clock menu more-arrow no yes mobile

Filed under:

The nuclear industry’s big bet on going small

Small modular reactors might be the nuclear industry’s best hope for a renaissance.

A mockup of NuScale’s power module.
NuScale currently is the only company with a small modular reactor design that has cleared regulatory benchmarks.
NuScale Power
Umair Irfan is a correspondent at Vox writing about climate change, Covid-19, and energy policy. Irfan is also a regular contributor to the radio program Science Friday. Prior to Vox, he was a reporter for ClimateWire at E&E News.

This year, the US nuclear energy industry did something it hasn’t done in more than 30 years: It started and completed new nuclear capacity as the first of two reactors located at Plant Vogtle in Georgia powered up.

At full tilt, the reactors would produce 2.2 gigawatts of electricity, turning the Vogtle facility into the largest single power producer in the US.

Initially approved in 2006, site preparation began in 2009 and construction of the actual reactors started in 2013. The original price tag was $14 billion, backed by $12 billion in loan guarantees from the US Department of Energy. But the cost of the project, which missed its planned activation in 2016, ultimately ballooned to more than $30 billion. A similar nuclear project in South Carolina was eventually canceled due to cost overruns, but still stuck the state with a $9 billion bill. In 2017, Vogtle’s manufacturer, Westinghouse, filed for bankruptcy. And as construction proceeded in Georgia, six other reactors shut down across the US due to age or rising operating expenses.

It’s a disheartening story at a time when advocates say the case for nuclear energy is the strongest it’s been in ages. The Biden administration has set a target of 100 percent clean electricity by 2035 and reaching net-zero greenhouse gas emissions across the whole economy by 2050. There is a long way to go. Right now, the share of clean energy on the power grid is 41 percent, and nearly half of that currently comes from nuclear power. The Department of Energy estimated that the US would need upward of an additional 770 gigawatts of new clean electricity generation to reach net zero. “[N]uclear power is one of the few proven options that could deliver this at scale,” according to a March report from the Energy Department.

So the US, and the world, would need vastly more nuclear energy to power a cleaner economy. But with such high costs and wariness around conventional giant reactors, the global nuclear industry is increasingly betting that the best way to reach those big targets is to go small.

Small modular reactors (SMR) have emerged as one of the most popular approaches for the next generation of nuclear power plants. Rather than designing giant, custom-crafted reactors at sprawling power plants that churn out gigawatts of electricity, industry stalwarts and startups are now developing smaller, factory-built atom splitters. In theory, they could be deployed cheaper and faster than current designs, meeting existing needs for power while filling new niches in the economy like hydrogen production. The hope is that SMRs could bypass or overcome some of the biggest obstacles to nuclear energy and the transition to clean energy.

Nuclear energy firms around the world are gearing up to test that theory. China has already powered up a plant with two 250 megawatt reactors. Russia has built a floating nuclear power plant producing 70 MW of electricity for a remote Arctic town. Four more SMRs are under construction in Argentina, China, and Russia. In 2022, Oregon-based NuScale earned the first SMR design approval by the US Nuclear Regulatory Commission.

And more are on the way. According to the International Atomic Energy Agency, there are around 50 SMR designs under various stages of development, from the drawing board to construction. In May, Westinghouse revealed its own plans for a 300 MW SMR. This year, EDF, the national utility of France — where nuclear makes up 70 percent of the electricity mix — created a subsidiary called Nuward to develop a 170 MW reactor. Rolls-Royce created a subsidiary to build SMRs in the United Kingdom.

Governments are stepping up their support as well. The Inflation Reduction Act folds in tax credits for zero-carbon energy sources, including nuclear. A bipartisan group of US senators released the ADVANCE Act in April, which would make it cheaper and easier for SMR developers to apply for licenses from regulators.

