A new nuclear project needs a modest-sized piece of land, transmission lines, water, access by road or rail, and earthquake-stable terrain. It needs something else too: public acceptance.

But nuclear plants often draw some unusual reactions. Some reactions are antique reflexes, and some are inversely proportional to the distance from the project. And some give clues about how developers should pick their sites for new projects.

Here are some “not-so-secret secrets” that explain why it’s easier in some places to get the public acceptance needed to build a new reactor. While every greenfield project has its own challenges, avoiding the most common pitfalls can make siting and construction easier.

Share

Old Reactors Have Old Foes

In the antique category, progress towards restarting Unit 1 at Three Mile Island (TMI) near Harrisburg, Pennsylvania, has awakened old fears. Opponents are reappearing like 70s rock band stars, grayer and maybe a bit hoarse, but singing the same tunes.

Jane Fonda has tried to re-enter the debate, with an opinion piece in the Philadelphia Inquirer opposing the restart. Fonda’s March 1979 film, China Syndrome, about a fictional accident at a fictional nuclear reactor, probably owes its commercial success to the well-timed TMI accident just twelve days after the movie’s release.

Fonda, who will turn 88 in December, gave a mélange of familiar arguments, including that nuclear power is too slow, too vulnerable to earthquakes, and too expensive. But, restarting Unit 1 at TMI would not encounter any of these issues. The work needed to re-start the reactor, now called the Crane Clean Energy Center, will take a few months, not years. Harrisburg is far away from any faultlines and has negligible earthquake risk. And the restart is budgeted at $1.6 billion, which is dirt cheap for 835 megawatts of round-the-clock power. But the op-ed illustrates the first secret of nuclear siting: if you propose making nuclear energy at an old nuclear energy site, you must deal with the same old opponents who have been fighting that site for years.

The re-emergence of the holdovers is a clue for companies seeking to build new reactors of the kind of reception they will get if they build them adjacent to operating units where there is a history of opposition. Avoiding controversial historical sites can avoid dealing with an opposition that is suspicious of legacy plants no matter how well they have operated, and that fails to see improvements in the technology, advances in safety, and the legitimate benefits to local economies and broader society.

Subscribe now

The Small-Radius Halo

A second not-so-secret secret is also apparent in the TMI restart case: that the popularity of a reactor is inversely proportional to the distance. People in adjacent communities, in a circle around the proposed site, may be very supportive. Those at a greater distance—notably, at the state capitol, or in Hollywood—may take the opposite position.

There are myriad other examples of the small-radius halo of nuclear acceptance. When Holtec went looking for a place to put dry casks of nuclear waste in centralized interim storage, two counties in eastern New Mexico, Lea and Eddy, quickly allied to find a piece of land near Carlsbad and facilitate the project. In a region not shy about making a profit from its geology, they welcomed the economic development. But the state legislature in Santa Fe stepped in to prohibit storage of nuclear waste without the state’s consent. The state law was opposed by the city of Carlsbad.

The Texas legislature in Austin took similar steps to block interim storage in West Texas, against the wishes of the local government and community.

Local communities often look at greenfield nuclear projects like manna. They are economic boosters that bring construction jobs and then well-paid, year-round jobs in operating the plant. Regionally, they can reduce energy costs for decades.

Kemmerer, Wyoming, for example, is counting on the TerraPower nuclear project to “save the town,” after the local coal-fired power plant was shut down. The mayor, Bill Thek, told the Washington Post that when the closing of the coal plant was announced, he thought, “we’re going to dry up and blow away.”

In Calhoun County, Texas, where Dow Chemical wants to break ground next year on a high-temperature gas graphite reactor to make steam for industrial use, the county commissioners quickly approved a tax abatement to help the project.

In fact, jurisdictions will squabble over how to share the benefits of a reactor. When Entergy considered building a second reactor at its Grand Gulf site in Port Gibson, Mississippi, and got to the stage of moving dirt to prepare, the state legislature stepped in to grab the money paid by the utility to the government and spread it around the state, rather than let it stay in Claiborne County. There was no opposition to the idea of a second reactor, but the project died as natural gas prices declined.

Share

The More the Merrier

While building a new reactor at a controversial site will elicit further controversy, there is merit to adding an additional unit to a place where a reactor is already operating without drama.

Building next to an operating reactor has some advantages—something Georgia Power realized when it picked the Vogtle site, near Augusta, with two operating reactors, as the place to add twin AP1000 units. The site had already been characterized for earthquakes and floods, and ample weather data was on hand. There is already a security plan, although it may need to be expanded. And when new plants are added to an old site, they can share human resources like instrumentation and control technicians, health physicists, electricians and engineers.

In the case of Vogtle, co-locating the new plants allowed the builders to draw on support from local communities that recalled the prosperity brought by construction of the first two plants in the 1980s.

Many of the plants now running were intended to host multiple reactors. Many have land and water sufficient for more units. And all have access for shipping in large components, and have grid connections.

New Kinds of Reactors Will Need New Kinds of Sites

All these lessons pertain to big reactors of the kind now in service, and modernized versions, like the Westinghouse AP1000 model built at Vogtle, that are under consideration around the country. But a different set of secrets may apply to some advanced reactor technologies.

Micro-reactors, for example, are all likely to be on “greenfield” sites, although the field may not be green; it may be a parking lot with a diesel generator, which makes a noisy, unreliable and smelly neighbor.

The siting decisions will fall to a different set of actors. Some will be corporate, like mining operations that are far from the grid. Others will be remote communities, which now use diesel generators and recognize the difficulties of getting diesel shipped in. In either case, the local benefits to reliability of electric supply and to air quality and noise will be obvious to local communities.

Larger advanced reactors provide benefits that may be more difficult to communicate. For example, they may use fuel forms that are extremely heat-tolerant, or safety systems that are mostly reliant on uninterruptible forces, like gravity and natural heat dissipation. The significance of the engineering details may be somewhat harder for neighbors to grasp.

But they, too, will be welcomed locally, for the clean energy and jobs they provide. They will get some opposition from more distant opponents, who are already arguing, for example, that we should not try to build something that no one has built before. But on that basis, we ought to go back to plants that burn pulverized coal.

Subscribe now

On July 3, 2025, workers at a nuclear weapons plant run by the Department of Energy (DOE) in South Carolina found a wasp nest near a tank of radioactive waste. After they used insecticide to kill the wasps, they discovered that the nest was radioactive. Soon they found three other radioactive nests.

So they bagged the nests as radiological waste. The nests were outside of regular cleanup operations, so they went back to work, stabilizing wastes from decades of nuclear weapons production. But the news was splashed over the web, newspapers, and broadcast media for days.

Why?

The DOE report on the incident does not have specifics, but wasps generally build nests from wood that they chew and digest. Some legacy waste materials, especially cesium and strontium, are absorbed into trees like two more ordinary elements they resemble, potassium and calcium, which are essential minerals for all living organisms. Nest building could have concentrated the isotopes even further. All the wasps had to do was choose one particular tree that had absorbed long-lived isotopes (Sr-90 and Cs-137 both last about 300 years before decaying to levels that are immeasurable), to make a bright little spot on an environmental survey.

In reality, the insecticide was probably more of a threat than the radiation. And the wasps themselves were a hazard; wasp, hornet and bee stings kill more than 60 people a year according to the Centers for Disease Control (almost entirely due to allergic reactions). Radiation, on the other hand, kills approximately zero people. But from the perspective of news and public interest, the story hits three key points: animals, radiation, and originality. It would still be a mistake, though, to ascribe much public health significance to the discovery.

Animal stories always play well, especially in periods when editors are looking for something—anything, really—to break the monotony of news of brutal wars, strife in the streets, heat waves, and economic uncertainty. Radiation always plays well, too. Combine them both and you have a phenomenon, like reports of genetic variations in wild dogs near Chernobyl, the site of the 1986 explosion of a Soviet reactor and significant release of radioactive materials. It also explains the continuing appeal of Blinky, the three-eyed fish that Bart Simpson, the cartoon character, caught in a pond near the Springfield nuclear plant.

One of us (Wald) has been seduced by the combination of factors, writing about rabbit feces with traces of radiation, located from a passing helicopter, and testing of tumbleweeds, which, before they tumble, have exceptionally deep roots that can pull radioactive materials out of the dirt. The Savannah River Site where the wasp nests were logged, is where weapons materials were formerly made. Checking wildlife for radioactive contamination isn’t new; hunters are allowed to take deer on the grounds, but they have to submit the antlers to Energy Department technicians who check for radiocesium. The technicians have never found much.

Significantly, this radioactive material is leftover from the Cold War production of weapons material. That stuff really is oozing goo, (but not green or glowing, as the Simpsons satirically suggest) unlike the mostly solid (ceramic, in fact) radioactive materials from civilian power operations. The waste was handled with the apparent haste needed in the 50s and 60s to stop Godless Communism. (If you’re under 40 years old, you may need to look that up. We won’t explain it here. It’s enough to say that the operations that led to the creation of this waste ceased in the 1990s.) Savannah River isn’t alone; at the contaminated Hanford site, workers have also logged a radioactive rabbit.

However, Thumper and Bambi can turn up with trace amounts of radioactive materials because of Cold War practices at government weapons sites similar to the environmental mismanagement which led to the toxic chemical dump at Love Canal in New York. That crisis affected hundreds of families and prompted the Superfund Act and ensured “cradle to grave” responsibility for waste. The weapons sites will require a lot of work, but the exposures are nothing compared to what the public faces from many commercial chemical sites.

Share

Why The Impact is Limited

Exposure to radiation is classified as either external or internal. External exposure, such as dental x-rays or cosmic radiation is just the energy coming from outside of your body and getting absorbed as it goes through. Internal exposure can come from ingestion, inhalation, or injection of radioactive materials such as getting a medical procedure with radioisotopes or radon gas in your home. Wasps aren’t a part of the human food chain, though, and very few people eat a lot of wild rabbits.

But it drives interest. The Associated Press wrote about the radioactive wasps nests, and the report was picked up around the country and abroad.

When is it time to start worrying about radiation? This is an open question, because for decades regulators have used a conservative assumption, still unprovable after seven decades of research, that there is no threshold of safe radiation exposure. The Trump administration’s recent executive orders instruct the Nuclear Regulatory Commission to re-examine that assumption. It may not actually be conservative, if it inhibits the use of nuclear technology and the result is substitution of other ways to make energy or deliver treatments that are demonstrably much worse. The official position of the Health Physics Society, the professional organization that represents the field of radiation safety, is that risk is not detectable below 100 mSv (10 rem). That is 100 times the dose that the NRC allows for the public. Assuming that the nest was just a chunk of Cs-137, instead of a wasp nest, someone would have had to have kept their face three feet away from it for almost 19 days to even reach the public dose limit. But if it was full of wasps, who would want to be next to it for even 10 seconds?

The NRC makes a distinction that can get confusing. It doesn’t limit exposure to radiation—it limits exposure to man-made radiation. So, you can build your house on a mountaintop, above much of the atmosphere that shields us from cosmic rays, and get as much dose as you like, but the dose from the americium in your smoke detector is regulated.

In the case of the contaminated wasp nest, maybe it’s the flip side of the old question: if a tree falls in the forest and there’s no one there to hear it, does it make a sound? In this case, there was no dose, and the tree fell silently. (For there to be a dose, you’d have to be foraging the wasps.) But compared to some environmental risks that are better documented, like lead contamination in the water, or mercury in fish or food, radiation is just, well, sexier and scarier. The immediate effects of lead poisoning and long-term radiation are both invisible, but only one of them can make a Geiger counter ping in a way that most people can recognize instantly. Leaded gasoline and paints caused irreparable harm to millions of Americans but if the same amounts of radiation were emitted by cars, soldiers would have been called to close the highway system.

The public focus on radiation isn’t always because of the perception of threat. The public follows radiation news partly because radiation is so easily detected, but that can also be an advantage. In South Africa, conservationists are injecting the horns of endangered rhinoceroses with radioactive Cobalt-60, so that if the animals are poached for their horns, the smugglers will get caught by radiation detectors at the airports. The radiation can be detected even in a horn hidden in a 40-foot shipping container but blood tests on the animals confirm that the exposure is too small to affect their health. Due to success, it is now being tested to prevent pangolin poaching as well.

That one, like the wasps, combines animals and radiation. If it stops the rhinoceros’ march toward extinction, it will certainly be more important than the news about the wasps. But the media won’t play it that way.

Subscribe now

Sometimes it seems that the political and economic system of the United States is simply not set up to take full advantage of the best available energy technology, nuclear reactors. But this is only a partial view of our problems; the mismatch between technology and governance is not, alas, confined to nuclear. As has become quietly but painfully obvious in the last few years, it is also true of transmission, specifically high-voltage direct current lines that could extend for hundreds of miles and make the grid cleaner, more stable, and more economical.

The failure to decarbonize promptly isn’t a technical issue, or a problem because of difficulty raising capital or making commercial arrangements. It’s an artifact of the regulatory system.

What gets built in the power system isn’t a function of what’s needed; it’s a function of what the regulatory system wants, or at least, what it will tolerate. That is what eventually determines the outlines of the energy system, including how clean, reliable, and expensive it will be.

Fifteen years ago, a Texas company set out to use modern technology to solve an obvious problem: the United States has a wind belt at its center, an area where the wind blows strongly enough to make electricity for many hours of the year. But the wind belt is sparsely populated, and demand for energy is to the west and the east. The company, Clean Line Energy Partners, planned to build four long-distance electric transmission, high voltage direct current (DC) electricity superhighways, each of which could securely deliver thousands of megawatts of power with low line losses.

And the last one of those sketched out lines may have just died, falling victim to a Republican U.S. Senator from Missouri who probably realized that there were votes to be had by joining the opposition and that the economic benefits of a transmission line crossing his state would go to other states.

The ability of a state to veto a transmission line is what gives us the fragmented grid we have today, a grid that becomes increasingly ill-suited to the job of moving electricity as needed, especially as we try to shift to zero carbon wind and solar.

This problem is clear to solar and wind developers but not so much to activists and advocates of renewables. Without a grid that can transport energy from the most efficient renewables—wind in the midwest, solar in the southwest—a wind and solar-focused decarbonization plan means higher prices and continued use of fossil fuel generation.

Share

Three Down, One Going Down

Productivity in a wind machine is measured by capacity factor, which is a comparison of the amount of energy actually produced in a year with what would have resulted from 100 percent performance, 24/7/365. The average capacity factor for onshore wind in the United States is about 35 percent. Kansas, where the power line would have started, is closer to 47 percent. So the same turbine built in Kansas would make about one-third more megawatt-hours in a year than the U.S. average.

That is one of the two big reasons that wind needs transmission, because the wind farms must be located where the wind is, often in remote places. The other is that the system benefits when wind farms are connected from many diverse locations. Geographic diversity reduces the variability in wind, and variability is wind’s great weakness.

The last of Clean Line’s transmission quartet was the Grain Belt Express, which was supposed to run from Dodge City, in southwest Kansas, on a route that ran north of Kansas City, Missouri, across Illinois south of Springfield, and to the border with Indiana, nearly 800 miles. The backers said it would cost $11 billion but would save $52 billion in energy costs over 15 years, and would provide 5,500 jobs directly, with more connected to the new generation that would be built to connect to the line, and in places that could use the electricity.

Clean Line, a company based in Houston, eventually bowed out of the business and sold its assets to others. One of those, the Grain Belt Express, seemed poised to succeed.Last November, Biden’s Energy Department conditionally approved a $4.9 billion loan guarantee for the first phase of the project. On July 23, the conditional loan guarantee was withdrawn. According to news reports, in a meeting with Republican Senator Josh Hawley of Missouri, President Trump—who doesn’t like wind, because he doesn’t like wind—picked up the phone to call Chris Wright, the Energy Secretary, and told him to cancel the loan. And the Missouri Attorney General says he is investigating whether the sponsors “relied on speculative and possibly fraudulent assumptions.”

The line could still be built, but it’s not looking good.

Transmission lines are usually fairly short, and Clean Line defied conventional thinking by proposing lines that crossed states and regions. The U.S. grid has hundreds of different owners, and is divided into multiple planning regions. A company like Clean Line that wants to build across state lines must be recognized as a utility in each state, and the public service commission in each state must recognize the project as being in the public interest. Only then can the company line up a right-of-way.

In pursuit of permission, Clean Line held dozens of community meetings on the route. But, there is a saying in the business: The public interest is different from the interested public.

The original proposal included a Plains & Eastern line, which would have carried 4,000 megawatts of power 720 miles from the Oklahoma panhandle to the Southeastern states; and a Centennial West line that would have carried 3,500 megawatts from New Mexico to California. The Rock Island Clean Line would have run 500 miles from the upper Midwest towards Chicago.

The problem for Clean Line and other developers is that they must win approval in each state, some of which may not benefit. And wherever they go, they are outsiders. There is no sense of national priorities in transmission.

“Silly Clean Line,” taunted one West Virginia opposition website. “It's not about whether the transmission line is transporting ‘clean’ or ‘dirty’ energy.”

"Wind energy must be transported to east coast states so that wind developers in Kansas, Oklahoma and Iowa can get rich? It’s not about farmers or landowners cashing in, it's all about private, for-profit corporations and their investors making a whole bunch of money off the backs of what they arrogantly consider to be uneducated rube farmers and environmentally conscious but sadly oblivious east coast consumers who are easily fooled by the green-washing Midwest ‘wind’ scheme.”

After years of work, the Grain Belt project was the last glimmer of hope for a turn to a technology that would establish superhighways for long-distance electricity transmission.

Subscribe now

Sub-Optimal Renewables

Building transmission is always difficult because people don’t like the towers and wire. DC transmission towers are smaller, and the towers are lower, but there is a particular difficulty for direct current. A DC line is a little like an interstate highway, and an AC line is like a city street. To add energy to a DC line or take energy from it requires building a large and expensive converter station, to change from DC to AC. For an AC line, all you need is a transformer to adjust the voltage. Hence in intermediate spots along the route, a DC transmission line provides no more benefit for a neighboring town than an interstate highway that passes through with no interchange. Politically, that’s a good way to foster opposition.

But DC has technical advantages. In a DC line, the electrons actually travel down the line, and back again. It’s the transmission system that was favored by Thomas Edison, but builders eventually decided that Alternating Current (AC), championed by Nikola Tesla, was more practical. In AC lines the electrons travel a very short distance, and alternate directions 120 times a second, on the familiar 60 cycle frequency. Household wiring, importantly, is AC.

High voltage DC lines deliver power with smaller line losses. DC lines can be buried more easily than AC lines, and if they are hung from towers, the towers can be smaller, reducing the visual impact. They can be used to deliver power from one grid to another when the grids are not synchronized, which is a major consideration at the borders of the three major alternating current grids in the United States, the Eastern Interconnection, the Western Interconnection, and Texas.

Even within those grids, connections are not nearly strong enough. The system, especially in the East, grew out of individual city grids, gradually knit together, for reasons of reliability and economy. Interconnecting the systems of multiple utilities allowed them to share capacity, a far more economical arrangement than having each system build enough to meet its own peak demand, while a neighbor’s system had idle generators. But that is different from the grid that is needed for pulling energy half way across the continent.

The contrast with the system for transporting natural gas is stunning. Gas lines are regulated by the Federal Energy Regulatory Commission, and are planned with a continental-scale market in mind.

The current electricity system requires frequent computer-run auctions to assure that at any given moment, the least expensive combination of generators is running. That often means shipping more megawatts more miles.

That, combined with construction of renewable energy farms distant from load centers, should be eliciting more construction of transmission lines. The Department of Energy’s 2024 National Transmission Planning Study calls for increasing transmission capacity by 2050 by a factor of 2.1 to 2.6, with a focus on “interregional,” long transmission lines. This implies construction of about 5,000 miles per year. Building more transmission could cut carbon dioxide emissions by 43 to 48 percent, the study found. But less than 900 miles of high-voltage transmission lines were finished in 2024.

For every dollar invested in transmission, $1.60 to $1.80 is saved, because plants that are cheap to run can operate more hours of the year, and expensive plants run less. The plants with high operating costs tend to be fossil-fired, which is why emissions also decline with more transmission.

There could be a silver lining of sorts; with less transmission, grid reliability will require more nuclear energy. It could mean that the only carbon-free way to serve the electric needs of big load centers is to build reactors near them, because continental-scale grid projects, or even inter-regional ones, are beyond our grasp. Being anti-transmission makes you, in effect, pro-nuclear. But most people don’t connect those dots.

The mismatch between public interest and interested public in the transmission world is the flip side of the nuclear case, where neighbors want new nuclear facilities of all kinds but politicians further away are skeptical. It mirrors the nuclear problem because transmission improvements are long-term solutions, and the market system is geared to quick fixes and short-term outcomes, not to putting large amounts of capital at risk for many years.

In both cases, the regulatory structure gets in the way of an energy system that is optimized for price, reliability, and cleanliness.

