Page 1794 – Physics World

Home » Page 1794

“ViriChip” shows up viruses

Henderson and colleagues tested their atomic force microscopy-immunosensor assay device on six group-B “coxsackieviruses”. The researchers applied just a microlitre of sample to the chip: if a particular virus was present in the sample it attached to a specific antibody region. AFM was then able to detect the presence of the virus at that region, indicating which type of virus it was.

The read-out does not require labelling or molecular or biological amplification and the researchers found it was easy to distinguish virus particles from debris. “The technique is currently being used by researchers and we hope it will be available for doctors and hospital pathology labs in the next two years,” said Eric Henderson, chief scientist at BioForce Nanosciences.

“This is the first time that scientists have been able to apply droplets of an antibody routinely on the micron to nanometre scale to a surface of a material like a silicon chip,” said Henderson. “In principle you can fit thousands of different antibodies on one chip and use it to test for thousands of different viral infections simultaneously, using just one sample from a patient. This means patients will not have to provide large blood samples, just a single drop will be sufficient.”

This story first appeared on nanotechweb.org

Closing in on black holes

The centre of our galaxy lies within a cluster of stars in the constellation Sagittarius. Though hidden behind a dusty cloud of gas, a bright radio source known as Sagittarius A* marks the spot where over three million solar masses of dark matter have collapsed to form a highly concentrated source of gravity. Some 10 million stars swarm within a light-year of Sagittarius A*, and at least three or four of them have been seen to swing within a few light-days of it.

Using one of the four dishes in the Very Large Telescope (VLT) in Chile, Reinhard Genzel of the Max Planck Institute for Extraterrestrial Physics in Garching, together with colleagues in Germany, Israel and France, was surprised to find that the infrared light produced by Sagittarius A* during an explosion is actually modulated with a period of only 17 minutes. This cyclic repetition is almost certainly associated with the orbital motion of the hot gas around the black hole.

For a black hole with a mass of 3.6 million solar masses, such an orbital period means the emitting region is only one-tenth the distance between the Earth and the Sun above the event horizon.

In the January issue of Physics World Fulvio Melia in the Department of Physics and Steward Observatory at the University of Arizona in the US describes these results in more detail.

Climate model predicts even hotter summers

Beniston used the HIRHAM regional climate model developed by the Danish Meteorological Institute to run two 30-year simulations. The “current climate” simulation was run for the period from 1961 to 1990, while the “greenhouse-gas climate” simulation covered from 2071 to 2100.

For both periods, Beniston first studied the fluctuations in the daily maximum summer temperatures in Basel – a town in northwest Switzerland that is close to the French and German borders – averaged over July, August and September. He also analyzed the number of days in which temperatures exceeded 30°C. He found that 2003 was clearly the hottest summer since 1901.

He then extended the model to different parts of Europe and found that in the latter half of this century, summer temperatures would increase across much of the continent leading to a “northward shift” in climatic zones. This means that Switzerland could have a climate similar to that in the South of France today.

Beniston observed a general increase of about 4°C in a band stretching across central Europe to the Black Sea, with greater increases over the Iberian Peninsula and the south west of France. Moreover, he found that the number of hot days would increase – particularly in the Mediterranean region and in Eastern Europe – with an additional 40 to 60 days or more above 30°C (see figure). In comparison, the period 1961 to 1990 saw an average of around 10 days.

“Perhaps the media attention given to these and similar results might motivate policy-makers to start thinking about the long term – and the implications of such climate change on health, water availability and quality, food security and so on,” Beniston told PhysicsWeb.

Condensate research heats up

A group of particles undergoes Bose-Einstein condensation when their de Broglie wavelength becomes comparable with their average separation and they all collapse into the same quantum ground state. Bose-Einstein condensation plays a central role in superfluidity and superconductivity, and has been a subject of intense research in low-temperature physics since an atomic gas was condensed for the first time in 1995. However, it should be easier to create a Bose condensate with excitons because they are lighter, and therefore condense at a significantly higher temperature than atoms.

Excitons are unstable because the electron and hole can recombine to emit a photon, or lose their energy in another way. In 2002, Daniel Chemla and co-workers at Berkeley and Santa Barbara made a semiconductor structure that was designed to confine the electrons and holes in the excitons in different quantum wells. The excitons were more stable in this structure and therefore could be cooled down more easily.