The big barrier is that the business landscape for energy in general and nuclear in particular is more challenging than ever. Nuclear energy has seen its operating costs rise over time while renewable energy prices continue to fall. And with interest rates rising during the fight against inflation, borrowing money to build new designs is becoming more expensive. Going small might be the nuclear industry’s best chance to overcome the longstanding problems that have stalled nuclear energy for decades — but it’s still a long shot.

How NuScale’s SMR design works

NuScale Power, with a market cap of $530 million, has received more than $600 million in grants from the US Department of Energy since 2014 to support the development of its small modular reactors. It began building components for its first power plant earlier this year in South Korea.

“SMRs are no longer an abstract concept,” said Assistant Secretary for Nuclear Energy Kathryn Huff in a January press release. “They are real and they are ready for deployment thanks to the hard work of NuScale, the university community, our national labs, industry partners, and the NRC.”

The design approved by US regulators uses 12 light-water reactors in a plant, each producing 50 MW, much smaller than most conventional reactors that range in the hundreds of megawatts. But the company has since shifted to a larger power output design. NuScale has now submitted a proposal for a higher capacity module producing 77 MW in a six-unit configuration based on results from early tests.

“We saw an advantage to uprating the power,” said Jose Reyes, chief technology officer for NuScale. “As we learned more about the performance of the machine itself, we realized we had quite a bit of margin.”

The new design will require another round of checks and approvals and has pushed back the timelines for NuScale’s projects. In both designs, the reactor module is about 15 feet in diameter and 76 feet tall. Each design would produce about 462 MW of electricity in total.

One key advantage to SMRs is that the reactors would be built at a factory before being shipped to sites around the world. That’s unlike conventional reactors, which are typically built on-site, albeit with large, pre-fabricated parts, which means they require specialized construction equipment and transportation infrastructure. Since they’re tailored to a specific customer, the builders can’t easily apply lessons from one plant to another, making it hard to achieve economies of scale and adding to the construction and operation costs of conventional nuclear power.

While NuScale’s reactor design is standard, the plants they’re installed in can be scaled up or down in terms of capacity by adjusting the number of reactor modules. They can also be built in more remote locations, unlike most conventional nuclear power plants that require a power input from the grid to run auxiliary systems like cooling or on-site backup generators.

Like many of the new generation of nuclear reactor concepts, NuScale’s reactor was designed with passive safety systems that can automatically stop it if something goes wrong. Unlike conventional reactors, “we don’t require any integration to the grid for safety,” Reyes said, reducing the risk of outages and larger potential failures like meltdowns. Another perk of using a handful of small reactors at a plant rather than a couple big ones is that when a reactor is down for refueling or maintenance, a smaller chunk of power goes offline.

Cut-away views of a conventional nuclear reactor and a small modular reactor.
The basic design differences between a conventional nuclear reactor and a small modular reactor plant.
US Government Accountability Office

NuScale is now exploring deployments in North Carolina, Wisconsin, and Missouri. Its first US plant, called the Carbon Free Power Project, will be built at Idaho National Laboratory and is scheduled to reach full production by 2030, generating 462 megawatts of electricity to be sold to a consortium of utilities. NuScale is also working to build its plants in countries including Romania, South Korea, and Poland.

Another advantage of NuScale’s design over conventional nuclear is that it can ramp power up and down more readily and has the built-in capability to follow power demand. “We can go from 20 percent power to 100 percent power in 96 minutes,” Reyes said. Conventional nuclear power plants are optimized to run at a high, steady rate, which makes them a poor fit as intermittent wind and solar power plug into the grid, bringing sudden crests and dips in electricity production.

The market is stacked against nuclear power

SMR developers may be going small, but they still face many of the same big headwinds as other energy companies, including supply chain disruptions, inflation, and rising interest rates that make financing and building more expensive. And while some SMRs are built on existing nuclear power designs, they are still first-of-a-kind in terms of their smaller scales and how they work together in a plant. Companies have to learn how to transport a nuclear reactor rather than building one on site, for example. This creates the potential for cost overruns as companies run into the usual initial snags.

