Fusion: 70 years of experiments are finally paying off!

Last updated Dec 2024. Anti-fascism update here. Update notes

Introduction

What is fusion

Stars form when a bunch of gas collects under its own weight, heating and compressing enough to merge hydrogen atoms together to form helium. This is how all elements up to and including iron are created. This is the primary source of energy in the universe, and on our planet. [1] [4]

The scale of the system, both in physical distance and in time, means the gas's own internal gravity is what causes such enormous pressures in a stellar nursery. There are even “forbidden transitions” (non-standard and statistically negligible atomic interactions) possible at this literally astronomical scale that are not possible in the laboratory case. This is the reason it isn't easy to recreate fusion in the laboratory. [1] [2] [a]

Impact of fusion

As big as the leap in produced energy between burning carbon based fuels and nuclear fission is, the leap from fission to fusion is even bigger. It's the kind of technological advancement that means we will never need to worry about energy scarcity again. Energy will be so easy to produce that charging money for access to it will be a thing of the past. We will be able to share this technology with countries still undergoing their industrial revolutions such that they can skip destructive, dangerous, and inefficient fossil fuels entirely and jump straight to abundant carbon-free energy. [2] [3] [22]

As our civilization becomes more advanced we require more and more energy to keep it running. [d] This is why climate change has become so out of control: Because the demand for power has historically been met by carbon based fuels, which produce CO2, causing the multitude of extreme effects we are currently experiencing. For the continued survival of humanity, industrial fossil fuels have to go. To fill the gap in energy production left after we cut out carbon-based fuels we need a more powerful energy source than even carbon can provide, with a reliability that renewable energy can't match. We have two options:

Considering what we know of the fission process, the first option is a very short term one simply due to the inevitable production of nuclear waste. This is a problem that fusion doesn't have, and so it really is the only sustainable solution to our exponentially climbing energy demands. Fission is the solution we needed to invest in decades ago. Fusion is the solution we know we will need tomorrow. [1] [5] [21] [b] [c]

Basics of laboratory fusion

Fusion can only occur when a gas has been ionized and energized into a very distinctive state of matter called plasma. A plasma is “a quasi-neutral gas of charged and neutral particles which exhibits collective behaviour” (Francis F. Chen 2016). While the overall population is close to neutral, there is a lot of charged particle interactions going on within the population of particles in a plasma that leads to some fascinating physics. Electricity and magnetism are related at right angles and a change in one causes a change in the other, as according to Maxwell's equations. This interconnection is what gives rise to “collective behaviour”, where interactions between particles can have remote reaching effects on the population of charged particles as a whole. Basically, if you poke the edge of it the whole thing reacts, instead of just that localized area you poked. [4]

Plasmas are confined in a laboratory through the use of various magnetic bottles. This is how we can use the cyclical feedback nature of electromagnetism to our advantage: Running any current will induce a magnetic field, and vice versa, so we use huge amounts of current to shape the plasma into following particular pathways that increase the heat and pressure until the plasma is able to actually begin fusion. [4] [5]

In the laboratory setting we need much more heat than a protostar nebula does to get started, because the pressures we can create on laboratory scales are so much lower than they are in a stellar system due to gravity and scale. This means that fusion reactors that are currently running are often many times hotter than the surface of the sun! [1] [4]

to top

Prototype Reactors:

Tokamak (ITER)

The name of this design, Tokamak, is a sort of portmanteau of Russian words meaning a toroidal chamber with magnetic coils. A toroid, or torus, is what math nerds call the shape of a doughnut or bagel. There are two versions of a tokamak, the original design and the spherical tokamak design, but the physics are very similar. The magnetic fields these two designs form are both rings, just that the spherical design is about as tall as it is wide, rather than the short wide loop of the first design. Think Jerusalem vs New York bagels. [6]

This design was proposed by Oleg Lavrentiev and first built in 1954. For a long time this design was only used by soviet scientists, and made the USSR the leader in fusion performance for decades. By the 1970s, dozens of tokamak designs were in use, most of them reaching some (none of them reached all) of the conditions required to reach energy break-even. In the late 1980s, however, we saw a sort of stall happening. The problems the tokamak design has at this stage are scale and money problems, the solving of which is only possible by collaborative efforts of multiple countries. [6]

The current biggest fusion project in the world is ITER, a collaboration of 35 nations (EUROfusion), which is a classic tokamak fusion prototype reactor currently under construction in France. It's basically a scale up of the older European tokamak project JET, which currently holds the world record for energy produced by fusion. ITER is expected to produce a tenfold power gain: This means they expect to get 10 times more energy out of it than it takes to run it. They are not yet aiming to harness all the heat that it would produce though, considering that a problem for later. We are still very much at the proof of concept stage, even for this most documented of fusion reactor designs. [7]

