Sun in a Bottle: The Strange History of Fusion and the Science of Wishful Thinking

  • Charles Seife
Viking: 2008. 304 pp. $25.95 9780670020331 | ISBN: 978-0-6700-2033-1

It is 50 years since the first international symposium on fusion energy research took place in Geneva, Switzerland, as part of the second United Nations 'Atoms for Peace' conference. There, the United Kingdom, the Soviet Union and the United States announced the declassification of controlled fusion research, raising the hope of clean and limitless energy for mankind.

Fusion force: inside the tokamak used to confine plasma at the Joint European Torus, Culham, UK. Credit: EFDA JET

In his new history of fusion research, journalist Charles Seife argues that such grand hopes push researchers to make unjustified claims of major advances. But in pursuing the controversies generated by a few isolated individuals, Sun in a Bottle neglects the more important story of the wider fusion community.

Fusion scientists have not lived quiet lives.

At the first Atoms for Peace conference in 1955, its chairman Homi Bhabha said: “I venture to predict that a method will be found for liberating fusion energy in a controlled manner within the next two decades.” But the proceedings of the second conference in 1958 remind us that the scientific leaders of the main delegations were much less optimistic. Edward Teller from the United States said that the state of controlled fusion was “similar to the stage at which flying was about 100 years ago”, and that the technical difficulties of fusion “are likely to make the released energy so costly that an economic exploitation of controlled thermonuclear reactions may not turn out to be possible before the end of the twentieth century”. Similarly, reviewing work in the Soviet Union, Lev Artsimovich stressed that “worldwide collaboration is needed for progress”.

Soon after, a network of collaborations under the auspices of the International Atomic Energy Agency, the International Energy Agency and Euratom (the European atomic-energy community) was established in the domain of magnetic fusion. Results were openly shared and, two decades later, a major European facility, the Joint European Torus (JET), was constructed at Culham, UK. Scientific progress since then has been impressive — the foundations of a new 'plasma science' have been established. We now have a fundamental understanding of the complex collective processes prevailing in the hot, electrically charged gases known as plasmas, and have made significant technological advances in magnets, materials and high-power electrical systems. Consequently, fusion machines have improved greatly in their performance, both in fusion power and plasma duration. The US Tokamak Fusion Test Reactor (TFTR) and its European competitor JET, both of which use deuterium and tritium fuels, demonstrated power exceeding 10 megawatts, but only for short periods set by the limits of the auxiliary systems, notably the magnets. Other, smaller tokamaks and stellarator devices use superconducting magnets to confine the plasma and can be operated stably for much longer periods — 5 hours in the case of the Japanese tokamak TRIAM 1-M. Plasma science also provides industrial spin-offs. Plasmas are now common in low-consumption light bulbs, television screens and, through plasma processing, in nearly all electronic equipment.

Although JET and the TFTR have produced large amounts of power, these were less than the power consumed to heat the plasmas initially. A larger experiment, ITER — meaning 'the way' in Latin — was therefore designed with the aim of having a fusion power output that is ten times greater than the input power. ITER was originally designed by scientists from Europe, Japan, Russia and the United States. Its dimensions were defined by scaling laws derived from data collected worldwide. In 2004 and 2005, these four partners were joined by China, South Korea and India. Soon after, the seven partners agreed to construct ITER at Cadarache in the south of France. The ITER international organization, which was established by treaty in 2006, is a collaboration of unprecedented scale. Its seven partners represent more than half of the world's population. The annual budget for ITER construction is about €0.5 billion (US$0.7 billion). This may seem high, but it is a tiny investment compared with the annual worldwide cost of electricity, which stands at around €2 trillion based on the average cost per kilowatt hour in 2007 in the European Union.

The fiftieth anniversary of international fusion research was marked in October 2008 with many lectures on the history of fusion (see, which showed that fusion scientists have not lived quiet lives during this time. Teller and Artsimovich were right — the physics and technology of fusion are challenging, but the fusion community can be proud of its progress.

Yet this community will not recognize its own history in Seife's book. As admitted in the title, it is a rather strange history. After relating at length the early years of fusion research, it concentrates disproportionately on two table-top fusion fiascos — cold fusion and bubble fusion — generated by isolated individuals.

The book leads without proof to the dubious conclusion that “Over and over again, desperate scientists have deceived themselves and their peers — and cheated — in hopes to keep their fusion quest alive”. Yet it focuses on outsiders to the field, who thought they had made a major discovery and who, after having been proven wrong when their results could not be repeated by others, did not have the courage to admit their errors. Thankfully, the book does not identify any scandal within magnetic fusion research, the main line for fusion energy. But as a result, it pays too little attention to this large international community. Whereas the dream of limitless energy may afflict isolated scientists, it is certainly not applicable to all fusion researchers as the author suggests.

The book identifies correctly that peer review is a necessary prerequisite for preventing fiascos. Scientists who call on journalists to make announcements without having had the traditional discussions with colleagues followed by publication in a peer-reviewed journal are in danger of damaging their reputation. Yet peer review may not always be sufficient. In addition to internal reviewing, in the late 1980s JET set up an internal database on which its experimental data were made available. Researchers within the organization can easily cross-check a scientific claim made by their colleagues. An open, international, multidevice database followed in the early 1990s, which has proven to be a sound basis for progress in fusion research. Such a system protects scientists from the insidious distortion of reality that can be provoked when they remain too isolated. Maybe Seife himself is a victim of 'wishful thinking' and should have sought peer review before publishing such a strange thesis.