Physicist Dennis Whyte is a professor of engineering at MIT and the former director of its Plasma Science and Fusion Center, where his research interests focus on accelerating the development of magnetic fusion energy systems. His biography page says that Whyte has published over 350 papers across the multi-disciplinary fields of magnetic fusion—including plasma confinement, plasma-surface interactions, blanket technology, plasma diagnostics, superconducting magnets, and ion beam surface analysis.

He also leads the overall MIT research team on SPARC, a private-sector-funded compact high-field tokamak presently under development to demonstrate net fusion plasma energy gain. In addition, Whyte leads the Laboratory for Innovations in Fusion Technology, which has energy company sponsorship and explores early-stage, disruptive fusion technologies.

In this interview with the Bulletin‘s Dan Drollette Jr, Whyte gives what may well be a unique, insider-like view of what other countries—especially China—are doing in fusion research and explains why he maintains a bullish attitude toward the development of fusion as an energy source. But, he warns, it “will not be a cakewalk.

(Editor’s note: This interview has been condensed and edited for brevity and clarity.)

Dan Drollette Jr: One of your physicist colleagues said that the best, neutral people to talk to about magnetic fusion energy research at ITER are those who are not directly involved with that project but are still knowledgeable about fusion—which is how I got your name. This issue of the magazine is not confined to just what is happening in France with ITER; we’d also like to find out what countries like Russia or China are doing—the bigger picture.

Dennis Whyte: Russia isn’t doing much anymore, unfortunately. I mean, Russia’s a member of ITER, and it had an outsized performance in the past—back when it was the old Soviet Union, they were critical in many fusion breakthroughs. It’s just that that country’s domestic R&D hasn’t really built any new devices in a while—even before the recent conflict. China clearly surpasses them in terms of national capabilities in this field.

As it happens, I just did an interview for the Wall Street Journal about what China is doing.[1]

Drollette: I know that China Daily said the Chinese government is calling their fusion project the “artificial sun”—a nice bit of public relations.[2]

Whyte: Well, China’s indeed pushing hard on fusion. The American Nuclear Society[3] reported that the Chinese government is already putting together the base of the supply chain necessary for a fusion power industry—and it’s important to remember that in China, the walls between the government and the private sector are lower; where one ends and the other begins can be blurry. But they’re basically making fusion a priority within their large-scale industrial sector—just like steel manufacturing, or nuclear components.

China’s on the way to building something like a hundred nuclear fission power plants—knockoffs of the AP1000s.[4] And they’re looking to the future, into what an entire fusion industry would look like in China. Their stated goal is to actually be the first to achieve commercial fusion.

Drollette: So your opinion is that it’s for real? It’s not just public relations fluff when the Chinese government essentially says: “We’re going to have commercial fusion before anyone else”?

Whyte: It’s good to be skeptical; that’s one of the reasons the Journal interviewed me.

Because until recently, I had a fairly good inside view on what was going on in China, because I was on their advisory and oversight committees. But due to the deteriorating relationship between China and the US, I’m not really allowed to do those anymore.

But I saw what they [the Chinese government] were doing, and it’s a very serious attempt. They have built real devices that are actually achieving real, relevant fusion experiments. They evolved in a short period of time to a leadership position in fusion science.

You mentioned ITER. So, China’s one of the member states in that organization, and they have very successfully leveraged that into building up their own technical prowess in fusion.

ITER is basically made of up in-kind contributions from each of the different participants. And the Chinese have been very clever about leveraging the buildup of their capabilities to provide specialized fusion components around things like magnets and so forth. They’ve basically leveraged superconducting magnets into their internal capability to build their own devices.

So, yes, their fusion effort is pretty real.

Drollette: That’s fascinating, because when I read the words “artificial sun…”

Whyte: Yeah, sometimes when you read the press releases, it’s hard to know what to make of them—is there a major breakthrough here or not?

But in general, they basically have caught up to some of the [fusion] experiments and the scientific accomplishments that happened first in the West.

And in a few areas, they are soon to surpass what we were doing. Because while everyone tends to focus on the achievements with respect to the plasma—where the fusion occurs—fusion power is about so much more than just that. It requires all of these other systems: To have successful commercial fusion, you have to capture the fusion energy, produce the specialized components to carry that out, have the ability to maintain the device, and then eventually convert that energy into making electricity, or whatever you want to do. That’s a rather complicated set of things, with the absolute minimal requirement being that you can make the fuel get into the right state so that you can make fusion energy out of it—which they’ve done, and which is what the press cares about.

The Chinese have made significant investments in real facilities and have built what we call fusion technologies in record time.

