What all these tokamak plans lacked was money to build them, but that was soon to change. Bishop knew that the time was right to move into tokamaks. Even Congress had got interested in the Russian results and was enquiring of the AEC what it needed to catch up with the Soviets. So Bishop arranged a meeting of his standing committee of fusion advisers in Albuquerque, New Mexico, in June 1969 and called for any researchers with tokamak proposals to come and present them. The Oak Ridge team came and proposed Ormak; Texas researchers put forward the Texas Turbulent Tokamak and Coppi and his MIT colleagues suggested their compact high-field machine, known as Alcator, derived from the Latin phrase for high field torus. What Bishop wanted but didn’t get was a proposal from the Princeton Plasma Physics Laboratory, or PPPL. They were the torus experts and Bishop figured that the fastest way to duplicate the Russian results in a similar machine would be to cannibalise Princeton’s Model C stellarator. Although the Model C was racetrack shaped, if you removed the straight sections and added an electromagnet you would get a tokamak of roughly the same size as Russia’s T-3.
But Princeton wasn’t playing ball. PPPL researchers insisted at the meeting that the Russian results were mistaken and this led to heated argument. They were, however, fighting a losing battle. After days of constant pressure from the standing committee, PPPL director Mel Gottlieb finally caved in and agreed that it was vital to test the abilities of tokamaks and that Model C should be sacrificed for the purpose. The confirmation of the T-3 temperature by the Culham team a few weeks later sealed the Model C’s fate. But the AEC’s standing committee was limited by its budget and could only give two projects the green light: the converted Model C and Ormak.
Once the decision was made, the Princeton researchers didn’t waste any time. By September they had come up with a design. The plan was to increase the radius of the plasma tube – and so reduce the aspect ratio – and also to cut down the straight sections of Model C’s racetrack shape to 20 centimetres. But they learned from the Russians that having a uniform symmetrical magnetic field was important, so the design was changed again and the 20-centimetre straight sections removed altogether. This provided the proposed machine with a name: the Symmetric Tokamak, or ST. Once all the new components were ready, the Model C stellarator was switched off for the last time on 20th December, 1969 and in little more than 4 months the team transformed it into a tokamak.
Unlike the Russian machines, the ST was bristling with measuring devices. It was an unwritten law at the Princeton lab: everything has to be measured. Once they applied those instruments to the plasma in the completed ST they found that everything the Russians had said was true. The temperatures and confinement times were higher than anything yet achieved in the United States. US fusion scientists fell head over heels in love with the tokamak. In 1971, just a year after ST started operating, Oak Ridge’s Ormak was ready to go, soon followed by the Texas Turbulent Tokamak. The following year saw the completion of MIT’s first machine – Alcator A – while General Atomics in San Diego built an unusual tokamak with a kidney shaped cross-section called Doublet II, and Princeton built a second, the Adiabatic Toroidal Compressor or ATC, to test methods for heating the plasma. More joined them in the following years.
The United States was not the only country to jump on the tokamak bandwagon. Researchers at Culham were in the middle of building a new stellarator called CLEO and, as happened to Princeton’s Model C, it was rebuilt as a tokamak. France entered the fusion race by building, straight off, the top performing tokamak of the early 1970s. The Tokamak de Fontenay aux Roses (TFR) had the strongest toroidal magnetic field of the time and could generate a plasma current of 400,000 amps. Germany built a smaller machine, the Pulsator, at its fusion lab in Garching near Munich and the Italians built the small TTF in Frascati. Japan too joined in with its first tokamak, the JFT-2. And the Russians, not keen to relinquish their lead, built a string of new machines to test different ideas.
For researchers outside Russia, the arrival of the tokamak transformed the field. Throughout the 1960s Bohm diffusion had frustrated all their efforts, leeching energy and ions out of the plasma and making high temperatures impossible to reach. Their work had become an effort to understand the behaviour of plasma, rather than a race to produce a power-producing fusion reactor. But the tokamak seemed to show them a way forward to higher temperatures and longer confinement. True, they didn’t really have a thorough understanding of how it worked or why it was better than other devices but that would come, they believed, as they got to know the machine better. The important thing was that the race towards a fusion power reactor was back on.
Not that everything was plain sailing. Once researchers started to push the tokamak to the limits, its own menagerie of plasma instabilities began to make an appearance. The potentially most serious was dubbed simply a ‘disruption.’ In a disruption, the plasma current which does most of the work confining the plasma suddenly collapses to zero in a tiny fraction of a second, leading to a loss of plasma and temperature. More serious than that is the fact that the sudden loss of that huge current (in the millions of amps for a large tokamak) induces strong eddy currents in the vacuum vessel which put it under huge mechanical strain – equivalent to hundreds of tonnes for a big machine. Researchers were obviously keen to avoid such events because of the damage they could potentially inflict on their precious devices. Experiments showed that high plasma pressure, high current and impurities in the plasma, increased the chances of a disruption. They were soon able to draw up a diagram showing the boundaries of safe operation but such boundaries were a problem because high pressure and current are desirable if you want to get to the conditions needed for fusion. Finding ways to push back those boundaries became a high priority.