Building wind and sun in the best possible locations is going to become even more important as federal subsidies disappear. But instead of optimally sited renewables made possible by a rational grid, we’re more likely to get sub-optimal wind and solar combined with continued fossil-fuel generation. It means that the last few years of engineering advances in those fields are not realized in actual construction. It means that electricity costs more than it should, because choosing the combination of generation resources with the lowest operating cost is often impossible because of electricity traffic jams. It also means that progress on emissions reduction will be slower, even if we build more and more renewables.

Subscribe now

On June 22, the United States struck three Iranian nuclear sites, entering the Iran-Israel conflict that began nine days prior when Israel launched an attack against Iran’s nuclear infrastructure. While the U.S. entry into the campaign is political and its merits can be debated, understanding the technical realities of Iran’s nuclear infrastructure could make for a more intelligent conversation.

There is no debating that Iran has a nuclear weapons program. But, as of yet, it is not building a nuclear weapon. Iran has enriched uranium far beyond what is necessary for either civilian energy or medical use, but has stayed short of the threshold of a weapon, leaving public uncertainty.

And that uncertainty has produced some confusion within the Trump administration. Tulsi Gabbard, the Director of National Intelligence, testified before the Senate Intelligence Committee on March 25 that the intelligence community “continues to assess that Iran is not building a nuclear weapon,” and President Trump told reporters in June that “I don’t care what she said, I think they were very close to having one.”

But, these two positions don’t actually contradict each other. Iran has intentionally stayed in a gray area on the road to a nuclear weapon—it’s already completed the heavy lifting required, but has not begun the final steps. Whether the U.S. and Israeli strikes have seriously delayed an Iranian effort to complete a warhead is still in question, but we do know the technical reality of Iran’s infrastructural capabilities, giving us some clues as to what might come next.

Share

Iran’s Uranium Campaign

Atomic weapons are a technology of the 1940s. Today the hardest part isn’t the physics, it’s getting the right material.

In nature, uranium is more than 99 percent U-238, a stable isotope that does not easily split. Less than 1 percent is U-235, which is the kind that can sustain a nuclear chain reaction. That’s the type that the weapons builders care about.

To build a weapon, you need to “enrich” uranium to raise the proportion of U-235. The uranium used in power reactors, like the one at Bushehr in southern Iran, only needs to be enriched to about 3-5 percent U-235. That is known as low-enriched uranium, or LEU, and it cannot be used for a nuclear weapon. Weapons-grade uranium, by contrast, is usually enriched to around 90 percent U-235

The International Atomic Energy Agency reports that Iran has enriched uranium to 60 percent, far beyond what’s needed for civilian energy or medical uses. There is a small research reactor in Tehran, but that doesn’t need uranium enriched to high levels. High-enriched uranium can be used to make medical isotopes, but it is not essential for that purpose. Iran does not have any civilian hardware that needs uranium enriched that high. This is either a program to build a weapon or reduce the time needed to build one if the country later decides to do so, or an effort to enhance the country’s power and influence by reaching the edge of the nuclear club.

Enrichment is not the entire story. The uranium that comes out of a mine is a metal, and is typically converted into a compound called uranium hexafluoride to be enriched. Uranium hexafluoride becomes a gas at relatively low temperatures. That gas is spun through high-speed centrifuges to separate out more and more U-235. Iran’s enriched uranium is in this chemical form. To actually make a weapon, that enriched gas must be converted back into solid metal, specifically shaped and machined for detonation.

So, when Gabbard testified that the U.S. intelligence community didn’t see signs of Iran building a weapon, this may have been a simplified way of saying that Iran is not yet taking those steps: pushing the enrichment up to 90 percent and then converting the uranium hexafluoride back into a metal. It’s not just a question of how much enriched uranium Iran has, it’s what physical form it’s in, and what’s being done with it.

But uranium isn't the only path to a weapon. Some countries, like India and Israel, used plutonium instead, a different element which is created in uranium-powered reactors. Unlike uranium, plutonium doesn’t need to be enriched. Certain isotopes of plutonium, especially plutonium-239, split very easily and can sustain a fast chain reaction, making them ideal for weapons use. If a country has a reactor capable of producing plutonium and a facility to chemically extract it from spent fuel, it can bypass the need for enrichment beyond the low levels needed for reactor fuel. The drawback is that this chemical plant is expensive and obvious.

Iran’s Progress: The Heavy Lifting is Done

When people ask whether Iran is “close” to a weapon, they’re usually thinking about the end result: a warhead on a missile. While Iran has not done that yet, by already enriching uranium to 60 percent U-235, Iran’s progress to building a simple nuclear weapon is nearly complete. Enriching uranium to 60 percent may not sound close to the 90 percent typically used in weapons, but in terms of the physics and engineering effort required, the difference between natural uranium and 60 percent is far more significant than the final push to 90 percent. Natural uranium is one part U-235 in 141 parts U-238. Low-enriched uranium is roughly 7 parts in 141, and the enrichment level that the International Atomic Energy Agency says that Iran has reached is about 85 parts in 141. To get from there to 90 percent, or 127 parts in 141, is not nearly as much work as has already been done.

Iran has had about two decades, though, to take that step, and hasn’t done it. This is not for lack of centrifuge capacity. Instead it has gone right up to the line, essentially creating a bargaining chip, and positioning itself to make a nuclear weapon quickly, while still being able to say for almost all the development period that it didn’t have uranium enriched enough to make one. Call it “implausible deniability.”

And combined with its extreme rhetoric against Israel, plus support for three militant groups that have inflicted substantial damage on Israel (Hezbollah, Hamas and the Houthis), Israel and the United States have taken the threat seriously. The Prime Minister of Israel, Benjamin Netanyahu, has been warning for decades that Iran was close, and that, too, does not contradict the facts as we know them at this time.

The International Atomic Energy Agency said that Iran had about 400 kilos of high-enriched uranium hexafluoride, and since the U.S. attack, there has been a lot of speculation about whether that material could have been scattered into the environment. The International Atomic Energy Agency says that there was no detected increase in radiation. There is also the question of whether the material was moved in advance of the bombers. But this may not be a mystery to the military.

Uranium hexafluoride is kept in heavy steel cylinders that can be moved by truck, which would be visible to satellites and aerial surveillance. If Iran moved the cylinders, the Pentagon may know where at least some of them are. Israel may too. So more bombing is a possibility.

An open question is whether Iran has spare centrifuges in some other location. Those might be hard to locate at first, but historically, Iran has had trouble keeping secrets like that.

Share

How Does Iran Get from Here to a Nuclear Weapon?

If Iran resumes enrichment and gets to the 90 percent threshold, the uranium hexafluoride would still have to be “de-converted” back to metal. Those steps could be done in days, depending on the capacity of the chemical plant and assuming that neither the US Air Force or the Israel Defense Forces haven’t taken out all the deconversion plants.

After that comes making a bomb. The secret of nuclear weapons is that there is no secret. The design isn’t difficult; in 1976, a junior at Princeton designed one for his physics homework. And that was for a plutonium bomb; uranium is easier.

The first nuclear weapon used in combat, the Hiroshima bomb, was a simple gun firing a wedge of uranium into a uranium target, into which it fit like a 3-D puzzle. When it hit the target, a critical mass—the minimum amount necessary to sustain a chain reaction—was formed. It was not tested beforehand, because the United States did not have enough U-235 for a test, and because the designers were certain it would work.

The Hiroshima bomb may have been fairly simple, but it was of a size and shape that required delivery by an airplane, letting gravity do its job. Japan at the time had limited air defenses, but Israel is more capable. Using a plane to bomb Israel is something Iran probably can’t do at the moment, because Israel’s air defenses are strong. Iran does, however, make sophisticated missiles, some of which evade Israeli defenses.

But building a uranium warhead that will fit on a missile is much harder. A warhead of the size and shape that would fit atop a missile needs sophisticated electronic switches to control the conventional explosives that compress the uranium into a critical mass. Various countries have those, including North Korea and probably Pakistan. Whether Iran could import them is another question.

And U.S. nuclear weapons have some highly sophisticated enhancements that get more yield out of the uranium. But for Iran, it’s probably not necessary to get the biggest possible bang out of a quantity of uranium. A simple multi-kiloton open-air test in the desert would make the point, no matter what the device's efficiency. Or they may prefer nuclear ambiguity.

Demonstration explosions are the way that India introduced itself into the “nuclear club,” with a “peaceful nuclear device.” Pakistan followed immediately with a series of tests. It’s also the way that the United States introduced the hydrogen bomb.

Subscribe now

What Does This Mean for Nuclear Energy?

Understanding the Iranian path towards a nuclear weapon should help us re-calibrate the nuclear proliferation calculation and its relevance to nuclear energy. Experience is showing us that the preferred path to weapons-grade nuclear material is centrifuges (Pakistan, Iran) or research reactors to make plutonium (India, and Israel).

India and Israel did not use power reactors to make plutonium. Iran has one power reactor, but Russia, which supplies the fuel, takes it back after use. Real-world experience is that power reactors do not seem to be attractive choices to would-be nuclear states.

A country that has its own enrichment capability can make weapons-grade uranium. But there is an irony here; if it has a civilian nuclear program, then developing weapons could put the civilian energy program at risk of sanctions. In that sense, developing civilian reactors may be a deterrent to weapons, not an adjunct. Nicholas L. Miller, a nonproliferation researcher and associate professor of government at Dartmouth College, made the case in 2017 that “Although such programs increase the technical capacity of a state to build nuclear weapons, they have important countervailing political effects that limit the odds of proliferation. Specifically, nuclear energy programs increase the likelihood that parallel nuclear weapons programs will be detected and face counterproliferation pressures; they also increase the costliness of nonproliferation sanctions.”

The distinctions are important for nuclear energy production in the United States and around the world, because they concern the fuel cycle for civilian reactors.

For example, one of President Trump’s recent executive orders seeks to encourage reprocessing of spent nuclear fuel, which means using a chemical plant to extract the plutonium produced in ordinary reactor operations, plus the uranium that was not consumed in the reactor.Presidents Ford and Carter banned that technology because they thought it would set a bad example for other countries, which might see that route to a nuclear weapon.

But reprocessing makes good use of components in spent nuclear fuel, and makes the remainder easier to dispose of. In the real world, as Iran shows, it’s not the method of choice for countries that want to enter the nuclear club. The easier route is uranium centrifuges or research reactors.

It follows that exports of power reactors do not appear to be relevant to proliferation. A policy of “energy dominance” that includes robust American exports does not provide a route to nuclear weapons for our friends—or even our friends who later become enemies, a category that includes Iran, which got a research reactor under President Eisenhower’s Atoms for Peace program.

In the short term, whether Iran has civilian nuclear power has little to do with the future of its weapons program. Iran’s choice to remain ambiguous when it comes to nuclear weapons capabilities likely produced some geopolitical benefits, until it prompted a strong military rebuttal from Israel and the United States. Whether or not Iran will choose to continue a strategy of nuclear ambiguity—one in which it remains close to a nuclear weapon, but far enough away to leave space for denials, however implausible—is the question of the day. But its relevance to nuclear energy is limited.

Nuclear construction is still too expensive. That central problem confronting the nuclear renaissance has come into sharp focus with recent announcements about projected costs for new plants.

Two recent developments involving concrete, though, might show the way to lower costs. But the industry needs to be more aggressive, and the Nuclear Regulatory Commission will need to re-think its rules to allow developers to skip the use of high-grade concrete in components that are needed for traditional, large light-water reactors, but are less important to some newer designs.

Using less expensive concrete is a pressing problem, because it is one of the most used materials in a nuclear plant. The recent financial data isn’t promising. Korea Hydro & Nuclear Power recently won a contract to build two 1,055 megawatt pressurized water reactors in the Czech Republic, for a little over $8,500 a kilowatt of capacity, or roughly $18 billion total. That’s expensive but seems like a bargain compared to Ontario Power Generation’s budget for four BWRX small modular reactors, at $12,500 per kilowatt. The SMR design was supposed to be less expensive to build than the big reactors. And those are the prices estimated before construction, not including any potential delays and cost overruns.

Finding ways to lower the cost of construction is the name of the game. Nuclear-grade concrete, being one of the most expensive inputs, is a prime target for innovation and improvement.

Subscribe now

A 50 percent premium

Concrete is pricey. A 2022 analysis estimated the cost of nuclear concrete at $527 per cubic meter, compared to $352 per cubic meter for non-nuclear concrete, a premium of nearly 50 percent. The price of each has gone up since then, but the ratio may be constant. A 2007 report by the Economic Modeling Working Group, for example, found that nuclear concrete cost $421 a cubic meter, vs. $281 a cubic meter for non-nuclear concrete.

For concrete to be nuclear-grade the mixture of water and cement must be more precise so it will not crack as it sets. That means more care in mixing. Too much water in the mix will leave voids in the finished product. Nuclear-grade concrete also has a thick network of steel reinforcing bars, far denser that is used in bridges and other civil structures. That re-bar is tedious and expensive to install and inspect.

The Department of Energy’s Idaho National Laboratory reviewed nuclear construction costs in 2017, and found that as measured by weight, three-quarters of a large pressurized water reactor is concrete. The concrete pours for nuclear concrete take a lot of time, about twice as long for foundations, and 50 percent longer for superstructures, compared to ordinary concrete. Cutting the installation time in half would reduce the cost by 28 percent, the researchers found. Construction experience and engineering analysis suggests that this is possible with better project management and better construction equipment. Strict quality-control standards represent another 23 percent of the cost of the concrete.

But a simpler strategy than reducing the cost of nuclear-grade concrete is to just use less of it. Two nuclear plants now under construction are trying this route.

One is the GE-Hitachi BWRX small modular reactor, with a basemat and reactor building that will be built from modules made of steel and concrete.

Researchers recently tested steel and concrete composite blocks at Purdue University to simulate loading conditions that they would see in an earthquake, and found that their strength exceeded expectations. Two steel plates are connected with adjoining steel plates with holes to allow insertion of concrete. The modules can be factory-fabricated and assembled on site, and then filled, allowing for quicker construction. With the composite-blocks, the re-bar typical in nuclear-grade concrete is no longer needed. GE-Hitachi hopes to cut construction costs by 10 percent with the technique.

Share

Minimizing the Critical Area

The Natrium project, in Kemmerer, Wyoming, is taking a different approach. It separates the reactor from the “energy island,” the part of the plant that produces electricity by using steam to spin a turbine and then a generator. In a conventional reactor, these components are adjacent, but Natrium, designed by TerraPower, uses the reactor to heat up a big external tank filled with molten salt. The salt’s main function is to be a battery, storing heat energy by varying its temperature. Operators can draw heat off the salt to boil water and make electricity, but at variable rates so they can time their production to match price changes on the grid.

A side benefit, though, is that the part of the plant that has to meet nuclear-grade standards is substantially smaller. Concrete for the area away from the reactor can be fabricated to normal industrial standards.

The NRC has recently endorsed the concept of a smaller “nuclear island,” separate from the “energy island” where electricity is made, although the Kemmerer plant does not yet have a construction permit for the nuclear portion.

In a letter sent to the company on May 7 of this year, the NRC gave TerraPower an exemption from its rules on the electricity-generating half of the plant. Normally the agency pays close attention to the turbine and generator because a problem there can trigger a reactor shut-down, but in the Natrium case, the reactor is designed to be immune to those upsets.

There are other opportunities to reduce the use of nuclear-grade concrete. One is by taking full advantage of “pebble” type fuel to redefine what constitutes a containment, and thus reduce the need for nuclear-grade concrete.

Today’s reactors are fueled by ceramic pellets wrapped in thin metal tubes. The entire reactor lives within the containment, which is a very strong building. If cooling stops and the pebbles overheat, they can leak radioactive materials into the cooling water and if that water boils it can raise the pressure in the primary cooling system, opening a valve and sending radioactive steam into the containment.

But the pebbles, more properly called Tristructural Isotropic, or TRISO, have particles of fuel wrapped in concentric layers of heat-resistant materials. Thus, the containment doesn’t have to be a concrete dome building surrounding the reactor; the containment can be the fuel itself. With this technology, there is one less significant reason to hermetically seal the building that houses the reactor.

For example, in Kairos Power’s SMR design the fuel floats in a salt mixture that can withstand tremendous heat without boiling away. If the fuel fails, the salt will absorb the radionuclides. This means the whole operation is low pressure, despite being very high temperature, making containment a lot less of an issue.

Even new pressurized water reactors could get by with less concrete in their containments. The requirement for nuclear grade concrete is to assure that a crack-free structure would not allow radioactive gases or particles to escape in an accident. In practice, reactor containments are lined with steel that can also do that job.

Share

Flattening the Nuclear Construction Cost Curve

While the price of nuclear construction needs to come down, it does not have to reach the levels of solar or wind installations. A kilowatt of nuclear capacity, after all, is worth more than a kilowatt of renewable capacity.

The standard way to measure construction cost is the price per kilowatt of capacity. Buyers can afford to spend more on a nuclear plant per kilowatt of capacity than on a solar or wind plant, because the nuclear plant will run more than 90 percent of the hours in a year. But the capacity factors for solar and wind (the amount of energy they will produce in a year, compared to what would result from 24/7/365 operation) is far lower; 30 to 40 percent for a good wind site, and 25 percent or less for solar. And the reactor will be designed to run for 60 years or more, while wind machines in the United States are often retired after 10 years, and solar after 20 at most.

The price requirement for new nuclear construction depends on three other factors that are more difficult to nail down. One is the future cost of natural gas, which has become the dominant competitor in the United States. The fuel cost to run a gas-fired plant has varied by a factor of three in recent years. The future price will depend in part on how much natural gas the United States exports, and thus how much the world price tends to pull the domestic price upward.

A second factor is the value that a utility puts on a zero-carbon generator. That was a federal priority in the last administration but isn’t now. It is still a priority in many states.

And the third is the confidence that the buyer can put in the estimated price and schedule. No matter what the estimate for a new plant, a company shopping for new capacity won’t be confident until several units have been built.

Flattening the cost curve of nuclear construction will require even more innovations in design, materials, and management. Developments in nuclear concrete are just the beginning.

There are a number of signs pointing to a nuclear renaissance. New reactor designs, growing global interest in firm and clean electricity generation, and increased investment in reactor developers and adjacent industries all signal a potential bull market for new nuclear. But, common misconceptions about the nuclear industry remain pervasive.

Last September we listed seven of those misconceptions; here are seven more.

Share

The key to success for advanced reactors is standardizing a design and building it over and over.

This sounds good if you’ve never built anything, but reality is trickier. No mass-produced product—not airplanes, not pick-up trucks, not laptops—comes out exactly right the first time. Engineering doesn’t work that way. The trick is to build it once, and make some tweaks to optimize subsequent models, with the Nuclear Regulatory Commission (NRC) recognizing the changes as refinements, and not entirely new concepts, so that the improvements are regulated as amendments, not wholesale do-overs.

The part that must be standardized is the core design. This doesn’t mean reactor core. “Core design” is an engineering term that enables innovation while maintaining the major characteristics. But it isn’t a “standard design,” which, for regulators, means identical design.

A chairman of the Nuclear Regulatory Commission in the 1990s, Ivan Selin, once observed that France, which has a successful and largely uncontroversial nuclear program, has one kind of reactor and 365 kinds of cheese; the United States, he said, has the opposite. There is an element of truth to this; our reactors are bespoke. A control room operator is licensed for two or three reactors at most, and analyzing and maintaining so many unique designs is inefficient. When one reactor operator finds a glitch in a component, reactor owners can’t even tell initially if they have that component in their plants.

But the cure isn’t identical units. It is allowing small changes and recognizing the difference between big differences and small ones. The system has to allow space for learning by doing, while still achieving some standardization.

But the nuclear industry has gotten itself into an odd position with the NRC, because of the way the licensing system has evolved. At the beginning of the nuclear age, a company that wanted to build a reactor would apply for a permit to start moving dirt, and then for a construction permit. As the plant was being built, the company would apply for an operating license, which was based on the design—a design that was being completed as the construction work progressed. The idea was to shorten the total time it took to complete the project, by letting design and construction overlap.

A parallel in the consumer world might be pouring a foundation for a new house before deciding what color to paint the interior walls, or perhaps some more serious details, like whether a bathroom door should lead to a bedroom or a hallway. If you had to take out a construction loan and pay interest on it until the job was complete, why waste time doing everything in sequence, instead of in parallel?

This turned out to cause problems. Without a finished design at the outset, workers would pour concrete or install pipes for various systems that later turned out to conflict with each other. The result was “re-work,” or tearing out nuclear-grade construction to do something different. At times, haste made waste.

So the industry and the NRC came up with a different system, design it first and then get the design licensed. Then build it. Conceptually, it was an attractive approach. It created the idea that the builders knew from the outset exactly what they would be doing, and that this would make it easier to predict the cost and schedule.

But doing something the first time turns out to be very hard. And if a design needs to be changed to move a piece of steel a few inches one way or the other, that means developers need to get a license amendment, which is a cumbersome change. This was a key problem that delayed the construction of Vogtle units 3 and 4, two AP1000 reactors, near Augusta, Georgia. Some changes were as simple as moving some steel studs inside a concrete wall a few inches, to make space for a door. The fact that this required something as serious as a license amendment has discouraged the industry from using the NRC’s newer licensing system, called Part 52. Instead they have gone back to the older system, because the designers know there will be changes as they build, and they want fewer regulatory slowdowns.