There’s a tendency to artificially separate fusion science—which is about getting the plasma to this particular state—from making fusion power and energy. Essentially, one is making the energy, the other one is extracting and using that energy in some way. But they’re closely intertwined; you need those terrestrial objects surrounding the fusion device to accomplish the things you want. And the Chinese have made about a billion-dollar commitment, called CRAFT [an acronym referring to the “Comprehensive Research Facility for Fusion Technology, scheduled for completion in 2025], to accelerate the development of these kinds of technologies.[5]

Investment in fusion has been heavily weighted to the science part, because that was the first necessary step; you need to get the science down before going further.

So, there was a tendency in the West for everyone to think, “We better wait for results from ITER before making big investments in ancillary technology.” That’s clearly true in the private sector in the United States. There’s been this mindset that we should just wait for ITER—which basically means that fusion power plants are, conservatively, 30 to 40 years away. In other words, “Why make investments in this technology for something so far away”?

But lately that’s stopped being the dominant narrative, and fusion companies in the United States are not waiting for ITER anymore. In fact, the US government’s stated goal is to try to accelerate the commercialization of fusion—the White House announced a so-called “Bold Decadal Fusion Plan” in 2022. The idea is that if we start on that date, then by the early 2030s we should be getting the first prototypes in place.

Meanwhile, the UK has a plan in place that would mean that their electricity-producing power plant would come into operation at the same time as ITER does.

And China’s goal is have their thing operating before ITER does, too.

This is a big shift that has happened in the last three to four years. At first, there were just a few companies, including one of the companies that came out of MIT—Commonwealth Fusion.

And it was met with some skepticism, which has since largely gone away for a variety of reasons that include progress in fusion science and technology, but for other strategic reasons as well, too.

Things are getting to the point where you have to making some kind of decision: Are you going to be active or stand on the sidelines in fusion’s development? Because fusion’s got a much better chance now of happening on a short timeline rather than a longer one.

And China has kind of voted with its feet about that.

Drollette: This may be going on a limb here—but back in January, I had interviewed [former Secretary of the Department of Energy] Steven Chu, and he said that it’s interesting that China is buying up all these rare earth metals and locking up a lot of contracts for their mining and production. He suggested they were trying to corner the market. Is that part of what you’re talking about?

Whyte: That’s really more related to the materials needed to make solar panels and batteries than to fusion. Actually, MIT had an extremely comprehensive assessment of different energy technologies which looks at aspects of fusion and other energy sources in light of the demands for decarbonization. And it found that fusion has the advantage of not needing any particularly critical elements to operate; fusion is not really sensitive to the scarcity of rare earth metals, by its very nature. That’s one of fusion’s attractive features: very high power density with essentially free fuel. And even the things that you build around it—those technologies I talked about—while they can be complex, they tend not to be heavy on raw resources, because you’re getting so much power out of a single object.

Drollette: It would be fascinating to know why China’s doing so well—and how long it can keep it up. For example, back in the 1980s, Japan was considered to be this monolith that was going to take over the world with its semi-planned economy and public/private consortiums. But now…

Whyte: Yes, things change. But I wouldn’t count Japan out. Japan is the one to ask about when it comes to fusion, rather than Russia. Japan has a very active fusion program—though in many ways, it has not yet quite made that transition from a pure science program to an active energy program.

But it looks like Japan’s coming very quickly, as well. It appears that they’re revving up a national plan to try to rapidly commercialize fusion.

Drollette: But to get back to China…

Whyte: China has a growing economy. And China’s government realizes that there are limits to coal; it’s got severe environmental and health consequences for their population, and the Chinese authorities can do the math. They realize that fossil fuels are not infinitely sustainable. That’s why they’re turning to nuclear power.

And there are other things that come about with the successful commercialization of an energy source: It’s all part of gaining the geopolitical influence that comes with dominating such a power source. It’s not some of that so-called “soft power”; it’s the influence that comes with being able to control the supply chain—as well as export the knowledge.

Because this technology would replace fossil fuels, which is going to change everything.

So China’s government is very keen to make sure that they are in the leading position with respect to this technology. It’s always dangerous to guess about what could happen decades down the road, but it could be that fusion—and AI—are the two most disruptive technologies of this century. Because one totally changes how you work and exchange information, and the other changes how you get access to the root stock of what you need, which is energy.

Drollette: Speaking of artificial intelligence, a Washington Post piece[6] talked about the possibility of harnessing AI to search for the best fusion energy design. The author—from Princeton Plasma Physics—was very bullish about it.

Whyte: It’s not bullish, it’s happening already—you blink, you miss it. In fact, our team at MIT and our colleagues, we’re using AI and machine learning already, because it turns out these are very attractive and important tools in a couple different ways. [Fusion is] fantastically suited to machine learning, especially AI.