Stellarator (W7-X, TypeOne)

The stellarator design was originally the West's response to the USSR's extremely successful tokamak design. It sort of faded into the tokamak's shadow until around the 1990's, when computational science and electronics construction progressed to a point where we could take a fresh look at old stellarator projects and make big leaps forward again. A stellarator is harder to build, but much more stable than a tokamak. [5]

A stellarator is a much more precise way of containing plasma because instead of imposing an enormous magnetic field topography on the system, it's a careful construction of a magnetic topography that makes slight deviations to the emergent drift behaviour we expect to see, which guide it gently towards a more stable configuration. It's a lot like the corners of a Formula 1 racetrack being angled up to keep the cars from flinging off of the circuit. It's an elegant solution that requires a precision that was not possible, either in physics modelling or electrical engineering, until fairly recently. [8] [9]

The magnetic field topology a stellarator aims for in the ideal case, has a type of continuous symmetry called quasisymmetry. Noether's theorem states, in basic terms, that any symmetry that exists has some type of stability associated with it. The stability gain from achieving (in the laboratory case, approaching) true quasisymmetry in a stellarator is better, more stable plasma confinement. Some examples of stellarators are the Max Planck Institute's Wendelstein 7-X prototype in Germany, as well as the prototypes being constructed by Type One Energy. [9] [10]

The tokamak and the stellarator are the current most popular fusion prototype designs, but they by no means form a complete picture of the fusion industry as it stands today. [b] So, considering that, let's talk about some wildcard designs:

Field Reversal Configuration (Helion)

The fusion company Helion, in the United States, uses a design consisting of two identical chambers that each create a toroid of plasma using a field reverse configuration (FRC). [11] This means that the plasma toroids are “self-organized, closed-field plasma configurations created in an open-ended cylindrical magnetic topology” Kirtley and Milroy, 2023 [12]. Some interesting properties of this type of system is that it can be well described by a two-dimensional magnetohydrodynamics model, which will save on both computational power and physicists' emotional well-being.

These toroids are then accelerated by magnets towards each other and further adiabatically [e] compressed together until fusion begins. Then the fusing plasma expands, counteracting the compressing magnetic field and inducing an anti-parallel current. This current gets recaptured as electricity, and used to skip the “heating water” part of almost every power plant (including fusion). [f] They are also breaking from convention by using helium 3 instead of tritium (hydrogen 3) to mix with deuterium, theorizing that this will outperform the standard fuel mix of deuterium and tritium. [11] [12]

Magnetized Target Fusion (General Fusion)

The Canadian company General Fusion's approach is to impose a magnetic field, then use mechanical steam driven pistons to literally compress the plasma simultaneously from all directions. The design also includes a liquid metal first wall (the plasma facing material). [13]

With a solid first wall, the material needs to be replaced quite often, as the plasma facing material will melt away (ablation) due to the neutron radiation it is directly exposed to. A liquid metal is naturally resistive to this kind of degradation, so this completely sidesteps the “first wall problem” other methods encounter. When the small amount of lithium in that liquid wall is exposed to neutron radiation, it forms a helium atom and a tritium atom, the latter of which can be used to continue fuelling the machine. As the first wall is liquid, they can alter the flow of it to direct heat flow elsewhere easily, also skipping the traditional 'boiling water' step of energy production. [14] [g]

This approach is really unusual and interesting, offers efficiency and low cost, an outlier even in the wildcard prototype category.

Projectile Based Inertia (First Light Fusion)

The UK's First Light Fusion team's method was directly inspired by the incredible pistol shrimp (alpheidae family). This shrimp can snap its claw so fast that it creates a vacuum in the water. When this vacuum bubble collapses, the resultant pressure and temperature is enough to actually create plasma in the water for a brief moment. [15] An uncited claim on wikipedia states that the shrimp uses this as a flash-bang in hunting or escaping danger, which is a bit like using a high end video-card to bash a home intruder over the head. [h]

The First Light Fusion team used this example in nature as inspiration for their fusion reactor designs, aiming to use compression waves to reach the pressures needed. This design also uses a liquid first wall, like General Fusion, however this one makes it into a game of clay pigeons. Using a railgun style electromagnetic propulsion system they launch a projectile at a target containing both the fuel needed to start the fusion reaction, and an amplifying material that increases the pressure of impact from the projectile and focuses shock waves back onto the fuel capsule to reach the temperatures and pressures required for ignition. [16]