“Building identical copies” is a talking point, not an engineering strategy. Developers know better. The next AP1000 will incorporate the changes from Vogtle, and thus it will be better. Continued optimization and learning yields significant cost savings and performance enhancements.

The NRC needs to embrace this reality or it will be a barrier to innovation.

After we get the regulatory problems out of the way, we’re well set up to build lots more reactors.

Optimizing the regulatory system is necessary but not sufficient. There are other issues, like the supply chain for hardware and people. After years of atrophy, our ability to turn out big structures like reactor vessels, and even smaller ones like specialized valves and pumps, is limited. So is the supply of welders, pipefitters, quality control specialists, and instrumentation and control technicians who can build and operate new reactors. We also have a bottleneck in uranium fuel. We have a whole supply chain to rebuild, but inducing private companies to invest in that means convincing them that there is going to be a sufficient market.

Consider a parallel in the aviation industry. Lots of companies will compete to produce parts for an airplane like the Boeing 737, because there are thousands in service, and even for a specialized new variant, there are likely to be hundreds built. But only 20 Concorde supersonic jet transports were built, of which two were pre-production prototypes, two were development aircraft, and only 14 were in commercial service. Private companies don’t like tooling up for production runs that short, or investing in nuclear-specific Quality Assurance programs. Some reactors today are running on parts that are no longer routinely manufactured, and a whole specialty has developed in figuring out what spares to keep in inventory, to avoid having a reactor shut down while an obscure part is special-ordered.

There is also a craft worker problem. In 2013, when work started on Vogtle 3 and 4, in Georgia, and on V.C. Summer 2 and 3, in South Carolina, the Energy Department was also trying to build a nuclear fuel facility at the Savannah River Site, also in South Carolina. The combination created a shortage of nuclear-qualified welders, pipefitters and electricians. The shortage was relieved only when work stopped on the South Carolina projects.

Share

The industry would be better off without the Nuclear Regulatory Commission.

Working with the NRC is frustrating, slow and expensive, but it has value. To prosper, the industry needs the NRC, but needs it to work better.

The NRC is supposed to set requirements and let applicants figure out how to meet them; instead, it is often prescriptive, telling them exactly what to do. Rather than, “Prepare a nutritious dinner for 10 guests,” the NRC tells developers what ingredients to use, no matter what is on the menu. And the NRC potentially takes years to do it, all the while charging the applicant $317 per hour that NRC staff spends on evaluating the menu. This is also known as the “bring me a rock” problem. The saying goes, bring me a rock and I will tell you if it is the correct rock.

The industry complains about inefficiency at the Nuclear Regulatory Commission, but also praises it. The Nuclear Energy Institute, the industry’s biggest trade association, calls the NRC “a strong and effective regulator.” The trade association lauds the Commissions for having “resident inspectors” at each plant who will order the plant shut down if they believe it is unsafe. Aggressive, intrusive regulation may be a burden, the industry is saying, but it is also a benefit.

A safety regulator is necessary to protect the public and the environment. In the absence of a regulator, the industry wouldn't even know what standards would be acceptable for them to self-implement.

That means we need a better NRC, not no NRC. Having a functioning regulator is essential. For one thing, nobody wants to live next to a nuclear plant that the NRC doesn’t say is safe. For another, nobody abroad wants to buy such a plant. And the alternatives could be worse; several developers and state attorney generals have brought a suit against the NRC, claiming that the Atomic Energy Act doesn’t give the NRC jurisdiction over very small reactors, and they should be regulated by the states. But the states aren’t set up to do that kind of work, and more lawsuits and delays seem inevitable if they take over, not to mention the fragmentation of a national market, as differing states impose differing standards.

Instead, the NRC should use regulations appropriate to the level of risk. That means alignment with Congressionally set risk thresholds, and scaling regulatory applicability to what is necessary to provide oversight. Many new designs have a lower risk than already licensed research reactors, which the statute requires the Commission to impose only the “minimum amount of regulation of the licensee” necessary. This will require NRC reform.

The NRC is what it is; it can’t improve

The NRC has recently completed some licensing actions ahead of schedule, and has shown increased concern about its own effectiveness. It’s been trying to reform for 30 years, with mixed success, but that doesn’t mean it can’t happen. What the NRC needs most to do is to get set up for evaluating and approving reactors that are highly similar, without starting from scratch each time. Most of the new designs are in the range of one third as large—in power capacity—as the reactors now running. That means that if the United States wants to double or triple its nuclear capacity, the NRC will be licensing dozens or scores of reactors every year. It can’t hope to do that taking them on a case-by-case basis.

The staff recently took a step towards rationalizing the licensing of projects that are highly similar. It said it would consider the license application of a company with a new technology for cleaning up mine waste. This was significant because the thing being licensed could be the technology, not the cleanup. One system, once licensed, could be used on hundreds of mines, vastly reducing the bureaucratic overhead involved in one category of nuclear projects.

Decisions like that will be essential or the wait to get a license will preclude a nuclear renaissance.

And the five-member commission has sometimes embraced key principles needed to re-make the agency, like telling the staff to form rules that set requirements without specifying the particular way that they should be met—known as performance-based regulation. Getting the staff to actually change its culture to work that way is a separate task. The NRC must take implementation of the ADVANCE Act in text and spirit to reform, but it doesn’t have another 30 years to do it if the U.S. is going to scale up nuclear energy.

Among other steps, the NRC should stop focusing on the most remote and improbable risks, and should integrate its risks assessments, to recognize that failure to permit deployment of a new, safe technology means continued use of older technologies that threaten human health and safety, like fossil-fueled power plants.

But it has made some progress. The NRC has modernized its emergency preparedness rules to take account of the characteristics of advanced reactors, permitting smaller emergency planning zones, sometimes ending at the plant fence.

To implement the ADVANCE Act, the Commission updated the NRC’s mission statement to fully embrace what the statutory mandate has always been—enabling the safe use of civilian nuclear energy for the benefit of society and the environment. The Commission must now commit to implementing that mission and make decisions, including efforts to modernize, that do not unnecessarily limit the benefits of nuclear energy. With that framing, a lot of changes become self-evident and justified.

Share

The Trump Administration’s shake up of government bureaucracy is good for the nuclear industry.

Some of it might be, but implementing government programs requires a dedicated, experienced civil service staff. And wholesale cuts to the Energy Department’s Loan Programs Office might not leave anybody around to sign the checks that will help get new nuclear projects built. Firings at the Tennessee Valley Authority board of directors have left the agency without a quorum, at least until President Trump can appoint replacements and the U.S. Senate can confirm them.

The NRC has been trying to hire staff to meet the expected wave of reactor applications. Many of the staff that are currently leaving are those most needed, the ones that will help the agency to innovate or fill critical roles. The impulse to cut staff, as DOGE and Trump have demonstrated across federal agencies, would outright damage the NRC’s ability to function well.

Orders requiring agencies to “sunset” outmoded regulations won’t help the nuclear scale up. Many of those rules pertain to obsolete reactor designs that nobody will ever build again. Getting rid of them will reduce the number of exemptions needed, but will mostly become a distraction for a very busy NRC staff that can’t keep up with more urgent work, like finishing rulemaking that solves the problem.

Shaking up the system is a problem, because uncertainty is a direct barrier in some cases. It creates a harder-to-quantify hesitancy to move forward. A more strategic approach is needed.

We can’t have a “Nuclear Renaissance” until we know what to do with spent fuel

We do know what to do with spent fuel.

Spent fuel, in the form of ceramic pellets wrapped in metal tubes, is stored in deep pools of ultra-clean water for a few years until the heat generation subsides. Then it is moved into steel capsules filled with inert gas, sealed tightly, and encased in concrete. These dry casks have no moving parts, and will last for many decades, if not centuries. In the meantime, a cluster of casks sufficient to hold decades of fuel from a giant electricity generator has a footprint the size of a tennis court or two. Unlike fossil generators, reactors are not putting waste into the air or the water.

Eventually, either the casks will be buried deep underground, where they will be isolated until the radiation levels have fallen to a level of radioactivity comparable to the natural uranium that the fuel came from, or the casks will go through a processing plant, where unused materials will be removed for use in making new fuel. Then the remainder will go underground.

At the moment, the balance of uranium supply and demand produces a fuel price that does not economically justify recovering useful materials from spent fuel, but because of the political impasse over finding a burial site, we have the luxury of testing the economic question in future decades.

Meanwhile, two factors give hope for a consensus on a site. One is the very substantial amount of money in a government account for waste disposal, money that could fund economic development and other benefits at a host site. The other is being able to follow the example of others. Three countries are moving rapidly towards burial: Finland, Sweden and Canada. All are using geologic structures similar to those present in the United States.

Nuclear is too slow and too expensive to make a contribution to stabilizing the climate

In late April, China approved construction of five twin-unit nuclear plants, or ten reactors, for $27.4 billion. The price, for a generator that will run more than 90 percent of the hours in a year, and can be located near load, is quite competitive. It’s worth noting that China could simply choose to only add more solar, since it already dominates global solar panel production. But it isn’t choosing to do that, because it recognizes the benefits of nuclear energy, including a lower-cost energy system.

And in China, nuclear certainly is not too slow. Nearly every Chinese reactor that has entered service in the last 15 years has been built in seven years or less, which is reasonable for an asset that has a lifetime in the range of sixty to eighty years. We could build them that fast, too, if we got the cobwebs out of the production system.

We must recognize that Federal priorities are changing, shifting away from control of climate-changing emissions, but there are excellent reasons in addition to climate to displace fossil fuels with nuclear. Advanced reactor deployment will cut emissions of soot, which research by the Clean Air Task Force finds may be causing 100,000 to 200,000 excess deaths in the United States every year. Nuclear deployment will also cut smog, a threat to human health, and acid rain, which damages forests and lakes. This is in addition to substantial national economic benefits.

China achieved cost reductions both on solar and nuclear energy by strategically building a supply chain to reach scale, not one-off demonstrations. In this country, the Advanced Reactor Demonstration Program is a good start for one-off demonstrations, but it’s not sufficient to build an industry; that will take gearing up a supply chain, a workforce and a regulatory system, to support a stream of orders. Of course, even with a strong strategy in place, there is no substitute for actually building plants.

By Matthew L. Wald and Deric Tilson

Donald Trump says coal is clean and beautiful, and on April 8 he signed an executive order to encourage its use. Whether or not you share his taste in fuels, the order appears to overlook some critical details of how the power system works, details that nuclear advocates should bear in mind when they talk about what will constitute a “renaissance.” The lesson is that government policy can help identify political priorities, but when policy conflicts with the market, the market will win.

Coal has some attractive attributes. It’s plentiful, here and abroad, its price is fairly stable, and a plant manager can look out the office window and see where the next few months of fuel are coming from. In contrast, natural gas is delivered just in time, wind and sun come when they want to, and even water for hydroelectric dams is variable.

But, the amount of coal we burn does not depend on how much we can mine or how many power plants we have to burn it in. It depends mostly on its cost per megawatt-hour of electricity, an area where it has been overtaken by natural gas. There have been substantial improvements in coal mining, and there are potential improvements in coal plant thermal efficiency, but they cannot match natural gas.

Nuclear energy is feeding into the same market. And depending on whether the market is traditionally regulated, with ratepayers compensating utilities for the cost of utility investments, or is “deregulated,” and converted to an auction system, success will come when a company that wants to sell electricity in a regulated market thinks that regulators will see nuclear as a reasonable choice for consumers, or, in a market system, when the price of production from the reactor is going to be lower than the average price on the wholesale market. The people who decide what kind of plant to build are driven by market signals.

It Won’t Work for Oil, Either

A corollary applies to the White House’s ambitions for oil. Just as leasing federal land to coal mines won't mean more coal production, opening up land to oil drilling won't increase oil production. That will depend on the price per barrel and the expectation of future prices compared to the price of production. Opening land could lower the cost of producing coal or oil. Firms must determine whether the projected decrease is enough to pursue investments and how those costs will measure against future fuel prices. If opening the land has any effect, it would not start for years to come.

There are too many moving parts in such a system; the government cannot possibly align all of the markets and incentives. This could only work if the government were both the buyer and seller of the entire coal supply chain from mines to purification to power generation to electricity consumption. Price signals matter. Price signals tell you the status of supply and demand without much delay or high costs of communication. Right now, price signals are telling us that coal is not valuable relative to other means of energy production.

In that context, Trump’s executive orders are a bit like building new runways that the airports will need on the day that pigs fly. One order seeks to open up more federal land for coal mining, but mines cost tens of millions or hundreds of millions of dollars to build, investments that would be stranded if the market for coal continues to decline. Another allows coal plants to delay retirement if their energy is needed, and it seems possible that a little bit of extra coal will be burned in the immediate future to meet an anticipated rise in demand. But there are cheaper ways to meet demand.

The big hurdle for coal, and for nuclear, is changes in natural gas. Fracking and combined-cycle gas plants have attacked the cost problem from both ends. Oddly, the growth of wind and solar on the grid has helped natural gas as well. Here’s how it works:

Fracking is a decades-old technology used in oil drilling, fracturing the rock to let the hydrocarbons flow to the well. But the natural gas industry began early in this century, using it on a type of rock that holds natural gas previously considered unrecoverable. And the industry has learned how to drill holes that twist from vertical to horizontal, to reach underground structures that hold gas. Annual gas production has more than doubled since then, to more than 40 trillion cubic feet, and increasing supplies have caused prices to collapse.

And a unit of natural gas goes farther than it used to at the power plant. Gas plants used to make electricity by using gas to boil water, and then running the steam through a turbine to create mechanical power; a generator turned that into electricity.

But now, utilities use a machine that resembles a jet engine, bolted to the ground, to create mechanical power. Then the exhaust from the jet is used to boil water, to make steam to make mechanical power. The older method made electricity with between 35 percent and 42 percent of the energy value of the natural gas. The newer method makes electricity with an energy value of up to 64 percent of the natural gas input. In other words, today, a utility can make about 50 percent more electricity with a unit of natural gas than it could with the kind of plant that was common in the 1970s and 80s.

Solar and wind helped natural gas because those generators are nimble. A grid with a lot of solar and wind, which can arrive and depart quickly, needs some dispatchable generation to compensate. Of the other assets on the grid, nuclear and coal move slowly. Batteries move faster, but their capacity is quite small relative to swings in production. Hydro, when available, can also help, but utilities have countered the volatility of wind and solar by relying on natural gas generators.

Don’t Argue with a Computer

But the White House does not appear to have accounted for how these factors translate into energy generation. A market system determines which power plants run at any given moment and how many hours a year each power plant will run. Working backward from that determination, utilities decide what to build and what to retire. Calling coal “beautiful” doesn’t change that. Neither does being attracted to nuclear because it is reliable, an aid to national security, or energy-dense.

In most of the United States, a computer decides, every hour or at shorter intervals, how much electricity is likely to be needed. And it ranks all the available generators, according to the bids they have submitted, from low to high. If a region needs 5,000 megawatts at 2 pm, the computer will go down the list until it has secured the five thousandth megawatt, and order those plants to run. The last megawatt ordered is the most expensive, and whatever that price is, that’s what all the generators are paid.

If that last increment costs $40 but the coal plant’s cost to operate is $45, the coal plant will not be ordered to run.

In fact, because the plant has limited operating flexibility, it may decide to bid in at $39 so that it is chosen, and to operate at a loss during some hours, because on average it will make money. Or the operators may shut the plant during mild weather, and only start it up during the cold or hot months, when demand is higher and they can make money. Letting the industry dig mines on additional federal land won’t change their calculus.

If plants are needed in continued operation, because of their location in a load pocket or because of the unreliability of natural gas supplies, the market will pay them a premium so that they remain profitable. That isn’t part of executive orders either.

In non-market areas, which are traditionally regulated, utilities will still dispatch their generators in price order, so that at any moment, the least costly combination of plants is serving load. If they see that coal plants are running very seldom and that their labor and maintenance costs are being spread over a very small number of hours of operation, they will apply to the public service commission to retire those plants.

How Nuclear Fits In

Most power reactors now in service have very limited flexibility. They will bid in at a low cost and, at some point, accept revenues that are below their operating cost, hoping to make money on average. When the price of natural gas was exceptionally low, several were retired, including Palisades in Michigan, Duane Arnold in Iowa, and Three Mile Island 1 in Pennsylvania, all of which the operators are now seeking to re-open.

For new nuclear, would-be builders face a complicated decision. If they think it will take ten years to get a plant online, the question is what the cost per megawatt-hour will be in, say, 2035, which is largely a function of construction cost, and what the average wholesale price of a megawatt-hour will be at that point, and in the early years of the plant’s lifetime. Market price will be important for the whole lifetime of the plant, which could be 60 or 80 years, but the first few years are far more important from a financial viewpoint.

For a plant in a traditionally regulated region, a builder has to convince the regulators that when the plant comes online, it is likely to be of more benefit to the consumers than an alternative would have been. The alternative could be gas, coal whose retirement was delayed, hydro, solar, and wind plus batteries, or money spent to reduce demand by improving end-use efficiency.

For a “merchant” plant, built in a market region, the builder needs to convince investors that the plant’s costs will be lower than market prices.

Clean energy tax credits, investment tax credits, and state mandates to reduce carbon emissions can all play a role. So can rationalizing regulation at the Nuclear Regulatory Commission, because it would reduce costs.

But the renaissance will dawn when a rational actor will calculate that a reactor—large, small or in between—is a better bet than any of the alternatives. That will move the technology beyond demonstration projects and back into the commercial mainstream.

Subscribe now

By Matthew L. Wald

Say “nuclear renaissance” and what comes to mind is new, advanced reactors, but radical innovation in the fuel supply chain would be crucial to a world with more nuclear power.

The Department of Energy, with a mandate to “re-shore” reactor fuel production, is facing a decision about the vast store of depleted uranium, left over from decades of low-efficiency enrichment work. If the DOE is bold, it could open the door to a third-generation enrichment technology. The moment is ripe as Western companies and governments seek to replace Russia as a supplier of enriched uranium.

That new technology is laser enrichment, which could be used on hundreds of thousands of tons of depleted uranium to scavenge the more fissile uranium isotope, U-235, left behind in the original enrichment process. Deploying laser enrichment technology could reduce the waste disposal burden on the Energy Department and expand critical enrichment capacity without further stressing an already bottlenecked supply of uranium hexafluoride, the chemical form of uranium used in the enrichment process.

Despite extensive work since the 1980s, laser separation remains commercially unproven. And the price for a first-of-a-kind project is not clear. Any project would be high risk—far too high for most private sector investors. But, by an accident of history, it is a risk that the United States is uniquely in position to take.

Share

Making Up for Past Wasteful Behavior

The immediate target for laser enrichment is an unusual stockpile of depleted uranium started by the Manhattan Project, continued by the Atomic Energy Commission, and now under the auspices of the DOE. This is material in which the content of uranium 235 has been reduced below the natural level of .7 percent. But the Energy Department’s stockpile has more U-235 left in it than most other stockpiles, because most of that uranium was enriched in an era before centrifuge enrichment using a more-expensive process called gaseous diffusion. In the 60s through the 90s, it was easier for the department to get the necessary quantities of enriched material by using a lot of uranium, and not being particularly thorough in plucking out the U-235. In addition, before the fall of the Soviet Union, the global market for enrichment was strong, so there was pressure to produce as much enriched material as possible, another reason that the government wasn’t very thorough. Some of the uranium “tails” have a uranium content in the range of .3 to .4 percent, meaning that the enrichment process captured only about half the available U-235.

Russian and European companies did much of their enrichment work later, in a period when demand for enrichment was weaker. Thus they had spare enrichment capacity, and nothing else to use it for, because the centrifuges cannot be turned off once they are running. So Russia and Europe ran the uranium hexafluoride through their centrifuges more thoroughly, processing the same uranium through more cycles, a technique called “underfeeding,” and produced tails at .2 percent U-235 or lower, which is less attractive for re-enrichment.

But as much as the change from gaseous diffusion to centrifuges was a technological leap, another may be coming.

Among the six companies approved by the DOE last October in an ”umbrella contract” as eligible to supply enrichment to the government was GLE, which uses lasers. It wants to start work on Government stockpiles at a former gaseous diffusion plant in Paducah, Kentucky. The material there is already compounded with fluorine, in a form called uranium hexafluoride, which is what is needed for processing with centrifuges or lasers. Eventually, the material will have to be “de-converted,” with the fluorine separated and sold for re-use. The uranium will be combined with oxygen, becoming a form of chemically inert rust. And then it will be buried, probably in Texas.

Processing it with lasers would modestly reduce the volume that must be de-converted and then buried.

“We are waiting for the DOE to issue task orders that have meaningful funding behind them,” said Nima Ashkeboussi, Vice President for Government Relations and Communications at GLE. Congress has allocated $2.7 billion for the DOE to establish an enriched uranium reserve to induce producers to make the material so that it will be ready if and when advanced reactors demand it. Most of those reactors want fuel enriched to nearly 20 percent.

