Until April 1st, 2025, roughly a third of Princeton’s government funding — and a fifteenth of its operating budget, if you care about that sort of thing — went to a place that the vast majority of its undergraduates have never been, toward a problem that very few of them have ever seriously considered. For 75 years, in a cluster of boxy buildings encircled by trees and kiosks and fence gates, the scientists and engineers of the Princeton Plasma Physics Laboratory have been trying to solve the many problems of commercially viable nuclear fusion. Princeton’s plasma workforce numbers in the hundreds. They have coined a gratuitous number of acronyms. They have won Nobel prizes. They have run enough simulations to put the machines from “The Matrix” to shame. They have pioneered methods in plasma nanosynthesis and built devices that can detect radiation from a dirty bomb. They have managed to get bright, undulating coils of energy — plasma, ten times hotter than the core of the sun — to dance and flicker in magnet-ringed chambers. One thing they have not done is figure out fusion.

The lab’s machines just aren’t efficient enough: none of them produce enough energy to outweigh what they cost to run. Plasma sputters, particles skip. To fix this, scientists will have to build devices that allow for smooth, sustained reactions. Two problems arise; both of them are heat. First, scientists have to figure out how they can get a chamber to roughly 150 million degrees Celsius without anything melting or breaking in the process. Next, they must build magnetic cages for the plasma itself — if the substance touches the walls that surround it, it gets contaminated by tiny flakes of metal and glass, which cool it down enough that fusion becomes impossible. 

This is all especially frustrating because fusion is, in theory, the most efficient form of energy out there. Unlike fission (its shunned and sometimes disastrous brother) fusion produces no radioactive waste. There are no carbon emissions. The fuel is hydrogen, one of the most abundant elements in the universe, and a handful of the stuff would produce a staggering amount of energy. If this all sounds a bit far-fetched, step outside and look up: our sun, like every other star, is powered by fusion.

Before earth’s cities can be powered by starfuel, though, scientists will have to navigate the most important obstacle of them all: funding. Unlike wind turbines and solar panels, fusion devices just don’t produce energy yet, and it’s unclear when they will. This uncertainty means that businesses and banks alike are wary of investing too heavily in plasma. Politicians, realizing that fusion is unlikely to power much of anything before their next election rolls around, also tend to deprioritize the technology, which is why less than a tenth of a percent of the almost $400 billion that made up Biden’s green energy bill had anything to do with fusion. 

In a way, this general lack of funding is what makes the PPPL so special. Of the Department of Energy’s 17 national laboratories, the PPPL is the only one that centers on plasma research. The lab is afforded a constant, dependable stream of cash, which it draws on to build house-sized machines and mouse-sized gadgets; and everything in between. The money allows the PPPL to hire swaths of builders and gaggles of researchers. It means that Princeton is able to attract talented graduate students and, with a consistency other labs can only aspire to, turn them into leaders.

As of late, though, the DOE’s funding seems less constant and dependable than it once did. The alarm bells first went off when Trump picked Audrey Robertson, a former oil executive, to lead the Department of Energy’s renewables office. In the months since, the ringing has only gotten louder. Dozens of employees — theorists and engineers, some of whom had worked at the PPPL for decades — have been laid off. Letters of protest have been written, panicked emails have gone unanswered, and a general sense of unease has spread through the little clearing in the woods. And then, on April 1st, it happened: the DOE officially started to retract existing PPPL research grants. The lab — so used to charting our circuitous path toward the energy source of the future — now finds itself in unfamiliar territory.

 

In between classes at the University of Arizona (where he would triple major with a perfect GPA), Liam David built a Farnsworth Fusor in his garage. With the help of an online forum — “[fusion] is God’s own way of powering the universe,” reads the website’s home page, which looks like it hasn’t been updated since the site was launched in 1998 — David was able to piece together a device that could achieve nuclear fusion. Dozens of enthusiasts have built this sort of thing since it was invented in the 50s. Still, without any real source of funding, and with all of the unpredictable problems of real-world plasma physics, assembling a fusor can be a long, painstaking process. It helps to have mentors.

Enter Richard Hull, a wry, retired septuagenarian whose homemade nuclear reactor was once featured in the Washington Post’s “style” section. “I conditionally accept your results,” Hull declared in 2017, responding to David’s application to the “Neutron Club” (to join, an enthusiast’s device must achieve nuclear fusion). “You have good detection gear! Give it a break and allow it to do the good work it is capable of; obtain some stable operation.” 

Five years later — unconditional results long since achieved — David was still working on his device. “My current record for silver activation is >13,000 CPM on a 2″ pancake,” he posted in 2022, alongside a picture of his fusor. In the image, a pair of thick metal rings sit above a boxy frame, which encases a forking accretion of wires, knobs, dials, and tubes. Many of the contraption’s inner workings seem to be wrapped in aluminum foil. “Fabulous silver activation!,” responded Hull. A day later, David came back to the post. “Edit: now up to 37600 CPM.”

