Deconstructing the Floodgates hypothesis: is nuclear recycling a proliferation threat?

Opposition to recycling used nuclear fuel in the U.S. centres around the assumption that if the U.S. resumes recycling, an activity it stopped doing in the 1970s, then that will open the floodgates worldwide. Countries that allegedly followed the U.S. lead in the 1970s will begin their own recycling programs. And, if recycling is based on aqueous (i.e., separative) reprocessing methods like PUREX, UREX+, or COEX, then—say those who oppose recycling—it is a proliferation threat.

This notion—I’ll call it the Floodgates hypothesis—is asserted as if it were a natural law. The U.S. resumes recycling, and others naturally follow the lead, like night naturally follows day.

 How resilient is this hypothesis to testing? Let’s see.

A brief look at the actual history of recycling reveals that the countries to whom recycling made economic sense made use of it regardless of what the U.S. did. France is of course the supreme example. Of the fifty-eight nuclear reactors in the French civilian fleet, 21 are burning reycled fuel as I write this. A further six are capable of it. Another one is authorized to burn it, and the Flamanville 3 EPR, which I visited last week, will also be capable.

Areva, the French company that, along with manufacturing reactors and their components, handles the front- and back-end of the nuclear fuel cycle, provides fuel recycling services for Belgium, Germany, Italy, Switzerland, and Japan. Under the service agreements with each of these countries, Areva transports used fuel from their reactors and recycles it in France. All of those countries are non-weapons states under the nuclear non-proliferation treaty (NPT).

Of these countries, only Japan has decided to build its own recycling plant, which, when it is up and running, will thereby obviate the need to ship used fuel to France. Areva is helping Japan build the plant, and is training its operators.

In other words, we have a Floodgates scenario right in front of us. What does this experience tell us? The Floodgates hypothesis would predict that Germany, Switzerland, Belgium, Italy, and Japan can, because they are engaged in nuclear recycling, turn that activity into weapons programs.

Well, every one of these countries except Japan would have to dramatically change the terms of their actual recycling arrangements in order to pull that off. As I mentioned, Areva does the actual recycling, at its La Hague plant, just east of Flamanville along the channel coast. I also visited that plant the other day, along with a delegation of nuclear experts from the U.S. While we were there we watched the fuel assembly from an Italian boiling water reactor being hoisted out of its shipping canister (don’t worry—we were behind a radiation-proof window).

If Italy decided to start a weapons program, it would have to figure out another way to get the plutonium out of its used fuel. It would first have to move its recycling capability onto its own soil. La Hague is pretty well protected: when you pass through the initial security perimeter and look out at the formidable barriers of barbed wire, you get an idea of what it feels like to be an inmate in a super-max prison. Video cameras, remotely operated by Euratom, a.k.a. the European Atomic Energy Community, monitor certain areas, such as the unloading bay where we saw the Italian BWR fuel. These circumstances make it difficult for anybody to involve La Hague in a proliferation scheme. Unless you think you could arrange some hell of an inside job, you’d be better off taking plutonium out of your used fuel on your own soil, in your own facility.

Proponents of the floodgates hypothesis actually slip in another assumption: that the international rules governing the civilian nuclear trade will relax as more countries take up recycling. This would have to happen in order for prospective proliferators to actually be able to extract plutonium from their used fuel. The prospective proliferator would have to either acquire this technology from another country that already has it, or develop the technology on its own. The second alternative is open to any country on earth, NPT signatory or not, so I’ll leave it alone. Let’s assume the aspiring proliferator opts to acquire recycling technology from another country. The Floodgates hypothesis assumes that other country will transfer the technology to the proliferator.

How likely is that? International civilian nuclear trade is governed by the Nuclear Suppliers Group (NSG). The NSG  is made up of 45 member countries. Decisions on the terms of trade are made by consensus. Today, trade in recycling technologies remains restricted to a small group of NSG country members. If a country outside of that group wants to acquire recycling technology and know-how, the NSG as a group would have to approve it. Good luck with that.

Proponents of the Floodgates hypothesis tacitly assume that a U.S. decision to resume recycling will also be accompanied by a general relaxing of the NSG rules. There is really no reason to assume that. In fact, going by the recent U.S.-India civilian nuclear deal, which brought about an NSG exemption in the case of India, I’d say there is no chance the rules will relax.

This leaves Japan as the only real current Floodgates candidate. As I mentioned, Japan is building its own recycling facility, with Areva’s help. Regardless of the actual technology that will be employed in the Japanese plant, the plant will be a civilian facility, subject to the same international monitoring and safeguards as the recycling and enrichment plants I visited in France last week. This means that the only scenario in which Japan could misuse the technology in a weapons program is the breakout scenario (see article).

