THE DEPARTMENT OF ENERGY will soon award matching grants totaling $452 million over a five-year period to help guide two small modular nuclear reactor designs through the licensing process to actual operation by the year 2022. Several firms are competing for the grants, and as important as the grants will be to the winners, the DOE has a much more ambitious goal. It hopes to spur a nuclear industry that will supply domestic needs and restore the United States to supremacy in a burgeoning world market for nuclear plants. This, of course, is hardly the first time a U.S. nuclear renaissance has been predicted, but this time three details suggest it could be real.
First, the DOE decided to focus on small and modular. Small means plants that generate less than 300 megawatts; modular can mean either that they are modularly built or that they can be deployed sequentially, starting with as little as 45 megawatts for a single unit and expanding over time to as many as 12 units at a single site.
Second, the DOE intends to choose the two winning designs based, in part, on actual teams of designer/manufacturers and utility operators. Because licensing is by far the most time-consuming part of building a nuclear plant, the DOE wants to make sure the process includes an end-user from the very beginning.
Third, while several fundamentally different forms of SMRs are in development here and abroad, including high-temperature gas-cooled reactors, molten salt reactors and molten metal reactors, the likelihood is that the DOE will focus on light-water reactors, because that's the technology the Nuclear Regulatory Commission is most familiar with, having already licensed large light-water reactors. "Light-water reactors are most likely to meet NRC licensing requirements within our time frame," says John E. Kelly, deputy assistant secretary for nuclear reactor technologies at the DOE. "We're open to other technologies, but how quickly they can go through the regulatory system is a concern."
The enormous up-front capital investment and the long construction time have been the greatest hurdles to the wider adoption of nuclear plants. SMRs address both issues. Large reactors have, traditionally, been built on-site. Precisely because they are smaller, SMRs are being designed to be built in factories and delivered in finished pieces for on-site installation. A large nuclear plant may take anywhere from eight to 10 years from first spade to first electron. Kathryn J. Jackson, senior vice president and chief technology officer for Westinghouse Electric, estimates construction of the company's factory-built SMRs could take just 24 months. She notes that time frame is competitive with the 18 to 24 months it takes to construct a combined-cycle gas turbine.
The Westinghouse design is modular in the sense that it will be built in pieces and assembled on-site. "The goal is to be as modular as possible, perhaps 90 percent modular," says Jackson. "Parts will be built in factories and shipped to the site by rail or truck. Building everything in a consistent way saves enormous time. Studies have shown that building this way, every hour spent in a factory equates to about three hours at an assembly facility on-site, and that one hour is equivalent to about eight hours of stick-built construction in the hole, the traditional way of building large nuclear plants." Because construction time is perhaps the largest factor in up-front nuclear costs, cutting that time by as much as 80 percent suddenly brings nuclear into the budgetary range of many more utilities.
Warner Baxter, president and CEO of Ameren Missouri, the utility that is partnering with Westinghouse in the DOE application, agrees. "The SMR, with its smaller footprint, shorter construction period, and factory-based concept mitigates, if it doesn't eliminate, the large up-front capital risk. It makes it more attractive to us and, I believe, will make nuclear more attractive to many more utilities," he says.
NuScale Power, which is also competing for the DOE grant, will manufacture its entire SMR in American factories and ship the finished parts for on-site installation. But unlike the Westinghouse design, NuScale's is modular in the sense that a utility can begin with one or two 45-megawatt units and add units as needed and as financing becomes available. Additional units would be controlled from the same control station as was originally installed. This modular approach, according to Paul G. Lorenzini, NuScale's CEO, greatly enlarges the potential market for nuclear. "Assuming our
unit costs, dollars per kilowatt, are approximately the same as for a large plant, we would put in an infrastructure and add modules one at a time," he says. "We're working through ways we can maximize the advantage of scaling up. At the very least, if the unit costs are comparable, as we think they are, and the construction schedules are shorter, you might be looking at less than a third to a half the total capital demand of a conventional, large nuclear plant. If you look at it from a national policy standpoint, it expands the ability of nuclear power to provide noncarbon solutions, which is part of what the energy policy people want to achieve."
The NuScale approach offers important flexibility. Because each module operates independently, one can be taken down for service or refueling while the others continue to operate. NuScale is partnering with SCANA and NuHub, an economic development consortium focusing on developing the nuclear industry in South Carolina, to build the first commercial NuScale plant at the DOE's Savannah River Site. NuScale already demonstrated its design in a one-third scale test facility in 2003 at Oregon State University.
While it would appear that SMRs are economically viable, not everyone is convinced. Even John Kelly of the DOE admits the department wonders about the economics. "You think it's going to be favorable, but we don't have enough information to say that for certain," he says. "By going forward with this program, we'll get more information about the design, reduce the uncertainty about the licensing and therefore be better able to quantify what the actual costs are going to be."
John Gilleland, CEO and nuclear program manager for TerraPower, calls himself agnostic about the up-front economics of SMRs. TerraPower is moving in a different direction, designing a 300- to 400-megawatt plant cooled not by light-water technology but by liquid metal sodium. Sodium-cooled reactors are already operating in Russia, France and Japan, among other places. Gilleland believes their greater energy density - they operate at much higher temperatures than water-cooled plants - and the fact that they can go much longer between refueling will ultimately prove more cost effective. Furthermore, so-called fast reactors, such as the design being pursued by TerraPower, can burn what other plants consider waste products or spent fuel, which reduces the volume of eventual nuclear waste. That's one reason Gilleland estimates that fuel costs are between $2 billion and $6 billion lower over the life of the plant than for a light-water plant. But because the DOE is at this point much better prepared to license light-water plants, Gilleland sees his primary market as overseas, at least for the next decade.
The other major concern, of course, is safety. What virtually all proposed SMR designs have in common is their reliance on passive safety systems. Rather than depending on pipes, valves, batteries, diesel generators and other active ways to circulate coolant, SMRs rely on gravity and convection. In the event of a total disaster, such as the tsunami that struck Fukushima, these systems require no power whatsoever to effect a safe shutdown of the plant. In this, Westinghouse has been a leader with its passive safety design for the large AP1000 reactor, which it is adapting to its proposed SMR. That design has already won NRC approval, but other companies use essentially the same system, so approval should be relatively routine.
Whichever designs the DOE decides to reward, there's no question that SMRs offer attractive options to utilities here and abroad. They allow smaller utilities, with less capital, to diversify their fuel portfolio with nuclear as a hedge against future carbon regulation and the inevitable volatility of natural gas prices. The ability to start small and add capacity as demand grows and as coal-fired plants are retired makes them that much more attractive. But the lead time remains a concern, at least for now, and no matter how safe passive systems are, there remains a large segment of the public that needs to be convinced that nuclear, along with renewable sources, is a safe, viable option to replace environmentally harmful carbon-fueled baseload generation. On the surface, SMRs look extremely attractive from an economic, safety and environmental perspective. As the DOE moves forward in helping at least two designs move through the long approval process, it remains to be seen whether the promise they offer will to be fulfilled.