Thursday, January 20, 2011

SMRs sail into headwinds at SRS

Licensing issues could push prototypes projects off course

An innovative program to build prototypes of multiple designs of small modular reactors (SMRs) at the Savannah River Site (SRS) is getting off to a rocky start. The anti-nuclear group Friends of the Earth (FOE) is charging that the Department of Energy (DOE) is trying to avoid having to submit the projects to the Nuclear Regulatory Commission (NRC) for licenses to build them.

The issue that FOE brings to the fore is that even research reactors need licenses and even if they never generate any electricity for use on the grid. This is the point where the DOE and the NRC part ways.

DOE is asserting its authority over R&D projects on DOE sites, like SRS, saying an NRC license isn’t necessary. On the other hand, a source familiar with the ways of the NRC tells this blog, “there are no special cases for prototypes.”

Here’s the pro and con in quotes from the Augusta Chronicle for Jan 19.

"SRS plans to fully engage the NRC throughout all aspects of any reactor project at the site, and that may include licensing," said Jim Giusti, an Energy Department spokesman at SRS. “

"However, DOE does have the authority to construct, operate and monitor safety of research reactors at its sites. NRC regulation is not necessary unless these reactors feed electricity to a commercial grid."

Scott Burnell, an NRC spokesman in the agency's Washington headquarters, also said the Energy Department has indicated it fully intends to comply with licensing guidelines.

"Everything the NRC has seen from those departments indicates they fully intend to use the NRC licensing process if they intend to pursue small modular reactors on Department of Energy facilities."

It’s DOE position that got socks in a knot for Tom Clements, the Southeastern Nuclear Campaign coordinator for Friends of the Earth.

"Construction of 'small modular reactors' that are not licensed by the Nuclear Regulatory Commission would violate U.S. law as well as endanger the public, and we will strongly oppose any attempt to avoid required licensing of such reactors.”

Early progress on SMR licensing would be a good idea

Everything this blog has heard, aside from the fantasies of some marketing types talking to new investors, is that future customers want the NRC licenses done and soon.

One of those customers is the Department of Defense (DOD) which would rather get electricity for tactical jet fighter squadrons from an SMR than reply on the grid in a potentially unfriendly country. DOD is not going to bet the tactical readiness of a front line fighter squadron, or the safety of our service men and women, on an unlicensed reactor.

That said FOE’s Clements sees his job as attacking anything that looks like a weakness in the nuclear world as a means to thwart the nuclear renaissance. The anti-nuclear agenda isn’t about licensing. It is designed to set the bar so high that the safest reactor is one that is never built.

The Department of Energy is helping this cause by tap dancing on the licensing issue. SMR vendors who encourage this type of conversation are not doing themselves any favors. Speculation about what investors in SMRs might want, or not, regarding the involvement of the NRC is more or less irrelevant.

There is a workable process available. Just look at how the Tennessee Valley Authority (TVA) and mPower are working to license a 125 MW SMR at the Clinch River reactor site. They are using the old Part 50 process and then plan to blend the resulting information into a Part 52 design certification. Maybe the folks at SRS should take a short trip Oak Ridge, TN, for a briefing?

The Atomic Energy Act is clear that the only exemption for licensing involves a presidential determination on civilian use of “nuclear materials. Maybe someone at SRS should call Sec. Chu to see how willing he is to carry that request to the White House?

SMR licensing challenges

SMRs are by their nature low power, usually in the range of 25-300 MW, compared to their bigger brothers which weigh in at 1,000 MW or more. The NRC has lots of experience licensing the large designs because they are all based on similar design principles. Basically, set off a nuclear chain reaction with enriched uranium (3-5% U235) bundled into long fuel rods, use the heat to make steam, drive turbines, and make electricity.

The problem for the NRC is that it has not, in recent years, conducted a safety analysis to certify a design for a “fast reactor” cooled by liquid metal such as sodium or a lead-bismuth mix. These reactors use enriched uranium at levels of 15-19% U235 and operate at much higher temperatures than the light water reactors the NRC knows so well.

The materials used in secondary loops to transfer heat from fast reactors to make steam for turbines can be exotic including helium, carbon dioxide, and liquid metals. There are design and materials challenges ahead before these reactors will hit an engineer’s desk at the NRC. That’s why the prototyping projects at SRS will be useful.

In addition to the Hyperion reactor (25 MW), with its so-called “bathtub design,” there is the GE-Hitachi 300 MW reactor that is designed to work with spent nuclear fuel. Two other SMR developers are also interested in building prototypes at SRS. They are General Atomics with a 240 MW reactor that uses plutonium and HEU and TerraPower’s 300 MW “Traveling Wave” reactor that uses depleted uranium for fuel.

Licensing SMRs will take time, and the NRC will have to go up a steep learning curve for the new “fast reactor” designs. Help is available. The American Nuclear Society and the Nuclear Energy Institute have published a series of technical papers on the multiple aspects of licensing SMRs. The NRC is listening to the ideas from both group as it develops its capability to craft a licensing strategy for the new designs.

