It’s the announcement that’s making quite a bit of a buzz in the controlled fusion news arena right now, and it accompanies a publication in the journal Nature. South Korean engineers and physicists are proud to have reached a temperature of about 100 million kelvins for about 30 seconds in their experimental tokamak with superconducting magnets: Korea Superconducting Tokamak Advanced Research (KSTAR).

The first impression is a shrug. So what ? Temperatures of more than 500 million kelvins have been reached in tokamaks by magnetic confinement for decades and the world record for stability of a heated plasma for controlled fusion exceeds six minutes.

This is correct, but it turns out that the performance of the Koreans lies in a double success, first the heating and the state of the plasma is done in the jargon of the physicists of thermonuclear fusion according to a mode called H (from l English High confinement, that is to say high confinement), precisely the one that will be used in the Iter reactor, and finally the temperature/duration couple obtained is very close to the record reached by the Chinese in 2021 with Experimental Advanced Superconducting Tokamak (EAST), namely 120 million Kelvin for 101 seconds.

The meaning of these two combined prowess is that in both cases, it is indeed two mini-Iters, so to speak, and which operate according to the same modality, namely the H mode of which we already have spoken and with superconducting magnets. This is therefore one more reason to hope to achieve the expected successes with the machine, the construction of which is progressing well in the immediate vicinity of the Cadarache nuclear study center in Saint-Paul-lez-Durance.

The royal road to fusion by magnetic confinement

Let us recall if necessary that the Iter (International Thermonuclear Experimental Reactor) project consists of the construction of an experimental nuclear fusion reactor, born of a long-term international collaboration between 34 countries, but the first plasmas will not be obtained before 2027 and the production of fusion reactions in a mixture of two isotopes of hydrogen only a few years later.

It will be a solar mini-core on Earth but, since we cannot achieve exactly the density conditions at the center of the Sun for the fusion reaction, the interior of Iter will have to be 10 times hotter, that is to say say reach about 150 million kelvins. Of course, no material can resist for long an ionized gas transformed into plasma at such a temperature and it is for this reason that the plasma will be confined in the powerful magnetic fields of a tokamak (Russian acronym for “toroidal chamber with coils magnets”), as these machines developed by the former Soviet Union in the 1950s–1960s are called.


The key problem to be solved with a tokamak is that of the stability of the fusion reaction which must be maintained by producing much more energy than it consumes in order to be able, by 2050, to be exploited by reactors for the global industrial scale production of electricity.

Plasma is potentially very turbulent with instabilities that can lead to the equivalent of solar flares but during experiments in the 1980s and 1990s it was discovered that there was a new way to build fusion plasma with magnetic fields according to a so-called H mode, replacing the so-called L mode which had made it possible to reach some of the previous fusion records.

For the record and as Robert Arnoux tells it on the Iter site in a text from which we draw inspiration, mode H was discovered by Friedrich (“Fritz”) Wagner on February 4, 1982 by serendipity on the tokamak ASDEX at the Max Planck Institute for Plasma Physics (IPP) in Garching, Germany.

The German physicist related about this discovery in the experiment of heating of the plasma by injection of particles in the form of a neutral beam which he led that: “It came out of nothing. It was not planned, it happened…”

The plasma has become much less turbulent and if other instabilities still exist, subsequent work, notably with the Jet tokamak, has suggested that by building a sufficiently large machine operating in H mode, the holy grail of abundant energy and with very little radioactive waste had to be on hand.

Nuclear fusion: a reactor pushed to 100 million degrees for 30 seconds

Korean physicists have just taken an important step for the future of work on nuclear fusion with their experimental reactor Korea Superconducting Tokamak Advanced Research center (KSTAR); for 30 seconds, he managed to maintain a temperature of 100 million degrees Celsius. Excellent news for ITER, the major international project based in France.

The KSTAR is not at its first attempt; since 2008, this reactor has served as an experimental platform to study the concepts that will one day be used to operate ITER. And this combination of very impressive figures represents a great progress.

This temperature, although close to 7 times greater than that of the Sun’s core, does not constitute a record in itself. Same thing for the 30 seconds of operation. But the fact of having succeeded in to achieve simultaneously is a great first, and a new step towards commercial nuclear fusion.

Don’t touch the wall

Very vulgarly, the objective of a tokamak, like EAST, KSTAR or ITER, is to force atoms carefully prepared in advance to collide at monstrous speed. To generate this vast nanometric upheaval, it is necessary to maintain an absolutely infernal temperature of several tens of millions of degrees.

However, generating such a temperature is not easy, far from it; the engineers constantly seek to push back the limits of the various prototypes to reach the famous threshold of 150 million degrees Celsius. It is from this temperature (variable according to the machines) that the conditions become ideal at the threshold of the enclave, and that the fusion reaction can therefore begin within the plasma.

This furnace, no material in the world is capable of supporting it. To confine this superheated plasma, the tokamaks are equipped with gigantic electromagnets; they generate a magnetic field which keeps the ionized material at a good distance from the walls of the reactor.

It’s very important for the stability of the reaction, and it’s not just about productivity. Admittedly, there is no risk of a Chernobyl-type disaster in this context; but if the plasma comes into contact with the internal walls of the reactor, it can still cause catastrophic damage inside this extremely expensive and very difficult to maintain device.

And at this level, researchers have no room for error. The smallest point of contact between the superheated plasma and the internal walls close to absolute zero, as stealthy as-it, immediately disrupts the system; this then triggers a snowball effect that causes the reaction to fall like a soufflé.

A new form of magnetic field

To prevent this scenario, researchers are experimenting with different forms of magnetic field. The goal is to trap the plasma as efficiently as possible. It is a very important subject of study in this discipline; we remember, for example, the work of DeepMind. The company specializing in artificial intelligence has gone so far as to develop an algorithm to optimize the shape of the magnetic field.

To achieve this impressive combination of stability and temperature, KSTAR physicists relied on a modified version of a form of magnetic field called the Internal Transport Barrier. The peculiarity of this model is that it tends to make the plasma denser in the center of the reactor. On the other hand, it is more sparse on the periphery, near the walls.

They got a slightly lower density than they expected. Usually this is not good news. The energy produced by a reactor depends directly on the temperature, density and confinement time of the plasma.

But in this case, the researchers explain that this modest density was not a problem. It was finally compensated by the temperature and by the presence of very energetic ions in the center of the plasma. These play an important role in the stability of the reaction.

The road is still long

Admittedly, these figures are very impressive; but in absolute terms, the KSTAR and the other tokamaks are still far from being able to maintain the conditions necessary to maintain a fusion reaction over a prolonged period. From now on, the challenge will be to learn how to push these tokamaks even further. This involves reaching even higher temperatures and above all longer confinement times, all without damaging the reactor.

And that’s just the tip of the nuclear fusion iceberg. There are still lots of other problems that the engineers will have to solve. For example, for the moment, nothing indicates that the information provided by these experimental tokamaks will also be valid for larger-scale reactors.

And sooner or later, the issue of energy efficiency will also have to be addressed. Because as it stands, it is not even a question of recovering the energy produced by the reaction. This means that in addition to that which is used to heat the plasma and cool the enclave, any energy eventually produced by the reaction is also sacrificed on the altar of experimentation.

Suffice to say that even if this progress is impressive, we will have to be patient. Admittedly, the underlying physics is beginning to be well mastered. But there are now immense engineering challenges awaiting specialists at the turn.

Categorized in: