Nuclear Energy Gets a Redo
Mining the past to rescue the future
THE MOONSHOT A global embrace of nuclear power could dramatically lower the carbon footprint of the energy sector. For that to happen, innovators need to succeed at developing new reactor designs that overcome the safety, security, and environmental concerns that so far have prevented the technology’s wider adoption.
THE PHILANTHROPY OPPORTUNITY Developing next-generation concepts in nuclear energy and moving them into prototype, demonstration, and ultimately operational and commercial phases is a costly prospect. A combination of government, investor, and philanthropic funding sources could accelerate this trajectory at a time when low-carbon energy production is a must-do on a global scale.
After a generation of near stagnation, the field of nuclear energy is undergoing a surge of innovation — with a distinctly retro flavor. It could be just in time.
Starting in the 2000s, “you had this new wave of nuclear engineers inspired by Silicon Valley and the emergence of the tech companies,” says Joshua Freed, who has been keeping close tabs on this movement as head of the clean energy program at Third Way, a non-partisan think tank in Washington, D.C. “They were all looking at the climate change crisis, realizing that there needed to be new solutions that moved coal and natural gas off the grid.” And, given nuclear energy’s proven ability to produce carbon-free energy in abundance, they were convinced that by far the best solution would be a radical reboot based on reactor designs from the 1950s and 1960s.
These technologies, which come with exotic-sounding names like “molten salt reactor,” “high-temperature gas reactor,” and “liquid-metal fast reactor,” were serious contenders once. They got sidelined only because of a premature rush to commercialize the “light water” reactors that still dominate the established nuclear power industry. But when reimagined with 21st-century sensibilities, the upstarts believed, these half-century-old designs could eliminate many of the problems that have made conventional reactors anathema to much of the public.
Instead of being vulnerable to catastrophic meltdowns, like the one that ravaged Japan’s Fukushima Daiichi plant when it was hit by a tsunami in 2011, the new-generation reactors would be safe enough to put in cities and factories. Meltdowns would be physically impossible: If one of these reactors were hit by a natural disaster, say, or a terrorist attack, it would just quietly shut itself down guided by the laws of physics — no emergency pumps or operator action required.
Instead of adding to the radioactive waste that nobody knows how to dispose of safely, many of the new reactors could actually consume it. The 300,000 tons of spent nuclear fuel that has already piled up at conventional reactors around the world would suddenly go from being a liability to being an invaluable resource.
And instead of opening yet more pathways for rogue nations or terrorists to obtain materials for nuclear weaponry, most of the designs would effectively eliminate the two biggest loci for proliferation risks: fuel reprocessing facilities and uranium enrichment plants.
This vision of rebooted nuclear power has taken hold with striking speed. When it began in the 2000s, the advanced nuclear movement included just a handful of established firms such as the defense contractor General Atomics in San Diego, California, and a slightly larger handful of startups such as NuScale Power in Corvallis, Oregon, and TerraPower in Bellevue, Washington. By 2018, the movement comprised 60 advanced fission-energy projects being carried out by companies, universities, and national labs in the United States and Canada alone. Other advanced-nuclear companies have launched in Europe, notably Moltex Energy in London and Seaborg Technologies in Copenhagen. And although secrecy makes details hard to come by, Freed says vigorous development is thought to be underway in Russia and China.
“They were all looking at the climate change crisis, realizing that there needed to be new solutions that moved coal and natural gas off the grid.” — Joshua Freed
True, none of these advanced concepts has been demonstrated yet; the developers are still doing laboratory experiments on materials for their reactors and optimizing the designs with high-powered computer simulations. The first prototype reactors won’t begin operating until the mid-2020s, assuming all goes as planned, and the first commercial power plants aren’t expected until the 2030s. Even then, these projects could be facing a stiff headwind in the form of natural gas from fracking. If gas continues to be as abundant and cheap as it is now, it could be hard for any new energy technology to get a foothold … unless it has very patient backers.
