Last month, David Leblanc made a kind of pilgrimage to Oak Ridge, Tenn. to celebrate the 50th anniversary of a very special moment in nuclear history. From 1965 until 1969, the United States Department of Energy’s Oak Ridge Laboratory was home to the world’s only functioning molten salt nuclear reactor (MSR). Based on an experimental design very different from what has become the mainstream, to many the Oak Ridge MSR symbolized a rare flourishing of innovation in what is now a very conservative nuclear industry. But for Leblanc, it represents something else: not the end of an era, but rather the beginning of a nuclear renaissance.
David Leblanc, Chief Technology Officer, Terrestrial Energy Inc. Photo credit: Terrestrial Energy Inc.
In the mid 1990s, Leblanc was working on a PhD in physics at the University of Ottawa. His research focused on superconductors with which he hoped to create ultra-powerful magnetic fields to confine the super-hot hydrogen and helium gases involved in a nuclear fusion reaction. Harnessing fusion, the same process that powers the sun, has long been a dream of alternative energy researchers and science fiction writers alike. But as he researched, Leblanc was surprised to discover that the world of nuclear fission — the opposite of fusion and the basis of all modern nuclear power — was more varied than he had previously realized.
“There were a lot of great ideas that were looked at in the 1950s and 1960s which unfortunately didn’t get a full try-out,” Leblanc says. Among them was Oak Ridge’s molten salt program, which was eventually cancelled in favour of light water reactors, the design used in most of today’s nuclear power plants. “They were the best fit for a nuclear submarine,” says Leblanc. “They got a head start on everything.” (A third technology, the CANada Deuterium Uranium or CANDU reactor, is used in a few dozen facilities around the world.)
Leblanc’s self-described “innovative mind” became fascinated with the nuclear industry’s brief experiment with novelty. After finishing his PhD, he took a teaching position at Carleton University and founded a consulting company, Ottawa Valley Research Associates Ltd. For the next decade, the physicist researched fission reactor design, using the Oak Ridge model as his starting point.
Before long, Leblanc found himself giving talks for the Thorium Energy Alliance (TEA). Since 2009, this non-profit group composed of engineers, scientists and concerned citizens has held an annual conference on alternative nuclear technologies. It was at one of these conferences that Leblanc met Simon Irish, a former Wall Street investment manager who had been bitten by the nuclear bug a few years earlier. “I asked myself, ‘what’s the biggest market need over the next 20 years?’ I didn’t need to think about it too long to realize that it’s energy,” says Irish. He rejected green alternatives like wind and solar power for their lack of scalability. “It has to be nuclear,” he says. “The question is, what type of nuclear?”
Terrestrial Energy’s plants are designed with two parallel bays. While one integrated core is providing power, the other can be cooling off for up to seven years. Overhead cranes are used to deliver new cores and to move spent cores to long-term storage. Terrestrial Energy Inc.
Burners versus breeders
Uranium-235 is the primary fuel for modern nuclear reactors. This isotope is fissile, meaning that upon collision with a neutron it breaks up — fissions — into other elements, releasing energy and more neutrons that can continue the chain reaction. But uranium-235 is rare, making up less than one percent of all natural uranium. The other 99 percent is uranium-238, which is not fissile but rather fertile; if it gains a neutron of the right energy, it turns into uranium-239, which after a couple of beta decays becomes fissile plutonium-239. The transformation of fertile materials into fissile materials is called breeding.
Both light-water reactors and CANDUs breed a lot of their own fuel but it’s not enough to replace the uranium-235 that is used up in the reaction; they are therefore classified as “burner” reactors. Even with the common practice of enriching uranium-235 content from one percent up to five percent, fuel bundles typically only last a year or two. Depleted bundles contain small amounts of fission products and fissile plutonium, which are dangerously radioactive. But more than 95 percent of the material is simply harmless unused uranium-238. Many alternative nuclear proponents are focused on replacing today’s supposedly inefficient “burner” designs with true “breeder” reactors, which create as much or more fissile fuel than they consume.
Some breeder reactors can do away with uranium altogether. Thorium-232 is also fertile; after capturing a neutron and undergoing a couple of beta decays, it can be transformed into fissile uranium-233. Moreover, thorium is three to four times more abundant than uranium and a lot harder to make into a nuclear weapon. The advantages of breeder reactors, especially those designed to use thorium, drive a lot of the enthusiasm that created groups like the Thorium Energy Alliance.
Leblanc sees things a little differently. “The breeder is an admirable goal but uranium is not in short supply,” he says, pointing out that current reserves are enough to last two centuries or more. In his view, the problem with conventional burner plants — typically designed to last 30 years and generate 1,000 megawatts (MW) or more — is not that they’re inefficient or unsafe but that they’re extremely expensive to build. And most of that cost comes not from the fuel itself but rather the complex systems needed to safely remove and replace the radioactive fuel bundles every 12 to 20 months.
Leblanc’s key insight was to realize that with molten salts, it is possible to dramatically shrink the reactor. His vision calls for an integrated molten salt reactor (IMSR) that only generates about 300 MW, but can do so for seven years without needing to be refuelled. At the end of that time, the entire core — not just the fuel but the pipes, heat exchangers and all other components — is replaced wholesale with a fresh one. “It is counter-intuitive to many,” says Leblanc. “They say, ‘my god, you’re replacing your whole reactor, how could that be economical?’ ” The answer, he says, is that his integrated reactor has a very simple design that makes up a small fraction of the overall plant cost. Replacing the entire core means no complicated refuelling and no need to ensure that components will last a full three decades before needing maintenance or repair. “It’s a remarkably small penalty to pay and the benefits that come from it are pretty enormous,” says Leblanc.