NuScale has already revised its cost estimates upward for its first plant, the Carbon Free Power Project. It was initially projected to produce power at $58 per megawatt-hour, but has now risen to $89 per megawatt-hour as costs of materials like steel have grown and interest rates surged.

The overall price tag has grown to $9.24 billion from an initial estimate of $6 billion and could still go higher, though including prior cost-sharing between the Energy Department and NuScale brings the cost down to $7.97 billion.

“The biggest issue that the nuclear [industry] has to tackle is the topic of risk of that investment,” said Bill Lacitiva, a partner at McKinsey who leads its nuclear energy work. While the upfront costs may be lower than conventional nuclear for utilities, SMRs will still need years, if not decades, to pay back their investment, raising worries that SMRs could fall into the same pits as their bigger brethren. “The history has not been positive in that respect when investors look at this,” Lacitiva said.

SMRs also have to contend with many of the high fixed costs that come with nuclear energy. Complying with nuclear energy regulations is expensive and limits where developers can build plants. Though some SMR designs incorporate new safety features, regulatory agencies have to readjust their processes to evaluate new technologies, and that on its own can be tedious.

It also takes a specialized, highly trained workforce to build and operate nuclear facilities, but the industry is already struggling to retain personnel as earlier generations of workers retire. And it’s hard to recruit new staffers when the country goes decades between building new reactors. Nuclear fuel also requires specialized processing. And most countries, including the US, still don’t have a permanent place to store nuclear waste.

At the same time, many electricity systems in the US have shifted to competitive markets, where power plants bid to provide electricity at the lowest possible cost. Grid operators can buy electricity a day ahead or in real time. When wind and solar power are available, they’re often the cheapest source of power and can undercut more expensive nuclear energy.

Nuclear does have a valuable trait in that it can produce a steady stream of electrons without emitting greenhouse gasses, providing reliable baseload power. But in some markets, it’s hard to reward this function, and as long as there isn’t a price on carbon, coal and gas plants can often perform this task cheaper.

That’s why SMR developers like NuScale are also pitching their plants as a way to power industrial processes, to produce hydrogen or to desalinate water, creating other revenue streams.

Governments may have to step up their support as well. That includes taxing greenhouse gas emissions, smoothing over the regulatory process, and providing more backstops to assure skittish investors. “The successful long-term deployment of SMRs hinges on strong support from policy makers and regulators to leverage private sector investment,” according to the International Energy Agency’s 2022 report on nuclear power.

But given the need for more energy and fewer greenhouse gas emissions, the potential for nuclear energy is hard to ignore. The US currently has about 95 gigawatts of nuclear capacity, much of it from reactors that are decades old and inching toward the ends of their lives, so the US will need to begin constructing more nuclear power just to maintain its 47 percent share of carbon-free electricity on the grid. As everything from cars to stoves to furnaces switches to electricity, power demand is poised to grow. And if nuclear is aiming to dethrone natural gas and coal — currently 60 percent of US electricity — it will take even more. “An aggressive case ... could be more than 300 gigawatts total of nuclear needed, which is roughly 250 gigawatts of new [additional power],” Lacitiva said. “Those are massive numbers, and construction on a scale that at least the nuclear industry has never seen.”

All these hurdles may be too tall for SMRs to vault on their own. For nuclear to truly clear the bar, the industry will need a decades-long commitment from policymakers to see this build-out through, including financing, research and development, and a coherent climate strategy that favors cleaner sources of energy over fossil fuels. The technology has to get much cheaper too. Nonetheless, SMRs could be a crucial tool to help fix one of the biggest problems the world faces.

Correction, July 11, 1:10 pm: A previous version of this story misstated the status of Plant Vogtle’s reactors and the project’s cost estimate. Reactor unit 4 at Plant Vogtle is still under construction and unit 3’s commercial operations are currently delayed, and the newer cost estimate for the project is $30 billion.

Sign up for the newsletter Sign up for Vox Recommends

Get curated picks of the best Vox journalism to read, watch, and listen to every week, from our editors.