That said, however, most if not all the advanced-reactor startups are embracing new ideas for being cost-competitive in the current price environment — a notable example being the integrated mass-production of reactors in factories and shipyards, which could represent a substantial savings over the current practice of building nuclear plants piecemeal to fit each specific site. There is also quite a lot of systems-level thinking being done on the design of integrated hybrid energy systems, which would use advanced nuclear reactor to compensate for the intermittent nature of renewable sources such as solar and wind.
In short, says Freed, “there’s a lot of people that have come into this space. There’s a significant amount of capital that’s come into this space. And the government is starting to clear away obstacles.” Indeed, with at least five advanced-nuclear companies already starting the long approval process for their prototype reactors, the U. S. Nuclear Regulatory Commission is gearing up with the expertise and simulation software it will need to evaluate them. There is strong bipartisan support in Congress for boosting federal research on advanced reactors — a tangible example being the August 2019 launch of the National Reactor Innovation Center, a facility at the Department of Energy’s Idaho National Laboratory that will give companies a place and technical support to demonstrate their technologies. It’s also been suggested that the federal government could spur reactor development by offering payments for achieving certain milestones, in much the same way that NASA spurred the development of a commercial U.S. rocket industry with milestone payments to companies such as SpaceX.
And, says Freed, thanks to the emergence of philanthropic funds focused on issues such as climate change, there are many more opportunities for donors to pool their resources in ways that will further these technologies most effectively.
“I think that there is a clear desire from both the private sector and the government to see multiple advanced-reactor prototypes running before 2030,” he says.
Troubled Childhood
The problems that plague today’s nuclear industry stem from a series of seemingly reasonable decisions made in the late 1940s, when the U.S. Navy was trying to harness nuclear energy to power ships and submarines.
The basic physics was straightforward. Since the eve of World War II, physicists had known that certain heavy atomic nuclei were inherently fragile. Hitting such a nucleus with a neutron could cause it to fission, or rupture into two lighter nuclei. Each fission event also would liberate at least two new neutrons from the nuclei and, in accordance with Einstein’s mass-energy equivalence equation, E=mc², a lot of energy. The new neutrons, in turn, could strike other nuclei and trigger additional fission events. The original neutron would thus beget 2 neutrons, then , 8, 16, 32, 64 and so on in an exponentially growing cascade of energy release.
The wartime Manhattan Project had proved that such a “chain reaction” could make a history-changing nuclear bomb. But the Navy engineers needed to put that exponential process into a reactor and turn it into a controlled, steady release of power. That required a whole new set of considerations.
First was the choice of fuel. Several types of heavy nuclei were known to be capable of sustaining a chain reaction, but only one of them, uranium-235, was found in nature and available in practical amounts. Uranium was not suitable for nuclear reactors straight out of the ground, however. Like most elements, it exists in multiple isotopic forms, each of which has the same number of protons in its nucleus but differing numbers of neutrons. Uranium-235 (92 protons plus 143 neutrons) comprises fewer than 1% of the atoms in native uranium ore; virtually all the rest is non-fissile uranium-238 (92 protons plus 146 neutrons). For use in reactors, the uranium-235 fraction would have to be enriched to something like 5% — a hugely difficult and expensive process. Even so, that was hardly an issue for the Navy: the U.S. nuclear weapons program had already created an extensive infrastructure for creating bomb-grade uranium enriched to 80% or more.
Next, the reactor needed to tame the explosive chain reaction so that it would release energy at steady and controllable rates. This required absorbing roughly half the neutrons, so that the reactor core could settle into an equilibrium in which each fission event produces, on average, just one other fission. As with just about every other type of reactor, the Navy’s engineers did this by equipping the reactor core with a set of control rods made of a neutron-absorbing material such as boron. Pulling the rods part way out of the core would allow more fission events and raise the reactor’s power output, whereas pushing them further in would reduce the number of events and lower the power output.