The same team has now found that it can confine tens of thousands of excitons in an area of about 10 microns squared by shining a laser of moderate intensity on the structure at temperatures around 10 Kelvin (see figure). It appears as if the particles have been “trapped” by naturally occurring defects in this region. According to Chemla and co-workers this is consistent with what is expected in a Bose-Einstein condensate but still does not provide definite proof of a condensate.

“These insights could lead to new strategies for engineering confined exciton systems and open up new opportunities for making Bose-Einstein condensates in solids,” team member Chih-Wei Lai told PhysicsWeb.

Wolf prize goes to particle theorists

There are four fundamental forces in nature: gravity, electromagnetism, and the strong and weak nuclear forces. Gravity and electromagnetism are both long range forces, whereas the strong and weak interactions only operate inside the nucleus. The Standard Model includes all of these forces except gravity. However, prior to the work of Brout, Englert, Higgs and others, the Standard Model could not explain why some particles have mass and others do not. In particular, it could not account for the fact that photons – the particles that “carry” the electromagnetic force – have no mass, while the analogous particles for the weak interaction do have mass. The particles carrying the weak force must have large masses to explain why it only acts over short distances.

In 1964, Brout and Englert proposed that the weak and electromagnetic interactions could be united by “spontaneous symmetry breaking”. This phenomenon was well known in condensed matter physics where it had been used to explain how, for instance, tiny unordered regions of a magnetic material could suddenly align themselves in a specific direction.

Higgs independently came to the same conclusion as Brout and Englert and showed that the particles that carried the weak force acquired their mass through interactions with an all-pervasive field that is now known as the Higgs field, and that the interactions occurred via particles that are widely known as Higgs bosons. Although these particles have not yet seen in an experiment, many physicists believe that they will be observed when the Large Hadron Collider (LHC) comes on-line at CERN in 2007.

The Prize, worth $100000, will be presented in Jerusalem in May, by the president of Israel, Moshe Katsav. The official citation states that the awards are made “for pioneering work that has led to the insight of mass generation, whenever a local gauge symmetry is realized asymmetrically in the world of sub-atomic particles”.

Argon accuracy reaches new limit

The “Segrè chart” in nuclear physics is a plot of proton number versus neutron number. Stable nuclei tend to have about the same number of protons and neutrons and form the so-called valley of stability, although heavier stable nuclei tend to have more neutrons than protons. The valley of stability is bordered by regions containing short-lived nuclei that decay into more stable isotopes through a variety of different mechanisms. In neutron-rich nuclei, a neutron can decay into a proton, an electron and an antineutrino. And in proton-rich nuclei – like argon-32 and 33 – a proton decays into a neutron, a positron and a neutrino.

When the argon nuclei undergo beta decay the positrons and neutrinos are not emitted isotropically in space, but at an angle to each other. This angular correlation can be characterised by a “beta-neutrino correlation coefficient”, which can be used to place constraints on aspects of the weak interaction that are not included in the Standard Model. However, this requires accurate values for the masses of the nuclei involved in the decay.

Previously researchers had to calculate this coefficient using masses derived from theory because experiments were not accurate enough. However, Klaus Blaum and co-workers have now measured the masses of argon-32 and argon-33 with an accuracy that is an order of magnitude better than previous results. The new results could therefore be used to calculate improved values of the beta-neutrino coefficient.

The ISOLTRAP mass spectrometer consists of a radiofrequency quadrupole ion trap and two “Penning” traps. The improvement in accuracy is mostly due to the addition of a new carbon-cluster source to the device. The atomic mass unit is defined in terms of the mass of a carbon-12 atom, so the new carbon source allows the absolute masses of the argon nuclei to be determined. The team now hopes to measure other “exotic” nuclei, such as magnesium-22 and gallium-62.

Supersolid is seen in the lab

All atoms are either bosons or fermions depending on whether their intrinsic angular momentum or “spin” is an integer or half-integer in quantum units. When bosonic atoms like helium-4 are cooled to near absolute zero, they can all collapse into the same quantum ground state to form a Bose-Einstein condensate. Fermionic atoms, on the other hand, obey the Pauli exclusion principle and cannot form such a condensate.

