A visitor to the Darlington Nuclear Generating Station in Clarington, Ont., looks out onto the facility grounds during a tour in February, 2016.Fred Lum/The Globe and Mail
Ontario Power Generation’s selection of GE Hitachi Nuclear Energy to help build a small modular reactor (SMR) at its Darlington station in Clarington, Ont., set in motion events that could shape Canada’s nuclear industry for decades to come.
OPG’s choice, announced in December, is the BWRX-300. It’s a light water reactor, the variety most popular in developed countries, and quite unlike Canada’s existing fleet of CANDU heavy water reactors. Though not exactly small – the BWRX’s 300-megawatt nameplate capacity is roughly equivalent to a large wind farm – it would produce only one-third as much electricity as traditional reactors. It would use different fuel, produce different wastes and possibly have different safety implications.
The Darlington SMR would be the first BWRX-300 ever constructed. By moving first, OPG hopes Ontario will become embedded in a global supply chain for these reactors.
“OPG ourselves, we don’t really get anything out of it – it’s a lot of work,” said Robin Manley, OPG’s vice-president of nuclear development. “Our goal is to have as many contracts signed with Canadian suppliers as we possibly can.” But that might not satisfy some critics, who’ve protested OPG’s selection of a U.S. design by GE Hitachi, which is based in North Carolina.
It does seem to confirm the end of Canada’s tradition of homegrown reactors. The BWRX-300 would be Canada’s first new reactor since Darlington Unit 4 in Ontario, completed in 1993. According to Mycle Schneider Consulting, the average age of the country’s 19 operational reactors is 38 years. Attempts to update the CANDU design proved largely fruitless; OPG and Bruce Power opted to refurbish reactors at Darlington and Bruce stations to operate another few decades, while sizing up SMRs as a possible next act.
Time is running short. This decade is widely regarded as crucial for building emissions-free generation capacity. SMRs will be late to that party even if this BWRX-300 is built on time. Delays and cost overruns, ever-present risks with any reactor, could kill its prospects.
The partnership with OPG represents a major coup for GE Hitachi, a U.S.-Japanese alliance that set up its SMR subsidiary in Canada less than a year ago. There are at least 50 SMR designs worldwide, but most exist only on paper; vendors compete vigorously to sell to experienced nuclear operators such as OPG because they represent an opportunity to build a bona fide reactor that might entice other clients. For the same reason, OPG’s decision is a blow to the losing candidates, Oakville, Ont.-based Terrestrial Energy Inc. and X-energy, an American vendor.
“There’s lots of enthusiasm among nuclear reactor designers, developers and national laboratories, and academic nuclear engineering departments” about SMRs, said Edwin Lyman, director of nuclear power safety at the Union of Concerned Scientists, who published a report on SMR reactor designs in early 2021. “There’s a lot of supply but there’s not much demand, because utilities don’t want to be guinea pigs.”
Nuclear industry executives and government officials hope the Darlington SMR will be the first of many deployed in Ontario and beyond. SaskPower is also shopping; it has collaborated with OPG since 2017, and said the BWRX-300 is among its candidates. Canada has a small population, so observers doubt the country could support supply chains for multiple reactor designs.
But OPG’s selection of an American SMR has drawn some sharp criticism. Some observers assumed Terrestrial enjoyed a home turf advantage, particularly in light of the federal government’s decision to invest $20-million toward its Integral Molten Salt Reactor (IMSR). The Society of Professional Engineers and Associates, a union representing engineers and others working on CANDU reactors, complained that “priority should have been given to Canadian design.”
“It is a slap in the face for Terrestrial,” said M.V. Ramana, professor at the University of British Columbia’s Liu Institute for Global Issues. “It is not a good sign for Canada’s nuclear industry.”
Prof. Ramana added that OPG’s decision may prompt a rethinking of government support to SMR developers. In addition to Terrestrial’s funding, Moltex Energy received $50.5-million from the federal Strategic Innovation Fund and the Atlantic Canada Opportunities Agency to advance the Stable Salt Reactor-Wasteburner it is working on in New Brunswick. ARC Clean Energy received $20-million from New Brunswick’s government toward its ARC-100 reactor.
“If these companies are not able to persuade OPG, then maybe we should stop funding them,” he said.
But observers said that of OPG’s three publicly-announced candidates, GE Hitachi was the conservative choice. Whereas most SMR vendors are startups that have yet to build a single reactor, GE Hitachi has been in the business since the mid-1950s. The BWRX-300 is billed as the 10th generation of the company’s light water reactor design.