“The forum… is the way I actually got into all of this stuff to begin with,” David recalled from his office at the PPPL. “[It was] the way I decided on fusion as the problem that I needed to help solve.” 

David is still an active member of the fusor forum — he’s become something of a mentor himself, helping newer enthusiasts build their own neutron generators. A fusor is a good start: it appeals to precocious teenagers, ambitious retirees, people who own a vacuum chamber and aren’t scared of a little radiation. But the garage has its limits. And if you want the real thing, the PPPL is the place to go. David began a five-year Ph.D. track in 2022. Since then, he’s been working with a set of resources that his younger self could only have dreamed of.

 

Conceptually, fusion is simple enough: when two atomic nuclei combine to form a new, heavier nucleus, an astonishing amount of energy is released. It’s like an atomic bomb, but backwards, and fickle — plasma may be starfuel, but here on earth, you can blow it out like a candle. And there’s no way around it: unless the original atoms are superheated into a flighty brume of plasma, a swarm of electrons remains to guard their atomic cores. Without bare nuclei, fusion cannot occur.   

Machines that consistently produce workable plasma tend to be big and expensive. For this reason, small companies and underfunded labs alike can only invest in one or two fusion schemes. The PPPL, in comparison, has been able to fund a wide, weird array of plasma devices, allowing its researchers a nuanced look at what works and what doesn’t. 

David works on one such device: the Princeton Field-Reversed Configuration, which uses radio waves and magnetic mirrors to create a plasma that is, by plasma standards, “relatively cool” — unfathomably hot, by any other standard. It cost almost a million dollars to build.

David works on the machine with Sam Cohen, a lauded lecturer at the lab. He is Cohen’s only grad student. “There’s a surplus of researchers,” David says, by way of explanation. At most plasma programs, finding a thesis advisor can be difficult and competitive. At the PPPL, the students are the ones in demand, and the professors are tasked with attracting them. In David’s case, working one-on-one with Cohen has allowed him to guide the PFRC’s development.

On a typical day, David and Cohen will decide what they want to test — magnetic field strength, maybe, or neutral gas pressure — and then boot up the machine. After that, says David, “it’ll go once a second, nonstop, for as many hours a day as we can stomach.” This produces a constant, overwhelming stream of data that must be sifted through. Thankfully, the lab’s funded an equally robust suite of diagnostic tools for David to work with: x-rays, probes, interferometers, the works. David is especially interested in the instabilities of man-made plasma, and he uses a high-speed camera (think 150,000 frames per second) to watch as the stream of particles wobble and split.

The PPPL’s graduate program is successful precisely because it allows for students to find this sort of niche. “You get two years to branch out and do little projects before you transition into thesis mode,” explains David. And there’s plenty to explore. “In my opinion, the sheer variety and scope of the projects has always been one of [the plasma lab’s] strengths. If we focus too much on one or two big projects” — David pauses, chooses his words carefully  — “I think we lose something.” 

It’s unclear exactly what the funding cuts mean — (“graduate students are pretty well insulated from… the consequences,” explains David) — but they’re certainly not a good sign for project diversity at the lab. “Apparently the DOE is applying a lot of top-down pressure to restructure funding and ‘improve efficiency’,” he says. In a world where a “small project” can cost a million dollars, the easiest way to cut back on funding might be to sunset machines like the PFRC.

 

Of the “big projects” that David mentioned, one looms particularly large. This is the National Spherical Torus Experiment-Upgrade, or the NSTX‑U. It’s a huge, complex device that, like the rest of Princeton’s campus, is currently under construction. “Today, the NSTX-U Recovery Project is 91% complete,” reads the PPPL website. It’s unclear when the lab will nail down those last 9 percentage points — all we know for sure is that the device won’t be ready this calendar year. 

At its core, a tokamak is a donut-shaped array of electromagnets that squeeze plasma into a ringed coil. Branching out from this core are tubes and panels and catwalks and ladders and wires thicker than your wrists and switchboards longer than your legs and stout red-orange tanks of hydrogen and tangled sets of slender hoses and it is a structure that looms, that surmounts, that makes the men in its midst look ever so small.

It’s all as expensive as it looks. If you want big results, you need a big machine: this is the basic premise of a tokamak like the NSTX-U. Princeton’s last tokamak — well, the one before the NSTX, which only lasted a couple of years before it was reworked into the NSTX-U — ran from 1982 to 1997, breaking an eye-watering number of world records in the process. This new one is supposed to be a model for commercially viable fusion.

As of now, the machine seems to have escaped the DOE’s most punitive spending cuts. But fusion devices worth hundreds of millions of dollars have collapsed before. Look to the 80s — “there were two leading labs,” explains Tony Qian, a researcher at the University of Wisconsin-Madison. “One was the PPPL and the other was the Mirror Fusion Test Facility… in California. For the last few decades, there had been competitions between the donut shaped one and the line shaped one. But in the battle for funding, Princeton won, and the line-shaped machine never turned on.”

