February 05, 2008
America’s Nuclear Future
Nobelist Burton Richter on Why the US is Falling Behind
Nobel Prize-winning physicist and Breakthrough Senior Fellow Burton Richter, pictured right, talks to the Breakthrough Institute about next generation nuclear reactors, and what it will take to build them in the United States. President Obama recently recognized Richter's lifetime achievements in the field of energy with a 2010 Enrico Fermi Award. Photo by Pete Souza/The White House.
When it comes to nuclear energy, Dr. Burton Richter is Mr. Credible. Winner of the 1976 Nobel Prize for discovering a new sub-atomic particle, Richter has advised presidents and policy makers for almost 40 years. Richter has been a Breakthrough Senior Fellow since 2011, and is technical adviser to the forthcoming documentary, Pandora's Promise, about pro-nuclear environmentalists.
Breakthrough interviewed Richter recently to get his opinion on next generation nuclear reactors, and why so many of them are being developed abroad and not by the Department of Energy in the United States. "The DOE is too screwed up to go into a partnership and do this in the US," the blunt Richter told us, referring to the Bill Gates-backed nuclear design pursued in China by Terrapower.
Is DOE really to blame? In the end, Richter told us it was partisan polarization that was the problem. "George W. Bush actually had a good thing on next generation nuclear," Richter said. "When the Obama people came in all the Gen IV activities were stopped. With a system that keeps changing its priorities every few years, the [National DOE] Labs are pretty demoralized. The French have a long-term plan. The Koreans, the Chinese, the Russians have it. We don't have it. That's not the fault of the labs, that's the fault of the administrations."
And that's the fault, we might add, of irrational environmentalist and progressive fears of nuclear energy — something "Pandora's Promise" hopes to change. Read the rest of our interview below.
What is the future of next generation nuclear reactors?
People need to get clear as to what they want from next generation reactors. Do you want multiple recycling of waste so that the waste repositories only have to hold things for 500 years? I personally don't think it's a problem to have to store waste for 100,000s of years, but reducing the toxicity of the waste to reduce its radioactivity would build confidence.
In other words, we should figure out the product attributes we want from advanced nuclear designs, like less waste, less proliferation risk, cost, and those sorts of things?
Yes, and DOE is trying to do that work now, and has nine criteria. I'm on the fuel cycle subcommittee, and we've been critical of some of it, because they are mixing technical and policy criteria. How much does having reactors be passively safe for two weeks versus one month really matter? How much does proliferation resistance matter? How much weight should we give to proliferation versus safety? These are policy questions, not technical ones.
What next generation designs do you find interesting?
There are advocates for high temperature, fast gas reactors who claim their design is inherently safe. What that means is that they could have a complete station blackout, and the temperature of the fuel would not rise so high to melt the core and release bad stuff into the atmosphere. With such a design, you couldn't have had a Fukushima-type accident. I haven't validated this design, and it hasn't been run through the Nuclear Regulatory Commission (NRC). But the claims are being made by people who are very experienced and who I respect.
Other people think molten salt reactors have promise, like Per Peterson at UC-Berkeley. He's going to China to test a 2 megawatt version of the molten salt reactor (MSR). One attractive feature for the MSR is its continuous recycling. The fuel goes through the reactor, takes fission fragments out, and runs the stuff back in. MSR is a bit further behind the sodium-cooled reactors because we haven't done much with it since the sixties.
What about the small modular reactors (SMR)?
I can see big advantages to smaller reactors in the US. If it's true that the price per kilowatt is the same, I can see distributed reactors in the developing world where you don't need 1.3GW like the AP1000.
I'd like to see more on the SMR front. I don't know why DOE only accepted one of the bids for the support and licensing process. The original idea was to take two.
What about the reactor designed by Nathan Myhrvold and backed by Bill Gates through Terrapower?
They had to change designs because the original design of a kind of slow-burning candle didn't work. The new version is supposed to have a core that would be sealed for 50 years. But it's not completely sealed because you have to shuffle the fuel rods. One advantage is that that at the end of 50 years, the waste is so impure that nobody would want anything to do with it for making a weapon.
Terrapower is being done in China because in the US there's no way he could get it licensed. And the DOE is too screwed up to go into a partnership and do this in the US.
We always hear from people that DOE is screwed up. But what exactly does that mean? Can it be fixed?
Consider the fact that the DOE can, at one of its labs, go ahead with an experimental fission system that is not approved by the Nuclear Regulatory Commission (NRC). After all, the DOE is supposed to develop new technologies, while the NRC is supposed to deal with things in the civilian nuclear world.
In other words, the labs don't need NRC approval to make a 5MW version of TerraPower's reactor. They could just go do it. But it's so agonizing to get [lab] approval for that kind of thing. So political. Ultra-greens would say too dangerous and NRC has to approve it, and NRC would say it will look into it and it would take a decade.
That's the reason Nathan [Myhrvold] and Bill Gates said, “Let’s build the first one in China.”
Is the problem with Congress or DOE?
Both. At DOE there are a lot of layers of bureaucracy and very little continuity. Everything changes with every new administration. The long-term goals change. The result is that the labs have become very conservative.
With a system that keeps changing its priorities every few years, the labs are pretty demoralized. We cannot get a coherent accepted long-term plan. The French have a long-term plan. The Koreans, the Chinese, the Russians have it. We don't have it. That's not the fault of the labs, that's the fault of the administrations.
Is this a problem of ideological and partisan polarization?
