How to repair hot cannons for the fusion temple

Amidst the cranes that dot the ITER (International Thermonuclear Experimental Reactor) work site, she stands tall – the Tokamak building. Some lovingly refer to this 60 metre tall steel-reinforced concrete building as the ‘fusion temple’. Arriving at the terrain, we watch out not to step on dropped screws. This is a site under construction. The boots we were given by the ITER organisation are not just an unnecessary luxury. It is humid and we are soon standing in pools of water in what looks like a giant concrete catacomb.

The building is as yet an empty shell. In several years’ time though, the nuclear fusion reactor – the Tokamak – will be placed right in the middle of this huge bunker, the heart of ITER’s experiment. From 2028 onwards, humans will no longer be allowed inside. The building will be sealed off and robots will be used to retrieve parts through openings for maintenance.

And what will then happen inside the temple? A spoonful of deuterium and tritium (two hydrogen isotopes) will be heated to millions of degrees centigrade, forming a giant plasma that rotates in the shape of a doughnut. The two nuclei will fuse and form a helium nucleus and a high-energy neutron.

Burning plasma

ITER, where scientists will study a ‘burning plasma’ that releases more energy than is used to produce it, or so it is hoped, must be a milestone on the path to fusion energy. It will be the world’s largest magnetic confinement plasma physics experiment. Delta was invited to join a tour for journalists to visit this project in the south of France, in Cadarache, that brings together scientists from Europe, China, Japan, India, South Korea, Russia and the United States.

Nuclear fusion seems promising for many reasons. One is that its hydrogen isotope fuels are relatively abundant. For example, deuterium can be extracted from seawater and tritium can be bred from a lithium blanket using neutrons produced in the fusion reaction itself. Second, a fusion reactor produces virtually no CO₂ or atmospheric pollutants, and most of its radioactive waste is short-lived compared to that produced by conventional nuclear reactors.

Through DIFFER, the Dutch Institute for Fundamental Energy Research situated at TU Eindhoven, the Netherlands is deeply involved in the project. The centre of gravity of Dutch research lies in plasma physics, creating the right conditions in the reactor to produce a plasma of the right intensity for fusion to occur. TU Eindhoven is an important member of DIFFER.

And where does TU Delft come in? We don’t have a professorship in fusion energy or in plasma physics. Look for ITER and TU Delft in the scientific literature and you will hardly find anything. But appearances can be deceptive. One of the contractors we meet during the tour is TU alumnus Cock Heemskerk. His company, Heemskerk Innovative Technology (HIT), works on the maintenance robots. Humans may not do the maintenance since the objects in and close to the Tokamak are hot and radioactive. Besides his regular staff, Heemskerk works with dozens of students, mostly mechanical engineering students, and PhD students from TU Delft. HIT’s office and laboratory are situated next to the Science Centre.

‘Hello’, says Pepper, ‘are you coming here for revalidation?’

“Hello, are you coming here for revalidation?” The HIT office welcome is a bit peculiar. But then, it is a friendly robot that does the greeting. “We also work on robots for health care purposes,” explains Heemskerk. “This robot is called Pepper and was developed by our colleagues at Diginova, a HIT spin-off company. Pepper will become a clerk in a hospital.”

Most of the robots that the researchers and students work on are haptically controlled robots. This means that they can be operated remotely using a joystick or other control device. “This allows robots that perform particularly difficult tasks like opening kitchen drawers – notoriously tough for robots – while doing household chores in nursing homes for instance, to be operated from a distance. The robots’ haptic feedback enables controllers to feel the forces the robot experiences on their devices. A well-known use of a haptic device is the force feedback on cars equipped with lane-keeping technology. If you are about to cross a lane, the steering wheel gives gentle resistance, helping you to stay in the lane.”

‘We cannot automate the robot as you would in the automobile industry’

Back to nuclear fusion. “There are many similarities between our robots for health care and the robots designed for maintenance at ITER. In both situations, the robots need to be remotely controlled to perform very precise tasks. At ITER, robots will handle huge objects, weighing tonnes, that have to be pulled out of the Tokamak with extreme accuracy to avoid damaging anything. Then there are the repairs that need to be done like welding, tightening bolts, you name it. Since we don’t know what to expect – there is no body of experience in nuclear fusion – we cannot automate the robot as you would in the automobile industry where robots perform the same task over and over again. So remote control (or telemanipulated task execution) and haptic feedback are key.”

In the HIT lab, researchers and students perform simulations. The robots’ manoeuvring capabilities are put in the computer and the researchers verify whether they can properly retrieve and handle specified parts of the Tokamak. “These kinds of simulations have enabled us to discover several flaws. Some parts of the reactor had to be redesigned because we discovered that their design meant that they simply could not be removed from the reactor. It is better to figure these things out before the reactor is actually built.”

Playing a game of Jenga

One of Heemskerk’s students, Jelle Hofland (mechanical engineering), seems to be playing Jenga, a game of physical skill in which you remove one block at a time from a tower constructed of 54 blocks. Each removed block is then placed on top of the tower, creating a progressively taller and more unstable structure. This tower is a virtual one on a computer screen and the student is using a joystick to steer an equally virtual robot arm with a gripper. “If I move left, I will touch the tower, as you can see on the screen”, says Hofland. “If I do touch it, I feel resistance on my controller.” Using his joystick, the student opens and closes the gripper, pulling out one of the Jenga blocks. This is the kind of technology that will be used to replace and repair parts from the Tokamak.

But the technology needs to be much more refined. “The haptic assistance needs to be extremely sensitive. Getting that right is a tough challenge,” says Heemskerk. “It is so tough that we have already had three PhD students working on it. Henri Boesenkool, Jeroen Wildenbeest, and Jeroen van Oosterhout. They have all just finished or are about to finish their PhD theses.”

For what parts of the Tokamak is all that sensitivity required? “We are focusing on the maintenance of the Electron Cyclotron Heating launcher, a kind of radiation cannon,” says Heemskerk.

The ITER Tokamak will rely on three sources that work in concert to provide the input heating power of 50 MW required to heat the plasma to the temperature necessary for fusion. One of the sources is neutral beam injection and two, the electron cannons, deliver high-frequency electromagnetic waves. These cannons deposit heat in very specific places in the plasma, a mechanism to minimise the build-up of instabilities that lead to the cooling of the plasma.

‘The cannon produces enough energy to vaporise a chicken in a split second’

“Each cannon weighs twenty tons and will have eight channels with about a megawatt of power each. That is comparable to the power of 8,000 microwave ovens. Enough to vaporise a chicken in a split second.

“The electromagnetic wave bundle is steered via a set of mirrors. These mirrors will be given a pounding. Our job is to find out how to remove the cannon from the Tokamak and how to repair the mirrors. We have to take absurd scenarios into consideration. For example, an earthquake cuts the power supply to the Tokamak, destabilising the plasma. The forces that could be unleashed are huge, millions of newtons. If that happens, the cannon, made of several centimetres of thick steel, will bend. Can we then still retrieve it from the Tokamak?”

While exiting the Tokamak building we pass a narrow 13 metre deep trench. At the bottom, the building is standing on hundreds of concrete pillars. Between them are thick pieces of rubber that – should an earthquake occur – dampen lateral movement. But there are no earthquakes in this part of France, in Cadarache, or not that anyone can remember at least. Think of the impossible, seems to be ITER’s mantra.