Momentum is building for a nuclear energy revival as power demand and climate concerns soar. Can the industry overcome its long-term pitfalls?
If a “foolish consistency is the hobgoblin of little minds,” as Ralph Waldo Emerson wrote, then ecologist Stewart Brand must be considered an unusually expansive thinker. Once a vocal opponent of nuclear power, the 86-year-old creator of the Whole Earth Catalog experienced an intellectual volte face with his 2009 book in which he argued for “ecopragmatism,” including the use of nuclear power to replace coal and address climate change.
It’s further evidence of the complexity of the politics around nuclear power. On one hand, there is an inchoate fear of radiation and nuclear waste among the public, not helped by the Chernobyl, Fukushima and Three Mile Island catastrophes. For those of a certain age, the Jane Fonda movie, “The China Syndrome,” which depicted a near-nuclear plant meltdown, didn’t help, either.
For investors, the fear stems from projects like the V.C. Summer nuclear plant near Jenkinsville, South Carolina, which burned through $9 billion and was never finished. South Carolina rate payers will be helping pay off the project for many years.
Now, however, concerns over global warming burn brighter than fear of nuclear for many leaders desperate to meet surging demand for power. “The vast majority of (environmental groups) see climate as having changed the game,” says Brian Murray, who heads Duke University’s Nicholas Institute for Energy, Environment & Sustainability in Durham. “We’re going through round two of expectations for a nuclear renaissance. Climate is becoming a core issue, accentuated by anticipated load growth.”
Because of that shift, Duke Energy has included more production from nuclear operations in its latest “all of the above” energy resource plan. It also calls for adding increased renewable capacity (wind, solar, and battery storage), along with more production from natural gas. New sources are needed partly because the company says it will close its remaining coal-fired plants by 2038, eliminating a source that accounted for 9.8% of customer energy use in North Carolina and South Carolina last year.
Duke has filed to extend the life of its nuclear plants, which now provide half of its Carolinas customer’s energy use. In March, it was granted a 20-year extension for its Oconee facility in South Carolina by the Nuclear Regulatory Commission.
“We strongly believe nuclear energy has and will continue to play an essential role in meeting the energy needs in the Carolinas,” says Kendal Bowman, president of Duke Energy’s utility operations in North Carolina. “As we continue to see strong economic growth, it’s important that we ensure we have the power to meet this growing demand. A lot of these new economic opportunities require energy 24/7. The best resource to meet these needs is nuclear energy.”
Duke says it is looking at all types of nuclear projects, including 1,000-megawatt facilities like the Westinghouse AP 1000. One megawatt is enough to power around 650-1,000 homes, depending on usage.
But the more interesting development may be the utility’s plans to roll out Small Modular Reactors (SMRs) over the coming decade. With a generating capacity of around 300 megawatts of electric power, SMRs are all the rage in the nuclear industry these days. According to advocates, advantages include factory built components which can be mass-produced, theoretically reducing costs with a shorter construction time, and potentially lower security and land costs as a result of a smaller footprint.
All this remains to be seen, however, as there are no commercial SMRs up and running.
“The problem (with SMRs) is you can’t get the economies of scale unless someone builds the first one. And the first ones are very expensive,” says Murrayl.
Duke Energy built its last nuclear power plant, Shearon Harris, 30 miles south of Durham, in 1987. Now, it’s dusting off the nuclear playbook, motivated by both forecasts for future demand and planning requests from the North Carolina Utilities Commission. Duke’s plans call for having two SMRs generating electricity by 2035, one at Belews Creek north of Winston-Salem at the site of an existing coal and natural gas plant, and a second at an undetermined location. Together, they are expected to generate around 600 megawatts of power.
Kelvin Henderson, chief nuclear officer at Duke Energy, suggests that as many as six SMRs could be located at Belews Creek. “It makes sense to continue to build [there],” he says.
At the same time, Duke is working to spread the development risk. In January it joined a public-private partnership led by the Tennessee Valley Authority (TVA) to apply for a grant from the U.S. Department of Energy to fund research into SMRs. “Somebody besides the first companies have to be willing to share in the risk,” Murray says. “If we’re doing it for public policy reasons, then the government has a role (to play).”
Some doubt the advantages ascribed to these smaller reactors are likely to materialize. “Everyone and their brother is selling one (SMRs), but unfortunately none of them actually exist,” says Allison Macfarlane, director of the School of Public Policy and Global Affairs at the University of British Columbia, and the former chair of the Nuclear Regulatory Commission. “You know how much they cost? You’ve never built one. They don’t have economies of scale. In fact, it’s the opposite. It’s cheaper to build one large one than 10 small ones. I would like this to be a solution, but I just don’t see it as one.”