When liquid helium-4 is cooled to below about 2 kelvin, it undergoes Bose-Einstein condensation to become a superfluid. Although theory predicts that superfluidity should also exist in solid helium-4, no such supersolid phase has ever been observed in an experiment.

Chan and Kim first compressed helium-4 gas into a small glass disk made of fused silica that contains atomic-sized pores. Next, they placed this disk in a “torsional oscillator” capsule (see figure) and applied a pressure of over 60 atmospheres. Then they rotated the capsule and measured its oscillation rate as the temperature was lowered to almost absolute zero.

The pair observed that the oscillation rate of the capsule increased suddenly when the temperature fell below about 175 millikelvin. According to Kim and Chan, this meant that the helium-4 had “decoupled” from the pores in the glass disk and had entered a supersolid phase. “In this supersolid, the individual helium-4 atoms are continually flowing – without any friction – but because all the particles are in an identical quantum state, it remains a solid,” Chan told PhysicsWeb.

To test their findings, the duo repeated their experiment with helium-3 atoms – which are fermions. As expected, they found no change in the oscillation period of the capsule. “If our results are confirmed, this means that researchers have now been able to observe Bose-Einstein condensation in gases, liquids and solid phases,” write Kim and Chan.

Lasers overcome the Lorentz force

The proposal builds on existing techniques in which the electric field in a high-intensity laser beam is used to ionize an atom, and then accelerate the free electron away from its parent ion. When the electric field changes direction, the electron is driven back to the parent ion. An ultrashort pulse of ultraviolet radiation is created when the two re-collide, and this pulse – or the electron itself – can be used to probe events on timescales of attoseconds (10^-18seconds).

At present, the re-collisions are only efficient if the electrons have energies that are lower than about 1 kiloelectronvolt. At higher energies, the Lorentz force due to the magnetic field in the laser beam pushes the electron in the same direction as the laser beam, which means that the electron “misses” its target.

Now, Brabec and colleagues have shown, in theory, that they can eliminate the Lorentz force if they replace the laser beam with a pair of circularly polarized laser beams travelling in opposite directions (see figure). Moreover, they have calculated that the combined electric field of the two lasers can focus electrons with energies as high as megaelectronvolts back onto their parent ions.

“The resulting electron currents are comparable with the currents achievable at large-scale accelerators,” Brabec told PhysicsWeb. “We have shown that these currents are sufficient to excite nuclear reactions. On top of that, the electron pulse is extremely short, and this opens the possibility of making time-resolved measurements on nuclear processes on an attosecond timescale.” In addition to nuclear physics, Brabec and colleagues believe that the idea could be used in any situation where the Lorentz force plays a role, such as in plasmas.

Fellow travellers with the Nazi regime

In his 1932 Nobel-prize lecture delivered on 11 December 1933, Werner Heisenberg declared that “the path so far traced by the quantum theory indicates that an understanding of those still unclarified features of atomic physics can only be acquired by foregoing visualisation and objectivisation to an extent greater than that customary hitherto”.

Philipp Lenard

He may not, of course, have foreseen that this kind of proposition – together with Einstein’s relativity – was precisely what was later defined as highly undesirable “Jewish physics” by the Nazi government in Germany. Indeed, Heisenberg was eventually investigated by SS chief Heinrich Himmler. Himmler decided that it was in the interests of the Nazi regime to leave Heisenberg alone, despite attempts by disreputable individuals such as the physicists Philipp Lenard and Johannes Stark – hypnotized as they were by Nazi obsessions – to dispose of Heisenberg, whom they saw as a “white Jew”.

Heisenberg is one of the central characters – perhaps the central character – in John Cornwell’s readable account of the history of German science, mathematics and technology during, and shortly before, the infamous Nazi era. As Cornwell observes: “Heisenberg’s pact with the [Nazi] regime was an implicit agreement to separate science from the scientist – theoretical physics from Einstein’s Jewishness. It was a shameful act, and it was done under duress.”

It would, of course, have been astonishing if a major European power such as Germany had not produced many scientists and engineers of great distinction or even genius during that period. Indeed, about one-third of all Nobel laureates in 1933 were German citizens. On the other hand, its national reputation had already been tarnished by events such as the first sinking of a civilian liner (the Lusitania) by a German U-boat in 1915 and by the introduction of poison gas as a tactical weapon during the First World War. Both episodes relied on German work in science and technology. As Cornwell asks: was this Germans behaving as Germans, or scientists in Germany behaving as scientists?

