ITER — a Major Step Towards Thermonuclear Fusion

(Transcript of the lesson commentary.)

Principle and History

The ITER Tokamak is an experimental facility designed to test the feasibility of running thermonuclear fusion as an energy source under terrestrial conditions.

Its name is an acronym of International Thermonuclear Experimental Reactor, but it also means “way” in Latin, as ITER could be the way to a new clean source of energy.

ITER will be a tokamak-type device. Inside it, charged particles are held by a magnetic field in the shape of a torus. This field is created by so-called toroidal coils and its shape is further modified by poloidal coils. However, for this magnetic cage to really work, the magnetic field lines must be twisted so that the charged particle runs alternately on the outer and inner sides of the torus. This will guarantee an additional magnetic field generated by the current flowing through the plasma. It is induced by the transformer principle. A change in the current in the primary winding causes a change in the current in the magnetically linked secondary winding. In the case of a tokamak, the secondary winding is the plasma itself. However, because of the use of the transformer principle, the tokamak is a pulse device.

The ITER tokamak is the culmination of many decades of thermonuclear fusion research. The first T-1 tokamak was launched in 1958 in Russia. Other facilities followed around the world, and today, dozens of tokamaks are in operation. They achieved many insights and successes. A major breakthrough was in 1982 with the discovery of a mode with significantly better particle confinement, called H-mode, on the ASDEX tokamak. Tokamaks that were designed to work with tritium, i.e. TFTR and JET, have repeatedly achieved fusion. A significant success was achieved in 1997 by the JET tokamak, which generated 16 MW of fusion energy. This reaction released 67 percent of the energy that was expended to ignite it. Because tokamaks have proven to be very successful and have advanced the furthest in their ability to function as a power plant, it was decided that the world's largest experimental fusion facility would be a tokamak.

Plans for its construction date back to 1985. During the Geneva Superpower Summit, General Secretary Gorbachev of the former Soviet Union proposed to US President Ronald Reagan international cooperation in the search for a new, clean energy source. After complex negotiations, seven members were established and committed to work together to build ITER: China, the European Union, India, Japan, Korea, Russia and the United State. In 2006, the ITER Agreement was signed, and a year later construction work began in Saint-Paul-lès-Durance in southern France. Construction of the tokamak itself began in 2020 with the installation of the cryostat bottom into the tokamak pit.

The ITER Tokamak will consist of a number of essential parts. Let’s introduce the most important ones. These are the plasma, toroidal coils, poloidal coils, divertor, vacuum vessel, heating, central solenoid and cryostat. 


The most important is undoubtedly the hydrogen plasma. Here, thermonuclear reactions will take place. The entire tokamak around it exists only to create, contain and heat the plasma to a temperature of 160,000,000 kelvin. The plasma in ITER will have a volume of 840 m3. Initially, it will be a hydrogen or deuterium plasma, but in the later stages of the experiments, a 50 : 50 mixture of deuterium and tritium will be used.

Since at temperatures of millions of degrees any matter will instantly vaporize, the plasma will have to be isolated from the chamber walls by magnetic fields. This principle takes advantage of the fact that plasma is made up of charged particles. When they’re exposed to a magnetic field, they spiral along its lines. The shape of the torus has no loose ends to escape through, so they can circle around endlessly. A magnetic field in the shape of a torus can be created with a set of magnetic coils.

Various potentially dangerous phenomena can occur in the plasma, such as ELMs or disruption.

ELMs are Edge Localised Modes. These are short repetitive bursts of plasma resembling solar flares towards the chamber walls. They carry a lot of energy and could damage the chamber.

Since they are important for the transport of impurities from the centre of the plasma towards the edges, and thus for the purity of the plasma, we do not want to get rid of them completely. Just mitigate them so that they no longer pose a danger.

Two ways of ELMs mitigation will be tried on ITER. The first is resonant magnetic perturbations. In-vessel coils slightly perturb the shape of the magnetic field and break up an oncoming ELM, instead of one potentially damaging ELM a set of smaller ones arise. The second method is pellet injection. It disturbs the edge regions of the plasma by the injection of small pellets of frozen deuterium. This will decrease the magnitude of energy fluxes in one ELM while increasing their frequency.

ITER is also prepared for unexpected disruptions. This is a situation where, for whatever reason, the plasma cannot be contained and the discharge ends prematurely. At that point, the tokamak can start to act as a particle accelerator and produce runaway electrons, high-energy electrons accelerated to nearly the speed of light. To prevent this, the electrons need to be cooled and slowed down in less than 5 milliseconds. This will be done by the injection of large pellets of frozen deuterium and neon. 

Magnetic Coils

The toroidal magnetic field will be created by eighteen giant D-shaped magnets called toroidal coils. They’re 17 metres high, 9 metres wide and weigh 360 tonnes each. Together, they can produce 41 gigajoules of energy and create a magnetic field of 11.8 Tesla.

