The entire budget of the U.S. Department of Energy branch that covers today's energy mix—from cleaning up energy generation's environmental aftermath to energy efficiency programs and renewable energy development—is $10 billion. That's enough to "either build one supercollider on the basic end or one nuclear power plant on the applied end," said Kristina Johnson, the undersecretary in charge of the branch, at the ARPA-E conference on March 3. In other words, nuclear power ain't cheap.

Although estimates vary, there is no doubt that the up-front cost of conventional nuclear power plants—so-called light water reactors that use lightly enriched uranium for fuel—is high compared with any alternatives. Hence, the nuclear loan guarantees from the federal government: no single electric utility can afford to, in effect, bet the company on one nuclear power plant.

Yet nuclear power supplies 70 percent of the electricity in the U.S. that does not directly release carbon dioxide, according to Daniel Poneman, deputy secretary of energy. So are there ways to make it cheaper?

One idea is to create enclosed, small "modular reactors," like the one developed at Los Alamos National Laboratory and now proffered by Santa Fe, N.M.'s Hyperion Power. Its $50-million product would be an enclosed reactor roughly 1.5 meters wide by 2.5 meters tall; generating 25 megawatts, it would be buried underground and good for at least seven years. In fact, the promotional materials, on display at a booth at the inaugural ARPA-E conference, show nothing but a green field with a single tree. Think large (hidden) battery, the company urges.

Of course, the reality is that a steam turbine, generator and cooling device of some size (likely larger than the reactor itself, just like the big boys) would need to be sitting on top of that green field, displacing the one tree from the marketing materials. And the reactor, though enclosed, actually operates at a higher temperature than traditional reactors, roughly 500 degrees Celsius, and would require cooling by a liquid metal to enable its fast neutron fission. Much of that heat would then be passed to water to turn the turbine and produce electricity.

These small reactors are equally capable of having a runaway chain reaction, a so-called meltdown, so they have control rods to shut down fission just like traditional reactors. In fact, such "enclosed" reactors have several sets of control rods, including ones that slowly pull back over time as the fissionable fuel is consumed to enable fission to continue.

Hyperion is not alone in advancing this concept. Though the designs vary, Toshiba, Babcock & Wilcox and other major companies have similar small reactors and there has been initial interest from some potential customers, such as the village of Galena in Alaska. Yet the U.S. Nuclear Regulatory Commission has in the past refused to evaluate these smaller reactors (pdf), preferring to concentrate its efforts on enabling a renaissance in conventional technology.

That may be changing. This February, the NRC issued a call for such would-be small reactor builders (anything less than 700 megawatts, per NRC criteria) to alert the commission whether they would apply for site permits, licenses and certification in the future—to enable the regulatory body to plan its workload.

But Hyperion isn't waiting for the NRC, says Deborah Blackwell, the company's vice president of licensing and public policy. According to her, Hyperion is grappling with "hundreds" of applications from other parts of the world and plans to deliver its first product by the end of 2013.

Of course, such reactors do nothing to solve the problem of nuclear waste, as the spent modules would presumably end up in a similar situation to the spent fuel rods of the nuclear industry today—waiting for a permanent repository. Other designs at the ARPA-E conference might address that—so-called traveling wave reactors that could conceivably transmute some highly radioactive waste as part of their fission process, minimizing the amount of radioactive waste though not eliminating it.

Regardless, of the 52 nuclear reactors being built or planned around the world, exactly none employ these alternative designs. In fact, the only thing small about nuclear may be the amount of energy it ultimately contributes to the world at present rates of construction. "Nuclear will still be a small portion based on what's seen today," said General Electric CEO Jeff Immelt at the ARPA-E summit, one of the would-be manufacturers of new nuclear power plants who notes that either new nuclear, coal with carbon capture and storage or both will be absolutely necessary going forward for reliable electricity with low greenhouse gas emissions. "Someday there's going to be some CEO of GE that makes money on nuclear. I can tell you honestly the last two have not."

Image: Courtesy of Idaho State University