Thinking Small: A Historical Perspective on the Current Miniaturization of Fusion


To many in the general public today, the concept of “nuclear fusion” evokes a sense of strange, distant, otherworldly awe. Even in recent pop culture, Marvel’s cinematic Iron Man franchise uses this property of fusion reactors to augment the sleek, futuristic feel of Tony Stark’s suit. Yet despite its long relegation to science fiction, a few small organizations–including one notably propelled by university students–are starting to reveal a new dawn for fusion as a power source.

Nuclear fusion isn’t a new concept; in fact, its history stretches back almost a century. The oft-cited industry joke goes, “fusion is the energy source of the future, and it always will be”.[1]Even as early as 1920, just 15 years after Einstein’s famous E = mc2paper suggesting energy from fusion as a possibility, astronomer Arthur Eddington proposed fusion as the fuel source for stars.[2]By 1958, true thermonuclear fusion–fusion caused by heating up molecules into a plasma–had been achieved in the laboratory.

These early attempts began in Los Alamos, with nuclear scientist Jim Tuck working on what he called a “z-pinch apparatus,” so called because it pinches gas along a single axis (the “z”-axis) until it becomes hot enough to turn into a plasma and cause fusion between atoms. Early calculations by the Los Alamos group claimed that fusion in this setting would require temperatures on the order of 100 million degrees Celsius.[3]The difficulty with fusion was not heating gas to that temperature–rather, it was containing it. Finding a material to withstand that level of energy would be impossible, so Tuck turned to electromagnetism to help him.

If you’ve ever taken a class on electricity and magnetism, you might remember the two being mentioned together fairly often. Electricity and magnetism, despite seeming very distinct–one powers your fridge, the other holds magnets to it–are actually two sides of the same coin. Because of this, whenever you make a current flow in a straight line, you generate a magnetic field that circles around it, like how your fingers curl around your thumb when you give someone a thumbs-up.

Tuck used this principle by running a current of electrons through a gas-filled tube. The magnetic field around the stream of electrons constricted the gas inward toward the current, causing it to heat up. Unfortunately, instability in his setup caused the gas to disperse before fusion could begin. Although this specific experiment didn’t work, it was the first of many magnetic confinement reactors, including the earliest workingreactor–called the Scylla I–demonstrated in 1958.[7]

So if working fusion has existed for so long, why aren’t we using it as a power source? The answer is simple: right now, it takes more energy to power the reactor (with its massive magnetic field) than the reactor can produce from fusion. One response to this has been to make a giant fusion reactor. ITER, or the International Thermonuclear Experimental Reactor, proclaims itself as “the world’s largest fusion experiment”–what a title! Incorporating efforts from 35 nations, ITER is currently under construction in Southern France, and states boldly that it “will be the first fusion device to produce net energy.” ITER plans to generate 500 megawatts of usable power from a mere 50 supplied to generate the magnetic field containing the plasma.[12]For context, 500 megawatts is enough power for nearly a quarter of San Francisco’s energy needs.[5], [6]

Yet, while ITER proclaims “go big or go home,” a laboratory at MIT has quietly been proclaiming the opposite: think small. With the inside chamber of their reactor only 3.3 meters in radius–that being roughly the width of a Stanford dorm room–MIT plans to match ITER’s rate of 500 megawatts for 50.[6]

“The design for the MIT Arc Reactor,” from


In this sense, MIT’s Plasma Fusion and Science Center is the David to ITER’s Goliath. In their sling is a handful of brilliant design insights to enable smaller plasma containment, the boldest of which is a set of superconducting magnets made available through recent technology. The incorporation of these into the MIT design allows for both reduced size and improved performance. As MIT Ph.D. candidate Brandon Sorbom stated, “any increase in the magnetic field gives you a huge win.”[3]

MIT’s reactor, called ARC, or the affordable, robust, compact fusion reactor, was designed off the tail of another MIT fusion project, this one a 23 year endeavor that set the worldwide pressure record inside any such reactor on several occasions. And the paper on ARC (available here) reflects the university’s ingenuity in the field. The work outlines everything from stability analysis and optimizations all the way down to how the superconducting material should be layered in the magnets.[9]It might sound like if anyone can take Tuck’s early dreams of turning fusion reactions into usable power, it’d have to be MIT.