They are made of a special superconducting material, Nb3Sn (niobium-tin), which is brittle and difficult to process, but retains its superconducting properties even in intense magnetic fields. The coils are enclosed in rigid stainless-steel packages and cooled with liquid helium to near absolute zero temperature, to as low as 4 kelvin. The entire superstructure of toroidal coils weighs 3,400 tonnes.

The toroidal coils are manufactured in Japan and shipped by sea to the French port of Fos-sur-Mer. From here, they’ll take the long 104 kilometres journey on a truck along a route called ITER Itinerary. The journey takes approximately four nights.

Because the plasma ring tends to expand, it is held at the correct distance from the walls by a set of poloidal coils. These are six giant hoops strung across the toroidal coils. They are 8, 17 and 24 metres in diameter and weigh between 200 and 400 tonnes. Because each one generates a slightly different magnetic field, each one has a different weight.

The coils are made of a Nb-Ti (niobium-titanium) superconductor and cooled with liquid helium to 4 kelvin. They are capable of delivering a magnetic energy of 4 gigajoules and creating a magnetic field of 6 Tesla.

The largest poloidal coils are so huge that they could not be transported. To manufacture them, a winding facility was built on the ITER site. It has a 250-metre-long hall and is equipped with the high-tech machinery and handling equipment required to carry out the successive stages of the coil fabrication process.

There are other types of coils inside the vacuum chamber. Correction coils are designed to compensate magnetic field errors. Eighteen coils are arranged in groups of six around the toroidal circumference above, at and below the mid-plane of the vacuum vessel. They are about 8 metres wide and conduct current of about 10 kA.

The so called in-vessel coils are two non-superconducting coil systems inside of the ITER vacuum vessel that provide additional plasma control capabilities. Two vertical stability poloidal coils are installed above and below the machine’s mid-plane and allow fast vertical stabilization of the plasma. 

Central Solenoid

However, the donut-shaped magnetic field generated by the toroidal and poloidal coils would not be sufficient to hold the charged particles of the hot plasma. An additional field is needed to twist the magnetic field lines into a so-called helical shape, so that the particles will alternately circulate on the inner and outer sides of the torus. This field will be created by a current flowing through the plasma, induced with the transformer principle.

The primary winding of the transformer in ITER represents the central solenoid. It is a 13-metre high and 4-metre-wide structure made up of six independently operating coil packs wound from niobium-tin superconducting cable. In the centre of the stacked modules a maximum field of 13 Tesla will be reached.

By using the transformer principle, ITER will be a pulsed device. It is planned that pulses at maximum fusion power could last around 400 seconds and pulses at half power could last more than 3,000 seconds.

During operation, extreme electromagnetic forces will develop in the central solenoid and try to tear it apart. In order to maintain the structural integrity of the solenoid, it will be enclosed in strong pre-compression support structure made from super austenitic stainless steel. The support structure will have to withstand forces in the range of 60 meganewtons (in comparison, the force behind the thrust of a Space Shuttle lift off is about 30 meganewtons).

As long as the plasma is only partially ionized, it resists the induced current and heats up. This so-called Ohmic or Joule heating works up to about 10 million kelvin. After that, the resistance of the plasma is low and this type of heating is no longer effective. 


ITER will use Neutral Beam Injection and Ion or Electron Cyclotron Resonance Heating to heat the plasma to 160,000,000 kelvin.

NBI is actually a particle accelerator that accelerates deuterium ions with an electric field. To keep them from being affected by the magnetic field, the particles are neutralized before entering the plasma and then transfer their energy to the plasma by collisions.

Neutral Beam Injectors are monstrous machines sized like steam locomotives — 25 metres long, 5 metres high and 5 metres wide — with a chimney-like bushing reaching up 9 metres to connect to the openings on the third floor. Each one will deliver a deuterium beam of 16.5 MW with particle energies of 1 MeV. The ITER will be equipped with two heating neutral beam injectors, one smaller neutral beam injector for diagnostic purposes, and has a place for a third heating device if needed.

Resonance heating works much like a microwave oven. Radio waves of the right frequency transfer energy to the particles in the plasma and heat them up. The particles then transfer energy to each other by collisions.

The Ion Cyclotron Resonance Heating will use a frequency of 40 to 55 MHz to heat the ions. Two transmitting antennas will be about 3.5 metres long, with a cross section of 2 × 2 metres and a weight of nearly 50 tonnes. The system will deliver 20 MW of energy to about 1 gram of plasma thus raising its temperature very high and very quickly.

Electron Cyclotron Resonance Heating will heat electrons using a frequency of 170 GHz and supply an additional 20 MW of energy to the plasma. Radio waves for radiofrequency heating are produced by gyrotrons in the Radiofrequency Building and fed into ITER via waveguides. 

Vacuum Vessel

160 million degrees of hot plasma gripped by magnetic fields cannot be in contact with the atmosphere. The whole process will therefore be enclosed in a vacuum chamber. It’s a giant stainless-steel doughnut whose cross-section is elongated vertically into a D-shape. Its outer diameter is 19.4 metres, it is over 11 metres high and weighs 5,200 tonnes. From an internal volume of 1,400 m³, the plasma will occupy 840 m³.