The world of uranium enrichment is small, and the financial details are often opaque. GLE has not said what this would cost, but Ashkeboussi said that a first-of-a-kind plant was an opportunity to learn how to bring costs down. GLE already has a test module running at a GE campus in Wilmington, NC. (GE used to own the technology, but it is now owned 51 percent by Silex, an Australian concern that holds the patent, and 49 percent by Cameco, the big Canadian mining company that also owns part of Westinghouse.)

Ashkeboussi said that his company has acquired 650 acres adjacent to the government’s Paducah facility, and would use 200 acres to build a plant that would raise the U-235 content back up to the natural level of .7 percent. From there the company could raise it to levels useful in a light-water reactor, or it could be sold to another enrichment company to do that work. Today’s reactors run on enrichments of around 5 percent, called Low Enriched Uranium, but advanced models are designed for High Assay Low Enriched Uranium, or HALEU, approaching 20 percent.

A Market Ripe for Change

If laser operation emerges as a commercial contender, it would be entering a market where the underlying economics depend on a complicated mix of history and politics. Russia has about 40 percent of the enrichment capacity globally, and since its invasion of Ukraine, western companies are scrambling to reduce dependency.

Laser enrichment has a long history, thus far inconclusive. Like gaseous diffusion, invented as part of the Manhattan Project, and centrifuges, which supplanted gaseous diffusion in the 1990s, laser enrichment works by exploiting the very small difference in mass between the two natural forms of uranium, U-235 and U-238. In all three methods, uranium is compounded with fluorine, into uranium hexafluoride, or UF₆.

Centrifuges work by spinning a few grams of the compound, in its gaseous phase, so that it is subjected to a force hundreds of thousands of times stronger than gravity. Centrifuges spit out two streams of UF₆ gas, one very slightly enriched and one very slightly depleted, but when hundreds of centrifuges are arranged in series, the final product can be enriched above 90 percent. The limit for civil uranium is 20 percent.

Lasers, proponents say, do a much more thorough job in a much smaller number of passes. The laser is tuned to a frequency that differentially excites the two kinds of molecules. GLE will not describe the precise mechanism, but published sources hint that it has to do with suppressing the condensation of the molecules with U-235 so they can be separated.

Lasers can work with a smaller physical footprint – and smaller capital expense—because they handle more material than a centrifuge does.

And just as centrifuges made the process smaller and less energy-intensive, laser enrichment would continue both trends, although the energy improvement would be modest compared to the jump from gaseous diffusion to centrifuges, which cut the electricity requirement by more than 90 percent. The two trends make laser enrichment commercially attractive but also raise proliferation concerns.

GLE got a license from the Nuclear Regulatory Commission in 2012 to build a plant that could enrich up to 8 percent, and produce 6 million separative work units, or SWU, a year. (The entire American consumption is about 15 million SWU a year.) But after Fukushima, market conditions were not favorable. Lately, though, the company has been gearing up for expansion.

Adding Flexibility

The industry has historically been hesitant to expand without firm contracts in hand, because the centrifuges are designed to start up and run their entire lifetimes— decades. Lasers, though, can be turned on and off.

GLE is not alone in the field. TerraPower, which is building a fast reactor that will run on HALEU, signed an agreement last fall with the South African firm, ASP Isotopes,, which is developing a laser system for separating isotopes of uranium and other elements. There are other laser technologies in various stages.

Moving towards lasers requires tolerance for risk. Over time, Democrats and Republicans have varied in their outlook on how much risk the government should take to advance technologies that would be of general benefit. Republicans pummeled the Democrats when Obama’s Energy Department loaned $535 million to an innovative solar company called Solyndra, and the company could not repay. Oddly, that ended up being a commercial risk as much as a technical one, because Solyndra’s product was delivered roughly on time and on budget, but was not viable because it had been overtaken by other solar technology.

Trump himself displayed some risk tolerance in 2018, when he signed the law creating the Advanced Reactor Demonstration Project, which heavily subsidizes the construction of two reactors that need higher enriched fuel. More recently, though, the current Trump administration has cast doubt on the value of the Advanced Research Projects Agency-Energy (known as ARPA-E, in imitation of the better-known Defense Department version, DARPA), although the new Energy Secretary, Chris Wright, supports it.

Trump also seems likely to end some work by the Energy Department’s Loan Programs Office, which has also nurtured emerging technologies.

Laser enrichment is the kind of multi-application technology that in earlier years, the Energy Department liked to nurture. In addition to reactor fuel, there are medical isotopes that can be purified with that technology, and the silicon used in computer chips can also be processed using lasers.

Laser enrichment could be another Solyndra. Or it could be like directional drilling, 3D seismic and supercomputing, all advanced by the Department of Energy, and all contributing to the fracking revolution. Looking for an advocate with the financial stamina to try it out, and the raw material that needs processing, the logical candidate is Uncle Sam.

And the stocks of uranium hexafluoride at the department’s former enrichment sites are the logical place to start. That would recover a valuable resource, fissionable uranium, from what will eventually be waste.

Subscribe now

By Matthew L. Wald

How is it possible that the United States, which invented uranium enrichment, now finds itself with hardly any capacity to process uranium, dangerously vulnerable to interruptions in electricity generation, and highly dependent on foreign sources, notably Russia?

The answer lies in the continuing debate about what is properly a government function, and what is better left to the private sector.

We are torn. One side says that whatever the task, the private sector can probably do it better; companies can modernize, economize and operate more efficiently than the public sector can.

The other is that some parts of the economy are too basic, and too important, to be operated on a least-cost basis, and that only government can do that work. On that list is infrastructure critical to electricity, which is the mother’s milk of the modern economy. But other traditional pillars of prosperity are also up for debate under the Trump administration, including a postal system with universal service, a national passenger rail system, air traffic control, and weather forecasting. Despite the tension between the two approaches, government used to follow the idea that it had to do the work the private sector could not or would not, if the work was needed to make the rest of the economy function. At times, though, we have drifted away from that.

The demise of U.S. nuclear fuel processing capacity demonstrates that politicians ought to make more rational assessments of what the private sector shouldn’t be asked to take on. Companies can design, build and operate factories, but there is a limit to the financial and geopolitical risk they can be asked to take on. If it’s critical to the economy, government will have to provide a backstop.

The technology to sort out uranium atoms into ones that are easy and not easy to split in a fission reactor was invented here more than 80 years ago as part of the Manhattan Project. The easy ones, U-235, comprise just 0.7 percent of natural uranium ore. The balance is U-238, which has important uses in a reactor but can’t be used on its own to run the kinds of power reactors we have in the United States. The two atoms are chemically identical but U-238 has more mass (as the number suggests). Today, centrifuges do the work of separating the two atoms by spinning so fast that the uranium mixture inside is subjected to forces about 1 million times stronger than gravity.

The need for enrichment is expected to soar because reactor operators are delaying plant retirements (and even reversing them) and because reactors that now run on 5 percent enrichment may soon want 7 or 8 percent, so they can go longer between refueling or can use fuel that is coated with materials that improve safety (but which demand more reactive fuel). The advanced reactors moving towards commercial deployment with federal help mostly want enrichments just below 20 percent.

Mathematically, this isn’t as challenging as it looks. It takes more effort to get from natural uranium to 5 percent enrichment (from one part in 141 to one part in 20) than it does to get from 5 percent to 20 percent (one part in five).

But the enrichment industry is inflexible. It takes years to add capacity, and the companies in that business are reluctant to do that, because once the machines are up and running, it is very difficult to stop production.

Subscribe now

“Can’t” or “Won’t” Should Be a Cue

Our current national nuclear strategy is asking enrichment companies to add capacity to fuel reactors that haven’t quite been invented yet, let alone deployed.

That is a tough ask for private companies, and that’s why the major enrichment companies aren’t private. The two leading companies are Orano, owned by the French government, and Urenco, a German, Dutch, and British consortium. Only the United States tried to turn the job over to the private sector.

It wasn’t always that way. The American civil nuclear age began with a government operation; the Atomic Energy Commission (AEC) opened three gigantic enrichment plants, which made high-enriched uranium for weapons and for the propulsion reactors in submarines and aircraft carriers.

The problem was that the government program inherited the original Manhattan Project technology, called gaseous diffusion. Originally, this was simply an activity of the AEC. Then it was re-organized into a separate program within the AEC (and then the Department of Energy, a successor agency) as the United States Enrichment Corp., or USEC.

USEC sowed the seeds of its downfall when it stopped offering enriched uranium to the nuclear energy programs of America’s allies, because it did not think it could keep up with projected demand. European customers went out and built their own, with the technologically-superior centrifuge system. Those use about 90 percent less electricity.

Even giving up on foreign customers, USEC needed to expand and modernize. The Nixon administration first proposed privatization, and the government hired Smith Barney to analyze the options. In 1990, the consultants came back with the idea that USEC could do the job better if it were freed from government restraints. As a government-owned, contractor-operated fleet of factories, all running on cost-plus contracts, there was no incentive to improve operations, the consultants said, and there was none of the flexibility that normally attached to private industry.

And in 1998, it was sold to investors.

But the privatized USEC botched the modernization. It dabbled with centrifuge technology and laser enrichment and failed at both.

Then diplomacy made the problem worse. In 1993, USEC became the “executive agent” for a 20-year deal between Washington and Moscow to have the U.S. buy Russian uranium that had been enriched for weapons use in Soviet times.

The Megatons-to-Megawatts deal was a lifeline for Boris Yeltsin, then Russia’s president, and for USEC, which could dilute the enriched uranium to meet the specifications of utility customers, using some of the massive stock of uranium that had been stripped of half of its U-235. But it put off the day of reckoning for how to meet American needs. And when the deal ended, the solution was to buy enriched uranium from the Russian state nuclear monopoly, which controls about 40 percent of the world enrichment market.

USEC eventually closed its obsolete enrichment plants, and went bankrupt. But it was re-organized into a new company, called Centrus, which is still in business.

Damage from USEC’s failure, though, lingers. With Russian capacity in the market, the world is oversupplied. Without Russia—which is the American goal after the invasion of Ukraine—the market would be “savagely undersupplied,” in the assessment of David Leistikow, vice president of Centrus.

Reactors are expensive to build and staff. They can only run on uranium, and it’s almost impossible for them to economize on fuel. Running out of fuel would be monumentally stupid.

In 2024, Congress has set up a timetable for banning imports of uranium enriched in Russia, although there is some flexibility, depending on when substitute sources enter the market. But there are reasons for other producers to be cautious about expansion, including the possibility of a Trump-Putin agreement that lets the Russians back into the market. And, above all, when will the hundreds of new advanced reactors actually come online as customers for an expanded domestic uranium industry?

Reversing the Decision

Centrus drew the federal government back into the business in the form of Department of Energy support for its fledgling centrifuge complex. And now Washington is moving more seriously into uranium enrichment, by providing a guaranteed market for higher-enriched uranium, at an intermediate stage of fuel production.

In December, before President Biden left office, the Energy Department announced that it had picked six companies that it could contract with to purchase enriched uranium, one of them using lasers, “in order to encourage the build-out of new uranium production capacity in the United States.”

Congress allocated $2.72 billion to the department to support enrichment. But the department has not actually signed contracts yet.

We’ve gone from public to private and now to private operations with an extensive public support structure, because private alone did not work.

Some privatizations have worked, notably rocket launches, and operation of the internet. Notably, both involve technologies that were once too complex for the private sector but are now well within commercial capabilities. Having a technology that is commercially accessible is necessary for successful privatization, but not sufficient.

The private sector may yet achieve improvements in centrifuge technology. Centrus has recently opened a centrifuge complex in Ohio that it says does the same work with substantially less energy. There is also the possibility that lasers could do the work even better. But, Centrus will likely still require the public sector for investment and to serve as a primary customer as they ramp up enrichment operations.

We have billions of dollars tied up in reactors that need enriched uranium, and an economy that won’t work without their contribution to the grid. We’re planning billions more in nuclear investments. Preferring the private sector to run certain aspects of our economy is fine, but some things simply will not be done unless the public sector acts.

The United States stockpiles crude oil, diesel fuel, antibiotics, and, reportedly, some rare earth materials. It stockpiles high-enriched uranium for the Navy’s propulsion reactors. It ought to be stockpiling low-enriched uranium for power production. Uranium keeps (half-life of U-238 is 704 million years), it doesn’t take up much space, and isn’t expensive to store. An LEU reserve’s operation would stabilize the production system and end a threat to the generation of electricity. And it would help bring about the transition to advanced reactors that is a bi-partisan priority.

Congress has allocated enough money to start, but the Energy Department needs to get contracts signed, and to expand the program.

Subscribe now

Most of the concern about President Trump’s tariff strategy is that it will raise costs and reduce international trade, but in the case of nuclear fuel, the opposite may be true. The United States has been reliant on Russia for enriched uranium since the early 1990s, and only decided that it needed to be self-sufficient after Russia invaded Ukraine in 2022.

The United States’ recent UN vote against a resolution to condemn the Russian invasion raises the prospect that the United States could similarly backtrack on its plan to wean itself off Russian imports, including enriched uranium. This is a particular problem for a capital-intensive, slow-moving industry like uranium enrichment. Even the possibility of rapprochement can hurt the fuel industry, and damage the commercial nuclear energy business more broadly.

A little background on making nuclear fuel will elucidate the problem:

Share

Nearly all uranium is one of two types, uranium-238 and uranium-235. In water-moderated reactors (the only kind in use in the United States at the moment), the kind that splits easily is U-235. But in nature, that is only 0.7 percent of all uranium.

So, to be useful in a water-based reactor, the uranium must be “enriched.” That is, the proportion of U-235 has to be raised to at least 3 percent, from 0.7 percent, a level called “Low-Enriched Uranium,” or LEU. The United States invented a technology for doing that as part of the Manhattan Project. For a long time, the U.S. had a monopoly outside of the Soviet Union, and initially when it made enriched uranium available to the utilities, it retained ownership of the fuel.

Over the years, the business went through major changes. Utilities started calling for fuel with higher enrichments, so the reactors could run for 18 months or two years between re-fueling. That required enrichments around 5 percent. And they began experimenting with “accident-tolerant fuel,” doped or coated with materials that made the fuel able to withstand higher temperatures before it began to melt. But the new materials tend to steal neutrons, the sub-atomic particles that are liberated when a nucleus is split and that go on to split other nuclei, sustaining the chain reaction. So, the fuel requires enrichment above 5 percent but below 10 percent. The industry calls that product LEU+.

Using accident-tolerant fuel could increase the safety margin or could allow reactors to operate without quite as many hair-trigger emergency systems to rush in cooling water in case the main cooling system failed. That could cut costs, but whether the savings would justify the extra cost of the fuel has yet to be demonstrated. Widespread use would raise demand for enrichment.

The other change is the approach of advanced reactors. Most of them are designed to use fuel near the top end of the definition of LEU, which is 20 percent. The industry calls this High-Assay Low Enriched Uranium, or HALEU. If that market for HALEU develops as advocates predict, demand for enrichment will rise even more.

The catch is that at the moment, the Russian state nuclear monopoly, Rosatom, is the only commercial source for HALEU, and the developers said after the invasion of Ukraine that they would not source their fuel there.

We can’t have “energy dominance” without advanced reactors. And we can’t have most of those reactors without a domestic supply of HALEU, or, at minimum, a world market in which the West, plus perhaps Japan and South Korea, are self-sufficient in enrichment.

But if the Russian state nuclear monopoly, Rosatom, is let loose in the world market, it will price enrichment to seize a huge slice, to gain influence and foreign currency. Since the fall of the Soviet Union, the United States has set a quota for Russian uranium imports, but arguably, it was too high. We should not make our nuclear industry dependent on Russian exports. The clear solution is either tariffs or outright market restrictions.

Setting Limits

A key economic decision is how much effort to put into scrounging U-235 out of natural uranium. The answer depends on the value of uranium. For most of its history, the Department of Energy and its predecessor agency, the Atomic Energy Commission, harvested roughly half of the U-235 from virgin uranium, but if the price of processed uranium ore dropped, enrichers could use more and recover a smaller portion of the U-235. If it rose, they could use more enrichment. To make a given quantity of fuel, a manufacturer could add extra enrichment to a smaller quantity of uranium, or add extra uranium ore and use less enrichment work. The decision rests mostly on the price of processed uranium ore and the market value of enrichment.

Thus, enrichment shows a high degree of what the economist call “supply elasticity.” The need depends on its price and the price of other segments in the supply chain.

The limit for Low Enriched Uranium is set at 20 percent because governments do not want material in commercial use that approaches the enrichment needed for a weapon. Anything above that is Highly Enriched Uranium. The number commonly cited for weapons-grade uranium is 90 percent, but if the delivery system—air plane or missile or ship—can accommodate a larger core, the enrichment level can be in the 80 percent range.

Along with growing demand for enrichment, as the number of reactors grew and they used higher enrichments, the technology changed. When the U.S. had a monopoly, it used a technology called gaseous diffusion. Uranium in a compound with fluorine gas, called uranium hexafluoride, was forced through a semi-porous barrier, but the two types of uranium pass through the barrier at different rates, so running the gas through repeated cycles allowed substantial enrichment.

But other countries modernized to use centrifuges, which use the same gas to do the same work, with about 90 percent less electricity. The U.S. government privatized its uranium enrichment enterprise but the privatized company, known initially as USEC, for United States Enrichment Corporation, did not modernize, and eventually went bankrupt. Its last plant for enriching uranium closed in May 2013.

The company is now called Centrus, and has developed a new class of centrifuges, but is a small-scale producer at the moment.

Stepping into the gap were two government-owned companies in Europe. One, URENCO, under joint British-Dutch-German ownership, built a centrifuge plant in New Mexico that serves some of the American market. It has announced plans to add capacity. Another, French-owned Orano, would like to build in the United States.

But there are problems, related to the technical details of enrichment. The product, uranium hexafluoride, is a commodity, but the economics of enriching uranium defies common logic.

The first problem is that adding capacity is very expensive. The second is that in other process industries, like oil refining, if demand declines or supply increases and prices fall, producers can respond by cutting production. But once a centrifuge is spinning, engineers do not like to stop it for any reason. Production will continue, and prices will decline further.

A Tough Risk for the Private Sector

There are three other obvious risks. One is that for years, engineers have been trying to figure out how to use lasers to differentially excite the uranium hexafluoride molecules, sorting U-235 from U-238 using less energy than centrifuges do. There is no commercial laser process today, but the Department of Energy recently issued a list of companies that would be eligible for major contracts it would issue for uranium enrichment, under a $3.4 billion program voted by Congress to strengthen the domestic fuel supply chain, and two laser companies were on the list. Technical improvements might let lasers do to centrifuges what centrifuges did to gaseous diffusion.

The second risk to new enrichment capacity is that it may be undercut by reprocessing firms that can flood the market with new fuel by recycling nuclear waste. They might do this because it is cheaper than using virgin uranium, or because governments decide the step is helpful because it reduces the volume of nuclear waste. Reprocessing means chopping up the fuel, dissolving it, and recovering the U-235 that was not fissioned, as well as plutonium-239. When U-238, the dominant isotope, is hit by a neutron, often it absorbs the neutron rather than spitting, and is transmuted into plutonium-239, a good reactor fuel. Most light water reactors in the U.S. can be adapted to use a plutonium-uranium mix. One plant, Palo Verde, was designed for it.

And the third is that another Chernobyl or Fukushima stifles the growth of nuclear energy. Each of those events led to delays in the completion of reactors under construction and gave ammunition to opponents of new projects. The risk of another mishap on that scale is very small, but hard to quantify.

The Non-Market Threat

For the enrichment companies, though, there is another risk, one that is controllable if the Trump administration recognizes the problem. That is that Russia will be allowed back into the U.S. market. Even re-entry into the European market would be a problem for Western enrichment enterprises, because their market share would shrink.

Russia does not mine a lot of uranium, but it does enrich it. In Soviet times, it took uranium from what is today the independent country of Kazakhstan, in central Asia, enriched it and shipped it out through the port then called Leningrad, which is today Saint Petersburg. Post-Soviet Russia can still be a major player in uranium, and is not quite a market-based operator. It will price the product because of a government need for foreign currency, or because it wants to maintain political influence, rather than the traditional economic calculation of whether revenue will exceed costs.

In May 2024, Congress passed a law that seeks to force a phase-out of imports from Russia. The Energy Department can waive the ban until 2028. Before the law, Russian imports were limited to a quota by law.

Russia has enormous capacity to enrich uranium, although nobody is quite sure how enormous. A year ago, the Energy Department estimated that Russia was supplying 44 percent of global enrichment, and 20 to 30 percent of the enriched uranium product used in the United States and Europe. The reason is clear; in the cutthroat electricity market in the United States and elsewhere, reactor operators have been eager to buy from the lowest-priced bidder.

For the twenty years ending in 2013, half of the United States’ enrichment needs were met by Russia diluting high-enriched uranium produced by the Soviet Union for nuclear weapons and nuclear submarine propulsion. The program was a success on a policy level, because it eliminated enough highly enriched uranium to produce 19,000 warheads. But it was devastating to America’s enrichment complex. Not only did enrichment suffer, but so too did the business of converting ore into uranium hexafluoride (the chemical form needed for centrifuges) and “de-converting” it to a solid that could be turned into ceramic fuel for power reactors.