Laser Heating (National Ignition Facility)

The National Ignition Facility in California uses the most powerful lasers in the world to compress and heat tiny fuel capsules to fusion ignition. This prototype reactor is very precise, requiring 192 lasers to be perfectly aligned and perfectly aimed. Advancements in optics (engineering tweaks on the micron scale), electronics, and computer modelling, have only been able to achieve this in recent years. The progress they are making now is mostly in terms of gathering and analyzing data from each 'shot' to continue optimizing this system to reach higher and higher efficiencies. Tiny adjustments like widening the chambers opening (where the laser beams enter the system) by a few millimetres, or adjusting the design on the (peppercorn-sized) fuel target are making huge differences in terms of energy output, just because of how much energy is produced through fusion. [17] [18]

to top

Comparisons:

Fusion Milestones

To qualify the success of these prototypes, we need to establish some important terms.

A plasma's 'Beta' is the ratio of plasma pressure over magnetic pressure. A plasma with a Beta value less than one means that the plasma resisting compression pushing out is much weaker than the imposed magnetic field pushing in. Most of our fusion examples are in this group. A Beta equal to one is exactly balanced. Neither exploding or imploding, but not stable. The other fusion examples we covered may approach a Beta of 1, but they do so asymptotically (a word which here means 'with diminishing returns'). A Beta greater than one is an actively exploding fusion reaction, like in some parts of the Sun, and areas of Earth's magnetosphere. [4]

The reason everyone agrees fusion is not dangerous is that if a laboratory fusion reaction ever reached a Beta larger than one, if the confined plasma was able to ablate (ionize and absorb) the equipment confining it, the immediate drop in pressure and temperature of the active plasma population resulting would make further fusion reactions impossible. Breaking confinement would plummet the pressure, making further reactions impossible. Ablating more of the containing walls than designed would plummet the temperature, making further reactions impossible. Fusion is so hard to achieve because it requires such specific conditions to maintain reactions. To put it in a nutshell, when you make a star in a jar, breaking the jar kills the star. [1] [5]

A burning plasma is a fusion reaction that is self heating. This occurs when the sustained fusion reactions are producing enough alpha radiation to heat the surrounding plasma more than the heat is radiated away, a process called alpha heating. A plasma that is burning but not igniting is a plasma that can maintain its own heat, but still needs energy input to keep going. [4] [18]

Ignition is the point at which a fusion reaction produces more energy than it requires to sustain itself. This is the win condition for fusion energy research. Ignition is what is meant by saying Q is larger than 1. Q is the fusion energy gain factor, so a Q larger than one surpassed break-even conditions, the energy we get is equal (break-even) or larger (Q>1) than the energy required to run the reactor. [18] [23]

ITER predicting tenfold energy gain in 2025

EUROfusion is currently operating the biggest, and probably most bet-on fusion project right now. This group specializes in the classic Tokamak design, and its last project JET holds the world record for energy produced through fusion: 59 megajoules, from reactions sustained for 5 seconds. This record-breaking experiment only had a Q value of 0.33. Their next project ITER is expected to reach a Q of 10 (well beyond ignition), but we won't know until it actually runs for the first time late next year (2025). [17] [22]

NIF claims 50% energy gain achieved in 2022

The National Ignition Facility is reporting that they surpassed ignition requirements on December 5th of 2022, reporting a Q of 1.5. This is the first fusion project in the world to reach that goal, and is so far the only one that has! [19] [21] [23]

Bill Nye, beloved science guy, discussed this on CNN, saying “If this really is the beginning of something huge, it would change the world.”

Something you will notice good science communicators do is be very careful saying “we did it” for sure, and that the news breaks slowly. We are only now talking about an experiment run in December 2022 because they needed until December of 2023 to even realize that it happened. We're talking about processing terabytes of data, possibly even more. What we have right now is a big big chunk of very complicated data that teams of scientists around the world are analyzing themselves as quickly as they can trying to figure out if this really is what we think it is. [i]

So far the outlook is very good! The news has broken and other teams are already looking at the data NIF gathered to try and verify or disprove their findings. It will be very interesting to see if NIF is an outlier, or just the first of many projects to reach ignition.

to top

Conclusions:

A lightly conspiratorial theory

It's interesting to note, not that I wouldn't agree with the decision if this is the case, but I wonder if it matters to point out that it will probably take several months for this peer review process to occur... and it's an American project... and this is an election year? There is no doubt that NIF is close. Their 2022 results of burning plasma are still standing after ongoing peer review, and I cast no suspicion on those. I'm only pointing out the advantage to reporting this ignition result early, and betting that it doesn't get caught until after November.