As of late, though, the world of venture capital has become interested in this line-shaped setup. The Wisconsin HTS Axisymmetric Mirror, which Qian’s been working on since he graduated from the PPPL in 2022, is a prime example: after an initial DOE grant ran dry, the project switched over to a private source of funding. The lab runs at a frenetic pace — “there’s a hunger to move quickly and get results,” says Qian. 

WHAM is not a beautiful machine. It looks like an abandoned submarine, or an iron lung. From a startup’s perspective, this is a good thing: the device was built to be cheaper (and, from a physics perspective, far less complicated) than its competitors.

Qian isn’t new to cost cutting. He spent most of his time at Princeton working on MUSE, a small-scale fusion reactor that ended up as maybe the cheapest device of its kind ever built. According to Qian, the project wouldn’t have been possible without Lab Directed R&D, a PPPL program that supplies small, promising projects with quick cash. “It’s like a startup within a big national lab,” Qian says.

At first, the analogy makes perfect sense. Startups like new ideas, and quick results. But without government backing or academic support, things can go very wrong very quickly. “Many of these companies are outright scams,” argues David. “The designs are in no way practical.” WHAM has scientific backing, but for every legitimate startup there is a completely unvetted company promising to do in a single year what the PPPL hasn’t been able to do in 70. When those projects crash and burn, warns David, “it screws up the funding for the legitimate efforts, too.” 

It’s a feedback loop — as DOE dollars wane, the PPPL is forced to fire its theorists and pause its smaller projects, extending the fusion timetable ever further into the future. Political capital is just a special sort of investor confidence: when it ebbs, Princeton’s lab becomes more and more like the companies that it’s supposed to counterbalance.

 

Here on earth, plasma can seem like a strange, arcane force. It’s the crack of the lightning bolt, the arc of the comet, the glow of the northern lights. It’s the phase of matter they don’t tell you about in elementary school. But, as professor Hantao Ji of Princeton’s astrophysics department is happy to explain, “99% of the universe is plasma.” Once you start looking for it, the stuff shows up everywhere.

At the PPPL, Ji works on FLARE, the “Facility for Laboratory Reconnection Experiments.” Ji’s experiment, manned by six scientists and an ever-shifting fleet of engineers, aims not to generate energy but rather to study a set of natural phenomena (including, of course, the eponymous solar flare). There’s still a utilitarian motive — Ji explains that solar flares can ravage our power grids and fry our satellites — and maybe there has to be, to justify the millions spent on the project. 

At one point, Ji lays his thesis bare: “plasma,” he says, “is at the center of everything. You have to look at the applications.” He is lively and affable and he sounds like he means it. But to listen to Ji talk is to get the sense that he would be studying plasma even if there were no earthly applications — that he finds something wondrous and worthwhile in a sort of science that lacks clear economic value. When the machines are this expensive, though, wonder is a hard sell. 

There was a time, in the 1950s, when the president of Argentina promised that liter-sized bottles of plasma would soon hit the market, which his citizens could use to power their cars and their homes.  Now plasma is bottled up in a different way: the only people that fervently believe in it are the ones currently working on it. The public, sick of the half-baked start-ups, unmoved by attempts at outreach, have given up on fusion, or else mistook it for fission, or else expected that it will fall fully formed from the sky, which is just another sort of surrender. The government, entrenched in a battle with higher education, breaking slowly from Biden-era energy policy, is giving up a little more each day.

The impact is already being felt. Dozens of PPPL technicians were fired this March, alongside a half-dozen theorists. Rumor has it that a group of postdocs who’d been working on a device for four years lost their funding. Had the lab not stepped in with its own discretionary spending, David would’ve lost his stipend, too. “If this happens to too many students,” he warns, “the plasma program might not have the budget to pay for all of [us].”

This is all unfortunate, but it’s not entirely new. Seen from afar, the story of plasma is a story of false starts and delayed funds, of half-promises and fraudulent startups. Lyman Spitzer, the founder of the PPPL, was sure that he’d see commercial fusion in his lifetime. Before his retirement in 1967, he designed nine stellarators — twisting the machines into figure-eights, messing with the configuration of the internal magnets. None of them produced usable energy. The lab reworked and rethought and reimagined his designs until 1995, when the U.S. government suddenly decided that enough was enough. It slashed the national fusion energy budget, and the PPPL was forced to fire a third of its workforce. “On average,” went a blog post written by a professor at the time of the firings, “those laid off have served 17 years at the lab.”

Seen from up close, the story of plasma is the story of those years. They are years of work — haggling for funding, tweaking designs, watching for hours on end as a machine spits out strands of ionized gas — but behind that, beyond that, they are years of deep, magnificent belief. Children who once squinted up at the sun in wonder have spent their entire lives trying to build it here on earth.

Do you enjoy reading the Nass?

Please consider donating a small amount to help support independent journalism at Princeton and whitelist our site.