Because if you sit down to design a fusion device, there’s usually something like 10 or 11 controlling parameters in place, and some of them are dynamically related to each other. So at the moment, if you’re trying to find optimizations within these spaces, you wind up having to kind of make intuitive guesses if you’re using the human mind. But it’s well suited for things like machine learning and AI. So it’s happening already.

Drollette: So this Washington Post article was not exaggerating when the author said that there’s something like tens of billions of possibilities when it comes to a fusion design?

Whyte: Yes. I’m guessing that scientist is probably speaking of one particular kind of configuration called a stellarator, which is one of the leading magnetic forms as to how you provide the containment.

There was an article [that] iterated through something like 10 billion different design configurations of the magnetic coils to get the optimizations, for example. And so AI and machine learning are critical tools already. It’s sort of continuing an earlier trend—high performance computing has already fundamentally changed fusion. And I teach a design class at MIT on fusion, which deals with these things, and what we keep coming up against is that you have to trade off one part of the design to get a benefit somewhere else, it’s enormously complicated.

But technology is changing that whole process, not just here but everywhere. I like to point out to my students that they have more computing power available to them in that one classroom than the entirety of the huge international team that was started ITER in the early 1990s to design it. So, I would argue that all these advancements give us reason to be a lot more upbeat about making fusion happen.

China already has sustained a very high-performance plasma for over—I can’t even remember now—1,000 seconds; something big like that. And at that large fusion experiment in England, they actually established a record amount of energy released by a fusion system: over 60 megajoules of fusion energy that was produced

And, of course, there’s the achievement of net energy gain at the NIF laser fusion experiment as well too, which is an enormous scientific achievement.

At the same time that there have been all these advancements in purely fusion-oriented technologies, there have been tremendous advances in other, related fields. The magnets required for magnetic fusion have basically undergone a quantum leap in performance, something akin to what happened when the invention of transistors hit electronics. It’s basically an improvement in performance by a factor of 40 to 50.

All of these things happened within the last three to four years. That is why there’s so much momentum going on now; it’s not by accident. And I’m just giving you a few of the big highlights, there are many other developments which I would consider very important as well.

Drollette: That’s interesting because I’ve read of how difficult it’s been for researchers at superconducting supercolliders to get their magnets to work properly.

Whyte: About 10 years ago, people managed to successfully commercialize a new kind of high-temperature superconductors called a “cuprate.”[7] You can think of them as a crystal that has surprisingly strong superconducting abilities at crazy high temperatures, like 75 degrees above zero Kelvin instead of just barely above absolute above like at the Large Hadron Collider. In the superconducting world, that’s incredible.

They also have extraordinarily high tolerance to magnetic fields. This was something that was glommed onto by my team at MIT; we realized that these could be revolutionary in fusion. We published seminal papers on this about 10 years ago on this, and what it showed was that it appeared feasible that we could about double the magnetic field that was produced compared to the previous generation of the superconductors and keep them still in the superconducting state.

Drollette: If someone were to summarize, it sounds like there have been enough advances in technology, on so many different levels, that researchers can now go back and take another look at old fusion experiments from the 1950s and give them another try. Maybe making some design changes through the use of AI…

Whyte: Or trying these new magnets or using these new computational tools about predictions of plasmas as well. Yes, that’s right.

And just being able to crunch the numbers in some kind of large quantity is good for a physicist. To give you an idea, you may have heard of the three-body problem[8] in classical or quantum mechanics—the idea that it’s so complicated, with so many variables, that there is no equation that always solves it all the time. It’s so sensitive to initial conditions that you can’t really solve it, but just say that numerically this is what happens most of the time. So, a three-body problem is hard enough—and a fusion plasma is about a 10-billion body problem.

So raw computing power has helped.

So, those are some of the things that are pushing forward the development of fusion—not to mention the need to solve the problem of how to produce enough energy that is carbon-free. And, of course, there is always the problem of energy security—just think of what happened to Europe’s energy supply after the Ukraine conflict started. All of a sudden, they lost their access to all that Russian natural gas.

Drollette: Just to play Devil’s advocate for a second: I interviewed [plasma physicist and former director of Argonne National Laboratory] Bob Rosner earlier, and he had a whole different take on things. In a nutshell, he said that any problem in fusion is going to be 10 times more difficult than that same problem in fission. We talked about things like embrittlement of the whole fusion reactor facility, and the supply of fuel.

Whyte: Well, I have to say that I disagree with Bob on that one.

Though I grant that there are problems. For example, while we have achieved fusion here in a laboratory experiment already on the MIT campus—300 trillion fusion actions per second—there is more to it to make it commercial.