But there is a chance that this is a political move to report a highly anticipated expected result from an already established well performing experiment just a little bit early, and considering where a laboratory's funding is coming from is always worth doing. That considered I wouldn't take this seriously until after the US settles back to democratic sanity again just to be sure. Stay excited though, this is amazing stuff and NIF has already proven itself to be leading the field. [j]

Anti-Fascism Update

Well, folks. Unfortunately, this bit of the article did not age well.

Fascism is bad. Dictatorships are bad. Voter suppression is bad. Internment camps, systemic cruelty, intentional revocation of bodily autonomy, and hate crimes are bad. Disagreeing on this is not "disagreeing on politics". Disagreeing on this is opting out of the implicit agreement that human life has value. It is opting out of participating in the cooperation that defines humanity. If someone believes that human lives can be ranked in order of value, then they are so fundamentally anti-human that I have nothing to discuss with them at all.

Unfortunately, the current election results lead me to believe that there will be absolutely no progress seen from United States laboratories on any kind of energy innovation for at least the next four years. Human rights aside (and I hate having to say that at all) we are seeing a doubling down of intentionally toxic fossil fuel production and consumption that will affect every person in the entire world, and there is only so much the rest of the world can do to mitigate the damage that is already happening and will certainly accelerate due to policies we can expect to see enacted going forward. NIF probably does have results very near if not over the brink of true commercially viable fusion that they will almost certainly not be given the funding to pursue, due only to the intensive lobbying of fossil fuel oligarchs in the United States. That answer, however, does not need to be the rest of the world's answer.

Breaking artificial fusion, which several labs across the world are very very close to doing, will make commercial energy so inexpensive, it will completely bankrupt the entire fossil fuel industry. This has always been my goal, and will continue to be so. We CAN make destroying the environment and poisoning the people who live in it unprofitable, and, as a species, we are on the brink of doing so. Progress in the United States has frozen. Progress in the rest of the world will and must only accelerate as a response.

Stay safe. Continue to be yourself, in whatever ways feel safe. Lean on your support networks. Protest, donate, and educate if you are able. Know that the death of fossil fuels cannot be avoided, and a future with abundant carbon free energy in harmony with our planet is inevitable.

So, perspective scaled back... I wanted to mention one more thing that further emphasizes how close we are to cracking fusion energy production:

Honourable mention: desktop fusor

This comes directly from the work of a fella named Jay, who runs the account called “Plasma channel” on YouTube. This specific video is called “Building A Nuclear Star In A Jar (Fusor).” An incredible small scale project, Jay calls it a “star in a jar”, not meant to be a prototype energy reactor, but a simple proof of concept that fusion itself can be possible in a system you can fit on a desktop, using vacuums and very high voltages (40+ kilo Volts). [19]

The inevitability of a post energy crisis world

I truly believe that fusion energy is not only required for our continued survival as a species, but based on everything I've outlined here I believe it is inevitable too. So what will our world look like post energy crisis? Not only energy would be effectively unlimited, but heat as well. Also, we would have enough energy to actively undo much of the damage that has been done to our climate, like through carbon capture technologies, not to mention opening up tremendous possibilities for vertical farming, recycling, and even interstellar travel! [k]

to top

Sources and Further Reading:

References

Introduction:

[1] “An Introduction to Nuclear Astrophysics”, Richard N. Boyd, 2007. google books

[2] “Structure and Evolution of the Stars”, Martin Schwarzchild, 1959. google books

[3] United Nations Climate Change COP28, accessed 2024. link

[4] “Introduction to Plasma Physics and Controlled Fusion” Francis F. Chen, 2016. google books

[5] “The Star Builders”, Arthur Turrell, 2021. google books

Tokamaks and Stellarators:

[6] “The Kremlin letter that started it all”, Robert Arnoux, ITER Newsline, 56, 2008. link

[7] EUROfusion website, accessed 2024. link

[8] Type One Energy website, accessed 2024. link

[9] “Bounce-averaged drifts: Equivalent definitions, numerical implementations, and example cases” Mackenback et al., 2023. link

[10] “Tungsten based divertor development for Wendelstein 7-X”, Fellinger et al., 2023. link

Wildcards:

[11] Helion website, accessed 2024. link

[12] “Fundamental scaling of adiabatic compression of field reversed configuration thermonuclear fusion plasmas”, Kirtley and Milroy, Journal of Fusion Energy, 2023. springer link