Worse was to come with the advent of the Nazi government in 1933. The “poison-gas scientists”, with their pseudo-logical excuses for the lethal consequences of their work, were followed by the even more pernicious practitioners of Nazi racial hygiene and eugenics. Disreputable voices began to emerge from the murky background, complaining about the alleged waste of resources on people who were mentally ill, which ultimately led to a programme to get rid of “life unworthy of life”. Anti-semitism – promoted by the central authorities – grew exponentially.

In physics, it is probably enough to list some of the famous names of those who were forced out of Germany in the 1930s. They included Bethe, Bloch, Born, Debye, Einstein, Franck, Gabor, Gustav Hertz, Herzberg, Hess, Hevesy, Schrödinger, Stern and Wigner, as well as the mathematicians Richard Courant, Emmy Noether and Hermann Weyl. The names will be familiar to anyone with an interest in physics or the physical sciences.

As Cornwell accurately points out: “In the lethal mix of power, fear, cruelty and dilettantism, pseudo-science began to flourish virtually unchallenged under the auspices of the SS in Hitler’s Germany. The direction of SS-sponsored ‘researches’ in time brought appalling suffering and death to their victims and degradation upon all those associated with their conduct and data.”

Perhaps those who remained (as far as was possible) outside the Nazi mainstream were later sustained by Planck’s advice to Heisenberg in 1933. “If we can guide even small groups of talented and right-minded young through these horrible times,” advised Planck, “we shall have done a great deal to ensure Germany’s resuscitation after the end.”

There are, however, profoundly irreducible difficulties with this opinion. For example, in 1943 Heisenberg visited the ancient Jagiellonian University of Krakow in Poland, to which he had been invited by the infamous governor-general Hans Frank, who was later hanged at Nuremberg after the war. There Heisenberg gave a lecture on modern physics to a lay “Germans-only” audience. As he spoke, 184 members of the professorial staff of the university were languishing in a concentration camp. They had been lured some months earlier to a meeting – supposedly to discuss the re-opening of the university – and then treacherously arrested by Frank’s henchmen.

More generally, the Polish intelligentsia was being systematically destroyed in accordance with detailed policies formulated before the Second World War. A special list was prepared in Berlin of 61,000 (sic) members of the Polish intelligentsia who were to be arrested and killed as German forces entered the country. Some 45 professors of the University of Lwow (now Lviv) – and even some members of their families – were identified by name and were executed on the first week of German occupation in 1941. Many of these people were well known to members of the German academic establishment.

Those scientists who, in effect, followed Planck’s advice and stayed in Germany must have known what was going on. Indeed, there is convincing evidence that they did know. Dubbing them “fellow travellers in the Nazi regime”, Cornwell points out that these scientists “remained on the face of it morally and politically aloof while arguing that science is an apolitical, neutral pursuit”.

Then there is the question of the bomb. Cornwell describes the famous period during 1945 when leading German scientists, including Heisenberg, were held under house arrest at Farm Hall near Cambridge by the British authorities. The author discusses the transcripts of the secret recordings that were made of their conversations, from which it seems reasonably clear that Heisenberg did not really understand the practical physics of the underlying bomb-making process. It is also clear that Walter Gerlach, who was among those detained at Farm Hall, was, in principle, willing to use Germany’s nuclear discoveries. “If Hahn has made this discovery, let us at least be the first to make use of it,” he said. Fortunately they did not succeed, and they all tried to hide behind their last refuge – patriotism.

The final 37-page section entitled “Science from the Cold War to the war on terrorism” is a kind of paternoster addressed to scientists and others. Those who have already been converted will welcome it. Those who have not will wonder how it really relates to what came before. The publishers describe this volume as an “epic and unique chronicle”. It is not quite that, but it is a useful, well presented and reasonably priced introduction to an important subject.

Controlled fusion: the next step

For the last 200 years the bulk of our energy has come from the burning of fossil fuels. However, this finite reserve is running out, and the only way to meet the world’s increasing energy demands is to develop alternative energy sources, such as renewable energy, nuclear fission and nuclear fusion. Fusion is the least developed of these but it has the potential to provide a virtually inexhaustible source of energy. It is also safer than fission, and would produce no “greenhouse gasses” such as carbon dioxide. A fusion reactor burning just 1 kg of fuel per day could produce a sustained power output of 1 GW.

