Tokamaks and Donuts
Nuclear Fusion Part 3: Magnetic Confinement Fusion Technology
Key Takeaways:
- Magnetic Confinement Fusion (MCF) approaches use strong magnetic fields to confine a plasma at the conditions needed for fusion.
- MCF devices are often shaped like donuts.
- The magnetic fields don’t heat the fuel directly. They just confine the hot plasma so that the target temperature and density can be reached. Hot fuel atoms are injected into the plasma to provide the heat.
- Decades of experimental MCF programs around the world have culminated in ITER, a 35-nation collaboration to build the world’s largest fusion research reactor. It is expected to begin operation in 2025.
- Several startups are developing improvements to research reactor designs, or entirely new concepts utilizing magnetic confinement. Some of these companies have been active for over 20 years.
We previously discussed inertial confinement fusion, where fusion occurs in brief bursts, like a Supernova. In contrast, much of magnetic confinement fusion seeks to confine a plasma that burns steadily like the sun.
How Does MCF Work?
In all types of fusion devices, we need to overcome the high energy barriers between atoms for fusion to occur. In MCF devices, the conditions for fusion are met by generating a plasma with a high electric current (lots of ions), confined to a small region (so they are very dense and close to each other), and that is also very, very hot.
Most of the magnetic fusion devices built to date share some common design features:
Vacuum Chamber. In magnetic confinement fusion, the shape of the chamber is key. The most common shape is a donut that confines the plasma in the interior, like the cheese in a stuffed-crust pizza. The goal is to confine the 100 million degree plasma inside this metal donut. The metal donut has to be kept under vacuum so other atoms or molecules don’t disrupt the plasma. Also, the chamber must be made out of materials that can survive getting slammed by high-energy neutrons during normal operation, and a massive release of heat and energy if the plasma becomes unstable.
Magnetic fields. Around the metal donut are coils that are electromagnets, producing strong magnetic fields. These coils, or solenoids, help “stiffen” the plasma and confine it to its donut-shaped track. Initially, these coils were copper wires, which constantly need power to produce a magnetic field. Today’s designs use advanced superconducting materials, which are able to generate strong magnetic fields with almost no power input after they are initially charged. [1]
Other electromagnets help create an electrical current in the plasma that drives it around in a circle — these work similarly to an electrical power transformer.
The plasma itself acts as a third source of magnetic fields. The electrical currents carried by the plasma create magnetic fields around it that “pinch” it into a narrower cylinder. When the magnetic fields generated by the external coils are much larger than the fields generated by the plasma itself, the device is called a tokamak. [2] This is the most common type of magnetic fusion device.
Importantly, the magnetic fields don’t actually heat the fuel to cause fusion. They merely confine the hot plasma so that the temperature and density can reach the levels needed for fusion to happen.
Fuel. The atoms used to fuel fusion reactions are the same whether lasers or magnets are used. A mixture of deuterium (D) and tritium (T) is the easiest to ignite. Unlike inertial confinement approaches, MCF devices need the fuel to exist as a large volume of plasma rather than a tiny target.
Heating. To understand how the plasma is heated, we need to know that a plasma contains a high fraction of charged particles- ions. A plasma is essentially a soup of ions: the particles in a plasma have an electric charge, so they feel and respond to electric fields. A gas transitions to a plasma when a significant fraction of the atoms in the gas become separated from one or more of its electrons. For example, a deuterium atom has one proton, one neutron, and one electron. Once it is ionized inside a tokamak, the proton and neutron remain together as a positive ion, and the electron goes off as a negative ion. The collective soup of ions — the plasma — can now be controlled by electric and magnetic fields.
To heat the plasma without disturbing it, a method called neutral beam injection was developed. Just like it sounds, beams of high temperature neutral particles (typically the same atoms as the fuel) are injected into the plasma chamber. Because these particles have no charge, they don’t feel the magnetic and electric fields that confine the plasma and zip happily in. Once inside the plasma, they collide with ions, giving up their heat and becoming part of the plasma themselves.
Energy Conversion. For the fusion plant to generate power, the particles produced in the fusion reaction must be safely captured and their energy converted to electricity. Designing the lithium blanket or other method to collect neutrons and breed tritium is a significant engineering challenge for all fusion concepts. In particular, developing suitable materials for the “first wall” between the plasma and lithium or other coolant is difficult. This is especially true given the complex geometry and need to maintain vacuum and thermal insulation.