Yet there are still more players in this game of miniaturization. Notably, in 2014 Lockheed Martin, a Bethesda, Maryland corporation known for their R&D in aerospace, announced plans for a nuclear fusion reactor “small enough to fit on the back of a truck.” The design differs from most other miniature reactor designs in that it stores the plasma in a “tube-like” region, whereas most other designs seem to prefer a toroidal–or donut-like–shape. Unfortunately, the aerospace giant recently announced that their goal of a 20 ton reactor was to become a 2,000 ton reactor–much larger than initially planned.[4]One notable blogger remarked that inside the plasma tube at its new size, one could fit two full school buses, side by side. So much for compact. (For an in-depth look at the project, see here).

Even earlier, a UK startup known as Tokamak Energy started their fight for the miniature prize as well–and they’ve had far more success. Their first reactor in 2012 was able to generate plasma for a mere 3 milliseconds. Now, their ST25-HTS, designed for continuous operation, holds the world record of maintaining a plasma–that difficult but necessary step for energy generation–for 29 whole hours.[10]Fascinatingly, a late 2017 paper from Tokamak Energy reveals that both they and MIT seem to use the same shape–a spherical tokamak, which looks like a donut that’s been squeezed into a ball–and material for their superconducting magnets.[11]Although Tokamak predicts that they will have a finalized fusion electricity generator by 2030, only time will tell who might break even first.

In the meantime, the actual fusion plant isn’t the only thing being miniaturized. A recent research paper by the Department of Energy’s Princeton Plasma Physics Laboratory revealed a way to simulate nuclear reactors in a much smaller timeframe than before by analyzing the physics in a more efficient way, using a mathematical tool called an “eigenmode.” Currently, principal research physicist Nikolai Gorelenkov claims that current simulations of one typically-sized reactor might take several months to give results, and that a reactor the size of ITER would take one million times more calculations.[8]As he notes, these simulations are helpful in finding inefficiencies in the reactors and fixing them quickly. A faster simulation–one using the “eigenmode” solution–would be an invaluable boon for the everyday nuclear engineer.

Thus, with new superconducting magnets, sped up nuclear simulations, and several independent groups pushing the front toward the new wave of energy generation, it sounds like we’re teetering over the edge into a brilliant new era. Besides the direct implication of a virtually endless energy supply from nothing but raw, unexciting atoms, many individuals have fantasized about the implications of compact fusion. Most common is that involving space travel: if your energy source runs on atoms and can fit on a tiny space vessel, you can go almost anywhere.

But how will you, the reader of this article, be impacted? Ideally, not at all. Instead, the Earth should be impacted–positively, as fusion is entirely emission free. The heavens will be impacted, slowly, as science is propelled into a new era in which we can explore deeper than we’d ever dreamed. Scientific knowledge and aspirations will come to reflect the new paradigm of fusion energy, and our understanding of fusion processes and energy generation will be greater than ever before. Promises of a better world–for everyone–echo in the change. So indeed, you may not be impacted. Not yet. But everything around you soon may be.


  1. Bergstein, B. (2019). Finally, fusion power is about to become a reality. Retrieved from
  2. Eddington, A. The internal constitution of the stars.The Observatory, 43, 341-358. Retrieved from….43..341E/0000345.000.html
  3. Phillips, J. (1983). Magnetic fusion.Los Alamos Science, (Winter/Spring), 64-67. Retrieved from
  4. Chandler, D. (2015). New design could finally help to bring fusion power closer to reality. Retrieved from
  5. Best, S. (2017). Trouble for lockheed’s fusion reactor? device that could ‘solve the world’s energy crisis’ is 100 times larger than first planned. Retrieved from
  6. List of united states cities by population. Retrieved from
  7. Mearian, L. (2016). MIT takes a page from tony stark, edges closer to an ARC fusion reactor. Retrieved from
  8. Phillips, J. (1983). Magnetic fusion.Los Alamos Science, (Winter/Spring), 64-67. Retrieved from
  9. Rosen, R. (2019). Found: A precise method for determining how waves and particles affect fusion reactions. Retrieved from
  10. Sorbom, B. N., Ball, J., Palmer, T. R., Mangiarotti, F. J., Sierchio, J. M., Bonoli, P., . . . Whyte, D. G. (2015). ARC: A compact, high-field, fusion nuclear science facility and demonstration power plant with demountable magnets.Fusion Engineering and Design, 100, 378-405. doi:10.1016/j.fusengdes.2015.07.008
  11. ST25» Tokamak Energy. Retrieved from
  12. Sykes, A., Costley, A. E., Windsor, C. G., Asunta, O., Brittles, G., Buxton, P., . . . Smith, G. (2018). Compact fusion energy based on the spherical tokamakdoi:10.1088/1741-4326/aa8c8d
  13.  What is ITER? Retrieved from