The column goes on to discuss the benefits of small-scale nuclear power — i.e. less than 300 mw — and how a number of companies (as well as Sandia National Laboratories) are pursuing the market. In fact, the president of the American Nuclear Society says small nukes are now a big focus of the U.S. industry, which is beginning to acknowledge that utilities want a product that’s modular, scalable, less risky, and potentially less expensive than big reactors. The key is to build them in a factory setting using robotic assembly, and then deliver them filled with enough fuel to operate for up to 20 years without refuelling. The reactors would be low maintenance, have passive safety features, and would be buried underground. I’m not a huge fan of nuclear, but if we are to have it, I’m in favour of this form of distributed nuclear generation.

My Clean Break column in today’s Toronto Star takes a close look at Hyperion Power Generation, the New Mexico start-up that believes the future of nuclear power will be built in 25-megawatt chunks. Most of us, when we think of distributed power generation, think of solar and wind and other renewables. Hyperion wants to add nuclear to that list. Based on technology developed at the Los Alamos National Laboratory, the company has developed a nuclear “battery” the size of a garden shed that produces enough power over 5 to 10 years to supply the equivalent of 20,000 homes. At a cost of $30 million, the reactor is ideally suited to oil-sand developments, military bases and other remote outposts, the company says.

It uses uranium hydride as fuel and the design is based on TRIGA reactors used to safely train young nuclear scientists. It’s self-moderating — when the reactor core goes above 550 degrees C the fuel begins to lose hydrogen atoms, which are collected in special storage trays surrounding the core. The loss of hydrogen atoms naturally cools down the reactor to a point where it begins to reabsorb the hydrogen atoms again. This cycle keeps the temperature fairly constant and makes a meltdown virtually impossible. No water cooling is required. Hyperion’s plan is to mass-manufacture these nuclear batteries, seal them at the factory, then transport them under heavy security to the customer site. There, the battery will be buried three metres underground and left there until the fuel is used up (about 5 to 10 years depending on the load intensity). At that point Hyperion will dig up the battery and bring it back to its central facility where it will be refueled, resealed and resold. It hasn’t really explained where the old fuel will be stored and how it will be managed.

Sounds cool on the surface, and compared to conventional “go big” nuclear power plants I kind of like the idea. That said, there’s the same old hangups about nuclear that give reason for serious pause. Imagine: thousands of small nuclear reactors scattered around the world. And while I have no doubt these nuclear “batteries” operate safely, my bigger concern is possible tampering and vulnerabilities in the supply chain, such as the transport of a spent reactor (which has toxic spent fuel sealed inside) back to Hyperion’s facility. Hyperion says all units will be kept with sensors and monitored both remotely and locally around the clock. It also says the fact it is buried underground makes it very difficult for tampering. Difficult, maybe, but not impossible. Also, three meters underground isn’t that deep. What’s stopping someone from shooting a missile at the site? Certainly this would affect the reactor.

I like the idea of distributed generation, so Hyperion’s technology does fit within that paradigm. Again, compared to big conventional reactors this is a great idea. Beyond that comparison, however, I’m not so sure. Hyperion says it is in discussions with the U.S. Nuclear Regulatory Commission about getting a manufacturing license for the reactors, but don’t expect regulatory approval anytime soon. In fact, back in mid-December the NRC put out a comment related to Hyperion that didn’t come across as favorable. The message to Hyperion and others: Don’t get your hopes up.

The NRC is aware that Hyperion and others have proposed building such reactors. Hyperion has advised the NRC it intends to provide technical reports on its proposal in the fall of 2009 as part of a pre-application review. That is only the first step in a process that could take years and years. The licensing of new, small reactors is not just around the corner. The NRC’s attention and resources now are focused on the large-scale reactors being proposed to serve millions of Americans, rather than smaller devices with both limited power production and possible industrial process applications. In our innovative society it is not unusual for firms like Hyperion and others to propose reactor designs that are radically different from the existing generation of technology. And examining proposals for radically different technology will likely require an exhaustive review before the NRC could approve them as safe for use. Until such time as there is a formal proposal, the NRC will, as directed by Congress, continue to devote the majority of its resources to addressing the current technology base.

Now, it’s possible that the Canadian Nuclear Safety Commission has a different point of view, but I doubt it. Canada’s regulator is also overwhelmed with conventional nuclear applications and wouldn’t see Hyperion as a priority. It would prefer to defer to the NRC, and since the NRC isn’t prepared to act quickly — or at all at this point — then I wouldn’t expect to see Hyperion’s reactor being used by its target date of 2013. That said, it may get approved in other countries with less rigorous nuclear rules. A scary thought.

The idea of small reactors for distributed generation isn’t new. A source at Greenpeace reminds me of the 3 to 10 MWt Slowpoke Energy System — i.e. also a nuclear battery — proposed by Atomic Energy of Canada Ltd. back in the 1980s. The concept never went anywhere. That said, 20 years later the appetite could now be there.