The World Nuclear Exhibition, a major industry event under way in Paris this week, has been called the sector’s Super Bowl. For AtkinsRéalis, formerly known as SNC-Lavalin, it was the moment to unveil a new reactor dubbed the Monark – which, if built, would be the largest-ever exemplar of Canada’s homegrown Candu reactor technology.
But on the same day as that announcement, Ontario Power Generation, the world’s largest owner and operator of Candus, made its own: it had entered discussions with Electricité de France to explore the possibility of building a European model known as the EPR in Ontario.
Kim Lauritsen, OPG’s senior vice-president of enterprise strategy and energy markets, said in an interview that the discussions aren’t exclusive. But the fact that they’re happening at all marks a departure from decades past, when building reactors in Canada from foreign makers would have been almost unthinkable.
Ken Hartwick, OPG’s chief executive officer, hinted at the utility’s reasoning in an interview earlier this year. “We love the Candu technology,” he said. “But the technology, outside of refurbishments, has not kept up to date.”
That widely held perception is a serious liability for AtkinsRéalis at a moment when OPG and other utilities around the world are suddenly shopping for new large reactors. But it is rooted in historical fact: no new Candu design has found paying customers since the end of the Cold War. Some observers say it may be too late to reverse the situation. If they’re right, Canada’s once-vaunted reactor technology may have reached the end of the road.
The Candu arose in Second World War’s aftermath. Atomic Energy of Canada Limited, or AECL, a federal Crown corporation, developed a unique reactor that played to Canada’s strengths. Local manufacturers couldn’t build pressure vessels – the large, heavy cylinders favoured in the United States to contain reactor cores. So the AECL team devised pressure tubes, a signature Candu feature. Enriching uranium was something typically done only in countries with nuclear weapons programs, so the Candu used natural uranium, which Canada possessed in abundance.
The first Candu demonstration unit started up in 1962, touching off three decades of reactor construction in Eastern Canada and abroad. In 1987, the Canadian Engineering Centennial Board named the Candu one of Canada’s top 10 engineering achievements of the 20th century.
But the Candu was doomed to remain stuck in that century.
In industry nomenclature, all existing Candus belong to “Generation II,” a vague classification encompassing commercial reactors built between the 1960s and the 1990s. So-called “Generation III” reactors, first built in the 1990s, were intended to be cheaper, quicker to license and construct, more powerful and efficient, longer-lived and safer than their Cold War predecessors. The quest to build them spanned decades and the financial fallout was often gruesome. Reactor development is a protracted, costly, technically demanding process that frequently ends in failure.
The EPR was proposed in the wake of the 1986 Chernobyl disaster. “The French and the Germans sat together and said, let’s develop something from your best model and from our best model,” said Mycle Schneider, a consultant who produces annual updates on the state of the nuclear industry.
In 1995, France adopted the EPR as its standard design – a coup. But it wasn’t until more than two decades later, in 2018, that the first EPR connected to the grid, in China. (Europe’s first only started up last year.) Areva, a nuclear power group that had been owned by the French government, “was substantially restructured due to huge cost overruns on two EPR projects,” according to the World Nuclear Association’s website. Westinghouse, an American manufacturer, developed its Advanced Passive 1000 reactor simultaneously, and met a similarly Pyrrhic fate.
AECL began developing a 935-megawatt reactor known as the Candu-9 in 1993. According to a paper by Gord Brooks, a former AECL chief engineer, it never progressed to detailed design work. Another model, the Candu-3, also fizzled.
In the early 2000s, AECL assigned a 300-member development team to its biggest bet, the Advanced Candu Reactor, or ACR, which would have burned enriched uranium. Blueprints were to have been completed, and regulatory approvals obtained, by 2006.
Ibrahim Attieh, a reactor physicist who joined AECL in 2004 and worked on the ACR, said the design team ran into insurmountable engineering problems. Internally, he said, “everybody knew that the reactor was not feasible.” AECL abandoned efforts to license the ACR in Canada and Britain. Mr. Attieh said work had ceased entirely by early 2011.
Afterward, the development team pivoted to another model, called the Enhanced Candu-6, or EC6. It was to have been an evolution of the Candu-6, a successful reactor developed in the 1970s. But that design, too, was never licensed or sold.
By 2009, AECL had a bigger problem: the federal government of prime minister Stephen Harper had lost patience with the high costs of designing and building reactors. In the name of protecting taxpayers, it offered AECL’s reactor division for sale. SNC-Lavalin paid just $15-million for it.
In public statements and interviews, AtkinsRéalis executives have said reactor development continued after the AECL purchase: Work proceeded on the Candu-9, the EC6 and the ACR.
“Hundreds of thousands of person-hours have been invested into each of these designs over many years, right up to the present,” the company said in a statement.
Gary Rose, an OPG veteran hired by AtkinsRéalis in June as its new executive vice-president of nuclear, said the Monark draws on elements of the ACR, the EC6 and the reactors at Ontario’s Bruce and Darlington stations. But unlike the ACR, the Monark marks a return to the Candu’s traditional natural uranium fuel. It is being marketed as a fresh “Generation III+” reactor. A brochure lists several desirable features, including 1,000-megawatt power output, “improved cost per megawatt-hour” and “a longer operating life of 70 years.”
Mr. Rose said he already has nearly 100 people designing the Monark, a number he said will grow to 250 in the years ahead. Canada’s nuclear regulator, the Canadian Nuclear Safety Commission, has already accepted the reactor into its pre-licensing examination. Mr. Rose said he hopes to advance the design to the point where AtkinsRéalis can apply for a license to construct the reactor within three and a half years – a highly ambitious schedule.
But AtkinsRéalis acknowledged Tuesday that the reactor is still in “the definition phase.” That means “it’s in the conceptual phase,” said Akira Tokuhiro, a professor at Ontario Tech University’s nuclear engineering department.
“The detailed analysis, especially in terms of safety, in addressing questions that the regulator will ask, has not been started.”
Asked about OPG’s talks with Electricité de France, Mr. Rose said he would expect any customer to study alternative designs. But he added that AtkinsRéalis has received feedback from both OPG and Bruce Power, Ontario’s other nuclear utility, that informed the Monark’s design. And he emphasized that his team is not starting from scratch.
“This is not a brand-new technology,” he said. “This is an evolution of existing, licensed technology.”
Mr. Schneider, the nuclear industry consultant, said that updating a design that is even a decade old – much less two or three – is “a huge job.”
“So to me, this whole discussion about development of new reactor lines, and new constructions, et cetera, et cetera, is just totally outside feasibility,” he said.