NEWS FEATURES REKINDLING THE FLAME After decades of decline, the U.S. government’s fusion lab seeks a rebirth Downloaded from By Adrian Cho, in Princeton, New Jersey oseph Winston, a technician here at Energy’s (DOE’s) fusion budget and shut power plant, smaller and cheaper than ITER, the Princeton Plasma Physics Labo- down TFTR. In 2003, the United States and PPPL would likely play a leading role in http://science.sciencemag.org/ ratory (PPPL), knew something joined the effort to build ITER, the giant building it. was wrong with the fusion reac- international reactor under construction Perhaps most important, in 2018 Princeton tor just by listening. In 2016, PPPL near Cadarache in France—a commitment University, which runs the lab for DOE, hired physicists had restarted their Na- that squeezed fusion research at home even a new lab director. Steven Cowley, a strap- tional Spherical Torus Experiment harder. In 2008, DOE canceled another un- ping 60-year-old Englishman with a shock of (NSTX), after a 5-year, $94 million finished fusion reactor at PPPL, leading to silver hair and a knighthood, makes no bones upgrade. During one of the ma- a reshuffling of lab leadership. Now, PPPL about his role as an agent of change. “My job chine’s runs, which last just seconds, millions employs 560 people. Its one large machine, as a director is not to be an administrator,” he Jof amps course through NSTX’s magnet coils, NSTX, sits idle 3 years after breaking down. says in his velvety baritone. “It’s about scien- creating fields that squeeze an ionized gas so tific vision. What should we be doing? What tightly that atomic nuclei can fuse. The cur- are the interesting questions? How do we get on February 8, 2020 rents also stress the coils, which emit a groan to fusion?” He already has a plan to diver- loud enough to be heard through more than sify the lab’s work, grow its staff, and start to a meter of concrete. But the sound was peter- build things again. ing out prematurely, Winston recalls. Physicists are watching PPPL as a bell- When Winston and his team traced the wether for the fortunes of the U.S. fusion problem to a short in one coil, the 50-year program, whose share of the world’s pub- PPPL veteran knew the reactor, or tokamak, lic fusion research has slipped to just one- would be down for a long time. The ma- sixth. And some observers who have been chine is “like a one-way street,” he says. “If critical of the lab’s previous leadership and anything happens to these coils, the whole culture think PPPL is finally on the right thing has to come apart just to get to it.” track. “Steve is the best person on the planet After running for just 10 weeks, NSTX was for the job,” says William Madia, former di- shut down again. rector of two other DOE national labs, who It was a body blow to a lab that was al- urged Princeton to hire Cowley. “I’m opti- ready staggering. In the 1980s, PPPL ran PPPL director Steven Cowley wants to grow the lab. mistic.” Yet the lab still faces obstacles on multiple machines, employed nearly 1300 His first task is to repair its main fusion reactor. the path to redemption. people, and led the worldwide quest to har- ness fusion, the energy source of the Sun. Yet things may be looking up for the lab. NSTX RESEMBLES an extraterrestrial space- “The action was almost frantic,” says Dale After years of DOE reviews, PPPL researchers ship. The two-story orb nestles in a cocoon Meade, a PPPL physicist emeritus. “We were expect to start to rebuild NSTX in April. And of pipes and cables, the red coils of its main taking risks, building one thing before the a year ago, a report from the National Acad- magnet arching up out of the chaos like fly- other was finished.” In 1994, PPPL’s largest emies of Sciences, Engineering, and Medicine ing buttresses. Within the orb—the reactor’s machine ever, the Tokamak Fusion Test Reac- (NASEM) urged the United States not only to partially disassembled vacuum chamber— tor (TFTR), briefly generated 10.7 megawatts stick with ITER—which is hugely overbudget copper plates and graphite tiles line the of power, still the record for U.S. efforts. and behind schedule—but also to prepare to silvery walls. One could imagine that some The good times didn’t last. Within years, build the machine after it (Science, 21 Decem- reptilian alien slumbered away the eons here Congress had slashed the Department of ber 2018, p. 1343). This would be a prototype while traveling to the Solar System. During LABORATORY PHYSICS PLASMA ELLE STARKMAN/PRINCETON PHOTO: 618 7 FEBRUARY 2020 • VOL 367 ISSUE 6478 sciencemag.org SCIENCE