Lockheed Martin's claim of fusion power "in a decade" has my Spidey-sense tingling. Is there any merit to their claim? It seems like fusion power is always just a decade away — is there reason to hope anyone is going to create workable fusion power in our lifetimes?
Illustration by Slug Signorino
Depends on how long you plan on living. At the rate things were going, the timeline for commercial fusion power was up there with the half-life of radium. Sure, Lockheed Martin’s bid could crash and burn, but current efforts don’t seem noticeably more promising and it’s not my money. So why not?
Lockheed engineers raised eyebrows worldwide when they announced last October that they were pursuing a new type of compact fusion reactor. They planned on testing their design in a year, they said, with a working prototype in five years. The skepticism stemmed from the lack of technical detail provided, and the feeling we’d heard this before.
However, enthusiasm in some quarters was also high — the reactor is being developed by Lockheed’s Skunk Works research and development team, responsible for among other things the SR-71 Blackbird (the fastest non-rocket plane ever built), the F-117 stealth bomber, and the F-22 that replaced it. Lockheed Martin is a public company with an image and stock price to protect, and you’d think they wouldn’t be foolhardy enough to promise a breakthrough without something to back it up. Then again, Microsoft seemed pretty confident about Windows 8.
The details released by Lockheed are sketchy, but apparently the company has decided to go with a smaller-is-better approach to containment design. In a hot-fusion reactor a mixture of deuterium and tritium, two heavy forms of hydrogen, are injected into an evacuated chamber and heated to millions of degrees to form a plasma in which atoms fuse together, releasing energy. This insanely hot plasma must be contained in a small space not only to keep the reaction going but also to allow safe extraction of the heat needed for power production.
To date most fusion reactor designs have been of a type called a tokamak (a Russian coinage), which suspends the plasma in a superconducting magnetic field shaped like a giant donut. The drawback of a tokamak is that it’s huge and complicated but can contain only a small amount of plasma. The Lockheed people claim that by shrinking the reactor they can hold more plasma relative to the energy required to maintain the magnetic field, resulting in ten times the power production. Furthermore, they say their system is safer and more stable than a tokamak — as the plasma pressure increases, so does the strength of the field, containing the plasma even more securely.
Beyond these efficiency advantages, there’s obvious benefit to having something powerful enough to run 100,000 homes but small enough to fit in a semitrailer. On paper at least, the compact and safe design could make it suitable for powering ships, airplanes, and even spacecraft.
Lockheed isn’t alone in breaking away from the tokamak herd. General Fusion, for example, uses a sphere filled with liquid lead and lithium to contain the fusion reaction. Others have redesigned the tokamak to look more like a cored apple than a donut. It’s hoped that, within a decade (a familiar-sounding timeframe, admittedly), these so-called spherical tokamaks will achieve the critical “net power production” point — that is, where they’re producing more power than they consume.
We’re not there yet. In 1997 the Joint European Torus set a record for producing 16 megawatts of power for a few seconds — an impressive number, but only 65 percent of the power that went into running it. In 2014 a laser fusion experiment at the Lawrence Livermore National Ignition Facility managed to generate “fuel gain greater than unity.” Is that good? Absolutely. Does it mean we’ve crossed the net power production threshold? Alas, no.
Still, it’s more progress than some fusion efforts have made. The current leader in money spent vs. watts produced — and that’s not a title you want to hold — is the International Thermonuclear Experimental Reactor, or ITER. A monster of a project at ten stories tall and costing more than $18 billion, ITER utilizes a traditional tokamak design and hopes to produce fusion energy sometime after 2027 — which is, I note, more than a decade away.
By reaching its goal of 500 megawatts of power from 50 megawatts of input energy, ITER would set the stage for the next phase, called DEMO, projected to start construction in 2030 and possibly finish by 2040. DEMO wouldn’t be one plant but rather a sort of joint venture in which multiple parallel efforts would somehow produce a single reactor to serve as the prototype for multiple commercial-grade utility reactors, which would in turn begin construction after 2050. Right after that, Jesus comes back.
The one fusion reactor of demonstrated practicality is the sun, one of your more plus-size phenomena, suggesting Lockheed’s small-is-beautiful approach is no sure route to success. On the other hand, you have to like the idea of a test design in a year. The tech world has taught us you learn from your wrong turns. Therefore, fail fast.
Send questions to Cecil via firstname.lastname@example.org.