Duke’s Henderson has a different view. He notes that while the BWRX 300, under development by GE Hitachi, is a new design, its fuel and a lot of the components are used in conventional units now. Another model, the Westinghouse AP 300 “is a smaller version of the AP 1000. It’s never been constructed but if you think about the supply chain, the fuel, it already exists,” he says.
A new golden age?
The Shippingport Atomic Power Station was the first commercial nuclear power plant to open in the U.S. in 1958. Most of the existing facilities were built in the 1970s and 1980s, in the “golden age” of the industry. Duke Energy was a key force, gaining a global reputation for engineering and operations prowess.
As of last year, the 94 reactors operating in the U.S. were 42 years old, on average. The latest to enter service was Vogtle Unit 4 in Georgia, which began commercial operation on April 29, 2024.
Meanwhile, China has 31 reactors under construction, which is half the total globally, and plans to add 40 more in the next decade or so, according to the World Nuclear Association.
Construction of Vogtle 4 and its sister unit, Vogtle Unit 3, was not without its problems, however. As a 2023 story from The Associated Press put it, “Georgia Nuclear Rebirth Arrives 7 Years Late and $17 Billion over Cost.” Westinghouse Electric, which provided the reactor technology, went bankrupt along the way.
“What happened at Vogtle and Summer traumatized the nuclear industry,” says Stephen Arbogast, an energy industry veteran who teaches finance at UNC’s Kenan-Flagler School of Business.” It showed that “a major nuclear plant is a ‘bet the company’ proposition.’”
Duke Energy’s Henderson says, “You can look at Vogtle one of two ways – the time and cost to get them online or as a significant opportunity to learn. They learned so much on Vogtle 3 that Vogtle 4 was 30-40% less (expensive).”
Either way, SMRs should represent less of a financial risk. “The whole idea of an SMR is you get away from the massive construction project in the field,” says Arbogast. “You bring it out from the factory and you hook it up.”
History has shown that nothing in nuclear energy is that simple. Duke Energy’s Henderson concedes that some skepticism is warranted when it comes to vendor pricing “because they really don’t know. You have to get the design further along and then bring in a constructor (to determine the cost). Sixty percent of the cost is the construction. The other 40% is engineering and procurement. That’s why Vogtle 4 went down the cost curve (lower construction costs).”
Of the possible sources of new electrical generation, nuclear power is by far the most expensive. But once it is running, it’s also the most efficient. The capacity factor – how much energy a plant generates relative to how much it can generate at its maximum potential – is around 90-95% for nuclear, 60-85% for natural gas, and 25% for solar, according to Arbogast.
A project getting underway in Ontario, Canada, should provide a first look at the cost of SMRs. The builders have estimated the first one will cost $4.3 billion, with subsequent reactors expected to be less. In comparison, the cost for a similar 300-megawatt solar facility would be about $650 million, according to David Neal, senior attorney with the Southern Environmental Law Center.
Given the difference in the capacity factors, this is not an apples to apples comparison, however. To get the same total energy from a utility-scale solar array as you would from a 300-megawatt SMR operating at 90% capacity, you’d need about 1,125 megawatts of solar (at a 24% capacity factor). At 2025 estimated prices, that would be about $1.677 billion, a lot less than an SMR (not including lower operating expense and no fuel costs for solar), though battery storage would add to the cost, Neal says.
Of course there are other issues besides price. The biggest may be 24-hour a day reliability, often cited by Duke.
Pay as you go
After decades of modest growth, utilities around the country are now facing new capacity demands driven by artificial intelligence applications, the corresponding need for massive data centers, vehicle electrification, and manufacturing expansion, among other factors. Data center construction alone added 1% to US GDP growth in the first quarter of the year, according to Apollo Global Management economist Torsten Slok.
Duke Energy estimates demand rising 2% year over year in the Carolinas for 2025-26, rising to 4-5% from 2027-29. That’s the fastest pace in decades. The company describes its projections as conservative and based on potential customers taking one or more concrete steps to show their interest.
“We’re having customers come and ask us for 500 megawatts or even a gigawatt of energy,” Bowman says. “One gigawatt is about the size of Shearon Harris. We also have phenomenal population growth (in North Carolina). We’re adding 340 new residents a day.”
Estimating load growth can be an inexact science, however. Neal agrees that the flat demand growth experienced since 2006-2007 is likely over, but he doesn’t think “we’ve hit an emergency yet. A lot of the load growth projections that Duke presented to the Utilities Commission were based on an econometric model on what they can expect. But there are a lot of unknowns there.”
Neal isn’t sure that SMRs are the answer. “Our biggest concern with SMRs is that it’s not clear they can be built fast enough to meet the load growth demand, it’s not clear they can be built affordably, and it’s not clear they are what is required to meet the moment.”