Published by AAAS Sun in a bottle Achilles’ heel In 2016, soon after an When it restarts in 2021, the upgrade, an upper repaired National Spherical Torus coil failed. The machine Experiment (NSTX) at Princeton was disassembled Plasma Physics Laboratory (PPPL) and has sat idle, will use magnetic fields to trap forcing a reckoning and squeeze a hot ionized gas, over PPPL’s future. or plasma, coaxing atomic nuclei to fuse and generate energy the same way as in the Sun. NSTX will test how efficiently a spherical shape can squeeze the plasma. Poloidal Toroidal It will also test using liquid lithium feld line feld line to protect NSTX’s chamber wall and help shunt out heat. Moving Poloidal magnetic coil plasma Field line Toroidal magnetic coil A crucial twist To trap a plasma and keep it away from the Downloaded from Rev it up! walls of the vacuum During a secondslong chamber, the total run, a current of magnetic field—the 24,000 amps quickly sum of the toroidal reverses in the tubelike and poloidal fields— central solenoid coil must twist like a candy to propel the plasma cane. That winding around the torus is produced by http://science.sciencemag.org/ 40,000 times per the current in the second. That motion plasma itself. helps generate the poloidal field. Divertor Tokamaks on parade PPPL’s star has fallen, along with the size of its fusion reactors. But a refurbished NSTX could revive the lab, and set the stage for a leading role on February 8, 2020 in building a fusion power plant after ITER. *reactors drawn to scale ? Tokamak Fusion National Compact National Spherical International Thermonuclear Compact Pilot Plant Test Reactor Stellarator Experiment Torus Experiment Experimental Reactor To be built in the 2030s, PPPL’s biggest machine Canceled in 2008, NCSX would Built in 1999, NSTX tests The $25 billion reactor, an the prototype power ran from 1982 to 1997. have generated a twisting how a spherical shape international effort under construction plant would leverage In 1994, it set a U.S. record field with asymmetric coils, boosts plasma pressures. in France, aims to produce more emerging technologies for power produced. enabling it to run continuously Tests will resume in 2021. energy than it consumes. It should and be smaller and with a stationary plasma. begin operations in 2025. cheaper than ITER. a run, something nearly as otherworldly fills at which colliding nuclei can fuse to form half-century. Yet, on a rainy Monday morn- the chamber: a wispy ionized gas, or plasma, helium. The reactions release energy, car- ing in October 2019, the lab is eerily quiet. heated to 100 million degrees Celsius— ried away by free-flying neutrons. Replac- Nestled among pines in a technology park hotter than the core of the Sun. Injected ing some of the deuterium with tritium, east of tony Princeton, PPPL grew out of the microwaves and churning magnetic fields an even heavier isotope of hydrogen, could university’s classified work on the hydrogen SCIENCE heat and squeeze the plasma and whirl it make the reactions self-sustaining. Such bomb and astrophysicist Lyman Spitzer’s around the chamber 40,000 times per second. fusion promises abundant, carbon-free en- parallel effort to tame fusion as a power The plasma is made of deuterium, a ergy with little of the radioactive waste gen- source. Founded in 1961, PPPL built a series heavier isotope of hydrogen, and the goal erated by fission-powered nuclear reactors. of ever bigger devices that in 1982 culmi- GRAPHIC: C. BICKEL/ C. GRAPHIC: is to bring it to temperatures and pressures That prospect has fired PPPL for the past nated in TFTR, a reactor three times as wide SCIENCE sciencemag.org 7 FEBRUARY 2020 • VOL 367 ISSUE 6478 619 Published by AAAS NEWS | FEATURES The plasma chamber of NSTX seen in 2014, before a short forced it to be disassembled. Downloaded from http://science.sciencemag.org/ as NSTX that set its power record while run- suffer from a limitation, however. To trap a of its magnetic field. An investigation re- on February 8, 2020 ning on deuterium and tritium. “We made a plasma, the magnetic field going around the vealed numerous problems in addition to the tremendous splash in the newspapers,” re- torus must twist like the stripes on a candy shorted coil, Hawryluk says. “We felt strongly calls Michael Zarnstorff, PPPL’s chief scien- cane. To generate that twist, the plasma itself that we really needed to understand what tist. “People would ask, ‘When are you going has to race around the doughnut to produce was going on,” he says, “and not just fix one to have electricity from fusion?’” a current.