusion that the complexity, large size, radioactivity, and radiation damage problems associated with what has been placed on the table before the government, the utilities, and the public results in an electricity production plant that is too unreliable, and too costly to compete even with the renewables,” say Kulcinski and Santarius.

The fourth problem is that the fusion community has not integrated well with the electric utilities, none of which consider fusion in their long term plans. And the fifth problem, the lack of private investment in fusion research, also reflects the relative isolation of the fusion community.

The final difficulty is the lack of useful applications of fusion research which could be utilised in the short term and help justify the pursuit of fusion’s long-term aims. If this problem can be addressed, Kulcinski and Santarius argue that support for the fusion programme will be maintained.

“Until the public sees some tangible, and positive, results from the fusion programme, it will always have difficulty selling itself as an energy programme,” say Kulcinski and Santarius. “It is simply not acceptable to talk about 50 year development scenarios costing 10s of billions of dollars without giving some early tangible benefits to the public.”

However there may be ways in which fusion can be sold before it produces electricity. Kulcinski and Santarius single out two possibilities in particular which may have major economic potential; these are the transmutation of long lived radioisotopes produced in fission processes and the production of commercially useful isotopes. Other possibilities include developing devices to detect contraband such as explosives and drugs, alteration of material properties and generating pulsed images for inspecting welds. The authors argue that the focus of research should shift to developing practical uses for low Q devices.

The demand for radioisotopes for medical, industrial and research uses is growing rapidly. In 1994 the worldwide demand was $102 million, $59 million and $4 million for each application respectively. One isotope in particular, 99mTc, produced through decay of 99Mo, is the most widely used radionuclide in medicine. The worldwide demand in 1994 was $43 million, with an annual growth rate of between 5 and 10%. Currently the isotope is produced in fission reactors, but the D-3He fusion reaction produces protons at the 15MeV energy level, which is the peak energy for direct 99mTc production. Kulcinski argues in another paper that work should be carried out to develop a 300-watt D-3He fusion reactor with the aim of producing 99mTc. Other isotopes which can be produced in fusion reactors also have potential economic applications.

The crisis in fusion research is particularly acute in the United States and it is likely to be scientists there that first start to develop alternate uses of fusion processes which show potential in the short term. With the issue of the disposal of high level fission waste probably the most pressing problem the nuclear industry faces worldwide, the development of processes which could reduce the level of the problem would clearly be of interest (See table on p29).

Fission and fusion are complimentary processes and it may be that the potential for integration of the two has not received the attention it should have. By using one to address the problems the other form of nuclear energy is facing, a synthesis may be possible, to the benefit of both.

Fusion physics

Fusion is the process which powers the stars. Light elements such as hydrogen and helium fuse under great temperature and pressure, forming heavier elements and releasing energy in the process. The most promising fusion reaction is between hydrogen isotopes deuterium and tritium, the fusion of which produces a helium nucleus, otherwise known as an alpha particle, and high energy neutrons.
Magnetic confinement fusion takes place within a plasma, a state of matter where the atoms are stripped of their surrounding electrons. Temperatures in excess of 100m °C are necessary to achieve this state, hotter than within the sun. This is because the pressure within the plasma does not exceed two or three atmospheres. No materials could withstand this temperature without vaporising and as a result fusion reactors have to create powerful magnetic fields to hold the plasma in empty space. The conventional tokamak design creates the electromagnetic forces through an electric current running through a central core and another current running through the plasma. The forces hold the plasma in a tube running around the central core, which in the conventional design is shaped like a doughnut. The ITER design requires superconducting materials within the central core to produce a strong enough magnetic field, holding for a long period of time.
The energy of the alpha particle produced in the fusion reaction can theoretically be used to maintain the plasma temperature, while the high energy neutrons provide the useful power. However this has not yet been achieved and all fusion reactors so far built need an external power source, such as a neutral beam injector, to heat the plasma. In conventional designs the neutrons released in the fusion reaction are moderated using a blanket made predominantly of lithium. The kinetic energy transfers to the lithium atoms, whose relatively low mass ensures that the transfer is reasonably efficient. The lithium blanket is permeated with cooling coils which transfer the heat to boilers which then generate electricity in the conventional way.
The reaction between neutrons and lithium results in helium and tritium atoms. Tritium, unlike deuterium, does not exist naturally and the process can also, therefore, act as a breeder for tritium: Other potential fusion reactions include those between deuterium and helium-3, and deuterium with itself. However both of these reactions require significantly more stringent confinement conditions: Other possible fusion techniques includes inertial confinement, in which deuterium and tritium atoms are struck by high power lasers, causing a series of mini explosions, fusion, and the release of a burst of neutrons. The focus of research into this technique is the the National Ignition Facility at the Lawrence Livermore National Laboratory in the United States. France is also working on this technique.