Then the reactor needed a “coolant”: a fluid that would circulate through the core, absorb the heat being generated by the fissioning uranium, and carry that heat out to where it could drive a steam turbine or electric generator. Engineers elsewhere were experimenting with options such as helium gas, molten sodium metal, or even molten lead. But the Navy’s designers went with the most familiar choice: water. Water did have an inconveniently low boiling point, just 100ºC at sea level. This would make the reactor hopelessly inefficient for power generation, where hotter is always better. But it was easy enough to raise the boiling point past 300ºC by pressurizing the reactor vessel to 100 atmospheres or so. And given water’s enormous upsides — it was cheap, abundant, well-understood, and carried a lot of heat — operating at 300ºC was good enough.
Finally, the reactor design had to manage the fact that neutrons emerge from the fission reactions at some 14,000 km/sec on average. At that speed, they would more likely zip past the target nuclei and hit the reactor walls than trigger the next fission in a chain reaction. So, to make a compact reactor that could fit on a naval vessel, the designers had to find a way to “moderate” the neutrons, or slow them down to just a few kilometers per second. At those speeds, a neutron is roughly a thousand times more likely to trigger fission in a U-235 nucleus than it would have been originally. And here’s where water happened to make life easy: The coolant also functioned as an excellent moderator.
So that was the naval reactor: enriched uranium as fuel, and pressurized water as coolant and moderator. It debuted in 1955, when the Navy launched the world’s first nuclear-powered submarine, the USS Nautilus; the reactor’s descendants are still the standard nuclear power source for U.S. Naval ships. But as history would have it, 1955 was also when the Eisenhower administration was pushing its high-profile Atoms for Peace program to harness nuclear energy for civilian use. Since the naval reactors were further along than anything else, their basic water-cooled design was pressed into service and scaled up to gigawatt size to serve as commercial power reactors. Over the next two decades, hundreds of these “light-water reactors” were built around the world. (The name distinguishes them from Canada’s CANDU reactor, which is cooled by “heavy water”: deuterium oxide made with the heavy isotope of hydrogen.)
Only in mid-1970s did the light-water reactor’s drawbacks really start to sink in.
A big one was safety — a concern that arises because the fragments of a fissioning nucleus don’t just vanish. They hang around as middleweight isotopes such as strontium-90, iodine-131, or cesium-137. These fission products build up inside the fuel and — since many of them are fiercely radioactive — their decay releases more and more heat that no control rod can stop. This opened up a nightmare scenario known as LOCA: a “loss-of-coolant accident” in which a broken pipe, a pump failure, or some other mishap interrupts the flow of water and traps the decay heat in the core.
Within a day, if operators couldn’t get the coolant flow restarted, the now-stagnant water would boil into superheated steam. The reactor’s fuel, which consists of uranium oxide pellets encased in a corrosion- and radiation-resistant zirconium alloy, would start reacting with the steam to produce explosive hydrogen gas. And soon enough, a pool of molten fuel would be eating its way through the bottom of the reactor vessel.
Thus, the term: “meltdown.”
To guard against this possibility, engineers had always equipped their reactors with emergency core-cooling systems that would supposedly kick in before things got dire. Even so, the nightmare scenario ceased to be hypothetical on March 28, 1979, when a faulty relief valve resulted in a partial meltdown at the Three Mile Island nuclear plant in Pennsylvania. The operators were able to regain control before superheated steam ruptured the reactor’s pressure vessel and spewed radioactive fission products across the countryside. Yet the scare redoubled an already growing public opposition to nuclear energy — and deepened the disillusionment of utility executives who already were feeling burned by nuclear plants’ multi-billion-dollar cost.
The result was a vicious circle that put nuclear innovation on an indefinite hold. As safety concerns led to demands for ever more extensive back-up coolant systems and ever longer delays for regulatory approval, the cost of new plants only grew. Construction of new nuclear plants slowed after Three Mile Island and would stop almost completely after a steam explosion destroyed the Soviet Union’s Chernobyl plant in 1986. This explains why most of the world’s currently operating nuclear power plants are well over 30 years old and are based on designs from the 1960s.
