E2: The nuclear price tag

Jacopo Buongiorno: There is a set of countries that have been building nuclear power plants continuously for the past 20 or 30 years, that continue to build nuclear power plants cost effectively and on schedule. In fact, the irony is that the Chinese have built American designs better and faster and cheaper than in the United States. 

Madison Goldberg: If you’ve kept up with energy politics in the U.S., you know that the conversation can be pretty divisive. But lately, nuclear energy has been getting support from both parties.

Joe Biden: That’s why we’re keeping existing plants open, restarting shuttered plants, and building America’s first new nuclear plants in decades.

Donald Trump: I would say we’ll be second to none, because we’re starting very strong. But we—it’s time. It’s time for nuclear.

MG: So with all this talk about expanding nuclear energy in the United States, we figured now is a good time to dig into it. I’m your host Madison Goldberg, and you’re listening to Ask MIT Climate.

Now, we ran an episode on the nuts and bolts of nuclear back in Season 2—you can find that in our show notes—so today we’re going to tighten our focus a little bit.

As momentum grows to build more nuclear power in the U.S., we’re asking: Is that actually doable? Because the thing is, while nuclear power is clean and reliable, it’s also costly. And its supporters in both parties are banking on those high costs coming down. 

Today we’ll be hearing from the same guest who joined us in our second season.

JB: My name is Jacopo Buongiorno. I'm a professor in the department of nuclear science and engineering at MIT. I spent all my career in analyzing and researching nuclear energy technologies.

MG: So, you might know that a nuclear plant creates much less climate pollution over its life than a fossil fuel plant. And Professor Buongiorno points out another useful aspect of nuclear power, as we work to clean up our energy system.

JB: It’s not intermittent. It's reliable, right? It's 24/7.

MG: Which means it goes well with wind and solar, since their output rises and falls throughout the day. But nuclear’s high price tag has undercut those benefits, at least in the U.S.

In fact, it’s hard to even pin down how much it costs to add more nuclear power, because the U.S. has only built three new reactors this century.

JB: The U.S. just finished building two units, two large reactors at a site called Vogtle in Georgia. And sadly, they have not been super successful, in terms of delivering these plants on schedule and on budget. The cost came in over twice as expensive as they promised, and the schedule slipped by quite a bit.

MG: And those construction costs are super important.

JB: The cost of nuclear electricity is primarily the cost of the plant. When you look at the cost breakdown for nuclear, roughly 65 to 70% is the construction of the plant. Then you have another 20% which is the operation and maintenance. And then the remaining less than 10% is actually the fuel. If you look at a natural gas plant, it is almost like the mirror image. The plant is fairly cheap to build. Then what you pay for is mostly the fuel. It's the gas itself. 

MG: In a way, this is good for nuclear. After it’s built, its day-to-day costs are low and predictable: the uranium market doesn’t cause big swings in energy prices the way we see with oil and gas. The fleet of older nuclear reactors in the U.S. has no trouble competing in electricity markets.

But if the cost of nuclear is all frontloaded, and everyone agrees that cost is too high—who would pay up for a new nuclear plant? 

JB: For 20, 25 years, we didn't build anything, and the industry very much atrophied, right? So their ability to manage large construction sites, a complex supply chain, evaporated because there was no market.

MG: Meanwhile, in some countries where the nuclear industry is more active, including China, Russia, India, and South Korea, costs are much lower. So as policymakers in the U.S. try to bring our own nuclear industry out of hibernation, it’s worth asking: What would it actually take to do that?

We’re at MIT, so you might expect us to say the answer is new technology. I mean—it’s in the name, right? But Professor Buongiorno says people should maybe temper their hopes about new designs bringing down costs.

JB: It actually, it pains me to say, because, as a nuclear engineer, you know, we love to, we love to, come up with new designs and new materials that will do a better job, and how about if we change the fuel form? Fuel is always uranium, but you know, you can have uranium dioxide in the form of pellets, you can have metallic uranium, you can have TRISO particles. There are so many different flavors of fuel, and every time you move away from what we have been using, and we know it works, and it's cheap, your cost will go up.

MG: In fact, he says some of the new designs aren’t really meant to produce cheaper power. For instance, maybe you’ve heard of microreactors. These are little, factory-built reactors that provide around 1% of the power of a traditional plant.

JB: It sounds like it's much better to go with the tiny little things because, you know, it takes less time to build and it takes less money to build. But the problem with the tiny machines is that the cost per unit energy is actually going to be higher than the large machines, right? So there is a trade off here.

You probably are not going to connect them to the grid, because that grid is a pretty cutthroat market. You really need to deliver at the least cost to be part of the game. And these reactors will not. So these microreactors are really meant to be co-located with the energy user. So generate the energy here, use it right next door. And so they're very good for things like mining sites or a desalination plant.

MG: Now, there’s another category of new designs you might have heard about: “small modular reactors,” or SMRs. Let’s put aside the “small” for now—unlike microreactors, even one of these is a serious power plant in its own right. Let’s focus on the “modular.”