Prior coverage on this blog

  • 09/04/10 – Stand up double for SMRs
  • 08/18/10 – NEI seeks consensus on licensing small reactors
  • 07/28-10 – ANS committee works SMR licensing issues
  • 06/22/10 – How to open running room for small reactors
  • 11/21/09 – Will the nuclear renaissance start with small reactors?
  • 06/26/09 – Change the NRC cost recovery rule for small reactors

# # #


Anonymous said...

Aside from the discussion in this excellent article, if the SMRs proposed for SRS ever do get built, and if they're intended to be prototypes for commercial plants, it makes no sense to try to avoid NRC licensing, even if were legal to do so. If DOE "self-regulated," the prototypes would still have undergo some sort of safety review, possibly the equivalent of an NRC licensing review. Then, when the time came for commercial deployment, they'd have to go through a new, separate NRC review, delaying the process even more. NRC would, in all likelihood, insist on an independent review, not simply recycle DOE's work. Perhaps a little time would be saved by virtue of the operational data the prototypes would have generated, but it still makes little sense--technically or economically--to do the same thing twice.

The folks at SRS need to get their heads on straight. One hopes that perhaps Assistant Secretary Lyons or Secretary Chu will take some time to "counsel" them about this issue.

Rod Adams said...


There is a middle ground. 10 CFR 50 includes two different classes of licenses, a class 104 and a class 103. The class 104 licenses are specifically designed for "medical therapy and research and development facilities". Class 104 licenses can be issued for

"(3) A production or utilization facility for industrial or commercial purposes, when specifically authorized by law."

Research and test reactors most often have power levels of less than 10 or 20 MW thermal, but that is not a firm requirement.

10 CFR 50.41 directs the NRC to permit the conduct of widespread and diverse research and development.

The testing and development required for the non light water reactor models is going to be huge and would be greatly inhibited if the vendors all had to wait in line for time at the ATR. A good case can be made that the first of a kind systems at SRS would be prototypes, research and development test reactors that should be licensed under part 104 vice under part 103.

There are a number of considerations that make that process more suitable for a facility like SRS that was specifically chosen for its low local population density and enhanced technical support capabilities.

Anonymous said...

In response to Rod's comment, I think there's an important caveat.

The important phrase in the NRC's regulations regarding Class 104 licenses is "where specifically authorized by law." The provisions of the Atomic Energy Act and the NRC's rules (implementing Sections 103 and 104 of the AEA, which is where the "class" designations originated) would appear to indicate that Congress would have to explicitly designate these commercial prototypes to be eligible for licensing as R&D facilities under Section 104, and--of course--the President would have to sign such legislation. It's impossible to predict the future, but it seems unlikely, at best, that the current Administration would do this. And such a move seems discriminatory, at best. Why favor Hyperion and GEH over, say, B&W, or NuScale (assuming NuScale goes back into operation), or other SMR vendors by allowing some to get Class 104 licenses but not others?

Other than renewing the licenses for exisitng research reactors (e.g., at universities), I'm not aware that the NRC has issued a Class 104 license for a reactor for many, many years. (In fact, it may not have issued any new Class 104 licenses for reactors since it came into existence in 1975.) It's not clear to me how a review for prototype commercial facilities would be handled without any specific guidelines for doing so, or if, in the end, such a review would be significantly different from one that the staff would do for a Class 103 license. Nor is it clear how a Class 104 review, if substantially different from a Class 103 review, would benefit a prospective commercial licensee.

The other question has to be, who would pay for these prototypes? Are the vendors willing to "donate" them to SRS? If not, how does SRS get the money to pay for them, build them, and operate them? Would DOE have sufficient funds in its budget to underwrite the licensing (whatever the license class), construction, and operation of these facilities? Or would the nuclear industry be expected to cost-share (as in, say, the old Clinch River Breeder Reactor model)? And if the facilities are going to be built at a national laboratory, what makes SRS a better choice than, say, INL or LANL? SRS does not have much history of civilian nuclear R&D, while the others do (INL more than LANL, of course).

Lots of questions here, not many definitive answers. Licensing is only part of the issue--a major part, no doubt, but DOE, NRC, Congress, and the industry need to give this some thought before rushing headlong into such a venture.

donb said...

Dan Yurman wrote:
That said FOE’s Clements sees his job as attacking anything that looks like a weakness in the nuclear world as a means to thwart the nuclear renaissance. The anti-nuclear agenda isn’t about licensing. It is designed to set the bar so high that the safest reactor is one that is never built.

FOE thinks it is doing the public a favor by trying to stop "dangerous" nuclear power. But we have over a half-century of experience showing that nuclear energy is the safest way to generate power with minimum impact on public health. Power plants will be built because we need the power. If nuclear power plants are NOT built, then some other un-healthful and more dangerous plant will be built instead. FOE is doing the opposite of their stated purpose.

donb said...

I would like to see the Department of Energy win on this one. We could then see some "new idea" reactors constructed quickly. I feel confident that they will be operated safely. The next step would be licensing. The almost certain bureaucratic delay would expose how NRC regulation has stifled the advancement of nuclear technology.