Meanwhile, much of the research money that might have been devoted to developing new, inherently safe reactors was diverted into ensuring the safety of existing plants. Robert Schleicher, a nuclear engineer at General Atomics, remembers leaving the nuclear field in the early 1980s to work on high-powered lasers. “When I came back to nuclear in 2009, I was stunned,” he says. “Nothing had changed! And I think that is the tragedy of the nuclear industry: They’ve gone for so long without any technology advances.”
Past Perfect
This near-moratorium on new nuclear plants has begun to lift in the past decade, mainly because fast-growing countries such as China and India have come to see them as one quick way to meet their insatiable need for energy. But to many advocates, this only makes the need for new-generation reactors more urgent: virtually all of the 60-odd plants now under construction will use slightly updated versions of the classic light-water reactor technology, and will thus have all the same inherent problems.
The first group of new-generation machines to become operational will almost certainly include a small modular reactor — a generic name for transitional machines that try to eliminate most of the cost and safety problems of present-day fission plants, while taking advantage of all the advanced fuel and materials technologies developed for them. There are more than half a dozen corporate and academic projects in this category, but the closest to actual operation is the reactor being developed by NuScale Power in Corvallis, Oregon.
The inspiration for the device hit in 2004, says NuScale CEO José Reyes, who was then a nuclear engineer at Oregon State University in Corvallis. He had taken a sabbatical to work with the International Atomic Energy Agency in Vienna and found himself in frequent conversation with delegates from member states such as Ghana and Indonesia. “I kept hearing the same thing over and over,” he says: They needed power, particularly for things like desalination plants for supplying clean potable water. “But they needed smaller amounts of power,” he adds. “They just didn’t have the capital or the electric grid to support thousand-megawatt nuclear plants.”
That was the “aha moment,” says Reyes. Just the year before, he and his co-workers had finished up a Department of Energy-funded research project to reimagine light water reactors from the ground up. The design they’d settled on featured a reactor module with an electric power output of just 35 megawatts, or less than one tenth the 500- to 1000-megawatt capacity of standard power reactors.
This did mean that customers would have to install 30 modules if they needed that full gigawatt — or maybe just one or two, if they needed less. But the smaller reactors would be a big win when it came to the total power plant cost, much of which stems from the years it takes to tailor and integrate all the components at the plant site. The idea was to build, test, and certify the reactor modules on a factory assembly line, then ship them to the site on ordinary trucks, rail cars, or barges and drop them into place. It could cut months, if not years, off the construction schedule.
At the same time, says Reyes, shrinking the reactors was a big win for safety. Because a 35-megawatt core produces far less heat than its 500-megawatt cousin, the reactor’s coolant water could remove that heat — whether from fission reactions or fission products — via nothing more than natural convection: no pumps required. In normal operation, water in contact with the core would simply form a hot liquid plume rising up the reactor’s centerline like a fountain. Then at the top, the water would transfer its thermal energy to the power plant’s electrical generators via components known as heat exchangers, and flow back down the reactor walls to the core.
And if the plant got hit with an earthquake or tsunami? That wouldn’t change the convection physics, says Reyes, so the water would just go right on extracting heat without any need for engineered backup systems. According to the team’s calculations, he says, “under the worst-case conditions, the reactor would shut itself down without any operator action, or computer action, without any outside AC or DC power. And it would remain cool and safe for an unlimited period of time without the need to add water.” Just to make sure, he adds, the team tested the convection-cooling process in a one-third scale model of the reactor that used an electric heating coil to stand in for the nuclear core. “It worked really, really well,” he says.