“I would bet money that that was the decisive factor,” Mr. Lyman said. “Its supply chain is probably better established than these other designs that have very little or no operating experience, and have a whole lot of other unknowns associated with them.”
Other experts have reached similar conclusions. Tractebel, an engineering firm that has worked on nuclear projects in 20 countries, assessed dozens of SMR designs for Estonia a couple of years ago. The BWRX-300 made its short list, favoured as a simple, “proven technology.” (Terrestrial also scored highly, but Tractebel concluded molten salt reactors such as its IMSR are further off.)
Unlike CANDUs, which consume unenriched uranium, light water reactors require fuel enriched to increase Uranium-235 content. Mr. Lyman said that by adopting any non-CANDU design, Canada will become dependent on enriched fuel imported from the U.S., Europe or elsewhere.
The industry would also need to learn how to dispose of unfamiliar wastes. The Nuclear Waste Management Organization (NWMO), which is in the final stages of selecting an underground storage site for Canada’s radioactive spent fuel, said spent BWRX-300 fuel would generate more heat and radioactivity than CANDU fuel, but could be stored in fewer containers, placed further apart.
“We will learn from our international partners who already have plans to permanently store this type of waste in a deep geological repository,” the NWMO said in a statement.
All this assumes OPG’s reactor gets built. To begin with, the BWRX-300 actually isn’t licensed to be built anywhere. GE Hitachi is participating in the Canadian Nuclear Safety Commission’s Vendor Design Review, through which it receives early feedback from the regulator on its reactor. Completing that process would confirm the CNSC didn’t find any features that would be offside from Canadian requirements.
After that, GE Hitachi would need a construction licence. Like other SMR vendors, GE Hitachi pitches its SMR as including “passive” safety features, meaning during an accident the plant would have sufficient water and electricity to operate without intervention for days, even weeks. A safer reactor might also be a cheaper reactor: For instance, the SMR might require less containment than traditional designs, and thus less concrete. GE Hitachi says the BWRX-300 occupies less than 10 per cent of its predecessor’s volume. But the CNSC would first have to agree the reactor has really earned lower safeguards.
“It’s pretty early on in the licensing process,” Mr. Manley acknowledged. “We as the operator are very confident that we will be able to license this reactor.”
But critics say completing the reactor by 2028 is a tall order. According to Mycle Schneider Consulting, one in eight reactors that have begun construction since 1951 were never connected to the grid. Many survivors, meanwhile, arrived years later than promised.
Mr. Manley said 2028 is “an aspirational goal” rather than a hard deadline. The project schedule will firm up over the next two years.
OPG has yet to publish a cost estimate, but according to a report published by PwC, the SMR project “is expected to spend $2-billion over seven years.” That’s already higher than the US$1-billion price tag GE Hitachi promised for a BWRX-300 in 2019. (In public presentations, GE executives declared that keeping the price below US$1-billion was crucial to its plans to exponentially grow its customer base.)
Even so, Mr. Manley said OPG’s intention is that the SMR will be cost competitive with other clean energy options.
Prof. Ramana said cost escalation is practically inevitable. The AP1000 reactor, a pressurized water reactor from Westinghouse that produces 1,110 megawatts, was originally supposed to cost US$2-billion. But the price of the two-unit Vogtle plant in Georgia was estimated at US$14-billion, and later rose above US$30-billion. Westinghouse’s marketing materials emphasized that the reactor has been tremendously simplified, making it less expensive to build, operate and maintain.
“Historically, the vendor estimates have always been far, far lower than what the actual costs are going to be,” Prof. Ramana said. “I don’t think there’s any reason to expect this to be different.”
Mr. Manley conceded that during the past two decades, large nuclear plants “have been challenged to get done on schedule and on budget.” But that’s one reason OPG decided to build an SMR. They “are lower capital cost, simpler, smaller and therefore faster and easier to get built,” he said.
This also won’t be the first time OPG has managed a complex, high-risk capital project. The company says the $12.8-billion refurbishment of its Darlington station, now half-complete, is on time and on budget. That experience has demonstrated that “when we take the time to plan the job properly, we are able to come up with a good estimate and a good schedule that we’re able to meet,” Mr. Manley said.
Interested in more stories about climate change? Sign up for the Globe Climate newsletter and read more from our series on climate change innovation and adaption.
Matthew McClearn