The idea behind controlled fusion is to use magnetic fields to confine a high-temperature plasma of deuterium and tritium. One way to do this is to use a tokamak – a doughnut-shaped vessel in which a strong, helical magnetic field guides the charged particles around it (see Further reading). The nuclei in the plasma undergo fusion reactions that convert some of their rest mass into energy – in the same way that energy is produced by the Sun. In order to overcome the mutual Coulomb repulsion experienced by the two nuclei, the plasma temperature, T, must be extremely high – typically about 10 keV, which corresponds to almost 10⁸K. However, the density of the plasma, n, can be relatively low at about 10²⁰ m^-3. The resulting pressure in the plasma is therefore only about one atmosphere.

Although a fusion reactor will use the deuterium-tritium reaction, for operational convenience most current experiments are based on plasmas that contain only deuterium. However, we do have experience of working with deuterium-tritium fuel mixtures from the Tokamak Fusion Test Reactor (TFTR) experiment at Princeton in the US and the Joint European Torus (JET) at Culham in the UK. In the mid-1990s the TFTR produced a peak fusion power of 10.7 MW, while JET – which is the world’s largest tokamak – reached 16 MW.

The next big step in fusion research will be the International Thermonuclear Experimental Reactor (ITER), which is designed to produce up to 500 MW of fusion power. The ITER collaboration – which consists of researchers from Canada, China, Europe, Japan, South Korea, Russia and the US – is currently negotiating where ITER will be built. Although ITER will not be used to generate electricity, it will allow us to explore the plasma conditions in a fusion reactor. A commercial fusion reactor would be only slightly larger than ITER and would produce a power of about 4 GW.

When a deuterium and a tritium nucleus undergo fusion they produce an alpha particle, a neutron and 17 MeV of energy. The aim is to use the energy of alpha particles to maintain the plasma at a steady temperature, thus allowing the reactions to be self-sustaining and leaving the neutrons – which carry 80% of the fusion energy – to boil water and drive steam turbines. For this “ignition” condition to be met, however, the triple product of the plasma density, plasma temperature and the energy confinement time – nTτ_E– must be greater than 3 x 10²¹ keV m^-3s. The energy confinement time, τ_E, is the characteristic time that it takes for the plasma to cool once the heating is switched off; a typical value for a fusion reactor is a few seconds.

Figure 1

This triple product, which is used to describe the performance of a reactor, has increased by almost four orders of magnitude since the first generation of tokamaks in the late 1960s. It is a little-known fact that this progress is slightly faster than that in other high-technology fields (figure 1). For example, the energy of particle accelerators has doubled every 3 years, and the number of transistors that we can fit on a piece of silicon has doubled every 2 years (Moore’s law). The performance of fusion plasmas, on the other hand, has doubled every 1.8 years.

Superconducting tokamaks

Most current fusion experiments run in a “pulsed” mode that lasts a few tens of seconds, but a fusion reactor needs to be able to operate continuously. The pulse length in these experiments is limited by the magnetic-field coils, which are usually made of copper and therefore consume a lot of electrical energy just to maintain the steady magnetic fields. A fusion reactor will use superconducting coils and will therefore be able to operate continuously with very low power consumption. Superconducting-coil technology has been used in several deuterium-deuterium plasma experiments, including the T7 and T15 experiments in Russia (T7 has been rebuilt as HT7 in China), and the TRIAM-1M tokamak and Large Helical Device (LHD) in Japan.

The Tore Supra tokamak at Cadarache in France is the largest operating superconducting device in the world, with a radius of 2.4 m and a toroidal magnetic field of 4 T. The superconducting magnets of Tore Supra, which are cooled by superfluid helium at a temperature of 1.8 K, have performed reliably for 16 years. Indeed, their innovative technology has formed the basis for the cryogenic systems of the CEBAF particle accelerator in the US and the Large Hadron Collider currently being built at CERN.