“Under the worst-case conditions, the reactor would shut itself down without any operator action, or computer action, without any outside AC or DC power. And it would remain cool and safe for an unlimited period of time without the need to add water.” — José Reyes
Reyes’s conversations during his sabbatical year in Vienna convinced him that there might be a market for these small reactor modules. “When I came back to Oregon in 2005,” he says, “we started on maybe a couple of dozen changes to make the design commercially viable.” The current specifications call for a reactor module roughly 23 meters tall by 4.6 meters wide — “small” is a relative concept in this business — with an electric power capacity of 60 megawatts. He and his colleagues launched NuScale in 2006 to continue that development work. The company greatly improved its financial stability in 2011 when the international engineering and construction giant Fluor became their lead investor. “That’s made a huge difference,” he says, “because they have a more long-term view.”
And then in 2015, NuScale partnered with Utah Associated Municipal Power Systems, a state-run utility, to build its first demonstration reactor at the Department of Energy’s Idaho National Laboratory. Assuming that the company’s final designs are approved by the Nuclear Regulatory Commission, that reactor could be producing power by 2026. And if that test goes well, full-scale commercial deployment could follow.
Generation IV
A few years behind NuScale but coming up fast is a diverse collection of advanced designs known as Generation IV reactors. (The first Navy reactors were Generation I; the first commercial nuclear plants were Generation II; and advanced light-water reactors such as NuScale’s are Generation III.) All of them approach the safety and cost conundrum with the same kind of factory construction and safety-by-physics thinking. But many of them also tackle the intertwining challenges of nuclear waste disposal and proliferation prevention, which light-water reactors like NuScale’s cannot.
It all goes back to those fission products. Even when the reactor is operating normally, the fission products tend to dampen the chain reaction by absorbing neutrons. After 2 or 3 years, these products build up to the point where the chain reaction stops entirely, and the fuel is “spent.” The spent fuel then has to be switched out for fresh fuel and submerged in a pool of water for a decade or so to allow the worst of the short-lived products to decay. And then …
… well, that’s the problem. The spent fuel still has most of the fission potential that it started with. But before that resource can be exploited in a standard light-water reactor, it would have to be chemically reprocessed to extract the reaction-hobbling fission products and refashion the leftovers into fresh fuel.
In fact, that was the plan back in the 1970s, when countries such as the United Kingdom, France, and the Soviet Union already had built the first of what then was envisioned to become a worldwide network of reprocessing plants. But that effort was rapidly running afoul of escalating concerns about proliferation, a problem that arises because of all that uranium-238 in the core. This isotope won’t undergo fission by itself, but it can absorb a neutron from the chain reaction going on around it and transform into plutonium-239, which is very fissile indeed. Plutonium-239 is a prime nuclear weapons material — it was used in the Nagasaki bomb — and workers at each of the reprocessing plants being envisioned at the time would be extracting plutonium by the ton. So a worldwide network of such plants, all operating under local control, would become a constant source of angst about rogue actors diverting even just the few kilograms required to make a crude but effective bomb.
This possibility was brought home in 1974, when India surprised the world by testing a nuclear bomb made with plutonium extracted in secret from spent reactor fuel. In 1977, President Jimmy Carter responded with a ban on commercial reprocessing in the U.S., and much of the rest of the world did the same. Except in those nations that already operated reprocessing plants, spent nuclear fuel would no longer be considered a resource to be exploited. It would be designated as waste, destined for long-term storage.
But that merely traded one problem for another: Since plutonium-239 has a half-life of 24,100 years, “long-term” storage meant at least 241,000 years. (The rule of thumb for safely disposing of radioactive material is 10 half-lives, by which time the radioactivity will have decreased more than a thousand-fold.) No one has yet figured out how to guarantee the isolation and safety of spent fuel for that long.
A good example of how Gen IV reactors could solve both the waste and proliferation problems — not to mention the cost and safety problems — is the Energy Multiplier Module (EM²) that Schleicher and his colleagues have been developing at General Atomics since 2009.
“We sat down to essentially rethink how you should design a nuclear reactor,” says Schleicher. The first and most fundamental decision was one shared by many other Gen IV designs — and many experimental reactors dating back to the 1960s: This would be a “fast” reactor. That is, the EM² would use neutrons coming out of fission reaction in their raw, high-energy state, instead of trying to slow them down with water or any other moderator.