JB: Traditionally, nuclear plants are built at a big construction site. You bring raw materials and you bring individual components to that site, and then you literally assemble the plant at the site. Modular construction is different. Modular construction means you prefabricate large structures and systems in factories, and then you ship off those modules, and then you connect them at the site.

MG: This is the same approach that has helped make solar energy so cheap: Solar panels can be made really efficiently in factories. Although, to be clear, there are limits to how far this can go with nuclear power.

JB: The better analogy, more than the solar panels, would be either large trucks or jet engines, something of that, you know, that level of complexity, with a lot of moving parts and instrumentation and control stuff.

MG: Still, if enough of these reactors get built, you could imagine that factory construction might bring costs down. It’s controversial, though. Some SMR designs use more expensive types of fuel. And every nuclear plant has some fixed costs, like security staff and certain big components. The smaller the reactor, the more those costs weigh on every bit of electricity it provides.

We won’t really know how this shakes out until SMRs start rolling off the factory floor. So far, China and Russia have built one each. In North America, the first commercial-scale unit recently started construction in Canada.

JB: So we'll see how it goes, right? They're building one unit, which, of course, is astronomically expensive, and they're going to test that one reactor, and if it works well, then you can start amping up production and making it efficient.

MG: So if new designs don’t get to the heart of nuclear’s cost problems… is there anything that could?

Well, first of all, to build a nuclear plant, you need investors. And the fact that most of the costs come right at the start has been a big barrier to attracting them.

JB: Because it's so capital intensive, that construction project, that you need to write a big check at the beginning. So if the industry is to go out there and raise money from Wall Street and other private investors, they're going to want to know, when am I going to get my money back? And is it cost competitive or not?

MG: It’s a chicken-and-egg problem. Nuclear plants need to show that they can be built cost-effectively, so they can raise money, so they can get built.

There might be some movement on this front. There are new, motivated customers for nuclear power—especially AI companies, which need lots of 24/7 electricity and often want that energy to be clean. The U.S. also offers tax credits for nuclear power, and the Trump Administration has been brokering loans for new reactors, with a mix of public and private funds.

But say you’ve got your financing. The next step is to actually build the power plant—which is a super complicated machine.

JB: A large nuclear power plant requires an enormous amount of materials and components. You know, concrete. Most, of course, is piping, wires, heat exchangers, pumps, vessels, instrumentation and control. When you add it all together, you're looking at tens of thousands of components.

MG: Some of these are really specialty parts. If something is needed to run the plant safely, then—for very good reason—it has to meet strict manufacturing standards.

JB: So because the U.S. did not build nuclear plants for such a long time, there was little incentive for these manufacturers to keep the nuclear quality assurance, because there was no market really, right? And so if you're at the mercy of one supplier for that particular component, they can name their price, right? But if you have three or four, you can shop around. And so we're not there with nuclear, and that's part of why nuclear is expensive.

MG: Professor Buongiorno says that, if it were clear the U.S. is building a large number of plants, that would probably entice more companies to make these specialized parts. But he adds that it also matters what kinds of plants get built: Manufacturers and construction crews want to know that each reactor is going to use the same plans and materials.

JB: We can't afford, going forward, to have 25 different designs, and everybody is doing their own thing. If you look at the successful nuclear power plant deployment programs overseas, in China now they have essentially two plants that they're standardized and they're building, okay? But not five, not six. They're making the same thing over and over and over again. And every time you do it, you get better, and the cost goes down.

MG: Of course, there’s a broader conversation to be had about the role we want nuclear to play in our power mix, and cost isn’t the only thing to consider.

But it’s important. And for those who want to see nuclear power expand, Professor Buongiorno says now is the time for the industry to prove it can make that power affordable.

JB: You know, this has been my life for 30 years, and I've seen so many ups and downs. But I got to tell you, I never seen so much support as we have now. Government support, attention by investment. Now it's, you know, the ball is very much in the court of the industry. They got to deliver. It's now or never. The real issues are financing, supply chain, and the ability of these companies to execute projects. You get those three right, then you're going to have the floodgates open, and we're going to have a ton of new nuclear capacity also in the United States.

MG: Ask MIT Climate is the climate change podcast of the Massachusetts Institute of Technology. Aaron Krol is our executive producer and the writer of today’s episode. David Lishansky is our sound editor and producer. Michelle Harris fact-checks our episodes, and the music is by Blue Dot Sessions. And I’m your host and associate producer, Madison Goldberg.

Thank you to Professor Jacopo Buongiorno for speaking with us, and to you for listening! You can find more Ask MIT Climate, and all our other climate learning resources, at climate.mit.edu. We’re also on TikTok, Instagram, and Youtube @askmitclimate. And we love hearing from our listeners; no question is too big or too small. Send us your climate question—whether it’s micro- or gigawatt-scale—at askmitclimate@mit.edu.