Almost all the fusion devices that are either planned or under construction rely on superconducting-coil technology, including Europe’s Wendelstein 7-X (W7-X), China’s HT7-U (now renamed EAST), Japan’s JT60-SC, South Korea’s KSTAR and India’s SST-1. When it comes online in 2015, ITER will be the first large superconducting facility that has the capacity to use deuterium-tritium plasmas under reactor-like conditions. The challenge between now and then is to learn how to control fusion plasmas for longer periods with a view to full, steady-state operation.

A plasma in a tokamak is initially heated by its own plasma current, which is induced by the primary transformer. But the heating power that can be drawn from this inductive source is limited because it depends on the resistivity of the plasma, which decreases as the temperature rises. In order to obtain the temperatures required for fusion we need to use external heating methods. A non-inductive plasma-current drive is also necessary for a fusion reactor to operate continuously, rather than in short pulses, otherwise the plasma duration is limited by the capacity of the primary transformer. Heating and current-drive methods have been successfully demonstrated using neutral-beam injection, in which high-energy beams of deuterium or tritium ions are injected into the plasma, and radio-frequency (RF) heating, in which the plasma is subjected to radio waves at resonant plasma frequencies.

Fortunately, a tokamak has a natural current known as the “bootstrap” current, which is driven by the pressure gradient in the plasma. A fusion reactor is likely to rely heavily on the bootstrap current, along with neutral-beam injection and RF heating.

Recent results have greatly improved our understanding of the role played by the plasma current in both stabilizing the plasma and confining its energy. In earlier tokamaks most of the current was driven inductively and the current distribution was determined by the resistivity, and therefore the temperature, of the plasma. By modifying the current distribution non-inductively we can reach regimes in which energy losses are reduced and confinement time is increased. The heating and current-drive systems for ITER require a pulse duration of at least 1000 s, and a vigorous R&D programme is under way to develop suitable systems.

Powerful negative-ion sources for neutral-beam injection have already helped researchers reach the long-pulse target. A beam of deuterium ions with a current of 80 A m^-2 – which were then neutralized in a gas cell so that they could penetrate far into a dense plasma – has been produced for 1000 s at Cadarache. In addition, RF power generators developed by Thales Electron Devices for W7-X have produced a power of about 0.3 MW for almost 22 minutes while operating at 140 GHz.

Keeping components cool

A further limiting factor for plasma duration is the need for the “plasma-facing components” (PFCs) to withstand very high fluxes of particles. These components include the exhaust systems, the vacuum-vessel walls, and the heating and current-drive antennae. The average power flux in a reactor might be about 1 MW m^-2, but the load is not uniformly distributed and some PFCs could receive a flux 10 or 20 times larger than this value. This corresponds to a surface temperature of thousands of degrees, which would cause serious damage to the machine.

In fusion experiments, however, the problems of heat removal are simplified when the plasma duration is of the same order as the thermal time constant of the PFCs (typically a few tens of seconds). Most current machines can therefore rely on inertial cooling – in other words the components are simply allowed to heat up adiabatically and then cooled slowly between pulses. A fusion reactor, on the other hand, will have to run under steady-state conditions, and will therefore need active cooling systems that can maintain the PFCs at a stable temperature. A prototype plasma-facing component for ITER has already been tested to withstand 2000 cycles at a power flux of 20 MW m^-2, which is almost one quarter of the heat flux on the surface of the Sun. But going from hi-tech prototypes like this to large-scale production presents a formidable industrial challenge.

During a recent upgrade, all the internal plasma-facing components in Tore Supra were replaced with ones that are actively cooled with water (see Garin et al. in further reading). A major part of this work was the installation of a “toroidal limiter” that has a heat-exhaust capability of up to 15 MW at a peak power flux of about 10 MW m^-2. The limiter is a continuous belt around the torus that is in contact with the plasma and is made up of 574 individually cooled elements called fingers (see box: “How a tokamak works”). The device had to be installed with a positional accuracy of about 1 mm to ensure an even distribution of the heat load. ITER will use an internal component called a divertor to handle similar fluxes of heat and particles.

Developing the limiter involved close co-operation with industry. Each of the 574 elementary components is armoured with 21 tiles made of a carbon-fibre composite, which are bonded to a water-cooled heat sink made of a chromium-zirconium-copper alloy (see Schlosser et al. in further reading). Such was the complexity of the components that it took four years to manufacture them.