Hence the American nuclear fuel chain was nearly wiped out.

Now the Energy Department is trying to rebuild the U.S. nuclear fuel supply chain. The industry has a chicken-and-egg dilemma because potential suppliers don’t want to invest in the hardware needed to do the work without assurance that the market will develop, and backers of advanced reactors are set back by the fact that there is no ready source for the fuel.

The fix, created by the Biden administration and Congress, was for the Energy Department to function as a strategic reserve, taking in enriched uranium to assure that it would be there when the industry needed it. . The department would stockpile the material as uranium hexafluoride, the intermediate form that could be turned into a variety of fuel forms, depending on the reactor for which it was destined.

The Inflation Reduction Act, provided $700 million for the Department of Energy to begin a “HALEU Availability Program.” And the United States is coordinating with Great Britain, France, Japan and Canada to strengthen supply chains. Collectively, the governments expect to spend $4.2 billion. The Department of Energy said the goal was to “establish a secure and resilient global nuclear fuel supply chain to ensure continued operation and support the growth of nuclear energy deployment around the world free of Russian influence.”

The Trump administration, however, has already laid off hundreds of Energy Department employees and summarily terminated a variety of government programs. And the administration’s approach to Russia has been mostly friendly. At a time of budget cuts, the future of the uranium enrichment program is not certain.

The program does include help for Centrus, which alone among enrichment companies could provide enriched uranium for defense use, unconstrained by international non-proliferation agreements. But its production ability so far is modest. And under Trump, the government is not in a spending mode.

The idea after the Russian invasion of Ukraine was to make the country self-sufficient in uranium fuel processing, but that effort could end up as collateral damage of other Trump policies. The Trump administration, assertively pursuing an “America First” strategy, could easily destroy a cornerstone of energy independence, and one of the key rationales for building more nuclear reactors.

Subscribe now

By Matthew Wald

The Trump plan to “unleash” American energy may run directly into the Trump plan to extract a better trade deal with Canada. This is especially true for nuclear energy. The immediate victim may be the GE-Hitachi small modular reactor that Ontario Power Generation wants to build. The sale may be crucial to the United States because the Tennessee Valley Authority is following the Ontario plan closely, and wants to be GE-Hitachi’s second customer.

Tariffs imposed by the U.S. on goods entering the country from Canada won’t raise the reactor’s cost to Canada in money, but may make it unaffordable because of public resentment.

Other trade related to nuclear energy may also be at risk.

The problem for the pending nuclear renaissance in the United States is that vast sums of money have gone into research and design work, but so far, orders have not followed. Two projects, intended to be half financed by the Department of Energy, are moving forward: TerraPower’s Natrium high-temperature fast reactor coupled with molten salt energy storage, in Kemmerer, Wyoming, and X-energy’s Xe-100 high temperature gas-cooled reactor for Dow in Seadrift, Texas, where it will displace natural gas in making steam for chemical processing.

But the traditional customers for reactors, the utilities, are watching and waiting, not ordering. The utilities could get federal subsidies for clean energy and for domestic content (if the new administration does not repeal them) and a non-money incentive: they have promised to cut their carbon emissions, and in some cases are under legal obligation from their host states to do so. But none of this is as generous as the federal program under which Natrium and the Xe-100 are to be built.

Canada is doing better. Ontario Power Generation wants four of the GE-Hitachi BWRX-300 models. These are small modular reactors designed to be simpler and easier to build, and to require far fewer safety systems, yet have an even lower probability of accident, because of inherent design features. They are also light water reactors, meaning that they are closer to what the United States operates today, and thus probably present a lower technical risk.

Making this model successful will require multiple orders. Four in Ontario, plus the possibility of another in Saskatchewan, would be an excellent start and would reduce the risk associated with construction of a first-of-a-kind reactor enough to let U.S. utilities follow.

That means Canada is not the best country to alienate when seeking to launch a nuclear renaissance in the U.S. While investor-owned utilities in the United States are nervous about ordering a first-of-a-kind reactor, with uncertain costs and schedule requirements, provincial utilities in Canada are a good alternative. Canada, which is probably more vulnerable to climate change and is certainly more committed to reducing its carbon dioxide output, has strong public support for new nuclear and is determined to build new reactors. It could take that difficult first step on our behalf.

There are two ways that tariffs can get in the way of this. A 25 percent tariff imposed by Trump on goods from Canada, announced in a February 1 proclamation but quickly paused for reconsideration at the beginning of March, or, in a second version, a 25 percent tariff on Canadian steel and aluminum, and 100 percent tariffs on the auto sector, seem certain to lead to retaliatory tariffs, and a general reduction in trade. And such tariffs have already led to a consumer backlash and a “buy Canadian” movement, mostly for consumer products, like supermarket items.

There is a risk that resistance to big-ticket items, like reactors, will follow.

A Bi-National Project

That spells trouble for the Ontario project, which would very much be a cross-border exercise. Ontario Power Generation has recently signed contracts with BWXT-Canada, a subsidiary of the U.S. company, to build the reactor vessel, and do other work. But much of the hardware would come from the United States, and it would make GE-Hitachi’s design vastly more valuable if it were proven with construction.

The fuel would probably be enriched outside Canada, and possibly in the United States. Canada has no uranium enrichment industry. But in a trade war across the border, it might have to come from somewhere besides the U.S. even if the reactors are built.

None of that seems to have gone into the new administration’s thinking about tariffs. “The administration isn’t making that much of a distinction between allies and competitors,” said Elaine K. Dezenzki, a trade expert and head of the Center on Economic and Financial Power at the Foundation for Defense of Democracies, speaking at a recent webinar on what to expect from the new administration on trade.

And Canadians are taking offense. One prominent Canadian nuclear advocate, Chris Keefer, wrote sarcastically on LinkedIn, “by all means let’s take FOAK (first-of-a-kind) development risk and pour hundreds of millions of Canadian taxpayer dollars into developing privately held IP on US origin reactor technologies” like the GE-Hitachi model.

Prime Minister Justin Trudeau has articulated both sides of the equation. He told delegates at a recent conference on AI, “Let me say this once and for all: as an environmentalist, the debate is over.”

“Large-scale nuclear reactors must be part of the solution for the future because if we’re not willing to embrace nuclear now, then coal-powered AI from other parts of the world will shape the coming decades for the worse,” he said.

But he also said, “We have been doing things together in an extraordinarily integrated way for many many years, and we will look to continue to work with our closest ally and partner. We should be doing more together, not fighting with each other. But Canadians will stand up strongly and firmly if we need to.”

And Canada could stand up firmly by picking its home-grown reactor, the CANDU, which is a source of national pride.

To understand Canada’s attitude towards the CANDU reactor, it helps to remember why it was developed in the first place. Its origins were in the 1950s, when the thinking was that any self-respecting country would have to build nuclear reactors. But Canada lacked (and still lacks) enrichment technology to raise the proportion of the kind of uranium that is easily split, U-235, to three to five percent, from its natura 0.7 percent. And it did not have the ability to fabricate metal parts as big as a reactor vessel.

It could, however, make strong metal pipes, and it has the world’s largest reserves of high-grade uranium ore.

So, Canada invented a reactor that can use natural, unenriched uranium, because instead of water, it uses “heavy water,” deuterium, which does not steal neutrons from the chain reaction as ordinary “light water” does. It is the Canada Deuterium Uranium reactor, CANDU for short, which is a word play on the slang phrase popular in the middle of the last century, “can do.”

Canada has sold reactors to China, Argentina, Pakistan, Romania, South Korea, and India—which has developed its own SMR version of the CANDU.

The National Pride Factor

Canada has always been proud of its nuclear energy capabilities. When cost overruns caused construction to falter at the Seabrook reactor, in New Hampshire, in 1984, New Brunswick Electric, already operating one CANDU at Pt. Lepreau, offered to build a twin to supply the New England market, Seabrook was 75 percent complete, but the New Brunswick utility said it could start from scratch and be ready before Seabrook was finished. Using a proven design and relying on an established supply chain, the company said it could build at a cost of approximately half the cost per unit capacity of Seabrook. (In the end, Seabrook’s main builder went bankrupt but its partners finished the reactor.)

A new model CANDU is a clear alternative to GE-Hitachi’s imported boiling water reactor technology. Canadian utilities may have thought that the CANDU wasn’t quite as good, but it is certainly native. Nuclear industry executives talk about the “social license to operate,” meaning public acceptance. At the moment there is a shrinking social license for anything imported from the United States.

The country has seen an explosion of patriotism and pride on social media. Canadians are posting lists of items sold in supermarkets that are made in Canada and can substitute for similar items that are imported from the United States. Posts on Facebook promote Sun-Brite Foods Brands, which makes Primo ketchup and pizza sauce, over Heinz and Frenchs, and Nivea skin care products over Oil of Olay. The company that makes Babybel cheese is reassuring customers that their product comes from Canadian milk.

Websites are making lists of made-in-Canada products, and inviting consumers to pledge to prefer them.

Provincially-owned utilities must be sensitive to public sentiment.

But it’s not just the GE-Hitachi reactor that is under threat. The nuclear industry has long touted that its fuel is secure because it comes from friendly countries, notably Canada, which is right next door. Did the industry contemplate that we could turn a friendly country into an unfriendly country?

Tariffs on uranium, which have been threatened as part of a general tariff, could in the long term stimulate production in this country, but their effect for years to come would be to raise prices for nuclear energy and make it less competitive.

Canada does not believe that its uranium mining is threatened, because demand for uranium is price-inelastic. “They don’t really have a lot of other options to go get that uranium from,” Devan Mescall, a professor at the University of Saskatchewan’s Edwards School of Business, told the Saskatoon StarPhoenix. “You’re not going to start and stop nuclear reactors,” he said

Tariffs could stifle the bi-national effort to build the 300-megawatt GE-Hitachi SMR, pushing Canadian utilities into using Canadian, rather than U.S. technology. Tariffs could also set back the micro-reactors that Canada wants for isolated communities and mining operations, and U.S. customers want for reliable, on-site power for computer centers and other installations that require high reliability In the past, the United States was so central to the world reactor market that a supplier could easily launch a model here and think later about selling it abroad. But now we have a “nuclear renaissance” in which there are no orders beyond two demonstration projects. The industry needs to get the ball rolling, and losing the Canadian market will not help.

SMRs and micro-reactors are going to work if they get economies of scale. Canada is a promising partner, because unlike other potential customers around the globe, it has money, technical know-how, and no desire to use U.S. equipment to make nuclear weapons material. If driving away this potential customer doesn’t constitute a runner shooting himself in the foot, it at least represents intentionally tripping his running partner.

*Note: the cover image is of the Canadian nuclear power station, Pickering Nuclear Generating Station. The station has eight CANDU reactors.

Subscribe now

John Muir famously explained that “when we try to pick out anything by itself, we find it hitched to everything else in the Universe.” Muir was a naturalist, remembered for his role in helping to establish Yosemite National Park, but he might just as well have been talking about technological progress. The idea that everything is interconnected also applies to nuclear energy, which is striving to better hitch itself to other technologies.

Interconnectedness, and piggybacking on each other, is how technologies advance, especially those that are no longer on the cutting edge of industrial progress. Nuclear energy, both in legacy plants and in the advanced reactors moving toward deployment, is currently trying hard to incorporate tools from outside the field—notably Additive Manufacturing (3-D printing), and Artificial Intelligence. But as with a lot of other nuclear projects, progress is somewhere between deliberate and just plain slow

There was a milestone of sorts in December, when Framatome, the French nuclear company, said it had put 3-D printed parts into Ringhals 4, a large pressurized-water reactor in Sweden, which began operations 41 years ago.

The technology may be important for sustaining the operation of legacy reactors, but also shows great promise for new ones. BWXT, working on NASA projects, and the University of Tennessee, as part of the Energy Department’s Transformational Challenge Reactor program, have both sought to marry cutting-edge computing and manufacturing technologies for the benefit of nuclear energy. Parts for old reactors are expensive or unavailable; new ones demand parts that conventional manufacturing can’t supply. So 3-D printing will likely play a key role in a nuclear renaissance.

Making Reactor Parts with a Printer

3-D printing builds objects or structures by layering material in a programmed pattern. It is the three-dimensional equivalent of a machine that turns on-screen documents into printed pages.

The kind of 3-D printer that they sell in computer stores uses a resin that is good for making an ornate object the size of a key chain fob, but in industrial settings 3-D printers can use concrete and build an entire building, or powdered metal or metal wire that is fused with a laser into solid stainless steel. The technology can make precisely shaped small parts, sometimes more precisely than conventional fabrication.

Often it is cheaper to manufacture small parts in a shop, especially if you need multiple copies. So, why bother?

Due to years of little or no production, the nuclear supply chain has withered, meaning specific parts are both more expensive and harder to find. If a part is available from somewhere in the world, it may be fabricated only when the order is placed, and getting it delivered takes time. The plant may be shut down until the part is in hand. 3-D printing can help solve this issue by producing intricate parts for critical uses quicker than with existing supply chains.

But it takes time to prove that the materials are durable enough. A nuclear reactor is a harsh environment. Neutrons liberated by splitting uranium or plutonium sometimes go on to split another atom, or are absorbed by uranium, creating plutonium, but if that doesn’t happen, they hit atoms of other materials in the core and can damage them. So, engineers start with small samples of candidate materials, introduce them to the swarm of neutrons in the reactor, and see how they perform over time. At Ringhals, Framatome installed 3-D printed anti-vibration supports in the spring and summer of 2024, and most recently, filters that catch debris present in the cooling water.

Ringhals is not the first. In 2021, at the Browns Ferry plant, the Tennessee Valley Authority installed brackets that hold the fuel in place. They were manufactured at the Department of Energy’s Oak Ridge National Laboratory in a joint project with Framatome. The parts will remain in place for three fuel cycles, or about 6 years.

A year ago, Westinghouse, which is a major fuel manufacturer, said that it had fabricated nozzles for reactor cores that were 30 percent better at filtering out debris from cooling water, and that these had been installed in the core at the Southern Company’s Farley plant. Water circulates fast through a reactor, and if it carries debris, the debris can damage the fuel rods. The company said it had recently built its 1,000^th component that was manufactured with additive manufacturing for fuel for VVER-440 reactors. These are Soviet-designed plants in Eastern European countries that have recently turned to the West for fuel.

Reactor managers lose sleep over the possibility of needing a replacement part from a company that doesn’t exist anymore. They like the idea of a part that can be fabricated on site from a computer file that can be sent over the internet.

But 3-D printing is not just useful for old plants that require new parts. Additive manufacturing is proving quintessential to the new wave of modular and advanced reactors.

Oak Ridge has been active in additive manufacturing techniques that it believes will be useful for advanced reactors. Micro-reactors, according to lab scientists, will need embedded sensors, like fiber optics, to measure temperatures and strain. The lab has embedded fiber-optic sensors in metals commonly used in reactors, including stainless steel alloys, copper, and aluminum. The sensors are put into machined cavities, which are then covered with foils that are zapped by microwaves to seal them in.

Last summer, NuScale Power, which has an NRC-approved design for clusters of small nuclear reactors, won a patent for using additive manufacturing to make steam generator components with embedded sensors. Battelle and the University of Idaho are seeking to patent for 3D fabrication of heat exchangers for nuclear plants.

Iowa State University researchers recently got a grant from the Department of Energy to explore using 3-D printing to make parts for reactor cores from tungsten. Tungsten works well in high-temperature environments and resists erosion. The experiments will begin with pure tungsten and move on to alloys, and the researchers will use artificial intelligence to analyze the results.

And China has used 3-D printing in a multi-stage process to make tritium, which can be used as a fuel in a fusion reactor. The printed product has embedded lithium, which is split by the neutrons to produce tritium. The conventional manufacturing process leads to a product that tends to crack when irradiated, Chinese researchers say.

Beyond What a Machinist Can Do

In addition to using 3-D printing to make specialty parts that are available commercially only with a delay, or not at all, the technique can also make parts that simply can’t be fabricated conventionally. BWXT has moved heavily into this area.

“We started looking at additive manufacturing to give our designers more freedom,” said Kate H. Kelly, director of Space and Emerging Programs at BWXT Advanced Technologies, which is developing nuclear rocket engines that are cooled by gas. The shapes include a three-dimensional lattice with complex twists and very thin ligaments. Commonly, such parts are made by machining (or carving out) the desired shapes from bigger pieces of material, but these are “things that you couldn’t machine out to get that level of precision,” she said.

The material used for the component is often a brittle alloy, making it hard to machine, but easier to 3-D print.

There are several techniques for 3-D printing. One is using electron beam welding on a bead of alloys that are chosen because they can withstand very high temperatures, called refractory alloys. These are key because some advanced reactors are intended to operate at far higher temperatures than the familiar light water reactors, like Farley or Ringhals. BWXT is working on a rocket engine that will operate at more than 1700 degrees C, or 3100 degrees F. Lasers can also be used, to heat a powder and melt it into a solid.

And the components’ shapes, often designed using artificial intelligence, can be tweaked to allow optimum heat transfer. That can allow a smaller reactor, which is important for applications like space travel.

Researchers at the University of Tennessee are using the same logic. They have designed a graphite-moderated, gas-cooled reactor that is designed by artificial intelligence and then 3-D printed. The goal is a core with minimal temperature variation over its volume, producing as much energy as possible but with no spots that are too hot. The A.I. can try out designs with cooling channels that vary in radius over their length, and do not have to be straight. Then a 3-D printer fabricates it.

The NRC is paying close attention. Is the powder being fused into solid parts sufficiently pure? Was the fabrication process interrupted and resumed, which could affect quality? Is the geometry of the finished product accurate? And of course, what is the durability of these parts? The Electric Power Research Institute points out that among the complications, materials used in nuclear plants are generally covered by the American Society of Mechanical Engineer’s code, and approval by the code is long and costly.

Computer-driven 3-D printing isn’t the only modern technology that the nuclear industry is importing from other industries. Notably, it needs to replace analogue dials and gauges and mechanical switches in the control room with modern computerized sensors and controls, although there are regulatory barriers. The industry also needs to take advantage of advances in materials science in the years since the current fleet was built.

It is moving quickly in the direction of integration, both in new reactors and in reactors that have been running for decades longer than the new technologies have been around.

By Matthew L. Wald

The world at large sometimes has trouble assembling facts into accurate narratives. This is especially true for facts relating to nuclear energy. The problem can extend to elected officeholders, career journalists, and activists who try to influence both groups.

For example, this BBC report sounded straightforward; a worker at the Fukushima nuclear plant after the 2011 tsunami and meltdown had died from lung cancer, because of his radiation exposure there.

But within the nuclear industry was a more detailed picture. The man, in his 50s, and worked at several nuclear plants over decades. He had accumulated a radiation dose in various places, and his dose at Fukushima was not particularly high. And he died of a cancer that might or might not have come from radiation exposure, but under Japanese law, lung cancer was one of a list of ailments that were presumed to have been caused by radiation exposure if the patient had worked in the nuclear industry. So yes, he had died of cancer, the cancer could have come from radiation, and he’d worked at Fukushima. All that remained was for the BBC to connect the dots, correctly or not.

Think of the facts as a Rorschach test. Hermann Rorschach was a Swiss psychiatrist who developed a series of inkblots, symmetrical from left to right, but with no intrinsic meaning. Researchers show them to subjects and ask what they see, and the answer tells more about the viewer than the object being viewed. (The joke is that the man who is shown the series of inkblots says, “who is this guy Rorschach, and why is he showing me pictures of my parents arguing?”)

Rorschach died 103 years ago, but if he were living today, maybe we’d call him a kind of media analyst; he measured the differing responses of various individuals to a single set of inputs.

Technically, it isn’t possible to get a failing grade on a Rorschach test, but effectively that’s what the BBC did. The Fukushima worker story shows a phenomenon that exists in many realms of human enterprise, but seems especially strong in nuclear. Here is the central rule that describes processing nuclear information: People make sense out of the facts by fitting them into previous preconceived notions, and close off all other interpretations. This gets easier when their frame of reference is limited, which is true for most people on this subject.

Share

In public perception, nuclear, and Fukushima in particular, create an enormous divide. On one side are the people think that after the giant earthquake and tsunami, the triple melt-down killed hundreds or thousands. On the other side are people who can rationally separate the tsunami’s death toll from effect of the reactors being wrecked.

In fact, the earthquake and tsunami killed about 20,000 people, but the World Health Organization says that the radiation doses weren’t large enough to cause a discernable change in cancer rates, and that no individual fatal cancers will be attributable to the accident. People died of cancer before Fukushima and have continued since then. Some smoked, some didn’t, some had a family history, some didn’t.

The BBC was hardly alone. Numerous TV news shows had an announcer reading dire pronouncements about Fukushima while the visual was a burning oil refinery. In the spring of 2011, if it was Japanese and on fire, it must be a nuclear plant, right?

There are other cases of nuclear-induced myopia. For example, there was the campaign of Norman O. Aamodt and Marjorie Aamodt, organic farmers in Pennsylvania, who campaigned against the re-start of Three Mile Island 1 after the accident at Unit 2.

