Such nuclear batteries could in principle be sealed, placed in the ground, and run for a decade before being swapped out for an entirely new modular reactor. And if manufactured in a factory, they could also be cheap. "There is no inherent reason why nuclear power needs to be expensive," Bill Gates, who has invested in the novel reactor proposed by TerraPower, told the ARPA–e summit on February 28, noting that nuclear's relative expense largely derives from building in safety features.
Evaluating the safety of such new reactors will take time, of course, and the U.S. Nuclear Regulatory Commission has yet to receive an application from any of the would-be vendors of small modular reactors, whether fast reactors or scaled-down light-water reactors. And staffing requirements, emergency planning and clean-up funds, among other issues, remain to be worked out between the reactor makers and the NRC—a key component of reducing the cost of such reactors. "The staff has contended pretty much all along that they will have to meet the same security requirements as all of the large reactors," says Michael Mayfield, director of the Division of Advanced Reactors and Rulemaking in the NRC's Office of New Reactors, noting that a timeline for licenses could be expedited if such reactors are simply scaled down versions of existing light water reactors that do not require new regulations. "Why would it take so long to review something that is substantially smaller with fewer parts? That however is based on the notion that vendors submit complete, high quality applications and address staff concerns more quickly than we have been able to do with some of the large [reactor] designs."
By eliminating human intervention—through passive safety features that kick in without the help of operators—staffing requirements might be cut. And if buried or otherwise hardened, the need to pay security guards might also be reduced. "If you need humans to do something, that is not a good design," Gates argued at the ARPA-e conference.
Ultimately, the success or failure of such scaled-down designs may relate to manpower. "If you need the same overhead to run 100 megawatts as you do to run a 1000, that's economically problematic," notes William Johnson, CEO of Progress Energy, a utility considering whether to build new nuclear power plants in future.
The Homer Simpson factor
Of course, human error has yet to be eliminated from either operating reactors or those that exist only on paper. And, much as in airplane or pharmaceutical development, government decisions will determine whether these reactors succeed. Small modular reactors "becoming a reality are dependent on government and the nuclear industry," said NRC commissioner William Ostendorff in a speech to the American Nuclear Society conference this past November. "With respect to new reactor licensing, 'the devil is in the details.'"
Regardless of how cheap such small modular reactors may allow nuclear to be in future, it is unlikely to be as cheap as natural-gas-fired turbines in the present. In fact, low natural gas prices stalled the U.S. nuclear renaissance outside Georgia and South Carolina, long before the reactor meltdowns at Fukushima Daiichi in Japan. "Because of an unanticipated abundance of natural gas in the United States, nuclear energy, in general, is facing tough competition," noted an analysis of the prospects for small modular reactors from the University of Chicago published last November. The analysis also suggested that small reactors would be more expensive than large reactors on a per-megawatt basis until manufacturing in significant quantities has happened. "It [is] unlikely that SMRs will be commercialized without some form of government incentive."