Another key element of tokamak design is the diverter, which sits at the floor of the vacuum chamber. It extracts heat and larger particles (“ash”) produced by the fusion reaction, reduces plasma contamination, and helps protect the surrounding walls from overly high heat and neutron strikes.
In short, magnetic confinement fusion devices are highly complex. These devices have been developed and refined over decades. Today’s cars look very different from the models from the initial Model-T, though some of the basic design features are the same. Magnetic fusion devices have also evolved significantly since the first research facilities in the 1960s.
Where are we today?
Mainstream fusion has meant magnetic confinement fusion for much of the last 50 years. In the two decades following the 1973 oil crisis, several tokamaks and other magnetic fusion research facilities were built in the US, Europe, and Japan. Notable among these are the TFTR at the Princeton Plasma Physics Lab (now closed), JT-60 in Japan, and the Joint European Torus (JET) in the UK. The JET still holds the record for fusion output at 16 MW from an input of 24 MW of heating in 1997. While most of these devices are tokamaks, the Wendelstein 7-X MCF device was recently completed in Germany based on the related Stellerator concept.
Gradually, magnetic confinement efforts and funding have been focused on fewer, larger facilities. The pinnacle of this is the ITER project, began in 2007 as a collaboration between the European Union, India, Japan, China, Russia, South Korea and the United States.
Despite delays, the first phase of ITER’s construction is more than half complete. It is now scheduled to begin experiments in 2025, and by 2035 it will be ready to conduct experiments with a tritium-deuterium mixture.[3] In particular, methods to control the plasma and extract the electricity-producing heat will be tested and developed at large scale. The learnings around materials development and energy extraction, even in the design phase, have benefited fusion programs, both public and private.
Several startups are also pursuing magnetic confinement approaches to fusion. Relative newcomer Commonwealth Fusion, in collaboration with MIT, is developing a smaller-scale tokamak design utilizing new superconducting magnet technologies. TAE Technologies, founded in 1998, has been working on a very different magnetic field-based approach — a “field-reversed configuration”. General Fusion is working on a “magnetized target fusion” system that simplifies energy recovery.
Research and development on magnetic confinement fusion technologies has produced advances in many areas, including large superconducting magnets, vacuum technologies, complex cryogenic systems, ultra-precise construction, and robotic systems to handle materials. There is likely more value to be generated in adjacent technologies as well.
Where is MCF headed?
Research facilities around the world, including ITER, continue to advance our understanding of both plasma physics and how to engineer physical systems.
A 2019 report from the US National Academy of Sciences identified the key development needs to enable fusion as: “the materials and technologies needed to extract the heat and recirculate tritium and to promote the industrial development of very-high-field superconducting magnets. Innovations should [also] be encouraged and developed to simplify maintenance and lower construction cost.” This is a significant shift from fundamental science to practical power plant considerations.
Framing development in terms of specific engineering challenges inspires optimism. Magnetic confinement fusion systems are undeniably large and complex. Many of the requirements around materials, welds, and maintenance are enough to give an engineer pause. However, progress in these areas has the potential to “lift all boats” — paving the way for a successful fusion plant, whether it is a direct successor of ITER or a totally different design.
So what lies ahead? Certainly continued engineering challenges. In the words of H. G. Rickover, former head of the US naval nuclear program, circa 1953:
I gratefully acknowledge the help and input of fusion scientists and engineers who informed this blog post. As always any mistakes are my own, and I would welcome the opportunity to correct them.
The first post in this series was a high-level overview of the fusion space. In the second post, we explored inertial confinement fusion technology. In the final post of this series, we consider what a fusion plant might cost, and how it might fit into the future energy landscape.
Notes
- Superconducting materials are called “superconductors” because they conduct electricity with zero resistance. This means that once an electrical current is started within a coil, there is nothing to stop it — it will persist forever unless something disrupts it. However, most known superconductors only have this property at very cold temperatures — near absolute zero (0 Kelvin, or -273 degrees C). They must be kept cold, or electrical resistance will become non-zero and the electrical current will be converted to heat. This is extra challenging when the superconductors need to be in close proximity to a high-temperature plasma.
- The name tokamak comes from a Russian acronym for a toroidal chamber with magnetic coils.
- The first phase of ITER’s construction does not include tritium handling capabilities (tritium is radioactive).
Prime Movers Lab invests in breakthrough scientific startups founded by Prime Movers, the inventors who transform billions of lives. We invest in seed-stage companies reinventing energy, transportation, infrastructure, manufacturing, human augmentation and computing.