Eleodor Nichita, a professor in the Faculty of Energy Systems and Nuclear Science at the University of Ontario Institute of Technology, agrees. “Banks of such small reactors can be built over time with new ones being added as the old ones are being amortized through the sale of electricity,” he says. The fact that the core only needs replacement every seven years could be especially attractive for remote communities. “Its small initial capital outlay also presents a compelling economic case,” Nichita adds.
How it works
The fuel used in Leblanc’s reactor is the same low-enriched uranium used in light-water or CANDU reactors but in a very different form: a uranium tetrafluoride salt, which is mixed with other fluoride salts like lithium fluoride, potassium fluoride or zirconium fluoride (Leblanc is understandably cagey about the precise formulation). At temperatures higher than about 500 C, these mixtures, called eutectics, become liquid.
The molten salt is pumped through a series of pipes surrounded in graphite. Although the uranium-235 atoms are constantly fissioning, the neutrons they give off are of such high energy that they are unlikely to be captured by other atoms and breed more fuel. Graphite acts as a moderator, slowing down the neutrons to the point where it makes them more likely to encounter other atoms. This sets off the fission chain reaction at the heart of the process, providing enough heat to keep the salts liquid and much more besides.
The hot salt flows back out of the graphite chamber into a heat exchanger, where it gives up its heat to a secondary fluoride salt that contains no uranium. It is this salt, and only this salt, which flows out of the core to exchange its heat with a conventional steam turbine system. This is a critical advantage; it avoids the challenges of pumping high-pressure water directly into the core as conventional reactors do. The hot salt then re-enters the graphite chamber to start the process over again.
After seven years, the graphite has degraded enough that it’s time to shut the reactor down. While a replacement core is started up in a parallel bay, the first one seamlessly transitions into being a container for any radioactive products of the fission reaction. It can remain where it is for another seven years — enough time for the radioactivity levels to drop significantly — before being moved to long-term storage in much the same way that current nuclear waste is handled. “The ultimate decommissioning we feel is a very tractable problem,” says Leblanc.
Liquid fuel reactors deal very differently with the central challenge of any reactor design: how to dissipate the intense heat of fission energy. Unlike solid fuel rods, liquid fuels can’t suffer a meltdown because they are already melted. In fact the salt acts as its own coolant, dissipating heat efficiently by convection, rather than relying on the heat exchangers to do all the work in all circumstances. In the IMSR design, decay heat is removed directly from the reactor vessel walls, aided by the natural circulation of the liquid fuel salt. Because of the volume of the molten salt and its high heat capacity, the IMSR has a tremendous natural ability to cool itself. “It’s a very elegant, simple reactor,” says Irish. “And there’s a direct linkage between simplicity and cost reduction.”
Terrestrial Energy’s integrated reactor is designed to contain the molten salt fuel during its seven-year lifetime and indefinitely afterwards; the only component that leaves the core is the secondary salt, which contains no uranium. By replacing the entire core, rather than just the fuel, the company hopes to avoid many of the complex and expensive systems required in conventional reactors. Photo credit: Terrestrial Energy Inc.
Making the case
In January 2013, after years of consulting experts and incorporating their feedback into his design, Leblanc decided to quit his job and go into business with Irish. Their company, Terrestrial Energy Inc., has attracted a number of major players in the nuclear industry. For example, their management team includes former executives with Atomic Energy of Canada Limited (Hugh MacDiarmid) and Westinghouse Electric Company (Robin Rickman), while their international advisory board includes a former administrator of the US Environmental Protection Agency (Christine Todd Whitman) and a former Chief Technology Officer of Lockheed Martin (Ray O. Johnson).
In laying out the business case, Irish points out that the potential market is quite different from the one facing conventional nuclear plants. “In the US, there was a big coal build in the 1960s and 1970s and a lot of those plants are at the end of their operating lives.” he says. “A small modular reactor like ours is the right size to replace these coal-fired power stations. It would be cheaper and better to replace that coal-fired power station with an integral molten salt reactor, rather than using clean coal or natural gas technology.”
Terrestrial’s goal is to get a pilot plant up and running by the middle of the next decade; Irish says it will likely be somewhere in Canada. That means it will need to be licensed and approved by the Canadian Nuclear Safety Commission (CNSC), which regulates nuclear power in Canada. Here, technologies like molten salts suffer from their shorter track record. “Since the majority of nuclear standards have been developed specifically for water-cooled technologies, new technology designers need to get involved in standards development to ensure that standards exist to meet their needs,” says CNSC spokesperson Aurele Gervais. “Generally, such information comes from research and development activities, including results from experiments.”
In the near future, the company will likely need to partner with a large engineering company to build a non-nuclear mock-up that can be used to demonstrate their safety claims to CSNC’s satisfaction. Still, Terrestrial believes it has an advantage in the existence of Oak Ridge. “We feel that our design is not really all that different from the main [Oak Ridge] reactor that ran back in the 1960s,” says Leblanc. Though it was meant to be part of a larger breeder reactor that was never built, the first MSR was a simple burner, of comparable size to Terrestrial’s system. Molten salts are also used in other, non-nuclear systems, such as steel and aluminum production, so there is a body of knowledge about how to work with them.
“There’s going to be a tremendous amount of work for us, but we feel that it’s a pretty well laid out technological development path that doesn’t require something new to be invented,” says Leblanc. “We’re not trying to be exotic, we’re trying to be as simple we can.”