Tore Supra’s cryostat – which contains the superconducting-magnet system – was also upgraded with small poloidal limiters called bumpers to cope with transient loads at the beginning and end of the plasma discharge. In addition, a set of 10 actively cooled neutralizers were installed below the toroidal limiter. These devices intercept the particles coming from the region between the plasma and the walls, and then neutralize them before directing them into pumping ducts. The peak heat flux on these neutralizers can reach 15 MW m^-2. The surface temperatures of the PFCs at Tore Supra are now monitored by actively cooled infrared endoscopes.

Steady-state operation

Tore Supra is now able to provide new information on long-pulse operation that will be relevant for a fusion reactor. Deuterium plasmas with durations of up to 390 s have recently been obtained with the machine, corresponding to a world record for injected energy of 1.1 GJ – four times higher than its previous record in 1996. The pulse length is mostly limited by the radio-frequency power emitters (or klystrons), which were only intended to operate for about a minute at a time. These klystrons will be replaced by devices designed to operate for a duration of 1000 s, which are currently being tested.

The world record for pulse duration is held by the TRIAM-1M tokamak in Japan, which ran for more than three hours. However, the machine performed with an injected energy of only 110 MJ and at a very low current, temperature and density. Real-time feedback control of plasmas has also been demonstrated in Tore Supra over a period of more than three minutes in steady-state conditions, where the plasma current was completely replaced by a non-inductive current driven by RF waves. This control has allowed researchers to constrain the relevant plasma parameters – such as current and density – to follow the optimum path towards steady-state operation. Real-time control will be essential for ITER.

Figure 2

One advantage of using the actively cooled PFCs is that their surface temperature remains fairly stable, as well as low. At Tore Supra the cooled elements remained at a temperature of 320 °C during 3 MW of plasma heating for four minutes, which meant that the density of the plasma was perfectly controlled (see figure 2a). This is one of the crucial issues for a fusion reactor, in sharp contrast to inertially cooled PFCs in which the plasma density often increases over time and hinders tokamak performance. Actively cooled PFCs also allow us to control impurities such as oxygen, carbon and metallic ions in the plasma, which dilute the main ion-plasma density and therefore the fusion reaction rate.

New physics

Figure 2

The complex interplay between physics and technology can only be seen in discharges with a long duration. Events in fusion plasmas take place on many different timescales, from milliseconds for the magneto-hydrodynamic (MHD) instabilities to several hours for the whole device to reach equilibrium (figure 2b). In between these timescales, the energy and particles in the plasma are transported over a period of seconds, while the plasma current reaches a steady-state distribution in a few tens of seconds. Most fusion experiments have a plasma duration time that is long enough to study MHD instabilities, and many also allow the various energy-transport processes to be studied. But only a few tokamaks have pulses that are long enough for the plasma current to reach a steady-state distribution.

Subtle effects can also occur because fusion plasma physics is complex and highly nonlinear. This means that a small change in the profiles of key plasma quantities, such as the density, temperature and current, can have a significant impact on the stability of the plasma and its energy confinement. Furthermore, current tokamaks cannot fully study the interactions between the plasma and the in-vessel components, such as the erosion of plasma-facing surfaces and hydrogen saturation. For example, when plasma-facing components get hot they release deuterium and other materials that enter the plasma and become ionized. This can cause undesirable effects such as increasing the plasma density and enhancing energy losses through plasma radiation.

As discussed earlier, full steady-state operation in a tokamak requires that the inductive plasma current is completely replaced by a non-inductive one. In other words, the toroidal electric field induced by the primary transformer must be made to vanish. At Tore Supra we have successfully achieved this, and discovered many important physics issues along the way. For example, an anomalous inward flux of particles that is responsible for the peaked radial profile of the plasma density has been seen for the first time. Such “peaking” is important because the fusion rate is proportional to the square of the density. Furthermore, a density profile that is strongly peaked helps to increase the bootstrap current, which will form an important component of the non-inductive current required for steady-state operation. The origin of this inward particle flux is still not fully understood, but the Tore Supra results suggest that it is related to turbulence in the plasma (see Garbet et al. in further reading). Particle transport is a subject of the utmost importance in fusion plasma physics, and it is now being studied in many other laboratories.