George W. Bush actually had a good program on next generation nuclear. We were part of the Generation IV International Forum, working closely with Japan and France. We had a program that was headed toward certain kinds of advanced reactors, including liquid sodium, and a high temperature gas reactor. When the Obama people came in all the Gen IV activities were stopped. Yucca Mountain was shut down. And we're off in totally new directions.
Partly, but there were even changes between the first George W. Bush term and the second. In first term, they were talking about reprocessing, and the second Gen IV designs. We have an on again off again program that changes too often. The next problem is the budget. The DOE nuclear budget is a complete mess. They are working off of a continuing resolution, and in that process you always take the lower budget line from either the Senate or House. This creates massive amounts of uncertainty in the programs.
Who can change that? Can Obama just tell the labs to build a next gen nuclear reactor?
No, it has to go to Congress to change. The whole structure has to change.
What’s your general impression of the integral fast reactor (IFR), the prototype of which ran at Argonne-West [which is now part of Idaho] National Lab, and is now being marketed by General Electric as the PRISM reactor?
The IFR is a sodium-cooled fast spectrum reactor with all the good and bad that come with it. The one sodium cooled reactor at Hanford ran for thirty years until we drilled a hole into it [after Congress ended funding for it in 1994]. France and Russia built versions as well.
What's new to the IFR is the on-site reprocessing, and the feeding back of the actinides [radioactive elements like uranium and plutonium] back into the fuel, so that nothing ever leaves it. The new IFR trick is in the electrorefining [sometimes called pyroprocessing] to reprocess the waste into new fuel, making it a continuous fuel cycle. So think of the IFR as a liquid sodium fast spectrum breeder reactor with a trick as to how to do the separation of actinides in an effective fashion.
Electrorefining is the most interesting new element in the IFR, but it has been hard to figure out how to get it working well enough to be used commercially.
Who is working on improving electrorefining?
South Korea is very interested in electrorefining and would like to do a joint program with the US. The question is whether we'll let them do it. The 123 agreement we have with them says that the US has to agree to any reprocessing. The Nuclear Energy Advisory Committee to the DOE has said that if you're going to do this, then having a Korean partner would be a great idea.
Wouldn't technologies like the IFR greatly reduce the amount of waste needed?
You need a geological repository anyway because you always have fission fragments, and that's the really radioactive stuff. So if pyroprocessing worked perfectly the long lived components would be removed to be used as fuel, and after 500 years you wouldn't have to worry about them any more because the radioactivity would be low.
So you'll still need a repository, though probably not for 100,000s of years. But there's a big if here. How efficiently can you separate these long-lived actinides from the fission fragments? If you allow only a few percent of the actinides in, then it will be for 100,000s of thousands years. It has to be really good. Right now, it's not that good. The people working on it say they have good ideas but they haven't fixed it yet.
Are IFRs proliferation-proof, because the plutonium is never isolated? How easy would it be to separate?
No fuel cycle is proliferation proof. There's a step in the cycle where you are producing actinides including plutonium. The IFR guys are right that since you don't move the plutonium out of the plant, there's less risk that bad guys will get it, but what if it's the bad guys who own the plant? Making the IFR the way they want to run it makes it harder to make a bomb, but not impossible.
GE has proposed building an IFR plant in the UK for plutonium disposal, not for commercial electricity generation. They say they can build it fast. Is this doable?
Yes, they can build a PRISM reactor. However, they are not doing on-site reprocessing but using UK plutonium blended with uranium for fuel and can run it until they run out of plutonium.
Is the IFR inherently safe?
The phrase "inherently safe" refers to the specific situation of full power blackout. You lose all power for a week, or a month. Inherently safe has to have temperatures not go up enough to get a core meltdown. So for that class of accidents, the IFR is inherently safe from the kind of accidents that happened at Three Mile Island and Fukushima.
But anything that relies on molten sodium is a danger, because of its reactivity with water and oxygen. They say they'll have double-walled pipes, and separate steam systems, but my question is, will it survive an earthquake? What if something falls on and breaks a pipe?
What about lead-cooled reactors that DOE is looking at?
That's the one I'm most skeptical of. The only lead cooled reactors to have been built were used to power Soviet Alpha-class submarines. Two of those subs vanished. Six are in port with reactors shut down, and the lead cooled and solidified. But nobody knows why those two vanished. Molten lead in pipes is very corrosive. They correct for this by adding bismuth to the lead which adds a corrosion resistant layer to the pipes. My suspicions may not be warranted, but I want to know why those subs disappeared, and I'd love to see a joint US-Russia program to take the pipes apart, and see if there's a corrosion problem.
What about thorium?
Some claim that thorium fueled reactors are much more proliferation resistant, but I have not seen a real analysis. Thorium itself is not fissionable and you breed U-233 in the reactor to keep it running. You can make bombs out of U-233, though it is said to be more difficult to do so.
India has a thorium breeder technology development program because they have lots of thorium but little uranium. But the fear of running out of uranium is clearly wrong. The history of mining is that you first mine the richest ores, then you get the less rich ores, but the technology is improving along the way, allowing you to extract more from the less rich ores, and the price stays constant as you go to less and less richness. We’re also looking at getting uranium out of seawater. This work started in the UK. The Japanese did a lot of it and the people at Argonne improved the process.
What’s the progress in other countries?
Gen IV international forum got together nations working on advanced reactors and picked seven as the most promising technologies. One of those is the liquid sodium. France is going ahead with a plain vanilla liquid sodium fast reactor and reprocessing in an aqueous solution and get out pure plutonium and that gets out the actinides. France predicts its first commercial reactor between 2040 and 2050.