Nothing is impossible, but there are things that are extremely improbable. So improbable, in fact, that we can safely just discount them. Japan breaking out of the NPT is one of them.

Germany, Belgium, Switzerland, and Italy are all examples of what we can expect in the current nuclear renaissance. Fuel recycling gives them fuel assurances that fit with their energy policies. The fact that the actual recycling is being done in France does not appear to be an issue of national pride in these countries. Recycling makes economic sense, as does having France do the recycling. It’s a simple economic decision.

If the U.S. does get back into the recycling game, there is no proliferation threat.

0 0 votes
Article Rating
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
13 years ago

Thanks for the post. It’s nice to see some rational thinking about fuel recycling.

But, I’ve got to add a couple of things:

1) You can not use plutonium from a civilian reactor to build a bomb. Fuel used for civilian purposes is left in the reactor for a good long time, which produces too much Plutonium 240, an unwanted isotope that makes the resulting plutonium totally useless for making a bomb. See for more detail. (Military plutonium is produced from fuel that is cycled in and out of the reactor quickly.

2) If educated, rich nations such as Japan and Germany wanted to get the atomic bomb, they could get it anyway. They could set up their own plutonium production reactors or bomb grade enriched uranium (95+% U-235), and once the materials are available, they’ve got the physics Ph.D.’s to make it into a bomb in about 6 months. All this non-proliferation talk is just that, talk. They don’t have the bomb because they don’t want it.

Steve Aplin
13 years ago


I have read a number of reports on the suitability of reactor-grade plutonium. Some agree with your assessment, and others (like this one that Richard Garwin prepared for the Council on Foreign Relations; see http: // say reactor-grade is sufficient.

I know the issue is not a quibble, since the U.S., Russia, and whoever else made plutonium bombs did not use reactor-grade material. And of course if you are correct and Garwin is wrong, then that would pretty much settle the question of whether civilian programs pose any proliferation threat at all. I just wanted to point out that even if it is technically possible, it is too difficult to do through a civilian program.

Which is exactly why your point #2 bears repeating. If advanced industrial countries that are NPT non-weapons states want weapons, they can get them. They don’t need to, and wouldn’t, pussyfoot around a civilian program and waste their time and money trying to screw Areva. The reason they don’t have weapons now is because they don’t want or need them.

I know the issue is not a quibble, since the U.S., Russia, and whoever else made plutonium bombs did not use reactor-grade material–

Um, you mean you know it is a quibble, right?

… Military plutonium is produced from fuel that is cycled in and out of the reactor quickly

That’s true, but it uses deceitful framing. Uranium that has been irradiated in a small low-temperature pile with no attached heat engine is not fuel. It is not spent fuel, it is not any kind of fuel.

An analogy that seems good to me (and one or more HuffPo editors may have agreed, because they censored it): weapon plutonium is to civilian plutonium as guns are to car motors.

Reprocessing can produce chemically pure plutonium, but cannot give any mixture of plutonium isotopes that wasn’t in the spent fuel to begin with, and the isotopic mixture in the plutonium from modern nuclear power plants is such that it has never been used in weapons. Some say it can’t be. This Google search — — will give some good introductions to the difficulties.

Some say it can’t be done. To my mind, this misses the point. It is like saying an old V8 can’t be made into an eight-barrel cannon, and therefore gun-control efforts should not focus on tightly securing all junkyards. Conceivably an old V8 could be turned to the dark side, but still, nonetheless, genuine gun-control efforts do not focus on locking down junkyards.

Similarly, spent nuclear fuel does not raise genuine weapon-proliferation concerns.

(Fireproof fuel, real-car range, no emissions: )

Steve Aplin
13 years ago

GRL: so that two-bit e-rag HuffPo censored you too huh… I like you already.

Quibble: I meant it is NOT a quibble, but I worded it poorly. I should have used “but it is nevertheless rendered moot by the fact that the U.S., Russia, etc.” instead of “since.”

I like your V8–cannon analogy, that’s a good way of putting it. While every anti-proliferation expert was wringing his hands over the V8s in junkyards, A.Q. Khan was selling centrifuges to customers who understood that V8s make lousy cannons.

[…] Deconstructing the Floodgates hypothesis: is nuclear recycling a proliferation threat? […]

12 years ago

Look up Thorium fission process. It is an amazingly green alternative to uranium or plutonium fission processes.