Mrs. Aamodt testified before the Nuclear Regulatory Commission That they had taken blood samples from 29 people in 1994 and 1995 and sent them for analysis from Pennsylvania to Moscow, to the Russian Academy of Sciences. She testified that the Russians concluded that their average dose to the people who gave the samples was 100 rems. This was at least 1,000 times larger than public health officials estimated was the maximum dose to a member of the public could have received, and even that level was unlikely because it required a person to stand at the fence line for several days. And it was about 17,000 times larger than the average estimated dose.

But aside from the enormous discrepancies in estimates of the amount of dose that someone near the reactor could have received, the Aamodts asserted that the 1979 accident was the cause of numerous cancer cases. But the Aamodts identified those cases earlier than a radiation exposure could have caused them, a court found. At the NRC, one official suggested privately that such a finding could be helpful to public health, because if there was, in fact, an unusually large number of cancers in the area, it suggested that it was from some cause other than the accident, and might be continuing; further research might uncover an ongoing health threat. But the Aamodts, unable to fit the information they thought they had into their desired framework, gave up. Their object, it appeared, wasn’t to protect public health, but something narrower and less useful, to protect public health from nuclear energy.

Not all the misunderstandings overestimate the potential hazards of nuclear technology, and some misunderstandings are more consequential than the BBC’s about Fukushima. A worrisome one was a case about nuclear safety, where the Department of Energy did not know what it did not know about a nuclear materials plant it owned and wouldn’t believe it when it was told.

The problem became public at a scene that could have been from a Netflix drama, but it happened a decade before Netflix and it was real, and it teaches something about why our energy and environment deliberations are so frustrating.

In September, 1988, Jerry Hulman, director of the office of Quality Programs at the department, was testifying before a joint House and Senate hearing when the chairman, Senator John Glenn (D-Ohio), held up a list created by DuPont, a government contractor, about the worst safety incidents in the almost 40 years that the company had operated the Savannah River Plant where the department made materials for nuclear weapons.

Hulman, a veteran staffer at the Energy Department and its predecessor, the Atomic Energy Commission, was adamant: these events had been concealed from the government. But a week later at the site, a DuPont engineer who was a co-author of the report told me that the company always told the government everything. There were two reasons: because the government owned the plant, which DuPont ran for a symbolic payment of $1, and because nobody in the government would complain, because they wouldn’t understand the report anyway.

The symmetry of differing perceptions was perfect. Senator Glenn and his staff said the list showed an alarming safety problem. But the engineer said that it showed precisely the opposite; the trend was that reportable incidents were becoming less frequent and less severe, he said.

The Energy Department later conceded that it had, in fact, been given the document, but it never found its way to anybody in a position of authority in Washington. And two of the 30 “suppressed” incidents on the list had been featured over the years in reports in the Atlanta Journal-Constitution.

The problem for the Energy Department was that these facts did not fit into an established framework and had to be rejected. And it couldn’t keep track of all the facts. In reality, it lived in an isolation chamber, unaware that it didn’t know information that was in the public domain.

It calls to mind the oft-repeated (and oft-bastardized) observation of the philosopher and writer George Santayana, that those who cannot remember the past are condemned to repeat it.

And memory isn’t the problem. Santayana’s formulation presumes we understand the past, or, more modestly, the present. Often, we don’t. And it does no good to tell something to someone who isn’t equipped to believe it.

Nuclear energy is prone to be misunderstood, for several reasons. One is that it is one of the few technologies that is explicitly designed with the worst case in mind. Hence nuclear power plants have advance planning for evacuation of everyone within ten miles, when no such evacuation has ever been ordered. Battery plants that store energy from intermittent sources are not designed with that in mind, although earlier this month, one such plant near San Francisco forced closure of a major roadway and evacuation of 1,200 nearby residents. A nuclear plant that caused such an evacuation would be in deep trouble, but there is no sign that California will stop building battery plants.

Another is the highly unusual way that regulatory agencies treat radiation. It is almost in a category by itself: a threat for which there is no minimum threshold. To draw an analogy: swallowing 350 aspirin tablets of 500 milligrams each will probably kill you. Radiation is regulated on a different methodology: Having 350 people each swallow one tablet will probably kill one of them. The analogy isn’t perfect, though, because aspirin doesn’t exist in nature and radiation does, unavoidably. And the regulatory theory, which has a thin scientific basis, isn’t applied uniformly. Airline flight attendants, for example, accrue more dose than almost all power plant workers, because they spend so many hours above the thick atmosphere that shields people on the ground from radiation from space. Their dose can be substantial. But their exposure is unregulated.

We do worry about the health risks of flying, but generally not because of the radiation.

All of this allows the public, from the BBC to various dedicated anti-nuclear activists, to connect the dots irrationally.

The upshot: The irrational will not inherit the earth. They already own it.

Subscribe now

Jimmy Carter held an engineering degree from the Naval Academy, served on diesel-electric submarines from 1946 to 1953, and worked closely under Hyman Rickover to develop nuclear submarines. But he described himself as a peanut farmer, which must say something about his attitude toward nuclear energy, and his assessment of what Americans wanted in a candidate.

Carter, who died in December and was buried on January 9 in Plains, Georgia, near his peanut farm, made three critical decisions about nuclear energy during his term as president (1977-1981). None of them is much remembered today, outside the industry itself.

1. He went to Three Mile Island (TMI) four days after the melt-down there in March 1979, but he refrained from making any broad statements about the need for nuclear energy.

2. He continued a ban, begun by his predecessor, on reprocessing used nuclear fuel to recover the unburned uranium and the plutonium created during operations for re-use in reactors.

3. And he presided over the early stages of the decline in the United States’ ability to produce nuclear fuel for power plants here and around the world, despite his pledge to do the opposite.

The TMI accident presented an unlikely intersection of technology and politics. A common complaint about political leaders in democracies is that they don’t understand the details of technical subjects well enough to make smart decisions. But in a twist that approaches the implausible, a reactor melted down and the president actually was a nuclear engineer—one of only two presidents who were engineers. (The other was Herbert Hoover.)

Carter, a graduate of the Naval Academy, walked through the control room at TMI and undoubtedly understood what he was seeing. Carter graduated in 1946, before the Academy offered degrees in nuclear engineering, and arguably before nuclear engineers can be said to have existed, but he can fairly be called one, given his role in propulsion reactor development.

And yet at TMI he said very little that added to public understanding; his contribution was mainly in simply showing up, with his wife, Rosalynn, four days after the event, making clear that he was confident enough about safety to make a visit.

It was a time of considerable uncertainty. The Carters and other visitors wore bright yellow booties, fastened tight with masking tape around the ankles, a precaution often used in reactor containments (but not in control rooms, which is what he toured). The booties assure that if any radioactive particles are picked up while walking around, they can be easily contained by stripping off the booties and turning them inside out, like dirty socks, so they can be disposed of as low-level waste.

His statement after touring the place was about as mild as could be imagined. “My primary concern in coming here this afternoon has been to learn as much as I possibly can, as President, about the problems at the Three Mile Island Nuclear Power plant and to assure the people of this region that everything possible is being done and will be done to cope with these problems,” he told reporters. He did, however, appoint a top-notch commission to investigate.

TMI Was Round 2

Even more unlikely, it was Carter’s second nuclear accident. A quarter-century earlier, he was sent by the Navy to lead a team of 23 people at the Canadian national nuclear lab in Chalk River, Ontario, to help clean up a research reactor that suffered fuel damage in an accident in December 1952. In that case, some control rods failed to descend into the core and were mistakenly withdrawn, causing the reactor power to surge to about three times its maximum safe operating power, melting a few of the rods. The United States was using the reactor to test submarine fuel, and wanted it fixed quickly.

It reopened 14 months later.

The story of Carter’s role at Chalk River has grown somewhat in the retelling, as well as the degree of risk involved. Cleanup work was shared among many Canadian nuclear technicians and the Americans who came to help, to reduce the radiation dose to any individual, and according to one estimate, the average dose was less than half the limit now set by the Nuclear Regulatory Commission in the United States for power plant workers each year. If the estimate is correct, average exposure was below the dose that, in this country, would trigger consideration of taking protective action for members of the public in a nuclear accident.

At TMI, one helpful part of his statement was that radiation levels were “quite safe for all concerned.” It might have carried more weight if he’d stressed his nuclear credentials, but he didn’t.

There hasn’t been a nuclear mishap in this country that was remotely worth a presidential visit since 1979.

Three Mile Island was a national fixation. Saturday Night Live did a skit about “Two Mile Island,” in which a control room operator brings back lunch for his colleagues and spills a soda on the control panel, triggering a melt-down. President Carter visits and immediately realizes that the plant has been through “the Pepsi Syndrome,” when cola shorts out the control panels. He goes into the containment, and the radiation exposure makes him grow into a giant, “The Amazing Colossal President.”

But the skit gave the public more presidential dialogue than the real event.

Despite not saying much on his visit to central Pennsylvania, Carter may have been the most consequential president for nuclear energy since Dwight Eisenhower, who pledged in his “atoms for peace” speech of December, 1953, to make nuclear technology “a great boon, for the benefit of all mankind.” But Carter’s influence wasn’t very positive for the technology. “There is no dilemma today more difficult to resolve than that connected with the use of nuclear power,” he said in April, 1977, a few days after the TMI accident. Carter cited nuclear energy’s benefit as a domestic energy source that would outlive oil and gas, but also highlighted the risk of the spread of nuclear weapons. He called for research into fuel cycles that he thought were less likely to facilitate nuclear weapons proliferation.

The Early States of a Nuclear Fuel Crisis

In that same statement just after TMI, Carter said he would support increasing the country’s capacity to make enriched uranium. But through his single term and beyond, the government failed to modernize the 1940s gaseous diffusion technology that had its origins in the Manhattan Project. The United States lost its dominance in the enrichment market because it failed to assure international customers that the U.S. would be a reliable supplier. That led the Europeans to develop centrifuges for their own use. Today that is the dominant technology and the U.S. is struggling to catch up, and wean itself off Russian enrichment services.

Carter is one of a series of presidents who led the United States into that problem, but he was probably the one who understood it best. And he made the fuel supply problem worse by banning reprocessing, which meant that all the fuel had to come from virgin uranium.

His successor, Ronald Reagan, reversed the ban on reprocessing, but by that time, the private sector had given up. The utility sector in general, and especially the nuclear industry, does not deal well with sudden shifts in federal policy.

And, ironically, the U.S. Government’s decision to let Europe go its own way on enrichment eventually led to nuclear weapons proliferation, because a Pakistani metallurgist stole the European centrifuge design and took it home, where his government used it to make bomb fuel. In contrast, nobody has used reprocessed fuel from light water reactors to make weapons fuel, largely because it isn’t well suited for that purpose.

President Carter, the Engineer

Carter was unusual for being in politics with an engineering degree; Engineers aren’t well represented in American politics. According to the Congressional Research Service, there are eight in the House and one in the Senate. (Martin Heinrich, D-New Mexico, has a B.S. in Mechanical Engineering.) Representative Bill Foster, an Illinois Democrat, worked as a physicist at the Fermi National Accelerator Laboratory.

Among the fifty governors, two have B.S. degrees in mechanical engineering: Glenn Youngkin of Virginia, and Bill Lee of Tennessee, both strong nuclear backers. The governor of Montana, Greg Gianforte, has a degree in electrical engineering. Governor Chris Sununu of New Hampshire, who supports the Seabrook reactor in his state, has a B.S. in environmental engineering. But the engineers are far outnumbered by the lawyers, consultants, and people who list their profession as “public service” or just plain politician.

And Carter, the engineer, showed some energy savvy and some aspiration in two decisions as president. Faced with the aftermath of the Arab Oil Embargo (which occurred during Richard Nixon’s presidency) and a high bill for energy imports, he declared a campaign to conserve that he called “the Moral Equivalent Of War,” although it was derided by some who noted the acronym, “MEOW.” And he spent $28,000 to put 32 solar panels on the roof of the West Wing of the White House, where they saved oil by heating tap water. This was in the days before solar panels were remotely practical for making electricity. But Reagan, his successor, had the panels removed.

Carter also signed the legislation establishing the Department of Energy, although much of that agency’s responsibility isn’t in energy at all; it is in nuclear weapons.

What Carter’s presidency did not include was any policy helpful to nuclear energy that was based on his unusual background. He clearly understood the essential nature of propulsion reactors, for the Navy’s submarines and surface ships, and must have understood that the technology could be operated to a very high standard. He himself had helped that happen.

And he was in a better position understand nuclear power and its benefits than his successor, Ronald Reagan, who had once worked as a a paid spokesman for a reactor manufacturer, General Electric, and who could be very persuasive on all kinds of topics, despite a lack of technical credentials.

But Carter stressed thrift and personal sacrifice, de-emphasizing material prosperity. He advocated turning down the thermostat, wearing a sweater, and driving slower on the highway. Before the TMI accident and after, Carter did not extend his deep understanding of the possibilities of nuclear to its obvious civilian applications, despite his unique qualifications for doing so.

In hindsight, the Fort St. Vrain reactor was a high-tech tragedy. Nobody was killed or injured, no company went out of business, and there were no mass layoffs, but there was a squandered opportunity that we are paying for decades later. And as we bring new kinds of advanced reactors into the world, we should be careful not to repeat the errors of that episode.

Ceran St. Vrain was a 19^th-century American fur trapper of French descent. He was not a saint; that was just his surname. But perhaps he should be named the patron saint of giving up too soon.

The nuclear project named after him, located in the Colorado town of the same name about 40 miles north of Denver, looked good on paper. It produced steam at 1,000 degrees F, hotter than water-based reactors, which gave it the potential to substitute for coal in industrial applications. It took non-fissile thorium and turned it into fissile U-233, essentially making some of its own fuel.

Its reactor vessel was made of concrete, which was potentially easier to fabricate than the kind used by water-based reactors, which is carbon steel with stainless steel cladding. It operated at lower pressure but higher temperature than a water-based reactor, a desirable combination.

Some glitches turned up during construction but nothing the builders couldn’t correct. And when fuel was loaded, neutron generation was almost exactly as predicted; no small feat in a first-of-a-kind reactor.

It looked like it had strong safety advantages. If it scrammed form 100 percent power, and all coolant circulation was lost after half an hour the temperature was still 500 degrees below the threshold for damage.

The project had top-notch participants; research by the Atomic Energy Commission’s Power Reactor Demonstration Program, Sargent & Lundy as the architect/engineering firm, and EBASCO as the general contractor.

But when it opened, it didn’t run very well. It had pumps to circulate the coolant material, helium, and the pumps had bearings that were lubricated with water. But the water leaked into the system even before the test program was complete. It spread through the reactor, and once inside, it corroded some parts. Helium often leaked too.

Power plant performance is measured with a metric called capacity factor. Compare actual energy production to what production would have been if it ran at 100 percent, 24/7/365. In the early days of pressurized water reactors, the only type of power reactor now operating in the United States, capacity factors were commonly in the 60 percent range, and after decades of operating experience, they now average over 90 percent, but St. Vrain was 14 percent.

Its owners were struggling through a period of economic slowdown—caused, ironically by high prices for another source of energy, oil—and electricity demand had slacked off. Natural gas was available and easy to understand. And the reactor was regulated under the system then common, with ratepayers reimbursing the company for the investment, and state officials were pressuring the owner, Public Service Company of Colorado, to give up.

So, after 18 months they pulled the plug on St. Vrain.

Share

Hindsight

That was in 1988. Fast forward to 2020, and the government, eager to revive the gas-graphite combination, promised a company in Washington, D.C., X-energy, that it would invest more than $1.2 billion to rush the first model into production, under the Advanced Reactor Demonstration Program. Few utilities have much appetite for a first-of-a-kind generator, but Dow stepped up to agree to buy four of them for a chemical plant in Texas that now relies on natural gas to make process steam.

At a time of growing energy demand and growing urgency to develop carbon-free sources, it is obvious that we would have been better off to nurture a project in the 1980s until it reached commercial competitiveness even if the first model had design flaws that made it an industrial lemon. The problem with first-of-a-kind projects is that the costs and the benefits don’t fall on the same people. So, the electricity consumers of Colorado were spared an expense, and we dropped the technology.

Now, we face a different tragedy: for the first time in a generation, there is strong demand for new nuclear, and the industry has no proven, mature advanced reactor ready to go. There is a cliché today that businesses should learn to “fail fast,” but that makes sense only if you draw the correct lesson from the failure, which in this case was to try again with a better helium circulator, and not to fall back on tried-and-true technology that would eventually become obviously troublesome, like fossil fuels.

St. Vrain isn’t the only example of a technology that was deployed early and died too soon. Fermi 1, south of Detroit, followed a similar trajectory.

Fermi-1 was a sodium-cooled fast reactor, a design that gives off high-energy neutrons. Add a blanket of uranium-238, the abundant form of Uranium, and it becomes a “breeder,” a machine that makes more fuel than it consumes, because the U-238 will capture a neutron and be turned into plutonium-239, an excellent reactor fuel. Capture is more likely when the neutrons are moving fast.

Sodium coolant was integral to the design because it was unlikely to interact with the neutrons, unlike water in other reactor types, which slows the neutrons down. Sodium is in many ways superior to water, because it does not want to expand as much when heated; thus, it can be used in a low-pressure reactor.

The plant was a commercial-scale follow-up to EBR-1, an experimental reactor in Idaho.

Like St. Vrain, which came a few years later, Fermi-1 also had operating troubles. The flow of the sodium coolant was interrupted by a blockage, and some of the fuel melted. The reactor shut down and there was no release of radiation beyond the containment building. The damage was repaired and the reactor re-started, although that took four years.

Detroit Edison ran it for another two years, and then shut it down. Its younger siblings, Fermi-2 and -3 are conventional light water reactors.

At both the Fort St. Vrain and Fermi-1, the issues were not so much reactor physics as they were materials science. Today, water does not degrade components very much in reactors, but only because the operators have learned from painstaking experience that tight control of water chemistry is essential. Moving to a different coolant poses new problems. Helium molecules, for example, are very small, so seals are tricky. Sodium poses a different set of problems.

And new problems are likely to turn up in new designs for advance reactors, now moving towards deployment. These will be unforeseen problems with equipment, the kinds of issues that it took years for current-generation reactors to work out. Or there could be difficulties with interactions between materials used in advanced reactors, materials more complicated than the plain old water we use in current designs.

Some of these problems could cause shutdowns that will last for a while, while engineers analyze what went wrong, redesign parts, and fabricate new ones.

Making Better Use of Failure

We could face problems that we haven’t seen before because the industry and the government missed the opportunity to discover the issues last time. They did not extract the maximum value from either St. Vrain, the gas-graphite design, or from Fermi-1, the sodium design, And now TerraPower is building a sodium fast reactor, under the same program that X-energy is using for its gas-graphite reactor.

Natrium has advantages over the builders of Fermi-1, including decades of engineering advances, but again, a reactor for which there is a market now will not be deployed for at least the next several years. Meeting our energy and climate challenges would have been easier if Fermi-1, although uneconomic, had been nurtured for a while longer.

The Dow and TerraPower projects might be better situated than Fort St. Vrain and Fermi were. Dow has deep pockets and doesn’t have a public service commission to criticize it; it wants the reactor so it can market its products as climate friendly and may have more patience than utility. TerraPower wants a product that will “scale,” a working model it can point to so it can sell a few dozen more.

The early problems will give people who never liked nuclear energy an “I-told-you-so” moment. But their verdict, right or wrong, will be premature if it comes when the reactors first start running.

Both companies might do well to follow the pattern of companies that build cars or airplanes; they may expect to lose money on the first few and make it up as the bugs get worked out and their commercial viability becomes clear. This maybe essential when the first project is a full-scale model. Another advanced reactor company, Kairos Power, is taking an entirely different approach, iterating with model after model, demonstrating component after component, before moving to a full-scale machine. But this is atypical for reactor development.

It may also be time for the government to learn a lesson: if you invest in something as complicated as a first-of-a-kind reactor, the endpoint isn’t when construction is finished, fuel is loaded, and operators achieve the first criticality. It’s when the model becomes a commercial success.

Subscribe now

There is good news for everyone concerned about the spent nuclear fuel stranded at scores of sites around the United States: another country, with a democratic and federal system of government somewhat like our own, appears to have found a political consensus on where to put a permanent repository.

Demonstrating anywhere that a voluntary process can work gives hope that the U.S. government can reduce its liability for spent fuel. The size of which Department of Energy (DOE) auditors state in surprisingly vague terms, but appears likely to be over $50 billion, perhaps well over. This year’s estimate is up about $4 billion from last year’s.

The recent success was in Canada, which is following a path that differs from the one that the United States recently started on. Nominally, both countries are pursuing “consent-based siting,” which means finding a willing host for spent nuclear fuel, rather than forcing it on reluctant towns, tribes or states. But the Canadians do three things we haven’t, at least not yet: they have moved the management outside the regular government bureaucracy, they have talked very openly with potential hosts about financial inducements, and they have promised the hosts substantial input in the process. Moving in that direction would improve our chances of success too.