Another puzzling instability was discovered in Princeton’s thoroughly instrumented Symmetric Tokamak. Using an x-ray detector, the PPPL researchers detected an x-signal coming from the plasma that went up and down in a regular repeating pattern; on a plot it looked like sawteeth. The sawteeth were caused by rapid heating and then cooling of electrons in the core and they were also seen in Russia’s T-4 (an upgrade of T-3) and Princeton’s ATC. Sawtooth instability soon came to be considered a sign that a tokamak had reached respectable operating conditions and its core was good and hot. The Russian researcher Boris Kadomtsev developed a theory that seemed to explain the oscillations and, as they were relatively benign, they were considered in the 1970s as a success story of plasma theory. But as machines got bigger, so did the sawteeth to the point at which they were causing turbulence that spoiled confinement.
But perhaps the most important upshot of the boom in tokamak building during the 1970s was that it allowed researchers to develop formulas that linked plasma properties to tokamak size, known as scaling laws. Even if you don’t have a detailed knowledge of how a plasma works, scaling laws allow you to predict what sort of tokamak will give you the best results. It worked because there were so many tokamaks of different shapes and sizes that physicists could operate with different sets of conditions. By plotting the results from many machines on a graph of, say, confinement time against the major radius of the tokamak, you can plot a line that extrapolates from those results to predict what confinement you would get with an even larger tokamak. Researchers did similar plots of confinement against plasma radius, toroidal magnetic field, plasma current and electron density. Some of the results were surprising: although plasma theory predicted that confinement time would go down as the density of electrons increased, real results from tokamaks showed confinement clearly growing the more electrons there were. Overall, there was one clear message: if you wanted to get closer to the temperature and confinement time needed for a plasma to burn with fusion reactions, then the larger the volume of plasma the better. It was time for tokamaks to get big.
CHAPTER 5
Tokamaks Take Over
IN SEPTEMBER 1958 PAUL-HENRI REBUT, A YOUNG GRADUATE from France’s prestigious École Polytechnique near P
aris, arrived for his first day of work as a researcher at the Commissariat d’Énergie Atomique (CEA). He had been hired to join the CEA’s new fusion department but as he walked in he found the labs and offices largely deserted. His new colleagues were all in Geneva for the second Conference on the Peaceful Uses of Atomic Energy, the meeting that revealed to the world the secret fusion programmes of Britain, the United States and the Soviet Union. Before that meeting there had been some plasma physics research going on in Europe but, beyond Britain’s Harwell laboratory, little that was aimed at fusion energy. The Geneva meeting changed all that. Fusion programmes started up in several countries and, in the same month that Rebut started his new job, European collaboration in fusion began.
A few years earlier, European governments had been debating how to build on the success of the European Coal and Steel Community, the precursor to the European Economic Community (EEC) which later became the European Union. At the time the community consisted of only six members: France, Germany, Italy, the Netherlands, Belgium and Luxembourg. Some of these wanted the community to cover sources of energy other than coal, including the new atomic energy which, because of its high development costs, was a prime candidate for international collaboration. Others wanted to create a single market for goods across the member states. Trying to accommodate these divergent goals in one body was thought too difficult so a compromise was thrashed out. On 1st January, 1958 the coal and steel community was joined by two others, the EEC and Euratom, a body to coordinate the pursuit of atomic energy.
Fusion didn’t fit naturally into Euratom’s remit because the prospects of fusion power were some way off, so Euratom managers asked CERN, the recently-formed European particle physics laboratory in Geneva, if it would take responsibility for fusion. CERN set up a study group to investigate what fusion research was going on in Europe and what role CERN could play. But the CERN council eventually decided that as the ultimate aim of fusion is to generate energy on a commercial basis, such research was outside the limits of its statutes, which restricted it to pursuing basic science. So the ball was back in Euratom’s court.
On 1st September in the halls of the Palais des Nations at the start of the Geneva conference the vice president of Euratom, Italian physicist and politician Enrico Medi, sought out his compatriot and fellow physicist Donato Palumbo and asked him to head up Euratom’s fusion programme. Born in Sicily, Palumbo was a talented researcher who had graduated from the respected Scuola Normale Superiore in Pisa and then returned to Sicily to teach at the University of Palermo where he was later made a professor. He specialised in theoretical plasma physics so he was scientifically well qualified for the fusion programme but he was a quiet and unassuming man and so not an obvious choice for the cut and thrust of European politics. Palumbo at first refused, saying he wasn’t ready for such a job, but Medi persisted and Palumbo finally agreed.
The fission department of Euratom had a head start on fusion and by far the larger budget. Palumbo was given just $11 million for his first five-year research programme. The fission effort had set about creating a series of Joint Research Centres in various countries where Euratom work would be carried out. With his limited resources Palumbo knew that he couldn’t create anything that would rival the national fusion labs, which were then growing rapidly. He also disliked bureaucracy and hierarchies, and so decided to take a different tack. He set out to persuade each national fusion lab to sign a so-called association contract to carry out fusion research agreed collectively within Euratom. It wasn’t hard to persuade them because Euratom was offering to pay 25% of the labs’ running costs. The first to sign up was France’s CEA in 1959 and over the next few years all six Euratom members joined the fusion effort.