[13] General Fusion website, accessed 2024. link

[14] “Development of Merged Compact Toroids for Use as a Magnetized Target Fusion Plasma”, Howard et al., 2008. springer link

[15] "Bioinspired mechanical device generates plasma in water via cavitation", Tang and Staack, 2019. Science V.5 NO.3

[16] First Light Fusion website, accessed 2024. link

[17] National Ignition Facility website, accessed 2024. link

[18] “Burning plasma achieved in inertial fusion” Zylstra et al., Nature, 2022. Nature link

[19] “US nuclear-fusion lab enters new era: achieving 'ignition' over and over”, Nature News, Dec 15 2023. Nature News link

Comparisons and Conclusions:

[20] “Building A Nuclear Star In A Jar (Fusor)”, Plasma Channel, YouTube , posted Oct 22 2022.

[21] Fusion Energy Insights website , accessed 2024. link

[22] “Could nuclear fusion energy power the future? - with Melanie Windridge”, The Royal Institution, YouTube posted Jun 15 2023.

[23] NIF's website, pursuit of ignition page, accessed 2024. link

to top

Notes to the reader

[a] I've referenced Chen's 2016 textbook on plasma physics, but I want to make a special note of it. My graduate work relied heavily on this book, so I reference this one all over the place. You can find a preview pdf of it on google books that includes a fantastic one page overview of stellar fusion on page 13.

[b] Turrell's 2021 book “The Star Builders” is an excellent accessible overview of the fusion industry as it stands, and was my first introduction to most of these fusion projects. I highly recommend this book, regardless of the reader's background in physics.

[c] I originally wanted to mention the Kardashev scale, as fusion ignition is often talked about as humanity reaching the Type 1 threshold on this scale. Kurzgesagt has a great video on this here.

[d] On the inevitable increase of energy demand: Ourworldindata.org has a great energy consumption graph that has a lovely visualization for this. It's got a really neat dip at 2020 as well, so that's interesting too.

[e] On the word 'adiabatically': Adiabatic is a thermodynamics term that means the system does not gain or lose heat. It usually refers to a compression of a gas causing it to increase the heat of the compressed gas, rather than letting that increased energy out of the system somehow. This is the cause of Calgary's chinooks (warm breezes during winter), for example.

[f] On the 'boiling water' step of energy production: Almost all energy plants of all kinds use water for regulating heat flow. It's to the point where it feels like the punchline of a bad joke that every kind of energy reactor design gets to the “generate heat” stage and goes “gosh let's store it in water I guess”. Bonus points if the steam from that water turns a mechanical turbine. The best nuclear fission reactors in the world do this, and I think that's pretty funny to be honest.

[g] On the steam pistons of General Fusion: I am reminded of the question and answer exchange I once had with a University of Calgary electromagnetics professor Dr. Brian Jackel, back in my undergraduate career. I asked how, in the limited current of a laboratory setting, do they make the most powerful magnetic fields in the world and he answered “they make a weaker one and compress it”.

[h] On the pistol shrimp: I am a physicist, not a biologist, so I want to do some more research before speaking with full confidence on this. The farthest I've gotten yet is here, which characterizes plasma creation just by light, which I'm not so sure about:

"Similar to what is observed with sonoluminescence, our light emission results indicate the presence of an inertially confined plasma generated by the imploding cavitation" Tang and Staack, Science V.5 NO.3 2019

I would love to dig into this topic thoroughly, and would be really interested in a marine biologist's opinion!

[i] Burying the lede: NIF's 50% energy gain result. I want to make sure to point out the distinction between the two sources I cited for NIF's results. The burning plasma result, from 2022, was published in “Nature”, the scientific journal proper. This requires stringent peer review, as well as lengthy revisions and scientific clarity. The second article, from 2023, was published in “Nature News”, which is NOT a scientific journal proper! This is a news article, and is subjected to journalistic standards, not scientific ones. The turn-around for this information is much much faster, and therefore much more tentative.

[j] Not sure how much stock to put in my little theory about why NIF is reporting this ignition result so early, but I wanted to bring it up anyway. I think it's part of my responsibility as a science communicator to be as clear as I can about things like this, especially when it involves following the grant money. Never trust a climate change study funded by big oil companies, for example. It's essential to remain critical, even when the reported result is exactly what we really really need right now.

[k] I also want to particularly recommend the talk that Fusion Energy Insight's CEO Dr. Melanie Windridge did for the Royal Institution. Her company is also a great resource for staying up to date on the entire field, and covers projects all over the world. I highly recommend signing up for the Fusion Energy Insights newsletter if you are interested in this field.

to top