That is why so many of the Gen IV reactors are cooled with substances that don’t slow the neutrons, such as molten sodium or lead. Drawing on General Atomics’ experience with two experimental reactors it had operated for the U.S. atomic energy program from the 1960s through the 1980s, Schleicher and his team opted to cool their reactor with helium. Unlike many coolants, helium has essentially no chance of absorbing neutrons. It also builds in an automatic safeguard against meltdowns: Because it’s already a gas, it cannot boil away the way water can.
Working with fast neutrons does mean accepting those thousand-times-lower fission probabilities. But the EM² design compensates with another well-known nuclear-engineering trick: Surround the core with a “reflector,” a layer of some heavy substance that will bounce escaping neutrons back through the core and give each neutron many additional opportunities to cause a fission event. Previously used reflector materials tended to absorb a lot of the neutrons, says Schleicher. “But we invented a new zirconium silicide material that proved to be far superior,” he says.
With this reflector in the design, the EM² enters the small modular range. The reactor vessel is about 12 meters long and 4 meters across. “We sized all the major components such that they could be truck-shippable,” all the while retaining a design that benefits from the many advantages of fast neutrons, says Schleicher.
One such advantage is that the very fastest neutrons can trigger fission events in uranium-238. This means that a fast reactor doesn’t have to burn enriched uranium. It can burn natural uranium or even depleted uranium — the uranium that’s left over from the uranium-235 enrichment process. And that, in turn, could reduce a major proliferation risk. Countries with nuclear energy programs would no longer need to run a network of enrichment plants, which can be modified to enrich uranium all the way to nuclear-weapons levels. (This is one of the biggest issues right now with Iran’s nuclear program).
“We could load the core and let it burn for 30 years without touching it.” — Robert Schleicher
Another advantage of fast neutrons is that fission products cease to put the brakes on fission. They will still accumulate, but the fast neutrons will zip through them with almost no probability of being absorbed. This would make spent nuclear fuel a thing of the past, says Schleicher: “We could load the core and let it burn for 30 years without touching it.” Better yet, this would make a resource out of all the spent nuclear fuel that’s currently piling up in “temporary” storage at conventional reactors: just put it into a fast reactor and burn it up.
The use of fast neutrons also could make long-term waste disposal a thing of the past. Thirty years of burn-up would be enough to consume not just all the fissile uranium in the fuel, but all the plutonium-239 and other long-half-life isotopes that form during the process. Nothing of any concern would be left beside the fission products, most of which have half-lives measured in hours or days. The longest-lived of them, strontium-90 and cesium-137, have half-lives of about 30 years. So, by the 10 half-life rule, says Schleicher, “after 400 years you could hold it in your hand.” While isolating something for that long isn’t trivial, it’s feasible — say, by encasing the short-term waste in concrete or glass, or putting it down a deep borehole.
“The next thing was that we needed to get the efficiency up high,” says Schleicher, since that’s a major factor in the reactor’s economics. So he and his colleagues designed the EM² to operate with its helium coolant at 850ºC, versus the roughly 300ºC maximum for pressurized water in conventional reactors. Calculations show that this higher temperature should allow the EM² to convert heat to electricity with a net efficiency of 53%, compared to just 30% to 35% for light water reactors.
The high operating temperature also opens up the possibility of skipping the electricity-generating step and locating the reactor right next to a factory, which could use its heat to drive high-temperature processes such as smelting metals, driving chemical reactions, or even extracting hydrogen from water via an efficient thermochemical reaction. This could have a sizable impact on carbon emissions: Generating industrial process heat currently accounts for about one-third of the total energy usage in the U.S. manufacturing sector — and almost all of it now comes from burning fossil fuels.