Share

The recent breakthrough is that the Canadian Waste Management Organization, a utility-owned nonprofit, says it has reached agreement with a tribe and a municipality to host a permanent repository. The site is in northwest Ontario, about 130 miles northeast of International Falls, Minn., and is located between the Wabigoon Lake Ojibway Nation, known as WLON, and the township of Ignace. It was one of two sites that were willing and had suitable geology. When completed, the repository’s underground tunnels will cover an area about 2 kilometers by 3 kilometers, or about 1.2 miles by 1.8 miles. Canada’s goal is for an “informed and willing” host. Canada embarked on that process in 2010. The United States is now following the same approach, as the path of last resort, following the stalemate at Yucca Mountain. Congress chose Yucca, 100 miles northwest of Las Vegas, over the objections of Nevada, but the effort was checkmated by Senator Harry Reid of Nevada after he became the Democratic leader in 2005 and then Senate Majority leader. Yucca is still technically the law of the land, but Congress has not appropriated significant money since 2012 to get it licensed.

With political deadlock, Congress followed the time-honored Washington path and ordered a study of what to do next. The Blue Ribbon Commission on America’s Nuclear Future reported back in January 2012 and called for consent-based siting.

In May 2024 the Department of Energy formally launched the process, in a less ambitious way. At the moment, it is looking for a location for consolidated interim storage, not an underground repository. “Interim” storage would probably comprise above-ground casks, or casks buried just below the surface. The casks that would be used for centralized interim storage are the same kind already used around the country filled with inert gas to prevent corrosion. They are steel with heavy concrete shielding and are guarded and inspected. They will be functional for many decades wherever they are. But moving them to a single federally operated site would allow the department to begin accepting wastes from the utilities they signed contracts with in the 80s and 90s. The contracts called for the government to start receiving shipments in January 1998, and the courts have ruled that the government must pay for breaching those contracts, reimbursing the utilities for all their extra costs.

A Multi-Billion Dollar Leak at the Treasury

The audit, in November, estimates the compensation that the government will have to pay until it starts taking the waste, but does not give a date for that event. Twelve to fifteen years is probably the earliest that can be hoped for. It says in a footnote that the government’s liability is “in the range of between $37.6 billion and $44.5 billion,” in addition to the $11.1 billion that has already been paid. The fact that operators are applying to extend their licenses to 80 years means the government’s tab is getting bigger. And the money does not come out of the Energy Department’s budget; it is paid for by the Treasury’s “Judgment Fund,” the same account you collect from if you sue because your car was rear-ended by a government car. The fund is automatically replenished, without a vote by Congress, which has reduced the visibility of the problem.

Opening a consolidated interim storage facility would get some economies of scale. The first casks moved to consolidated interim storage, if such a place can be built, would probably be from the growing number of sites where reactors have been decommissioned, leaving nothing at the site but a cluster of casks. The casks require 24/7 protection, and effectively preclude re-use of the sites, even though the area has been cleaned up to meet Environmental Protection Agency standards. Those standards are meant to protect future generations from drinking or eating contaminated water or food. It requires that the soil be cleaned up to an extremely high degree, so that a hypothetical future family of subsistence farmers, getting all its food and water from the site, even if it drilled a well on the most polluted spot on the site and used that water to irrigate the fields (so that radioactive materials could be absorbed into the food, and then into their bodies) the radiation dose would be so low that it would not pose a significant health threat. The scenario is somewhat fanciful because the United States does not have significant subsistence farming. The plant sites are in places that are unlikely to ever be used for farming anyway because they are prime industrial land.

But no matter how clean it is, re-use for industry or farming is unlikely while the casks are present.

Getting Started

The Energy Department has a long list of tasks ahead, including developing a working definition of what “consent” means and who must buy in. Several localities have already sought to host the interim site, but activists from distant parts of the same state have persuaded governors and legislatures to say no.

To get started, in June 2023 the Department set up 13 regional consortia and gave them each $2 million to initiate discussions about volunteering. That is supposed to be a two-year process. At some point, the department will ask for expressions of interest and will then enter further discussions with a smaller group.

How many communities in the U.S. will volunteer is an open question. Canada initially got 22 expressions of interest. The development in November was the Nuclear Waste Management Organization’s selection of one of the two finalists. Wabigoon Lake Ojibway Nation has not quite agreed to all the details but said it had “voted in a willingness decision/referendum to determine if Nation will progress into a site characterization process.” The vote count was not disclosed. The announcement said that the final project would have to uphold the tribe’s “laws and values.”

It is notable that the other party to the agreement is not the Canadian federal government, but rather a corporate group. Oddly, an enduring relationship and a long-term policy is more likely to come from the private sector than the public, since government policies can change with each election. The demise of Yucca is an example of how government plans can change.

The process requires policy stability because it is so slow. Even if all goes as Canada plans, regulatory approval will not come until 2032, and it won’t open until the mid-2040s

Finland and Sweden have also delegated the process to non-governmental entities, although repositories must still be licensed by government safety organizations, as would a Canadian or a U.S. repository. Finland and Sweden have simpler governmental structures, with a strong central government and local governments. They do not have states or provinces that could block plans despite local acceptance.

Lake Barrett, a former head of the Office of Civilian Radioactive Waste Management, the part of the Energy Department that was developing Yucca Mountain, said in an e-mail, that the countries that have created private-sector entities to manage the waste are the world leaders—Finland, Sweden, and now Canada. The strength of such a waste management organization is that “it is fairly insulated from political swings,” Barret wrote.

“We need to get the U.S. program out of DOE and, in my view, into a non-profit organization, like [Canada’s] Nuclear Waste Management Organization,” he said

The Canadian experience shows that the issue really is political rather than technical. In fact, the very geologic structure that Canada plans to use is not even unique to Canada, although it is called the Canadian Shield. It extends as far south as central Iowa and is more than 2 billion years old.

Barrett and others believe that with U.S. government cost of maintaining the current nuclear waste paradigm running at more than $2 million a day and rising, there is reason for Washington to to offer enough in benefits to a host location and still minimize total costs for taxpayers and electricity users. The key is “to establish a respectful business-like partnership hosting agreement that benefits all parties,” Barret argues.

The Blue Ribbon Commission’s report made a similar point, calling for “a new, independent, government-chartered corporation focused solely on carrying out that program.” The Energy Department, a descendant of the Atomic Energy Commission, with its legacy of Cold War environmental atrocities, starts at a trust deficit when negotiating with tribes and states.

Another reason that the waste issue remains unresolved in the U.S. is the varying level of emphasis that the parties put on the problem. In some places, like San Onofre in California or the Prairie Island Indian Community, adjacent to the reactors of the same name, the locals are adamant that the waste should be moved. In other places, there is not much fuss over the matter. The most politic\ally powerful entities in the game, the utilities, are divided. Those with operating reactors find the dry casks as an annoying add-on, but those with reactors that have been decommissioned are more eager. Advanced reactor developers have other priorities on the table but would welcome progress that would remove a talking-point against deploying new reactors.

Canada has one advantage that strategy changes can’t fix: all but two of its reactors are in Ontario, the province that will host the site. In contrast, Nevada pointed out over and over that it had no reactors. (Although Yucca was intended for both civilian and military waste, and Nevada did host a major weapons asset, the Nevada Test Site.)

But the underlying point, that there are jobs for decades, and lots of local spending, would seem to be a strong selling point here as well.

Subscribe now

• This article has been corrected to clarify that the government’s estimated liability of $47.6 billion to $44.5 billion is in addition to the $11.1 billion it has already paid out in damages.

By Matthew L. Wald

Right after Thanksgiving, mailboxes are stuffed with fund-raising letters from various charities, seeking year-end contributions. What are they raising money for?

You wouldn’t guess from their names, but some of them want to stop nuclear energy. The list includes the League of Women Voters, the American Association of Retired Persons (better known as AARP), the Union of Concerned Scientists, the Bulletin of the Atomic Scientists, and various smaller organizations.

On the flip side, the Nature Conservancy, which sends out a calendar with dazzling flora and fauna, actively favors nuclear energy.

The extent of these organizations’ influence on energy policy isn’t clear, but they can gum up the works by entering the cumbersome licensing and permitting processes at the federal and state level. And some potential builders of reactors don’t like technologies that come with opposition built in.

The League of Women Voters is an interesting case. It is better known for sending questionnaires to candidates for state and local offices and compiling the answers. It does this with scrupulous symmetry, giving equal prominence to Republicans, Democrats, minor party candidates, and independents. That is how it maintains its status as non-partisan.

But that doesn’t mean that it is neutral on issues. Some are unsurprising; it favors protecting citizens’ right to vote, fair drawing of district lines for legislative seats, and transparency in political funding. But it takes positions on a lot of other issues, including limiting access to guns, ending the death penalty, and improving childcare.

And then there is nuclear energy. The League wants to “minimize reliance on nuclear fission.” State and local chapters can oppose licensing for new plants, or, with permission from the national organization, call for closing existing plants. It refers vaguely to risk, and its policies include a lot of contradictory statements; it is concerned about climate change, air pollution, and drilling for hydrocarbons in sensitive locations. Its listed solutions are efficiency and renewable energy.

Share

In August, 2021, the League signed on to a letter to the Democratic and Republican leaders of the House and Senate opposing provisions in the bipartisan infrastructure bill that “would provide billions of dollars in subsidies to aging and uneconomical nuclear power plants.”

“We must marshal our national resources to address structural inequities and injustices that undermine our safety, health, economic security, and sustainability,” said the letter, which called nuclear energy a “false solution.”

“Nuclear power is too dirty, too dangerous, too expensive, and too slow to solve the climate crisis, and the industry is rooted in environmental injustice and human rights violations,” the letter said. The groups called for renewable energy, batteries, and microgrids.

The letter was signed by more than 100 individuals and groups, including several anti-fracking groups, Catholic Charities of the Diocese of Gallup, New Mexico, a Methodist women’s group in New York, Physicians for Social Responsibility, a coffee house in New Orleans, and a speech therapy company in Sarasota, Florida.

Nuclear isn’t the only energy on the League’s hit list; it also dislikes fossil fuels and biomass.

The League’s policy office did not respond to an e-mail asking for details about the reasons for its opposition, or what it would take to make the group reconsider, or whether the league believes that its position is in step with the opinions of its membership.

The AARP says it has over 37 million members, which means it has more members than the Republican party, and almost as many members as the Democratic party. Like the League, it is non-partisan, but it takes positions on a variety of issues, sometimes at the national level and often at the state or local level. And like the League, it doesn’t like nuclear.

AARP worked in Columbus to oppose the 2019 legislation in Ohio to save the two nuclear power plants. It called the program, designed to let the reactors survive the very low natural gas prices of that period, “a new, unfair and unnecessary annual $300 million nuclear bailout tax” of a business that it insisted was profitable. Referring to the companies that own shares in New Jersey reactors, it also opposed legislation in Trenton to help the three reactors in New Jersey. “PSEG, Exelon, and other energy corporations are waging a campaign to increase our electric bills by forcing ratepayers to pay subsidies to increase the profitability of their aging nuclear power plants. “Lower prices are not a problem for consumers,” the group said.

In places with traditional cost-based rate setting, the expense of preliminary planning of a new plant is usually considered a cost of doing business, and thus chargeable to ratepayers, but in Iowa, AARP placed ads opposing legislation that would have let a utility there bill customers for exploring the feasibility of building a cluster of small modular reactors.

In all cases, the organization said it was acting to protect the pocketbooks of older consumers. The result is that our energy supply depends in part on hardware and technology that was new about the time that the AARP members were born. This is not a forward-looking strategy.

AARP also did not respond to an e-mail asking for details.

On the other side, The Nature Conservancy is calling for raising nuclear’s share of total world energy generation to 33 percent by 2050, as opposed to about 7.8 percent that it projects in a “business as usual” case. TNC’s overall goal is climate stability, but it favors a bigger nuclear share partly to avoid watching nature be swallowed by renewable sprawl.

The state of Georgia, it points out, gets 2 percent of its electricity from solar, but expanding that to 20 percent would require 75,000 acres of land.

One of TNC’s board members, Shirley Ann Jackson, is a former chair of the Nuclear Regulatory Commission. TNC supported the ADVANCE Act (Accelerating Deployment of Versatile, Advanced Nuclear For Clean Energy), which Congress passed in July.

Two organizations whose names might hint at support are actually opposed, but do not say so explicitly. The Bulletin of the Atomic Scientists describes itself as “the authoritative guide to ensuring science and technology make life on Earth better, not worse.” It is mostly concerned with weapons proliferation, but it intermittently publishes articles that are highly critical of nuclear energy. Most recently, it published a piece by Victor Gilinsky, a former member of the Nuclear Regulatory Commission, complaining about Congress’ instructions to the NRC, in the ADVANCE Act, to change its mission statement so that it does “not unnecessarily limit the benefits of nuclear energy to society.”

He writes, “Congress wants the commissioners to make clear to safety reviewers that every hour they will take is an hour that society will be deprived of nuclear energy (and someone’s grandmother will sit in the dark).” Another article on the Bulletin of Atomic Scientist’s website decries nuclear energy as “a distraction on the road to climate solutions”

Share

The Union of Concerned Scientists also opposes nuclear energy, but never explicitly makes a blanket statement in that direction. It is a vigorous participant in NRC hearings, and its comments on safety, current reactors, and advanced models, range from critical to dyspeptic.

It has shown some signs of internal strain, though. In a blogpost in 2018 titled “Why We’re Taking a Hard Look at Nuclear Power Plant Closures,” Ken Kimmell, former president of the organization, wrote that “we keep an open mind about all the tools in the emissions reduction toolbox—even ones that are not our personal favorites.” But the Union’s policy remains undeclared war.

In an ambiguous spot is the Audubon Society, which worked hard to try to close down the Indian Point reactors thirty years ago, and opposed nuclear energy for years afterwards. Audubon attributed damage to birds from the radiation releases at Chernobyl and Fukushima. But it now notes, without comment for or against, that various clean energy bills in Congress contain provisions to help nuclear.

In 2021, the Audubon Senior Vice President for conservation policy appeared at a nuclear industry convention. Audubon has opposed some wind development as a threat to endangered birds, but still supports wind and solar. Its transition towards a neutral stance is a step in the right direction.

As with other old-line environmental groups, opinion in the membership may be softening. In September, an attorney for the Orlando Utilities Commission, who identified himself as a 'card-carrying member of the Sierra Club and the Audubon Society,' told the Public Service Commission in Tallahassee that he supported small modular reactors, according to a local news site.

Audubon has opposed some wind development as a threat to endangered birds. Audubon’s transition towards a neutral stance is a step in the right direction.

Is the position of these organizations important?

It’s not the first consideration for a company contemplating building a reactor. At the top of this list is probably the estimates of cost and schedule, and the confidence that is possible in those estimates.

But for utility companies, which are the traditional buyers of reactors, public attitudes probably play a role, and especially the attitudes of vocal, activist groups. Utilities are probably the category that first comes to mind when thinking of the market for new reactors, but, at the moment, they don’t’ seem to be on board. They worry about commitments to novel projects. The over-budget, long-delayed Vogtle project did not help in that regard.

And in the first round of nuclear construction, hundreds of small utilities thought they should invest, to be part of a bold new era of technology. In the 90s and the 00s, though, lots of those companies got out of the business, ceding it to a handful of operators who had accumulated the expertise to operate reactors effectively. It’s not clear how many small companies will jump back in, despite being able to choose from a menu that includes small reactors.

But other parties, less subject to public opinion, can build. Dow is the sponsor of the first Xe-100 high-temperature gas-graphite reactor, at a chemical plant in Texas. Amazon wants energy from that model reactor and is taking an equity position in the builder. Oklo plans to build and own its reactors, so the NRC will have input, but no public service commission will pass judgment on the prudence of the investment.

But to the extent that lobbying, testifying, and writing letters to the editor set the atmosphere in which companies decide about nuclear energy, these organizations aren’t helpful. And if you and your checkbook want to give year-end support to voting rights, issues that affect the elderly, and nuclear non-proliferation, there are plenty of places without this additional baggage.

Subscribe now

By Matthew L. Wald

The world has a crying need for a new class of high-tech products that companies in the United States have designed and are trying hard to sell. But so far, it isn’t happening.

The products are advanced reactors, which could help meet a potential doubling of electricity demand globally without defaulting to coal.

The nuclear export business is going mostly to countries that have active domestic construction programs, with proven products, and have the ability to offer package deals—financing, fuel, spent fuel removal, and extensive construction assistance. Those countries are prowling the developed and developing world, signing memoranda of understanding, and pouring concrete.

In contrast, American firms have traditionally sold a “nuclear steam supply system,” not an entire reactor complex. Some companies, like Westinghouse and GE, also fabricate fuel, but apart from the Department of Energy supplying fuel to research reactors, nobody takes the fuel back.

The world may be excited about advanced reactors designed here, said Craig Piercy, the executive director of the American Nuclear Society, at a recent webinar, but “it still seems like we’re at a disadvantage. We’re bringing a knife to a gunfight, competing against countries that have an entirely integrated nuclear industry.”

And other countries have many recently built reactors suitable for export, which potential buyers can tour. “Until we have a functioning reactor that we can give them and build in other countries, they’re looking to other countries,’’ said Brad Williams, the lead for energy policy and strategic analysis at the Idaho National Laboratory.

In that sense, we ought to turn the old environmental dictum on its head. We can think globally and act locally, but in fact we should be thinking about local steps that will allow global action. The United States’ share of global emissions in 2021 was about 13.5 percent, down by nearly half since 1980 and certain to decline further. Electricity consumption will nearly double by 2050, not counting new demand from artificial intelligence, according to Third Way and the Energy for Growth Hub, which recently updated their map of the global market for advanced nuclear. Globally, 98 countries “could be markets for advanced nuclear power by 2050.” Ten are viable now, and another ten by 2030, the groups projected, and the global market for nuclear could triple by 2050.

Share

Thus, America’s contribution to stabilizing the climate may turn out to be helping the rest of the world reduce its emissions, especially the developing world. But for now, other countries that want carbon-free power are not looking here. As DJ Nordquist, vice president of the Economic Innovation Group, notes in a recent web post, “Russia has 45 MOUs (memoranda of understanding) to develop nuclear in other countries. China has 13 MOUs with a goal to sell 30 overseas nuclear reactors to Belt and Road partners by 2030.”

“Both Russia and now China are poised to clean our clocks on nuclear exports,” according to Nordquist.

If selling your product abroad requires demonstrating construction proficiency at home, the winners are probably China, Russia, and India. China has 31 reactors under construction, which means it has a large cadre of highly-qualified welders, pipe-fitters and technicians, backed up by foundries, metal fabrication shops, turbine and generator manufacturers, and all the other supply chain elements.

Russia is busy at home, and in India, Turkey, Hungary and Egypt. Rosatom recently started pouring concrete for the fourth of a four-reactor complex in Egypt, that is expected to supply about 10 percent of the country’s electricity. Egypt is a country that the United States has tried hard to lure into its orbit. But Rosatom offers a package deal, including fuel and financing and building a product that it already has experience with.

And Rosatom has MOUs in various African countries, the kinds of places that during the Cold War, we assiduously sought to keep in an American orbit, or at least neutral.

Our hemisphere is not exempt. Forget the Monroe Doctrine, promulgated by the fifth president of the U.S., that the country was going to keep Latin America free from overseas influence. Today much of that influence has its roots in commercial relationships and imported technology. When Argentina went looking for a vendor for a fourth nuclear reactor in 2022, it ended up signing a contract with China. It’s the Hualong One model, China’s adaptation of the pressurized water design. (But Argentina is in financial straits, and the timing of the project is not clear.)

Other Competitors

Other competitors may emerge. India has plans for a fleet of 220-megawatt pressurized heavy water reactors, derived from Canada’s Candu design. While China and Russia build big machines for big grids, India is building smaller generators that can be dispersed across its electric system, making up for a shortage of transmission. In that regard, India’s reactors suit many third world countries that are struggling to meet growing electric demand but do not have a strong grid. If you place a large reactor on a weak grid, every time the reactor trips offline, it can cause a widespread blackout.

But nuclear exports are barely on the American agenda. Last month the Center for Strategic and International Studies’ Energy Security and Climate program held a panel discussion called “Powering Progress: Deploying U.S. Clean Technologies in Emerging Economies.” Several panelists used the now-clichéd phrase, “all of the above,” but the above was wind, solar, batteries, heat pumps and geothermal, not nuclear. There was talk of how to make sure that the United States participates in the value chain of global decarbonization, but in a one-hour discussion, nuclear was mentioned only once, by the moderator, as a topic of an upcoming CSIS event.

It's not that this isn’t on the minds of the nuclear start-ups. NuScale Power, for example, notes that its reactor modules can also desalinate water, using reactor heat to boil seawater without carbon emissions or fuel supply problems. NuScale could have pointed out that this was a solution for Texas or California, but its website notes that a four-module plant could meet the water needs of a city like Cape Town, South Africa.

In a rare step towards success, a company in the United States that wants to develop nuclear plants recently signed an agreement with a firm in Ghana for a project that would use NuScale technology. That puts the U.S. in the lead, for now, against Électricité de France, the China National Nuclear Corporation, Korea Hydro Nuclear Power Corporation, and Rosatom, the Russian state monopoly.