The walls of the chamber must be resistant to high temperatures and act as a shield against strong neutron fluxes. On the inner side of the chamber there is a so-called blanket, consisting of massive neutron shielding modules on which replaceable first wall panels are mounted. They will directly face the plasma. Piping will be routed through the blanket to circulate the cooling water. ITER will be the world's first tokamak with an actively cooled blanket. Some blanket modules will be used for testing tritium breeding options. The outside of the chamber will be covered with a silver coated thermal shield to protect the superconducting coils from heat transfer.

The vacuum vessel wall will be coated with 440 removable modules of blanket. There will be 180 types of blanket modules designed for different locations in the chamber. Generally, each blanket module measures 1 × 1.5 metres and weighs up to 4.6 tonnes. They consist of a main shield block mounted on the vessel wall and a detachable first wall that directly faces the plasma.

They are made from beryllium tiles bonded to a copper alloy heat sink mounted on a stainless-steel structure. There will be two types of them — 215 normal heat flux first wall panels designed for heat fluxes of up to 2 MW/m², while the rest, 225 enhanced heat flux panels, will be designed to survive harsher plasma conditions and heat flux up to 4.7 MW/m².

The inner surface of the vacuum chamber, facing the plasma of 160 million degrees, will be heated to hundreds and sometimes more than 1,000 degrees Celsius. But just a few metres away, superconducting coils cooled to just four degrees above absolute zero will be located. To protect them from heat transfer from the chamber, the outer surface of the chamber will be provided with a silver coated thermal shield. The highly reflective silver will reduce thermal radiation. Moreover, the shield will be actively cooled by liquid helium to 80 kelvin.

Tritium is extremely rare on Earth and therefore cannot be mined as fuel for fusion power plants, but must be produced by capturing fusion neutrons in lithium. The ITER Tokamak will test the possibility of tritium production and test different designs of tritium breeding modules. A breeder containing solid lithium pebbles or a liquid mixture of lithium and lead, and two types of cooling, water and liquid helium, will be tested. The liquid-filled breeder has the advantage that the tritium produced can be extracted during reactor operation, but the liquid metal also interacts with magnetic fields. This undesirable interaction does not occur with solid breeders, but it is necessary to remove them from the reactor to extract the tritium. Water cooling is a proven method, but it must be ensured that the coolant remains in a liquid state and does not boil. This problem is eliminated with helium which, however, has a much lower heat capacity compared to water. 


To achieve the conditions for thermonuclear fusion, it is essential that the plasma is clean and free of impurities such as helium or ions of higher atomic number elements. It is also important to avoid the impact of high-energy particles on the inner wall of the vacuum chamber.

In ITER, the divertor takes care of all this. The particles from the peripheral regions of the plasma will be guided by magnetic fields that cross in an X-shape next to the divertor and end directly on the divertor walls. From there, the impurities will be extracted by powerful vacuum pumps.

One divertor cassette will weigh 8 tonnes. Its stainless-steel structure will be over three metres long and over two metres high. Three plasma facing components made of tungsten — the dome and the inner and outer vertical target — will be placed on it. These components will be actively cooled by water flowing in pipes beneath them. During operation they will experience a heat load of 10 MWm² to 20 MWm².

ITER will have 54 cassettes arranged in a ring at the bottom of the tokamak. They will be inserted into the chamber after it is completed through the port and seated with sub-millimetre precision. 


To minimize the heat transfer between vacuum vessel and superconducting coils, the entire tokamak will be enclosed in a giant evacuated vessel called a cryostat. In principle, this is the largest thermos in the world. It’s a giant steel cylinder 30 metres wide and 30 metres high, weighing 3,850 tonnes. Its internal volume is 16,000 m3.

Vacuum will be provided by a combination of mechanical vacuum pumps and cryopumps. Mechanical pumps are using the rapidly rotating blades to give the particles a momentum and direct them out of the pumped area. In the cryopumps, particles condense and stick to the frozen surface.

There are six torus pumps, four pumps for the neutral beam injection assembly used for plasma heating, and two for the cryostat. To reach a vacuum pressure of 1 × 10−4 Pa will take about 24 hours. 

The Future of ITER

ITER construction is expected to be completed in 2026, when the first plasma will be ignited. Gradually, individual systems will be tested and commissioned. The temperature of the plasma will be increased, the discharge length will be extended, and the first scientific findings will emerge. In 2035, experiments with a mixture of deuterium and tritium and the first ignition of thermonuclear fusion should take place.

This should, if all goes to plan, release 500 MW of fusion power and produce up to ten times more energy than will be used to ignite it. Subsequently, a series of experiments should accumulate enough scientific knowledge to enable the world’s first commercial thermonuclear power station to be constructed and started up.

With ITER, thermonuclear fusion energy is one step closer.