By Calla Cofield
Two relatively new nuclear companies, NuScale Power and TerraPower, are cooking up new reactor designs, and meeting new challenges along the way.
Modern light water reactors generate, on average, 1000 megawatts of energy. Medium reactors can dip down to 700 MW. Ideas for smaller reactors have always been around, but never made it past the drawing board, as they seemed reasonable only for small, isolated markets. But in the late 2000s, the cost of large nuclear power plants began to grow unwieldy. Even large buyers were forced to make drastic financial bets on new reactors. So around 2009, the market changed its mind about small reactors.
Paul Lorenzini is CEO of NuScale Power, which is aiming to have its first small modular reactor (SMR) up and running by 2020. Close on NuScale’s heels is Babcock & Wilcox Modular Nuclear Energy LLC, with the mPower small reactor. Lorenzini says the two major factors in turning the market around were the need to build nuclear reactors without taking a major financial risk, and in turn demonstrating that small reactors could be built economically.
“And I am not bashful in saying that our entry into the market, followed by B&W,” said Lorenzini, “were the two major events that triggered that shift.”
NuScale formed in 2007, but it already had six years of R&D data to support its small reactor design. Lorenzini says the response to the design from all different branches of the industry was overwhelmingly positive. The cost of small reactors alleviated the growing cost of large reactors, while also offering scalability, that is, the option to add more modules to one facility if the energy demands grew.
The NuScale design is based on light water reactor designs, but the NuScale reactor units are only 45 megawatts. The reactor is scalable, and a single facility can host between one and twelve units. The reactor is cooled by natural circulation, so there are no pumps or pipes, which can potentially fail. The entire plant, including the containment, sits in a pool of water, so that no systems need to be running to remove heat. Lorenzini describes the technology as revolutionary, but also emphasizes its simplicity.
Work on the NuScale design began in 2000, and emerged out of a collaborative project led by Idaho National Environment & Engineering Laboratory (INEEL) with support from Oregon State University (OSU), and funded by the U.S. Department of Energy. The project ended in 2003, but OSU continued to support R&D on the reactor design. By the time the company was officially formed in 2007, the organization had six years of strong R&D data to support the design.
It is possible that NuScale could have sold the design to a larger nuclear company, but each meeting with a potential buyer also revealed NuScale’s design to a potential competitor. Eventually, the decision was made to start an independent company.
“We believed right from the beginning that you couldn’t sustain yourself in this business without establishing both market credibility and a financial balance sheet,” said Lorenzini. “The buyers of these plants want to know that the seller has got the capability to deliver and is going to be there. So you’ve got to have people behind you who are going do that.”
In 2011, Fluor Corporation agreed to invest in excess of $30 million in NuScale, which gives the company the financial security it needs to attract future purchasers. The next step will be gaining approval from the U.S Nuclear Regulatory Commission (NRC) to start construction. The company plans to submit its application to the NRC this year.
NuScale has a major advantage in its pursuit of approval from the NRC, because its design is based on current light water reactors. This may not be the case for companies working with more innovative designs, such as the traveling wave reactor (TWR) design by TerraPower.
The TWR reactor requires a small amount of enriched uranium to start the fission process, but the majority of its fuel is natural or depleted uranium 238: the most common isotope of uranium found in nature, and a waste product from the production of LWR fuel. Inside the TWR reactor, uranium 238, which is not fissile and cannot support a chain reaction by itself, turns into plutonium 239 which is also used as fuel. This would mitigate the threat of nuclear proliferation because the plutonium 239 is never separated from the uranium, and is used immediately. The TWR reactor can operate on one fuel supply for sixty or more years.
A major hurdle for a new and innovative nuclear technology is proving that it is safe. That’s the responsibility of the NRC. The majority of designs that come through the NRC are based on light water reactor technology, and in those cases “the staff here at the NRC expects it to take about 5 years to go through all the work necessary to show that any given design is acceptable for use in the United States,” said Scott Burnell, a spokesperson for the NRC. Burnell says the Commission is working to expand its knowledge base to keep up with more innovative designs on the horizon. But right now, the NRC may not have the expertise to evaluate all new technologies in the desired time frame.
“We have had conversations with vendors where we’ve said, ‘you’re going to need to do a lot of work to beef up the supporting case for this particular technology’,” said Burnell. “It’s not enough to simply run a computer model if you’re going to offer some innovative feature. To some extent the NRC is going to have to see real- world empirical data to say that that particular new feature is going to do what you say it’s going to do.”
This appears to be the case with TerraPower, which, without the ability to build a test reactor, can’t gather enough data to satisfy the NRC in the time frame they’d like. So the company wants to gain approval to build a reactor in a country that has the expertise to approve the TWR design. TerraPower will then return to the US with data to demonstrate the safety of the design.
©1995 - 2017, AMERICAN PHYSICAL SOCIETY
APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.
Editor: Alan Chodos