Instead, his group favors expansion of renewables, which can be added to the grid more quickly. Solar and batteries made up about 81% of the new utility-scale resources added to the grid nationally in 2024, with wind contributing another 13%, according to the U.S. Energy Information Administration. While cost projections for SMRs are steadily rising, Neal adds, there’s a clear downward trajectory for renewable resources.
A recent study from Duke University argued that more efficient management could “add nearly 100 gigawatts of large loads to the grid with minimal impact” by using onsite generators, shifting workload to other facilities or reducing operations.
“Whatever is being consumed has to be generated or injected (into the grid) in real time,” says Dalia Patino-Echeverri, a Nicholas School of the Environment professor and coauthor of the study. “The more responsive the load, the lower the need for new generation and transmission capacity.”
While acknowledging that the variable availability of wind and solar can be a problem (some have blamed Spain’s April nationwide blackout on an over-reliance on renewables), Neal argues that can be managed. “Match the intermittency of the resource with the intermittency of the demand,” he says. “We have the technology to do that. It’s a matter of deploying it.”
Duke Energy is considering “ways to get creative with curtailment and interruptability,” Bowman says. But, she adds, “Some of our customers don’t want that. We still need to invest. That (Nicholas) report made a lot of different assumptions. I don’t know that they’re all correct.”
The long interregnum in nuclear plant construction was driven by safety concerns, but more so by the failure of any major project to come close to finishing on time and on budget. That may be a similar problem for SMRs, Murray says. “Safety is not the issue,” he says. “Cost is the issue,” noting that while “nuclear is safe, it’s not safe for free.”
Enter North Carolina Senate Bill 261, introduced in March as a way to facilitate more power generation and give Duke more flexibility. The first section of the bill provides for the elimination of the carbon reduction mandate that would require Duke to reduce CO2 emissions to 70% of 2005 levels by 2035. Section Two allows “an increase in the base rates of an electric public utility for costs of construction work in progress for baseload electric generating facilities … if the Utilities Commission determines there is an overall cost savings for customers over the life of the generating facility.”
There are further conditions, but essentially what is proposed is a kind of pay-as-you-go billing, also known as “construction work in progress,” or CWIP. This type of financing contrasts with traditional “rate of return” policies that pay off for the utility after a project is completed.
(Provisions of the bill were rolled into another Senate bill that passed in June on a 29-11 vote in the Senate and 75-36 vote in the House. Three Democratic senators and 11 Democratic House members supported the measure along with all but one Republican lawmaker)
CWIP shifts some risk from the company and its stockholders to customers. It was part of the financing for the Summer plant, which is proving costly for ratepayers, Neal says. The South Carolina “experience is a good reason to be skeptical about relying on construction work in progress. It exposes customers to a lot
of risk.”
Duke’s Bowman says that there are differences between the North Carolina bill and South Carolina’s Baseload Recovery Act. “There is a strong review and prudency process in North Carolina and heavy consequences (for missing milestones) that I do not believe were present in South Carolina,” she says. Duke Energy says SB261 will save customers billions of dollars over time, while providing more predictable energy prices.
Competitive advantage
In this new world of growing load demand, providing easy access to power can critical to attracting tech companies and manufacturing jobs. Arbogast says North Carolina “has a significant advantage – the combination of reliability and power prices.” Another positive: more than half of the kilowatt hours of electricity consumed in North Carolina comes from carbon-free nuclear power. Duke has about 26,000 megawatts of generating capacity in the state.
Bowman cites another piece of legislation that has been helpful to the state in attracting new businesses. “The passage of House Bill 951 that set North Carolina on the path for net zero by 2050 has been very helpful in attracting customers and new development. A lot of the companies looking at the state have their own sustainability goals and the fact that North Carolina has that out there has helped from an economic development standpoint.”
Nuclear power has cycled in and out of favor since the days of “energy too cheap to meter,” as Lewis Strauss, then chairman of the U.S. Atomic Energy Commission, called it in 1954. More recently, it has been tarred by a reputation as too risky to build. But concerns over climate change have scrambled the equation, and nuclear is now the least-bad option in the minds of many.
There may not be a groundswell of excitement about a revived nuclear future, but there is a resigned acceptance that some nuclear is necessary to meet the growing demand for electricity and reduce the state’s carbon footprint. This is clearly in line with policy goals emanating from the White House. In late May, President Donald Trump said “it’s time for nuclear” and declared a goal of 300 gigawatts of new generation by 2050 and 10 large nuclear reactors under construction by then.
As the nation’s second-biggest operator of nuclear plants by generating capacity, Duke Energy likely has a major role to play in any such expansion. Still, meeting these goals will be a heavy lift. Whether it’s finding a home for nuclear waste or land for utility-scale solar or windmill installation, tradeoffs will have to be made. As the Nicolas School’s Murray says, “Any form of energy produced at scale has an environmental impact.” That’s one truth that is unlikely to change. ■