Palumbo may not have been a bureaucrat but he was a natural diplomat. Committee meetings at each lab always dealt with science issues first, with a bit of business discussed at the end. The central coordinating committee was delicately named the Groupe de Liaison, because France’s proud CEA and Germany’s Max Planck Society would not have liked anything that sounded too controlling. This low-key approach made him popular in the growing fusion community and reassured the national labs that Euratom was not planning to take over fusion research.
During the first decade of his tenure at Euratom, Palumbo’s fusion programme was mostly focused on understanding the plasma physics of confinement and heating. The labs built a variety of mirror machines, pinches and other toroidal devices. They experimented with heating using neutral particle beams and radio waves. The young Rebut found that the field was so new that there was no one there at the CEA who could teach him about plasma theory so he found what literature he could and taught himself, eventually making important contributions to the understanding of plasma stability. But trained as an engineer as well as a physicist, he soon got involved in designing, building and operating small fusion devices. He built a so-called hard-core pinch, a linear device that relied on a copper conductor along its central axis to carry the current to create the pinch rather than a plasma current. He moved onto toroidal pinches, again with a copper conductor along the axis, and became convinced that only toroidal devices would work because mirrors lost too many particles at their ends. This made him somewhat marginal in the French fusion effort since the CEA’s largest device at the time was a mirror machine at its laboratory in Fontenay aux Roses, a Paris suburb.
As a whole, European researchers were suffering the same frustrations as their colleagues in the United States: their machines were plagued by Bohm diffusion, confinement was poor, and funding was slowly dwindling. It was a crisis in Euratom’s fission section, however, that put the future of the programme in doubt. From the outset the organisation’s aim had been to collaboratively develop a prototype fission reactor that member states could then develop commercially. The favoured design was a heavy water reactor with organic liquid coolant, known as ORGEL. But the commercial appeal of separately developing their own reactors was too much for the member states and in 1968 the project collapsed, prompting savage budget cuts in all parts of Euratom. Palumbo was left with just enough money to pay the staff that Euratom employed in each of the associated labs. For a couple of years the programme stumbled on with a small amount of money from the Dutch government.
Just as the programme was at its lowest ebb, Russia announced its astonishing results with tokamaks at the 1968 IAEA meeting in Novosibirsk. Palumbo was in the middle of drafting a proposal for the next five-year programme but realised that the tokamak results changed everything. He started from scratch and arrived at the June 1969 meeting of the Groupe de Liaison with a new proposal. He suggested that the programme should concentrate its efforts on tokamaks and some other toroidal devices, that an extra 20% funding should be provided by Euratom for the building of any new device, and that a group should be set up to investigate building a large device collaboratively by all the associated labs. There was a heated debate over putting so much emphasis on tokamaks, but they reached an agreement and sent Palumbo’s proposal to the Euratom council for approval. Euratom, still smarting from the collapse of ORGEL, enthusiastically endorsed the fusion plan and even gave Palumbo slightly more funding than he had asked for. The laboratories set about drawing up plans for new machines.
The CEA, with its emphasis on mirror machines, was thrown into confusion when fusion fashion switched to tokamaks. Rebut, who had focused on toroidal devices, became the man of the moment. At the time of the Novosibirsk meeting he had been in the middle of designing a large pinch device, but he immediately abandoned it and began working on a tokamak design. When the Groupe de Liaison met again in October 1971, five new machines were approved with the new additional funding. At the same time, a small group was set up to investigate building a large multinational tokamak. This would-be machine was given the name the Joint European Torus (JET) – not ‘tokamak’ because German delegates thought it sounded too Russian.
Of the machines that were given the go-ahead at that meeting, the mos
t ambitious was Rebut’s Tokamak de Fontenay aux Roses (TFR). The design was roughly the same size as Russia’s T-3 and Princeton’s Symmetric Tokamak but the plasma tube, with a radius of 20cm, was larger so that it contained more plasma and had a lower aspect ratio. What made the TFR stand out was the huge amount of electrical power that was put into containment. Rebut designed a large flywheel that was accelerated up to a high speed over a long period and then acted as an energy store for each plasma shot. Connecting the flywheel to a generator created a huge electrical pulse which, via the tokamak’s electromagnet, drove the plasma current that pinched the plasma in place. The TFR was able to generate plasma currents of 400,000 amps for up to half a second, world records at the time.
While the TFR was still being built, the JET study group had to decide what the next generation of tokamaks would be like. They didn’t have many practical details to work with because few tokamaks had yet been built outside Russia. The Symmetric Tokamak had started operating in 1970 and Ormak in 1971; in Europe there was only Britain’s CLEO, converted from a part-finished stellarator. And they couldn’t rely on theory either; there simply was no conceptual understanding of how tokamaks worked. What they did know was that tokamaks got great results and that if you made them bigger their operating conditions would probably get better. The JET study group knew that they wanted a machine that got close to reactor conditions and produced a significant amount of fusion power – and that meant a plasma that would heat itself.
Piece of the Sun : The Quest for Fusion Energy (9781468310412) Page 12