Figure 3

There were more surprises when the electron temperature was measured at the centre of the plasma during a long-pulse run (figure 3). The temperature of a very narrow central region was observed to oscillate in a sinusoidal manner for over two minutes. These oscillations – which were present with a variety of different current-drive combinations – cannot be explained by MHD instabilities. Computer simulations indicate that they might be caused by a nonlinear predator-prey interaction between the local heat transport and the current-density profile.

Furthermore, some plasma numbers did not add up. By carefully accounting for the particles injected into the machine and for those exhausted in the pumping system we found a deficit, indicating that a large fraction of deuterium gets trapped in the walls and components inside the torus. Carbon plasma-facing components are able to trap deuterium particles by different mechanisms, but the direct implantation of energetic deuterium ions does not explain the large retention observed since this mechanism should eventually saturate.

It could be that complex chemical reactions coming from the erosion and subsequent redeposition of material in the PFCs are involved. A vital clue to this was the observation of deposited carbon layers and carbon flakes in the machine. Plasma-wall interactions such as these have far-reaching implications for next-step devices, especially for the lifetime of PFCs due to carbon erosion or the build up of tritium in the walls. We need long-pulse operation to investigate such long-timescale issues. The next major challenge for the fusion community is to integrate various fields of physics and technology to produce a steady-state fusion plasma. The performance of the plasma could be further improved by suppressing turbulence in the plasma to improve the energy-confinement properties. Tokamaks with such an improved energy confinement would be smaller and could therefore significantly reduce the cost of a commercial reactor. The physics of fully non-inductive current plasmas – relevant for fusion reactors – is being studied intensively in current tokamak devices.

A rich harvest of new results is expected from several major projects around the world, especially from ITER and also from further upgrades to Tore Supra. This will open the way to the science and engineering of future fusion power plants. The next major challenge for the Tore Supra team, for example, will be to upgrade the heating and current-drive systems in order to operate at a higher injected power of 19 MW for 1000 s.

After several decades of research, we are confident that fusion energy is scientifically feasible. Plasma conditions that are very close to those required in a fusion reactor are now routinely reached in experiments. ITER will be the next major step forward. Its primary role will be to demonstrate a fusion power gain – i.e. the ratio of the fusion power produced to the power used to heat the plasma – of 10 for very long periods of time. ITER will also bring together all the necessary physics and technology required to build a fusion reactor. While it is under construction, however, many other issues will continue to be addressed in smaller-scale experiments that can operate with very long pulses.

Developing fusion as a viable, clean and unlimited energy source has taken a long time – much longer than early researchers envisaged – but the next few decades will see the realization of their dream.

Box: How a tokamak works

Future fusion

The main challenge in fusion is to create a self-sustaining reaction in which deuterium and tritium nuclei undergo fusion and release useful energy. One way to achieve this is to use magnetic fields to confine a hot plasma in a torus-shaped vessel called a tokamak. The strongest field (typically about 5 T) points in the direction that goes “the long way” round the torus and is called the toroidal magnetic field (red lines). This field is generated by external coils. The second component of the magnetic field points “the short way” round the torus and is called the poloidal field (green). This is typically less than 1 T and is generated by an electric current flowing through the plasma in the toroidal direction. The combination of these two fields creates a helix that twists round the torus (blue). Additional, weaker magnetic fields are used to control the shape and position of the plasma.

The plasma current, which is typically several mega-amps, is usually driven by a toroidal electric field induced by a transformer. Most of the available flux from this arrangement is used in configuring the magnetic fields, and in driving the plasma current in the early stages when the plasma is cold and resistive. Thereafter, relatively little flux is consumed in maintaining a steady current in a highly conducting hot plasma. This allows tokamaks to be driven inductively for long pulses – up to 1000 s for a fusion reactor. However, the current could be maintained indefinitely using non-inductive plasma currents. The current also heats the plasma by resistive (or ohmic) heating, but the maximum temperature that can be reached is too low to enable fusion to take place. A fusion reactor will be self-sustained by internal heating from alpha particles that are produced in the deuterium-tritium reaction. However, it will require some external heating and current-drive systems for start-up and plasma control. Various heating and current-drive methods, such as radio-frequency heating and neutral-beam injection, have allowed experimental tokamaks to achieve temperatures well above the fusion requirement.

Browse articles by content type