To make that 850ºC temperature possible, Schleicher and his team needed a new kind of cladding: the material that encases the uranium oxide fuel pellets to hold them in place and keep fission products from escaping and contaminating the coolant. In conventional reactors the cladding is an alloy of corrosion-resistant zirconium metal, which does not absorb neutrons. “But most metals, like zirconium or stainless steel, have almost no strength after 800ºC,” he says. “So we invented a silicon carbide composite material that basically … doesn’t melt. It has the same toughness as carbon fiber composites, but almost full strength up to 1800ºC.”
Combined with the un-boilable nature of helium gas and a reactor design that emphasizes convective cooling, he says, this high-temperature fuel makes the EM² all but immune to meltdowns.
Nuking into the Future
If and when General Atomics takes the next step, says Schleicher, “we would first build a 50-megawatt prototype to qualify and prove a lot of the innovations that we have instituted here. And then, upon successful operation of the prototype for a couple of years, we would begin sales of our full-scale 240-megawatt modules,” which could power roughly a quarter-million homes apiece.
By then, though, General Atomics should have plenty of company — because the others are not waiting around. TerraPower, to take perhaps the most prominent example, is pursuing not one but two advanced nuclear projects.
The most advanced is the Traveling Wave Reactor (TWR), which the company has been developing since it launched in 2006 with funding from investors such as Microsoft founder Bill Gates. Like the EM², this reactor will use fast neutrons and a wide variety of fuels, including natural or even depleted uranium. And it will burn its fuel for decades at a stretch, until all the fissionable isotopes are consumed completely. Unlike the EM², however, the TWR will seek to maximize the burn-up by equipping the reactor with a remote fuel-handling system, which will reshuffle the fuel rods about once per year in somewhat the same way that you might use a fireplace poker to move unburned wood into the hottest part of the flames.
Another key difference is that the TWR will use molten sodium metal as a coolant. Sodium reacts violently with water, which will be used in the steam turbines, so there’s a premium on keeping the coolant pipes sealed tight. But that’s far easier than sealing the pipes in a light water reactor, since the sodium doesn’t have to be pressurized. And the payoff is that sodium in its elemental and molten form is non-corrosive, doesn’t absorb neutrons, and can carry lots of heat.
Recently, however, the TWR suffered a non-technical setback. “We were pursuing a roadmap that was going to begin in China with building a 300-megawatt demonstration reactor, and eventually would have moved to a larger plants,” says TerraPower CEO Chris Levesque. But on October 11, 2018, TerraPower’s cooperative agreement with China went up in smoke when the Trump administration tightened export controls with that country. Since then, says Levesque, the company has been looking at options for building that prototype in the U.S. And it has also been rethinking the initial plant’s size, he says, “because the U.S. has a different environment both in terms of funding and in terms of electricity markets.”
Fortunately, says Levesque, this turmoil has had no effect on TerraPower’s other reactor project, which it launched in 2015. Known as the Molten Chloride Fast Reactor, it’s a variant of the molten-salt concept that was pioneered in the 1960s at the Oak Ridge National Laboratory in Tennessee. The idea is to get rid of the solid fuel and instead fill the reactor with “salts” — a mix of fluoride and chloride compounds in the same chemical family as sodium chloride, a.k.a. table salt. Then you mix in a fuel-containing salt such as uranium tetrafluoride, heat the whole thing until it melts into a clear greenish liquid, and let the fission take place right there in the pot.
Molten salt reactors like this one have a number of inherent safety features, starting with the fact that they can’t melt down: They’re already molten. But the Molten Chloride Fast Reactor’s main appeal for TerraPower is that it can operate at much higher temperatures — in the at 800ºC range — than the TWR, which gives it a shot at the market for supplying industrial heat.
This is just a partial list of the scores of advanced nuclear energy projects now underway around the world. The arena is wide open, says Freed, and any one of the projects could surge into the lead.
Who’s going to get those first prototypes built? Who is going to become a major player in the industry? “That’s the exciting thing about advanced nuclear,” says Freed. “We don’t yet know.”
M. Mitchell Waldrop is a freelance writer in Washington, DC.