If small modular reactors are as smooth to construct as promised, they will fit well into more settings, and may be particularly well suited for markets in countries without a highly sophisticated technology base. This is because they are in the “some assembly required” category, rather than built from scratch. They have two other advantages. Their scale is more appropriate to smaller grids, and their ability to raise and lower their output promptly will be particularly helpful on such grids. On a big, strong grid, operators can cope with local shortages or surpluses by moving electricity over vast areas. On a small grid, that isn’t possible.  

Share

Many of the small modular reactors also sidestep an obstacle that faces conventional light-water reactors: they need outside electricity supplies. Today’s reactors draw power from the grid to run their control rooms, pumps, valves and other vital equipment, so that if the reactor “trips,” the equipment to keep the plant safe doesn’t stop working. The plants have emergency diesel generators, but a principle of good design is that they have strong grid connections. Designs like NuScale, BWRX and the AP300 can lose offsite power and still shut down safely, with no overheating from the residual heat production in the core. That makes them good candidates for remote areas that have local demand but limited connections to the outside.

Big Reactors Might Work

There are other routes to success. Reactors of a traditional size, but with more modern designs, are attractive in certain contexts, notably in Europe. Westinghouse may succeed in selling more copies of the AP1000, the design used for the new twin reactors in Georgia, although the delays and cost overruns in that project do not make for good product advertising. Poland is spending significant resources on the idea, and Bulgaria and Ukraine are interested.

The United States notably lost out to South Korea on the contract to build the four-unit Barakah plant in the United Arab Emirates, which now stands as a kind of benchmark for cost and schedule. But the UAE hired several American experts in reactor construction and operation, showing that this country still has expertise to offer.

But sales of any kind of reactor need support from the U.S. government. Countries that want civil nuclear energy have to sign agreements that foreswear military uses. China, notably, does not require such pledges. And financing, which China also provides, is also a big issue for the U.S.

The federal government is only intermittently concerned with promoting high-tech exports of any kind. Among the symptoms: Congress let the charter of the Export-Import Bank expire in 2015, and the five-seat board of directors dwindled to two members. The bank loans up to $100 million to foreign entities so they can buy goods manufactured here, but without a quorum, the loans are limited to $10 million each. Boeing, the U.S. Chamber of Commerce and the National Association of Manufacturers, among others, lamented the lack of a quorum.

The World Bank and the International Finance Corporation could help with loans to help Western companies export reactors.

The World Bank does not finance nuclear plants, partly because of opposition from one large shareholder, Germany. Indirectly, it still finances coal projects.

Exports are front and center for some policymakers in Washington. The ADVANCE Act, Accelerating Deployment of Versatile, Advanced Nuclear for Clean Energy, signed by President Biden in July, allows the Nuclear Regulatory Commission to establish a “Nuclear Reactor Export and Innovation Branch,” but does not require it to do so, and thus far it has not. The chairman, Chris Hanson, said that the agency is already carrying out many of the functions that such a branch would take on. The agency says it is “generally ready” to license the export of non-light water reactors.

But pre-requisite for exporting them is licensing them for use here, and building them. The NRC is still struggling to establish an advanced reactor licensing framework.

“At the end of the day,” said Amy Roma, a nuclear expert at a Washington law firm, Hogan Lovells, speaking at the American Nuclear Society forum, “for a lot of U.S. origin technologies, we still come back to, ‘please build it in your country first.’”

Subscribe now

By Matthew L. Wald

Nuclear energy doesn’t usually figure prominently in Presidential elections. It doesn’t rank high on the list of concerns for most voters—like inflation, reproductive rights, or managing the Mexican border—and a candidate who promises to get more reactors built won’t necessarily win a lot of extra votes. On the other hand, there are votes to be lost, among the “we’re-all-gonna-die” anti-nuclear crowd that still turns out at demonstrations now and then.

The nuts and bolts of implementing laws on the books that would help nuclear energy–that is, the administration’s actual business of administering–may be a more important issue.

Neither Kamala Harris nor Donald Trump has said a lot about nuclear energy. In 2020, the  Washington Post attempted to list the position of each candidate in the Democratic primaries on nuclear energy. It put Harris in the category of “Unclear/no response.” As a Senator, Harris voted against the Nuclear Energy Innovation and Modernization Act in committee, citing safety concerns about the San Onofre reactors. More recently, at a September 25^th campaign event in Pittsburgh, Harris listed nuclear among other clean energy technologies.

Trump’s position is also vague. In an interview with Elon Musk in August 2024, Trump confused nuclear power with nuclear weapons. His campaign website states:

 President Trump will support nuclear energy production, which reached a record high during his administration, by modernizing the Nuclear Regulatory Commission, working to keep existing power plants open, and investing in innovative small modular reactors.

It also calls for domestic uranium mining. But carbon dioxide emissions are not a factor; Trump often says that human-caused climate change is a “hoax.”

There are reasons that each candidate should like nuclear. Harris may like it as part of a climate program, and Trump as part of a nationalistic drive towards energy independence, although the United States has largely achieved this with fracking.

Senator J.D. Vance, Trump’s running mate, has acknowledged “all these crazy weather patterns,” and said, “if you really want to make the environment cleaner, you've got to invest in more energy production. We haven't built a nuclear facility, I think one, in the past 40 years.” Governor Tim Walz, Harris’s running mate, favored lifting Minnesota’s moratorium on new reactors.

But effective government is different from attempts at public persuasion. The administration of government programs, especially government contracting for procurement programs and subsidy programs, is governed by a welter of laws and procedures. There are opportunities for both expediting and slow-walking the process. Only time will tell if the next administration is up to the task of modernizing the U.S. nuclear sector for the 21^st century.

Share

Once Upon a Time, on the Campaign Trail

But there is not much indication that either candidate is enthusiastic about nuclear.

The last time that nuclear energy figured prominently into a presidential campaign was in 2008, when Senator Barack Obama of Illinois promised Nevada Democrats that he would kill the Yucca Mountain waste repository in exchange for support in his race against Hillary Clinton. He won and he did.

Congress is a different case, and candidates often have something to say about nuclear energy in the areas in which they are running. As the Huffington Post recently pointed out, Democrats running for U.S. Senate in Arizona, Florida, Michigan, and Texas have spoken favorably of nuclear energy, something that is more often heard from Republicans.

But a president doesn’t always have a strong influence over nuclear energy. In the late 1980s, when the Long Island Lighting Company finished the Shoreham nuclear reactor, local governments said it was impossible to meet evacuation requirements and they wanted it shut down. President Reagan, a former paid spokesman for General Electric, which had designed the reactor, worked hard to assure it would open. That didn’t work. The plant operated for a few days of start-up tests and then was decommissioned. New York consumers are still paying the more than $5 billion bill for the project.

There are some policy questions on the horizon. One is the future of the Advanced Reactor Demonstration Program, which received $2.5 billion under the Bipartisan Infrastructure Law. The program was created to pay half the cost of two advanced reactors, and smaller sums for reactors not as close to commercialization. But since that time, the cost of steel, concrete and labor have all gone up. So too has the cost of borrowing money. The industry is hoping that the Energy Department will “re-baseline” the amount that the government will match and that Congress will appropriate more.  

There are some other presidential decisions ahead. An executive order by Biden, now in force, requires that the Federal government’s operations run on clean electricity by 2035, which would create a market for new nuclear. Trump, if elected, seems likely to rescind that order.

But the administration, aside from urging Congress to pass or kill legislation, does more; it also administers the laws that Congress has already passed. And in the last few years, Congress has passed many laws that now require the Department of Energy to issue contracts or write checks to assist nuclear projects, and the Nuclear Regulatory Commission to reform its operation.

Among the initiatives:

HALEU

Most advanced reactors are designed to run on fuel enriched to nearly 20 percent, in contrast to the 5 percent enrichment that is commonly used now. The fuel is known as High Assay Low-Enriched Uranium, or HALEU. But fuel producers have been reluctant to invest in making that fuel because they are not sure that the advanced reactors will be built. So, Congress told the Energy Department to buy the higher-enriched uranium in an intermediate form suitable for various kinds of reactors to get the ball rolling and then sell it to the owners of advanced reactors.

In January, the Department of Energy announced that it wanted proposals for $2.7 billion in uranium enrichment services. It recently issued a list of four qualified contractors. This is a meaningful step forward, but it has yet to award contracts, negotiate terms, and take delivery. The next administration could slow-walk these steps or speed them up. The pace at which the Energy Department issues requests for proposals, evaluates submissions and makes decisions can be highly variable.

Gen 3+ and SMR

Almost all contemporary reactors are known as Generation 3, but there are more advanced models that still use low-enriched uranium and ordinary water but are designed to rely more heavily for safety on natural forces like gravity and heat dissipation instead of pumps and valves. Those are known as Gen 3+. Some of these designs are Small Modular Reactors, known as SMRs.

The Energy Department recently announced that it would accept applications until January 17, 2025 to share in $900 million available for 50/50 matching grants to support such projects. It will have to analyze the submissions, choose among them, possibly defend against lawsuits from disappointed applicants, and negotiate terms. The grants will be milestone-based, meaning that the recipients will have to demonstrate, to the Energy Department staff’s satisfaction, that they have met interim goals.

Money for this program has been authorized but not funded. Congress would have to vote to supply the money, which would be easier with support from the White House. Neither candidate has specifically addressed this question.

NRC Modernization

The ADVANCE Act (Accelerating Deployment of Versatile, Advanced Nuclear for Clean Energy) prompts the NRC to speed up the licensing of new reactors, including those with technologies that it is not as familiar with. The act also requires the NRC to develop a regulatory framework for fusion, issue guidance on licensing micro-reactors, and increase staff.

The NRC is an independent agency and the changes do not appear to require complicated bidding and contracting, as Energy Department mandates do. But it is notoriously slow to modernize. The commission would probably do better at modernizing if the White House rides herd on the commissioners, pushing, for example, for a workable licensing framework for advanced reactors. 

And, with one of the five commission seats becoming vacant every June 30^th, the next President will have to decide which nominees to back. Currently, there is one vacancy. The party that holds the White House designates the chair, and usually has dibs over three of the five seats.

With one exception—a rogue chairman—the White House has historically left the NRC to manage its own affairs. It isn’t clear that a Harris administration would break that pattern. And what Trump would do is even harder to predict.

Who Will Do the Work?

Laws are sometimes harder to implement than to pass. For one thing, it takes an agency that is fully staffed with competent bureaucrats—a real challenge.

Although Trump is proposing to move large numbers of civil servants into a category where he could dismiss them easily, a less obvious problem is filling top jobs that are already in the President’s purview. The Partnership for Public Service and the Washington Post track 817 important jobs that are filled by the President, with Senate confirmation. By their count, in Biden’s first six months, he nominated 304; Obama nominated 348 and Bush nominated 308. Trump, in contrast, nominated 213.

Anecdotal evidence is that lower-level jobs, many not subject to Senate confirmation, were filled more slowly in the Trump administration than in those of the presidents who preceded him or followed him.

Trump has already opted out of the government’s usual transition process, in which both major party candidates send over personnel who get security clearances and are briefed by incumbent officials on major issues. Some of the Department of Energy’s civilian nuclear energy work involves classified information.

But Democratic administrations have trouble getting things done too, and the obstacles to getting money out the door aren’t confined to nuclear. Congress voted massive stimulus bills in 2020 to keep the United States out of recession as the Covid pandemic set in with the CARES act. But two years later, more than $100 billion hadn’t been spent yet. By April of 2024, nearly $92 billion still hadn’t been spent. This was more than a year after President Biden declared that the Covid emergency was over.

It is also true that some of the demand for nuclear energy, current or future, doesn’t come directly from Washington. The electricity industry predicted a nuclear renaissance around 2008, not because of Congress, but because the price of natural gas had risen to $12 per million BTU. Many plants were proposed, but only two, Vogtle 3 & 4, made it across the finish line, partly because the price of natural gas fell to $2 per million BTU with the commercialization of fracking in shale.

That technique, which has changed the shape of the grid, is based on technologies nurtured by the Department of Energy for years, including supercomputing, directional drilling and 3d-seismic, but this certainly wasn’t a policy decision.

Now, the country is facing sharply higher estimates of load growth. Some of that is from policy initiatives, like subsidizing building owners to switch their heating systems to electric-driven heat pumps from natural gas, oil or propane, or programs to encourage electric vehicles. Some of it comes from the growth of data centers, which is a commercial trend, not a government program.  

And tech giants including Amazon, Google and Microsoft have all announced that they plan to put money into nuclear energy. So has Dow, the chemical company.

The commercial and policy ducks are in a row; an important task for the next president is to get the administrative ducks to line up too.

Subscribe now

Twelve astronauts have ever walked on the moon. As of this writing, just four are still living. The youngest is Charles Duke, who turned 89 on October 3. What have we been doing lately?

Putting a man on the moon isn’t the most important engineering accomplishment of the modern era; that’s probably electrification. But the moon landing is the flashiest, and people often refer to the greatness of a country that put a man on the moon.

But we didn’t put an astronaut on the moon; our grandparents did.

To be fair, the world has made lots of technical progress in the half-century since then, especially in medicine and communications. But there is a problem here. Analysts worry over the Baby Boom generation, a bulge in the population, passing into retirement and then into the years of peak health care needs. There’s another bulge in the system, though, and it’s not people. It’s the infrastructure, the built environment we live in, and rely on.

Some old stuff is famous and not anything to worry about. The Statue of Liberty’s 138^th birthday is a few days after Charles Duke’s 89^th. The Brooklyn Bridge is three years older than the statue.

But while we’re in the neighborhood of lower Manhattan, I’m more concerned with the Holland Tunnel, which will soon turn 100 years old.  Nearly 15 million vehicles a year use the tunnel. Traffic in the tunnel is stop-and-go at either portal most hours of the day, and often at a crawl through the old cast-iron tubes 93 feet under water. Traffic in the slightly younger Lincoln Tunnel and George Washington Bridge, the only other crossings from Manhattan to the west, is similar, and it has been driven up by another infrastructure failure, a fire in 1974 on a rail bridge built over the Hudson at Poughkeepsie, N.Y., in 1998. That bridge is now open only to pedestrians. The result is that rail freight bound for New York City or Long Island from west of the Hudson gets transferred to trucks in northern New Jersey, which then clog the river crossings and the city streets.

But we don’t do infrastructure any more, so getting the money and the political will together to fix the problem has been incredibly slow. Engineers and even bulldozers started work on a pair of new rail tunnels under the Hudson River twenty years ago, but Governor Chris Christie of New Jersey cancelled it in 2010. There is talk of reviving it.

Physically smaller technologies, like next-generation microchips, march ahead, but for the big stuff, there is a stop-start stutter to the idea of adding new infrastructure. It’s fine to add factories that make computer chips, but at some point the health of the infrastructure depends on deploying cement mixers, for highways, bridges, tunnels and runways.

Share

Old, Old, Old

Lots of critical stuff is older than 100 years. According to the Army Corps of Engineers, 55 dams still in use were completed before 1800, the oldest from 1640.

None of these compares, though, to the lacuna in the nuclear industry. We are facing a market and R&D failure. The stars are finally in alignment for nuclear rebirth, with sharply higher demand growth forecasts in the United States, enormous shortages in the developing world, a looming environmental nightmare of human-caused climate change, and a public that now has a receptive view.

What’s missing is a proven, demonstrated product that the industry is ready to sell. Hence the odd stopgap of trying to re-start half-century-old reactors. Palisades, in Michigan, opened in 1973 and shut in 2022, is coming back into the limelight like an old rockstar. Fans are lining up, money in hand. In addition to $1.52 billion from the Department of Energy, it’s also getting a loan of $1.3 billion from the Agriculture Department. Three Mile Island unit 1, now the Crane Clean Energy Center, is also being revived. {link to Double Symbolism piece?}

A comparison with the automobile industry is instructive. It has lean years and fat years, and offers a range of products to fit various market contingencies. Its development cycle is long enough that engineers are designing vehicles that will hit the market at a time that isn’t predictable when the car or truck is on the drawing board. Maybe the market in the 2030 model year will be demanding hybrids, or pure electrics, or big vehicles, or tiny vehicles. Like a restaurant that offers steak, vegan, and sushi, the car industry has something for almost everyone, even if it doesn’t know who will walk in the door and what menu they will ask to see.

And customers who want to buy a 2030 Impala or Taurus or Camry have enough confidence to do so even if what they’re buying is a morphed, modified version of what hit the roads in previous years. Detroit does not bring back the 1975 Dodge Dart, even if it puts the Dart nameplate on a new model.

In contrast, any small modular nuclear or advanced reactor that a utility buys now will be first-of-a-kind, which means the cost and amount of time needed to build it are highly uncertain. Contrast this to plants that burn natural gas. These can be ordered off the shelf, so to speak. They are tweaked from iteration to iteration, but the buyer has a clear idea of cost and schedule.

Adding New Infrastructure is an Epic Slog

Important parts of our electricity infrastructure don’t need replacement, just expansion, and occasional refurbishment. But that’s not going well either. Take the case of the Cardinal-Hickory Creek transmission line. (Power lines are usually named for places nobody outside the electricity industry has ever heard of. The Hickory Creek substation is in Dubuque County, Iowa, and the Cardinal substation is to the northeast, in Middleton, Wisconsin.)

The system for approving transmission, always cumbersome, has gotten more so in places that have switched to an Independent System Operator method of grid operation, the so-called “deregulated” areas. These are not really deregulated, but just differently regulated, and they now cover more than half the United States, electrically speaking.

A few days ago, the line was energized for the first time, thirteen years after the regional grid operator, the Midwest Independent System Operator, ran it through a lengthy analytical process and then approved it. The goal is to pull about 4 gigawatts of power from 160 wind and solar projects (about four big nuclear reactors’ worth, except that the line will seldom be fully loaded) from South Dakota and Iowa north into Minnesota. Sponsors say it will lower electricity bills there by more than the line’s $655 million price. The issue isn’t money.

It took only three years to build the 102-mile line. The project needed approvals beyond the grid operator, from, for example, the Wisconsin Public Service Commission, the Iowa Utilities Board, the United States Department of Agriculture’s Rural Utilities service, the Army Corps of Engineers, and the Fish & Wildlife service. Then there were the legal challenges, by the National Wildlife Refuge Association, the Driftless Area Land Conservancy and the Wisconsin Wildlife Federation. These worked their way up through the courts.

The striking fact about the Cardinal-Hickory Creek line is that the delay in that project is not unusual. Grid additions should be booming right now, if we are going to build wind and solar farms where the wind blows and the sun shines best. Instead, in the 2020s, the system is adding only about 20 percent as much new transmission capacity as it did in the first half of the 2010s, according to a study by Grid Strategies, a company that consults mostly for the wind industry. The study found that from 2010 to 2014, the system was adding 1,700 miles a year, and from 2015 to 2019, 925 miles a year. In 2023, it added just 55 miles. (The whole system, built over more than a century, is about 160,000 miles.)

This isn’t remotely compatible with President Biden’s noble idea of decarbonizing the grid by 2035.

As Dan Yurman, the energy blogger, says, the problem with getting anything done is that the system has too many umpires, and not enough players.

One reason that the grid and the rest of the energy infrastructure is so hard to modernize is that we are trying to solve national problems with local decisions. This problem isn’t unique to the grid; another case that makes our energy system more expensive and less clean than it should be is management of nuclear waste.

On the grid, inadequate transmission means that the system operator can’t use generators in price order, instead, technicians run more expensive generators while cheaper ones lay idle, because of congestion. It also means better sharing of generating resources in a way that minimizes the need to build expensive generators and reduces the possibility of running short in any given area.

The way the electricity system is now regulated, with a division between the generators and the companies that provide electricity to homes, business and factories, implies that energy will be competitively sourced from many different places. That means it will need to travel more miles on a bigger grid. And if those sources are wind and solar, the producer has to go where the resource is, not simply where rail cars full of coal or pipelines full of gas can make deliveries.

And the electric system is being told to take the place of pipelines, as we try to electrify transportation and home heating. Energy formerly transmitted through pipelines for natural gas, diesel or gasoline will now go over high-voltage transmission lines. It’s not just that the system is old, it’s that the system wasn’t designed for the modern paradigm.

But while the transmission system isn’t keeping up, the problem is worse on the generation side. Running the system on the technology and construction legacy of our grandparents only works if the load being served isn’t changing.  But it is changing.

Consider the idea of running data centers on small nuclear reactors. The data centers are being built today, and the projection is that in the late 2020s there will be a lot more of them. When can we break ground on ten or twenty small modular reactors? Maybe the mid 2030s, if we’re lucky. We dropped the ball on developing new products and today we’re simply not ready.

It’s hard to know what the 2030s and the 2040s are going to be like, but it’s a good bet that they will be tumultuous. A prudent, conservative approach is to hedge our bets. And that’s going to mean doing something new. We didn’t get where we are by not doing something new.

And it is clearly time to think about what infrastructure we will leave to our grandchildren.

Subscribe now