It’s not about making fusion, and in some sense, it’s not even just about making fusion electricity—although, of course, when we got net energy from fusion, that was an important one.

It’s actually about making economic energy, because if you don’t do that, then it’s not going to replace fossil fuels. And I agree with Bob on this, it is harder. There’s a reason why we haven’t done it yet.

But we’ve taken it such a long way, and now we see pathways that basically argue that the investment about turning it from a science-only endeavor into one that makes integrated energy products is worth it.

And by the way, you know, fission energy went through the same thing too. There were probably only a handful of people in the world who even understood what fission was in the 1930s, and Fermi’s pile experiment had an output less than one watt of total power. But it basically demonstrated the key aspects of a sustained chain reaction. Look at what came out of it: In 15 years, fission went from Fermi’s pile to a commercial power plant putting power on the grid.

So, it’s a similar process with fusion. And fusion is in an even better place right now—those advancements in magnets that I talked about are real game-changers. The rules of physics tell you that the increase in the amount of fusion power you get in a fixed volume increase the strength of the magnetic field to the fourth power. What this means is that by doubling the magnetic field, it improved the cost profile by, what, a factor of 40 to 50.

Overall, the situation looks exactly like fission did when first deployed. Fission, when it first deployed, had horrible available capacity, or capacity factor. A lot of the reason was because, basically, there just wasn’t enough experience or quality control in the assembly of the fuel rods. So you would operate your device and, darn it, it failed, and we’ve got to replace it.

I’ve got a great chart from the 1970s, showing that the capacity factor for fission back then was something like 35 or 40 percent of standard—because it was a pretty new technology. And then by the mid-90s to the early 2000s it was up past 90 percent—all because they improved, they got experience, they improved the maintenance scheduling. They got way better at just the act of refueling, and they got way better quality control on the materials that they use—because when you put materials in a fission environment, it’s also very harsh around all those things.

So to me, that seems like the pathway that we’re on—the transition that fusion is going through, on the road to commercialization.

Drollette: On what kind of time frame? The year 2035? 2039?

Whyte: That’s what many of the companies—and now nations—have said they’ve got as goals, for starting to put some kind of energy on the grid. But of course, that means it’s still at prototype stage.

From my point of view, there’s not many other choices on the table, when it comes to generating massive amounts of carbon-free energy at high-power density.

Drollette: I think we covered just about everything on my list, it was nice talking with you. And I think it’s great to have this sort of point/counterpoint about fusion, with Bob Rosner taking one position and you taking another.

Whyte: I’m totally happy with the skeptics, because I also wonder if fusion can actually meet that target, right? Because if the goal is economic fusion, then that is not a cakewalk; it is not an assured thing at all. Everyone should remember that.

But now I would say we’ve got a fighting chance, for two reasons. One—that set of advances that I talked about. And two—we actually have a set of entities, private companies and now national governments, that are pursuing this in earnest, right? And those are the combinations which basically have gotten all the other major technological advancements over the finish line. So it’s not a guarantee, but at least you’ve got the components of what is needed to get there.

Endnotes

[1]  See July 28, 2024 Wall Street Journal article “China Outspends the U.S. on Fusion in the Race for Energy’s Holy Grail” https://www.wsj.com/world/china/china-us-fusion-race-4452d3be?st=qb1dqx97krdqnj2&reflink=desktopwebshare_permalink

[2] See July 2023 China Daily, “Controlled nuclear fusion emerges as new frontier for China’s venture capitalists” at https://www.chinadaily.com.cn/a/202407/23/WS669f5ca3a31095c51c50f7ad.html

[3] See NuclearNewswire, January 8, 2024, “China launches fusion consortium to build ‘artificial sun’ ” https://www.ans.org/news/article-5668/china-launches-fusion-consortium-to-build-artificial-sun/

[4] The AP1000 is a pressurized water reactor originally designed and sold by the Westinghouse Company https://en.wikipedia.org/wiki/AP1000).

[5] For more about CRAFT, see September 19, 2024 CNN article “The US led on nuclear fusion for decades. Now China is in a position to win the race.” https://www.cnn.com/2024/09/19/climate/nuclear-fusion-clean-energy-china-us/index.html)

[6] See August 20, 2024 Washington Post op-ed “The world-saving potential of nuclear fusion just got a huge boost” https://www.cnn.com/2024/09/19/climate/nuclear-fusion-clean-energy-china-us/index.html

[7] For more on cuprate crystals, see September 21, 2022 Quanta magazine “High-Temperature Superconductivity Understood at Last” https://www.quantamagazine.org/high-temperature-superconductivity-understood-at-last-20220921/

[8] See Wikipedia entry on the “Three-body problem”  https://en.wikipedia.org/wiki/Three-body_problem